1HXCOMM See docs/devel/docs.rst for the format of this file. 2HXCOMM 3HXCOMM Use DEFHEADING() to define headings in both help text and rST. 4HXCOMM Text between SRST and ERST is copied to the rST version and 5HXCOMM discarded from C version. 6HXCOMM DEF(option, HAS_ARG/0, opt_enum, opt_help, arch_mask) is used to 7HXCOMM construct option structures, enums and help message for specified 8HXCOMM architectures. 9HXCOMM HXCOMM can be used for comments, discarded from both rST and C. 10 11DEFHEADING(Standard options:) 12 13DEF("help", 0, QEMU_OPTION_h, 14 "-h or -help display this help and exit\n", QEMU_ARCH_ALL) 15SRST 16``-h`` 17 Display help and exit 18ERST 19 20DEF("version", 0, QEMU_OPTION_version, 21 "-version display version information and exit\n", QEMU_ARCH_ALL) 22SRST 23``-version`` 24 Display version information and exit 25ERST 26 27DEF("machine", HAS_ARG, QEMU_OPTION_machine, \ 28 "-machine [type=]name[,prop[=value][,...]]\n" 29 " selects emulated machine ('-machine help' for list)\n" 30 " property accel=accel1[:accel2[:...]] selects accelerator\n" 31 " supported accelerators are kvm, xen, hvf, nvmm, whpx or tcg (default: tcg)\n" 32 " vmport=on|off|auto controls emulation of vmport (default: auto)\n" 33 " dump-guest-core=on|off include guest memory in a core dump (default=on)\n" 34 " mem-merge=on|off controls memory merge support (default: on)\n" 35 " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n" 36 " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n" 37 " suppress-vmdesc=on|off disables self-describing migration (default=off)\n" 38 " nvdimm=on|off controls NVDIMM support (default=off)\n" 39 " memory-encryption=@var{} memory encryption object to use (default=none)\n" 40 " hmat=on|off controls ACPI HMAT support (default=off)\n" 41 " memory-backend='backend-id' specifies explicitly provided backend for main RAM (default=none)\n" 42 " cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]\n", 43 QEMU_ARCH_ALL) 44SRST 45``-machine [type=]name[,prop=value[,...]]`` 46 Select the emulated machine by name. Use ``-machine help`` to list 47 available machines. 48 49 For architectures which aim to support live migration compatibility 50 across releases, each release will introduce a new versioned machine 51 type. For example, the 2.8.0 release introduced machine types 52 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures. 53 54 To allow live migration of guests from QEMU version 2.8.0, to QEMU 55 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8" 56 and "pc-q35-2.8" machines too. To allow users live migrating VMs to 57 skip multiple intermediate releases when upgrading, new releases of 58 QEMU will support machine types from many previous versions. 59 60 Supported machine properties are: 61 62 ``accel=accels1[:accels2[:...]]`` 63 This is used to enable an accelerator. Depending on the target 64 architecture, kvm, xen, hvf, nvmm, whpx or tcg can be available. 65 By default, tcg is used. If there is more than one accelerator 66 specified, the next one is used if the previous one fails to 67 initialize. 68 69 ``vmport=on|off|auto`` 70 Enables emulation of VMWare IO port, for vmmouse etc. auto says 71 to select the value based on accel and i8042. For accel=xen or 72 i8042=off the default is off otherwise the default is on. 73 74 ``dump-guest-core=on|off`` 75 Include guest memory in a core dump. The default is on. 76 77 ``mem-merge=on|off`` 78 Enables or disables memory merge support. This feature, when 79 supported by the host, de-duplicates identical memory pages 80 among VMs instances (enabled by default). 81 82 ``aes-key-wrap=on|off`` 83 Enables or disables AES key wrapping support on s390-ccw hosts. 84 This feature controls whether AES wrapping keys will be created 85 to allow execution of AES cryptographic functions. The default 86 is on. 87 88 ``dea-key-wrap=on|off`` 89 Enables or disables DEA key wrapping support on s390-ccw hosts. 90 This feature controls whether DEA wrapping keys will be created 91 to allow execution of DEA cryptographic functions. The default 92 is on. 93 94 ``nvdimm=on|off`` 95 Enables or disables NVDIMM support. The default is off. 96 97 ``memory-encryption=`` 98 Memory encryption object to use. The default is none. 99 100 ``hmat=on|off`` 101 Enables or disables ACPI Heterogeneous Memory Attribute Table 102 (HMAT) support. The default is off. 103 104 ``memory-backend='id'`` 105 An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options. 106 Allows to use a memory backend as main RAM. 107 108 For example: 109 :: 110 111 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on 112 -machine memory-backend=pc.ram 113 -m 512M 114 115 Migration compatibility note: 116 117 * as backend id one shall use value of 'default-ram-id', advertised by 118 machine type (available via ``query-machines`` QMP command), if migration 119 to/from old QEMU (<5.0) is expected. 120 * for machine types 4.0 and older, user shall 121 use ``x-use-canonical-path-for-ramblock-id=off`` backend option 122 if migration to/from old QEMU (<5.0) is expected. 123 124 For example: 125 :: 126 127 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off 128 -machine memory-backend=pc.ram 129 -m 512M 130 131 ``cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]`` 132 Define a CXL Fixed Memory Window (CFMW). 133 134 Described in the CXL 2.0 ECN: CEDT CFMWS & QTG _DSM. 135 136 They are regions of Host Physical Addresses (HPA) on a system which 137 may be interleaved across one or more CXL host bridges. The system 138 software will assign particular devices into these windows and 139 configure the downstream Host-managed Device Memory (HDM) decoders 140 in root ports, switch ports and devices appropriately to meet the 141 interleave requirements before enabling the memory devices. 142 143 ``targets.X=target`` provides the mapping to CXL host bridges 144 which may be identified by the id provided in the -device entry. 145 Multiple entries are needed to specify all the targets when 146 the fixed memory window represents interleaved memory. X is the 147 target index from 0. 148 149 ``size=size`` sets the size of the CFMW. This must be a multiple of 150 256MiB. The region will be aligned to 256MiB but the location is 151 platform and configuration dependent. 152 153 ``interleave-granularity=granularity`` sets the granularity of 154 interleave. Default 256 (bytes). Only 256, 512, 1k, 2k, 155 4k, 8k and 16k granularities supported. 156 157 Example: 158 159 :: 160 161 -machine cxl-fmw.0.targets.0=cxl.0,cxl-fmw.0.targets.1=cxl.1,cxl-fmw.0.size=128G,cxl-fmw.0.interleave-granularity=512 162ERST 163 164DEF("M", HAS_ARG, QEMU_OPTION_M, 165 " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n", 166 QEMU_ARCH_ALL) 167 168SRST 169``sgx-epc.0.memdev=@var{memid},sgx-epc.0.node=@var{numaid}`` 170 Define an SGX EPC section. 171ERST 172 173DEF("cpu", HAS_ARG, QEMU_OPTION_cpu, 174 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL) 175SRST 176``-cpu model`` 177 Select CPU model (``-cpu help`` for list and additional feature 178 selection) 179ERST 180 181DEF("accel", HAS_ARG, QEMU_OPTION_accel, 182 "-accel [accel=]accelerator[,prop[=value][,...]]\n" 183 " select accelerator (kvm, xen, hvf, nvmm, whpx or tcg; use 'help' for a list)\n" 184 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n" 185 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n" 186 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n" 187 " one-insn-per-tb=on|off (one guest instruction per TCG translation block)\n" 188 " split-wx=on|off (enable TCG split w^x mapping)\n" 189 " tb-size=n (TCG translation block cache size)\n" 190 " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n" 191 " eager-split-size=n (KVM Eager Page Split chunk size, default 0, disabled. ARM only)\n" 192 " notify-vmexit=run|internal-error|disable,notify-window=n (enable notify VM exit and set notify window, x86 only)\n" 193 " thread=single|multi (enable multi-threaded TCG)\n" 194 " device=path (KVM device path, default /dev/kvm)\n", QEMU_ARCH_ALL) 195SRST 196``-accel name[,prop=value[,...]]`` 197 This is used to enable an accelerator. Depending on the target 198 architecture, kvm, xen, hvf, nvmm, whpx or tcg can be available. By 199 default, tcg is used. If there is more than one accelerator 200 specified, the next one is used if the previous one fails to 201 initialize. 202 203 ``igd-passthru=on|off`` 204 When Xen is in use, this option controls whether Intel 205 integrated graphics devices can be passed through to the guest 206 (default=off) 207 208 ``kernel-irqchip=on|off|split`` 209 Controls KVM in-kernel irqchip support. The default is full 210 acceleration of the interrupt controllers. On x86, split irqchip 211 reduces the kernel attack surface, at a performance cost for 212 non-MSI interrupts. Disabling the in-kernel irqchip completely 213 is not recommended except for debugging purposes. 214 215 ``kvm-shadow-mem=size`` 216 Defines the size of the KVM shadow MMU. 217 218 ``one-insn-per-tb=on|off`` 219 Makes the TCG accelerator put only one guest instruction into 220 each translation block. This slows down emulation a lot, but 221 can be useful in some situations, such as when trying to analyse 222 the logs produced by the ``-d`` option. 223 224 ``split-wx=on|off`` 225 Controls the use of split w^x mapping for the TCG code generation 226 buffer. Some operating systems require this to be enabled, and in 227 such a case this will default on. On other operating systems, this 228 will default off, but one may enable this for testing or debugging. 229 230 ``tb-size=n`` 231 Controls the size (in MiB) of the TCG translation block cache. 232 233 ``thread=single|multi`` 234 Controls number of TCG threads. When the TCG is multi-threaded 235 there will be one thread per vCPU therefore taking advantage of 236 additional host cores. The default is to enable multi-threading 237 where both the back-end and front-ends support it and no 238 incompatible TCG features have been enabled (e.g. 239 icount/replay). 240 241 ``dirty-ring-size=n`` 242 When the KVM accelerator is used, it controls the size of the per-vCPU 243 dirty page ring buffer (number of entries for each vCPU). It should 244 be a value that is power of two, and it should be 1024 or bigger (but 245 still less than the maximum value that the kernel supports). 4096 246 could be a good initial value if you have no idea which is the best. 247 Set this value to 0 to disable the feature. By default, this feature 248 is disabled (dirty-ring-size=0). When enabled, KVM will instead 249 record dirty pages in a bitmap. 250 251 ``eager-split-size=n`` 252 KVM implements dirty page logging at the PAGE_SIZE granularity and 253 enabling dirty-logging on a huge-page requires breaking it into 254 PAGE_SIZE pages in the first place. KVM on ARM does this splitting 255 lazily by default. There are performance benefits in doing huge-page 256 split eagerly, especially in situations where TLBI costs associated 257 with break-before-make sequences are considerable and also if guest 258 workloads are read intensive. The size here specifies how many pages 259 to break at a time and needs to be a valid block size which is 260 1GB/2MB/4KB, 32MB/16KB and 512MB/64KB for 4KB/16KB/64KB PAGE_SIZE 261 respectively. Be wary of specifying a higher size as it will have an 262 impact on the memory. By default, this feature is disabled 263 (eager-split-size=0). 264 265 ``notify-vmexit=run|internal-error|disable,notify-window=n`` 266 Enables or disables notify VM exit support on x86 host and specify 267 the corresponding notify window to trigger the VM exit if enabled. 268 ``run`` option enables the feature. It does nothing and continue 269 if the exit happens. ``internal-error`` option enables the feature. 270 It raises a internal error. ``disable`` option doesn't enable the feature. 271 This feature can mitigate the CPU stuck issue due to event windows don't 272 open up for a specified of time (i.e. notify-window). 273 Default: notify-vmexit=run,notify-window=0. 274 275 ``device=path`` 276 Sets the path to the KVM device node. Defaults to ``/dev/kvm``. This 277 option can be used to pass the KVM device to use via a file descriptor 278 by setting the value to ``/dev/fdset/NN``. 279 280ERST 281 282DEF("smp", HAS_ARG, QEMU_OPTION_smp, 283 "-smp [[cpus=]n][,maxcpus=maxcpus][,drawers=drawers][,books=books][,sockets=sockets]\n" 284 " [,dies=dies][,clusters=clusters][,modules=modules][,cores=cores]\n" 285 " [,threads=threads]\n" 286 " set the number of initial CPUs to 'n' [default=1]\n" 287 " maxcpus= maximum number of total CPUs, including\n" 288 " offline CPUs for hotplug, etc\n" 289 " drawers= number of drawers on the machine board\n" 290 " books= number of books in one drawer\n" 291 " sockets= number of sockets in one book\n" 292 " dies= number of dies in one socket\n" 293 " clusters= number of clusters in one die\n" 294 " modules= number of modules in one cluster\n" 295 " cores= number of cores in one module\n" 296 " threads= number of threads in one core\n" 297 "Note: Different machines may have different subsets of the CPU topology\n" 298 " parameters supported, so the actual meaning of the supported parameters\n" 299 " will vary accordingly. For example, for a machine type that supports a\n" 300 " three-level CPU hierarchy of sockets/cores/threads, the parameters will\n" 301 " sequentially mean as below:\n" 302 " sockets means the number of sockets on the machine board\n" 303 " cores means the number of cores in one socket\n" 304 " threads means the number of threads in one core\n" 305 " For a particular machine type board, an expected CPU topology hierarchy\n" 306 " can be defined through the supported sub-option. Unsupported parameters\n" 307 " can also be provided in addition to the sub-option, but their values\n" 308 " must be set as 1 in the purpose of correct parsing.\n", 309 QEMU_ARCH_ALL) 310SRST 311``-smp [[cpus=]n][,maxcpus=maxcpus][,drawers=drawers][,books=books][,sockets=sockets][,dies=dies][,clusters=clusters][,modules=modules][,cores=cores][,threads=threads]`` 312 Simulate a SMP system with '\ ``n``\ ' CPUs initially present on 313 the machine type board. On boards supporting CPU hotplug, the optional 314 '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be 315 added at runtime. When both parameters are omitted, the maximum number 316 of CPUs will be calculated from the provided topology members and the 317 initial CPU count will match the maximum number. When only one of them 318 is given then the omitted one will be set to its counterpart's value. 319 Both parameters may be specified, but the maximum number of CPUs must 320 be equal to or greater than the initial CPU count. Product of the 321 CPU topology hierarchy must be equal to the maximum number of CPUs. 322 Both parameters are subject to an upper limit that is determined by 323 the specific machine type chosen. 324 325 To control reporting of CPU topology information, values of the topology 326 parameters can be specified. Machines may only support a subset of the 327 parameters and different machines may have different subsets supported 328 which vary depending on capacity of the corresponding CPU targets. So 329 for a particular machine type board, an expected topology hierarchy can 330 be defined through the supported sub-option. Unsupported parameters can 331 also be provided in addition to the sub-option, but their values must be 332 set as 1 in the purpose of correct parsing. 333 334 Either the initial CPU count, or at least one of the topology parameters 335 must be specified. The specified parameters must be greater than zero, 336 explicit configuration like "cpus=0" is not allowed. Values for any 337 omitted parameters will be computed from those which are given. 338 339 For example, the following sub-option defines a CPU topology hierarchy 340 (2 sockets totally on the machine, 2 cores per socket, 2 threads per 341 core) for a machine that only supports sockets/cores/threads. 342 Some members of the option can be omitted but their values will be 343 automatically computed: 344 345 :: 346 347 -smp 8,sockets=2,cores=2,threads=2,maxcpus=8 348 349 The following sub-option defines a CPU topology hierarchy (2 sockets 350 totally on the machine, 2 dies per socket, 2 modules per die, 2 cores per 351 module, 2 threads per core) for PC machines which support sockets/dies 352 /modules/cores/threads. Some members of the option can be omitted but 353 their values will be automatically computed: 354 355 :: 356 357 -smp 32,sockets=2,dies=2,modules=2,cores=2,threads=2,maxcpus=32 358 359 The following sub-option defines a CPU topology hierarchy (2 sockets 360 totally on the machine, 2 clusters per socket, 2 cores per cluster, 361 2 threads per core) for ARM virt machines which support sockets/clusters 362 /cores/threads. Some members of the option can be omitted but their values 363 will be automatically computed: 364 365 :: 366 367 -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16 368 369 Historically preference was given to the coarsest topology parameters 370 when computing missing values (ie sockets preferred over cores, which 371 were preferred over threads), however, this behaviour is considered 372 liable to change. Prior to 6.2 the preference was sockets over cores 373 over threads. Since 6.2 the preference is cores over sockets over threads. 374 375 For example, the following option defines a machine board with 2 sockets 376 of 1 core before 6.2 and 1 socket of 2 cores after 6.2: 377 378 :: 379 380 -smp 2 381 382 Note: The cluster topology will only be generated in ACPI and exposed 383 to guest if it's explicitly specified in -smp. 384ERST 385 386DEF("numa", HAS_ARG, QEMU_OPTION_numa, 387 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 388 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 389 "-numa dist,src=source,dst=destination,val=distance\n" 390 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n" 391 "-numa hmat-lb,initiator=node,target=node,hierarchy=memory|first-level|second-level|third-level,data-type=access-latency|read-latency|write-latency[,latency=lat][,bandwidth=bw]\n" 392 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n", 393 QEMU_ARCH_ALL) 394SRST 395``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 396 \ 397``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 398 \ 399``-numa dist,src=source,dst=destination,val=distance`` 400 \ 401``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]`` 402 \ 403``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=type[,latency=lat][,bandwidth=bw]`` 404 \ 405``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]`` 406 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA 407 distance from a source node to a destination node. Set the ACPI 408 Heterogeneous Memory Attributes for the given nodes. 409 410 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and 411 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a 412 contiguous range of CPU indexes (or a single VCPU if lastcpu is 413 omitted). A non-contiguous set of VCPUs can be represented by 414 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is 415 omitted on all nodes, VCPUs are automatically split between them. 416 417 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a 418 NUMA node: 419 420 :: 421 422 -numa node,cpus=0-2,cpus=5 423 424 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option 425 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to 426 assign CPU objects to a node using topology layout properties of 427 CPU. The set of properties is machine specific, and depends on used 428 machine type/'\ ``smp``\ ' options. It could be queried with 429 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ ' 430 property specifies node to which CPU object will be assigned, it's 431 required for node to be declared with '\ ``node``\ ' option before 432 it's used with '\ ``cpu``\ ' option. 433 434 For example: 435 436 :: 437 438 -M pc \ 439 -smp 1,sockets=2,maxcpus=2 \ 440 -numa node,nodeid=0 -numa node,nodeid=1 \ 441 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 442 443 '\ ``memdev``\ ' option assigns RAM from a given memory backend 444 device to a node. It is recommended to use '\ ``memdev``\ ' option 445 over legacy '\ ``mem``\ ' option. This is because '\ ``memdev``\ ' 446 option provides better performance and more control over the 447 backend's RAM (e.g. '\ ``prealloc``\ ' parameter of 448 '\ ``-memory-backend-ram``\ ' allows memory preallocation). 449 450 For compatibility reasons, legacy '\ ``mem``\ ' option is 451 supported in 5.0 and older machine types. Note that '\ ``mem``\ ' 452 and '\ ``memdev``\ ' are mutually exclusive. If one node uses 453 '\ ``memdev``\ ', the rest nodes have to use '\ ``memdev``\ ' 454 option, and vice versa. 455 456 Users must specify memory for all NUMA nodes by '\ ``memdev``\ ' 457 (or legacy '\ ``mem``\ ' if available). In QEMU 5.2, the support 458 for '\ ``-numa node``\ ' without memory specified was removed. 459 460 '\ ``initiator``\ ' is an additional option that points to an 461 initiator NUMA node that has best performance (the lowest latency or 462 largest bandwidth) to this NUMA node. Note that this option can be 463 set only when the machine property 'hmat' is set to 'on'. 464 465 Following example creates a machine with 2 NUMA nodes, node 0 has 466 CPU. node 1 has only memory, and its initiator is node 0. Note that 467 because node 0 has CPU, by default the initiator of node 0 is itself 468 and must be itself. 469 470 :: 471 472 -machine hmat=on \ 473 -m 2G,slots=2,maxmem=4G \ 474 -object memory-backend-ram,size=1G,id=m0 \ 475 -object memory-backend-ram,size=1G,id=m1 \ 476 -numa node,nodeid=0,memdev=m0 \ 477 -numa node,nodeid=1,memdev=m1,initiator=0 \ 478 -smp 2,sockets=2,maxcpus=2 \ 479 -numa cpu,node-id=0,socket-id=0 \ 480 -numa cpu,node-id=0,socket-id=1 481 482 source and destination are NUMA node IDs. distance is the NUMA 483 distance from source to destination. The distance from a node to 484 itself is always 10. If any pair of nodes is given a distance, then 485 all pairs must be given distances. Although, when distances are only 486 given in one direction for each pair of nodes, then the distances in 487 the opposite directions are assumed to be the same. If, however, an 488 asymmetrical pair of distances is given for even one node pair, then 489 all node pairs must be provided distance values for both directions, 490 even when they are symmetrical. When a node is unreachable from 491 another node, set the pair's distance to 255. 492 493 Note that the -``numa`` option doesn't allocate any of the specified 494 resources, it just assigns existing resources to NUMA nodes. This 495 means that one still has to use the ``-m``, ``-smp`` options to 496 allocate RAM and VCPUs respectively. 497 498 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth 499 Information between initiator and target NUMA nodes in ACPI 500 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can 501 create memory requests, usually it has one or more processors. 502 Target NUMA node contains addressable memory. 503 504 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is 505 the memory hierarchy of the target NUMA node: if hierarchy is 506 'memory', the structure represents the memory performance; if 507 hierarchy is 'first-level\|second-level\|third-level', this 508 structure represents aggregated performance of memory side caches 509 for each domain. type of 'data-type' is type of data represented by 510 this structure instance: if 'hierarchy' is 'memory', 'data-type' is 511 'access\|read\|write' latency or 'access\|read\|write' bandwidth of 512 the target memory; if 'hierarchy' is 513 'first-level\|second-level\|third-level', 'data-type' is 514 'access\|read\|write' hit latency or 'access\|read\|write' hit 515 bandwidth of the target memory side cache. 516 517 lat is latency value in nanoseconds. bw is bandwidth value, the 518 possible value and units are NUM[M\|G\|T], mean that the bandwidth 519 value are NUM byte per second (or MB/s, GB/s or TB/s depending on 520 used suffix). Note that if latency or bandwidth value is 0, means 521 the corresponding latency or bandwidth information is not provided. 522 523 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory 524 belongs. size is the size of memory side cache in bytes. level is 525 the cache level described in this structure, note that the cache 526 level 0 should not be used with '\ ``hmat-cache``\ ' option. 527 associativity is the cache associativity, the possible value is 528 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy 529 is the write policy. line is the cache Line size in bytes. 530 531 For example, the following options describe 2 NUMA nodes. Node 0 has 532 2 cpus and a ram, node 1 has only a ram. The processors in node 0 533 access memory in node 0 with access-latency 5 nanoseconds, 534 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access 535 memory in NUMA node 1 with access-latency 10 nanoseconds, 536 access-bandwidth is 100 MB/s. And for memory side cache information, 537 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB, 538 policy is write-back, the cache Line size is 8 bytes: 539 540 :: 541 542 -machine hmat=on \ 543 -m 2G \ 544 -object memory-backend-ram,size=1G,id=m0 \ 545 -object memory-backend-ram,size=1G,id=m1 \ 546 -smp 2,sockets=2,maxcpus=2 \ 547 -numa node,nodeid=0,memdev=m0 \ 548 -numa node,nodeid=1,memdev=m1,initiator=0 \ 549 -numa cpu,node-id=0,socket-id=0 \ 550 -numa cpu,node-id=0,socket-id=1 \ 551 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \ 552 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \ 553 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \ 554 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \ 555 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \ 556 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8 557ERST 558 559DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd, 560 "-add-fd fd=fd,set=set[,opaque=opaque]\n" 561 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL) 562SRST 563``-add-fd fd=fd,set=set[,opaque=opaque]`` 564 Add a file descriptor to an fd set. Valid options are: 565 566 ``fd=fd`` 567 This option defines the file descriptor of which a duplicate is 568 added to fd set. The file descriptor cannot be stdin, stdout, or 569 stderr. 570 571 ``set=set`` 572 This option defines the ID of the fd set to add the file 573 descriptor to. 574 575 ``opaque=opaque`` 576 This option defines a free-form string that can be used to 577 describe fd. 578 579 You can open an image using pre-opened file descriptors from an fd 580 set: 581 582 .. parsed-literal:: 583 584 |qemu_system| \\ 585 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 586 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 587 -drive file=/dev/fdset/2,index=0,media=disk 588ERST 589 590DEF("set", HAS_ARG, QEMU_OPTION_set, 591 "-set group.id.arg=value\n" 592 " set <arg> parameter for item <id> of type <group>\n" 593 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL) 594SRST 595``-set group.id.arg=value`` 596 Set parameter arg for item id of type group 597ERST 598 599DEF("global", HAS_ARG, QEMU_OPTION_global, 600 "-global driver.property=value\n" 601 "-global driver=driver,property=property,value=value\n" 602 " set a global default for a driver property\n", 603 QEMU_ARCH_ALL) 604SRST 605``-global driver.prop=value`` 606 \ 607``-global driver=driver,property=property,value=value`` 608 Set default value of driver's property prop to value, e.g.: 609 610 .. parsed-literal:: 611 612 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img 613 614 In particular, you can use this to set driver properties for devices 615 which are created automatically by the machine model. To create a 616 device which is not created automatically and set properties on it, 617 use -``device``. 618 619 -global driver.prop=value is shorthand for -global 620 driver=driver,property=prop,value=value. The longhand syntax works 621 even when driver contains a dot. 622ERST 623 624DEF("boot", HAS_ARG, QEMU_OPTION_boot, 625 "-boot [order=drives][,once=drives][,menu=on|off]\n" 626 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n" 627 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n" 628 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n" 629 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n" 630 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n", 631 QEMU_ARCH_ALL) 632SRST 633``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]`` 634 Specify boot order drives as a string of drive letters. Valid drive 635 letters depend on the target architecture. The x86 PC uses: a, b 636 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p 637 (Etherboot from network adapter 1-4), hard disk boot is the default. 638 To apply a particular boot order only on the first startup, specify 639 it via ``once``. Note that the ``order`` or ``once`` parameter 640 should not be used together with the ``bootindex`` property of 641 devices, since the firmware implementations normally do not support 642 both at the same time. 643 644 Interactive boot menus/prompts can be enabled via ``menu=on`` as far 645 as firmware/BIOS supports them. The default is non-interactive boot. 646 647 A splash picture could be passed to bios, enabling user to show it 648 as logo, when option splash=sp\_name is given and menu=on, If 649 firmware/BIOS supports them. Currently Seabios for X86 system 650 support it. limitation: The splash file could be a jpeg file or a 651 BMP file in 24 BPP format(true color). The resolution should be 652 supported by the SVGA mode, so the recommended is 320x240, 640x480, 653 800x640. 654 655 A timeout could be passed to bios, guest will pause for rb\_timeout 656 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will 657 not reboot, qemu passes '-1' to bios by default. Currently Seabios 658 for X86 system support it. 659 660 Do strict boot via ``strict=on`` as far as firmware/BIOS supports 661 it. This only effects when boot priority is changed by bootindex 662 options. The default is non-strict boot. 663 664 .. parsed-literal:: 665 666 # try to boot from network first, then from hard disk 667 |qemu_system_x86| -boot order=nc 668 # boot from CD-ROM first, switch back to default order after reboot 669 |qemu_system_x86| -boot once=d 670 # boot with a splash picture for 5 seconds. 671 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000 672 673 Note: The legacy format '-boot drives' is still supported but its 674 use is discouraged as it may be removed from future versions. 675ERST 676 677DEF("m", HAS_ARG, QEMU_OPTION_m, 678 "-m [size=]megs[,slots=n,maxmem=size]\n" 679 " configure guest RAM\n" 680 " size: initial amount of guest memory\n" 681 " slots: number of hotplug slots (default: none)\n" 682 " maxmem: maximum amount of guest memory (default: none)\n" 683 " Note: Some architectures might enforce a specific granularity\n", 684 QEMU_ARCH_ALL) 685SRST 686``-m [size=]megs[,slots=n,maxmem=size]`` 687 Sets guest startup RAM size to megs megabytes. Default is 128 MiB. 688 Optionally, a suffix of "M" or "G" can be used to signify a value in 689 megabytes or gigabytes respectively. Optional pair slots, maxmem 690 could be used to set amount of hotpluggable memory slots and maximum 691 amount of memory. Note that maxmem must be aligned to the page size. 692 693 For example, the following command-line sets the guest startup RAM 694 size to 1GB, creates 3 slots to hotplug additional memory and sets 695 the maximum memory the guest can reach to 4GB: 696 697 .. parsed-literal:: 698 699 |qemu_system| -m 1G,slots=3,maxmem=4G 700 701 If slots and maxmem are not specified, memory hotplug won't be 702 enabled and the guest startup RAM will never increase. 703ERST 704 705DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath, 706 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL) 707SRST 708``-mem-path path`` 709 Allocate guest RAM from a temporarily created file in path. 710ERST 711 712DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc, 713 "-mem-prealloc preallocate guest memory (use with -mem-path)\n", 714 QEMU_ARCH_ALL) 715SRST 716``-mem-prealloc`` 717 Preallocate memory when using -mem-path. 718ERST 719 720DEF("k", HAS_ARG, QEMU_OPTION_k, 721 "-k language use keyboard layout (for example 'fr' for French)\n", 722 QEMU_ARCH_ALL) 723SRST 724``-k language`` 725 Use keyboard layout language (for example ``fr`` for French). This 726 option is only needed where it is not easy to get raw PC keycodes 727 (e.g. on Macs, with some X11 servers or with a VNC or curses 728 display). You don't normally need to use it on PC/Linux or 729 PC/Windows hosts. 730 731 The available layouts are: 732 733 :: 734 735 ar de-ch es fo fr-ca hu ja mk no pt-br sv 736 da en-gb et fr fr-ch is lt nl pl ru th 737 de en-us fi fr-be hr it lv nl-be pt sl tr 738 739 The default is ``en-us``. 740ERST 741 742 743DEF("audio", HAS_ARG, QEMU_OPTION_audio, 744 "-audio [driver=]driver[,prop[=value][,...]]\n" 745 " specifies default audio backend when `audiodev` is not\n" 746 " used to create a machine or sound device;" 747 " options are the same as for -audiodev\n" 748 "-audio [driver=]driver,model=value[,prop[=value][,...]]\n" 749 " specifies the audio backend and device to use;\n" 750 " apart from 'model', options are the same as for -audiodev.\n" 751 " use '-audio model=help' to show possible devices.\n", 752 QEMU_ARCH_ALL) 753SRST 754``-audio [driver=]driver[,model=value][,prop[=value][,...]]`` 755 If the ``model`` option is specified, ``-audio`` is a shortcut 756 for configuring both the guest audio hardware and the host audio 757 backend in one go. The guest hardware model can be set with 758 ``model=modelname``. Use ``model=help`` to list the available 759 device types. 760 761 The following two example do exactly the same, to show how ``-audio`` 762 can be used to shorten the command line length: 763 764 .. parsed-literal:: 765 766 |qemu_system| -audiodev pa,id=pa -device sb16,audiodev=pa 767 |qemu_system| -audio pa,model=sb16 768 769 If the ``model`` option is not specified, ``-audio`` is used to 770 configure a default audio backend that will be used whenever the 771 ``audiodev`` property is not set on a device or machine. In 772 particular, ``-audio none`` ensures that no audio is produced even 773 for machines that have embedded sound hardware. 774 775 In both cases, the driver option is the same as with the corresponding 776 ``-audiodev`` option below. Use ``driver=help`` to list the available 777 drivers. 778 779ERST 780 781DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, 782 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n" 783 " specifies the audio backend to use\n" 784 " Use ``-audiodev help`` to list the available drivers\n" 785 " id= identifier of the backend\n" 786 " timer-period= timer period in microseconds\n" 787 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n" 788 " in|out.fixed-settings= use fixed settings for host audio\n" 789 " in|out.frequency= frequency to use with fixed settings\n" 790 " in|out.channels= number of channels to use with fixed settings\n" 791 " in|out.format= sample format to use with fixed settings\n" 792 " valid values: s8, s16, s32, u8, u16, u32, f32\n" 793 " in|out.voices= number of voices to use\n" 794 " in|out.buffer-length= length of buffer in microseconds\n" 795 "-audiodev none,id=id,[,prop[=value][,...]]\n" 796 " dummy driver that discards all output\n" 797#ifdef CONFIG_AUDIO_ALSA 798 "-audiodev alsa,id=id[,prop[=value][,...]]\n" 799 " in|out.dev= name of the audio device to use\n" 800 " in|out.period-length= length of period in microseconds\n" 801 " in|out.try-poll= attempt to use poll mode\n" 802 " threshold= threshold (in microseconds) when playback starts\n" 803#endif 804#ifdef CONFIG_AUDIO_COREAUDIO 805 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n" 806 " in|out.buffer-count= number of buffers\n" 807#endif 808#ifdef CONFIG_AUDIO_DSOUND 809 "-audiodev dsound,id=id[,prop[=value][,...]]\n" 810 " latency= add extra latency to playback in microseconds\n" 811#endif 812#ifdef CONFIG_AUDIO_OSS 813 "-audiodev oss,id=id[,prop[=value][,...]]\n" 814 " in|out.dev= path of the audio device to use\n" 815 " in|out.buffer-count= number of buffers\n" 816 " in|out.try-poll= attempt to use poll mode\n" 817 " try-mmap= try using memory mapped access\n" 818 " exclusive= open device in exclusive mode\n" 819 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n" 820#endif 821#ifdef CONFIG_AUDIO_PA 822 "-audiodev pa,id=id[,prop[=value][,...]]\n" 823 " server= PulseAudio server address\n" 824 " in|out.name= source/sink device name\n" 825 " in|out.latency= desired latency in microseconds\n" 826#endif 827#ifdef CONFIG_AUDIO_PIPEWIRE 828 "-audiodev pipewire,id=id[,prop[=value][,...]]\n" 829 " in|out.name= source/sink device name\n" 830 " in|out.stream-name= name of pipewire stream\n" 831 " in|out.latency= desired latency in microseconds\n" 832#endif 833#ifdef CONFIG_AUDIO_SDL 834 "-audiodev sdl,id=id[,prop[=value][,...]]\n" 835 " in|out.buffer-count= number of buffers\n" 836#endif 837#ifdef CONFIG_AUDIO_SNDIO 838 "-audiodev sndio,id=id[,prop[=value][,...]]\n" 839#endif 840#ifdef CONFIG_SPICE 841 "-audiodev spice,id=id[,prop[=value][,...]]\n" 842#endif 843#ifdef CONFIG_DBUS_DISPLAY 844 "-audiodev dbus,id=id[,prop[=value][,...]]\n" 845#endif 846 "-audiodev wav,id=id[,prop[=value][,...]]\n" 847 " path= path of wav file to record\n", 848 QEMU_ARCH_ALL) 849SRST 850``-audiodev [driver=]driver,id=id[,prop[=value][,...]]`` 851 Adds a new audio backend driver identified by id. There are global 852 and driver specific properties. Some values can be set differently 853 for input and output, they're marked with ``in|out.``. You can set 854 the input's property with ``in.prop`` and the output's property with 855 ``out.prop``. For example: 856 857 :: 858 859 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000 860 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified 861 862 NOTE: parameter validation is known to be incomplete, in many cases 863 specifying an invalid option causes QEMU to print an error message 864 and continue emulation without sound. 865 866 Valid global options are: 867 868 ``id=identifier`` 869 Identifies the audio backend. 870 871 ``timer-period=period`` 872 Sets the timer period used by the audio subsystem in 873 microseconds. Default is 10000 (10 ms). 874 875 ``in|out.mixing-engine=on|off`` 876 Use QEMU's mixing engine to mix all streams inside QEMU and 877 convert audio formats when not supported by the backend. When 878 off, fixed-settings must be off too. Note that disabling this 879 option means that the selected backend must support multiple 880 streams and the audio formats used by the virtual cards, 881 otherwise you'll get no sound. It's not recommended to disable 882 this option unless you want to use 5.1 or 7.1 audio, as mixing 883 engine only supports mono and stereo audio. Default is on. 884 885 ``in|out.fixed-settings=on|off`` 886 Use fixed settings for host audio. When off, it will change 887 based on how the guest opens the sound card. In this case you 888 must not specify frequency, channels or format. Default is on. 889 890 ``in|out.frequency=frequency`` 891 Specify the frequency to use when using fixed-settings. Default 892 is 44100Hz. 893 894 ``in|out.channels=channels`` 895 Specify the number of channels to use when using fixed-settings. 896 Default is 2 (stereo). 897 898 ``in|out.format=format`` 899 Specify the sample format to use when using fixed-settings. 900 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``, 901 ``u32``, ``f32``. Default is ``s16``. 902 903 ``in|out.voices=voices`` 904 Specify the number of voices to use. Default is 1. 905 906 ``in|out.buffer-length=usecs`` 907 Sets the size of the buffer in microseconds. 908 909``-audiodev none,id=id[,prop[=value][,...]]`` 910 Creates a dummy backend that discards all outputs. This backend has 911 no backend specific properties. 912 913``-audiodev alsa,id=id[,prop[=value][,...]]`` 914 Creates backend using the ALSA. This backend is only available on 915 Linux. 916 917 ALSA specific options are: 918 919 ``in|out.dev=device`` 920 Specify the ALSA device to use for input and/or output. Default 921 is ``default``. 922 923 ``in|out.period-length=usecs`` 924 Sets the period length in microseconds. 925 926 ``in|out.try-poll=on|off`` 927 Attempt to use poll mode with the device. Default is on. 928 929 ``threshold=threshold`` 930 Threshold (in microseconds) when playback starts. Default is 0. 931 932``-audiodev coreaudio,id=id[,prop[=value][,...]]`` 933 Creates a backend using Apple's Core Audio. This backend is only 934 available on Mac OS and only supports playback. 935 936 Core Audio specific options are: 937 938 ``in|out.buffer-count=count`` 939 Sets the count of the buffers. 940 941``-audiodev dsound,id=id[,prop[=value][,...]]`` 942 Creates a backend using Microsoft's DirectSound. This backend is 943 only available on Windows and only supports playback. 944 945 DirectSound specific options are: 946 947 ``latency=usecs`` 948 Add extra usecs microseconds latency to playback. Default is 949 10000 (10 ms). 950 951``-audiodev oss,id=id[,prop[=value][,...]]`` 952 Creates a backend using OSS. This backend is available on most 953 Unix-like systems. 954 955 OSS specific options are: 956 957 ``in|out.dev=device`` 958 Specify the file name of the OSS device to use. Default is 959 ``/dev/dsp``. 960 961 ``in|out.buffer-count=count`` 962 Sets the count of the buffers. 963 964 ``in|out.try-poll=on|of`` 965 Attempt to use poll mode with the device. Default is on. 966 967 ``try-mmap=on|off`` 968 Try using memory mapped device access. Default is off. 969 970 ``exclusive=on|off`` 971 Open the device in exclusive mode (vmix won't work in this 972 case). Default is off. 973 974 ``dsp-policy=policy`` 975 Sets the timing policy (between 0 and 10, where smaller number 976 means smaller latency but higher CPU usage). Use -1 to use 977 buffer sizes specified by ``buffer`` and ``buffer-count``. This 978 option is ignored if you do not have OSS 4. Default is 5. 979 980``-audiodev pa,id=id[,prop[=value][,...]]`` 981 Creates a backend using PulseAudio. This backend is available on 982 most systems. 983 984 PulseAudio specific options are: 985 986 ``server=server`` 987 Sets the PulseAudio server to connect to. 988 989 ``in|out.name=sink`` 990 Use the specified source/sink for recording/playback. 991 992 ``in|out.latency=usecs`` 993 Desired latency in microseconds. The PulseAudio server will try 994 to honor this value but actual latencies may be lower or higher. 995 996``-audiodev pipewire,id=id[,prop[=value][,...]]`` 997 Creates a backend using PipeWire. This backend is available on 998 most systems. 999 1000 PipeWire specific options are: 1001 1002 ``in|out.latency=usecs`` 1003 Desired latency in microseconds. 1004 1005 ``in|out.name=sink`` 1006 Use the specified source/sink for recording/playback. 1007 1008 ``in|out.stream-name`` 1009 Specify the name of pipewire stream. 1010 1011``-audiodev sdl,id=id[,prop[=value][,...]]`` 1012 Creates a backend using SDL. This backend is available on most 1013 systems, but you should use your platform's native backend if 1014 possible. 1015 1016 SDL specific options are: 1017 1018 ``in|out.buffer-count=count`` 1019 Sets the count of the buffers. 1020 1021``-audiodev sndio,id=id[,prop[=value][,...]]`` 1022 Creates a backend using SNDIO. This backend is available on 1023 OpenBSD and most other Unix-like systems. 1024 1025 Sndio specific options are: 1026 1027 ``in|out.dev=device`` 1028 Specify the sndio device to use for input and/or output. Default 1029 is ``default``. 1030 1031 ``in|out.latency=usecs`` 1032 Sets the desired period length in microseconds. 1033 1034``-audiodev spice,id=id[,prop[=value][,...]]`` 1035 Creates a backend that sends audio through SPICE. This backend 1036 requires ``-spice`` and automatically selected in that case, so 1037 usually you can ignore this option. This backend has no backend 1038 specific properties. 1039 1040``-audiodev wav,id=id[,prop[=value][,...]]`` 1041 Creates a backend that writes audio to a WAV file. 1042 1043 Backend specific options are: 1044 1045 ``path=path`` 1046 Write recorded audio into the specified file. Default is 1047 ``qemu.wav``. 1048ERST 1049 1050DEF("device", HAS_ARG, QEMU_OPTION_device, 1051 "-device driver[,prop[=value][,...]]\n" 1052 " add device (based on driver)\n" 1053 " prop=value,... sets driver properties\n" 1054 " use '-device help' to print all possible drivers\n" 1055 " use '-device driver,help' to print all possible properties\n", 1056 QEMU_ARCH_ALL) 1057SRST 1058``-device driver[,prop[=value][,...]]`` 1059 Add device driver. prop=value sets driver properties. Valid 1060 properties depend on the driver. To get help on possible drivers and 1061 properties, use ``-device help`` and ``-device driver,help``. 1062 1063 Some drivers are: 1064 1065``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]`` 1066 Add an IPMI BMC. This is a simulation of a hardware management 1067 interface processor that normally sits on a system. It provides a 1068 watchdog and the ability to reset and power control the system. You 1069 need to connect this to an IPMI interface to make it useful 1070 1071 The IPMI slave address to use for the BMC. The default is 0x20. This 1072 address is the BMC's address on the I2C network of management 1073 controllers. If you don't know what this means, it is safe to ignore 1074 it. 1075 1076 ``id=id`` 1077 The BMC id for interfaces to use this device. 1078 1079 ``slave_addr=val`` 1080 Define slave address to use for the BMC. The default is 0x20. 1081 1082 ``sdrfile=file`` 1083 file containing raw Sensor Data Records (SDR) data. The default 1084 is none. 1085 1086 ``fruareasize=val`` 1087 size of a Field Replaceable Unit (FRU) area. The default is 1088 1024. 1089 1090 ``frudatafile=file`` 1091 file containing raw Field Replaceable Unit (FRU) inventory data. 1092 The default is none. 1093 1094 ``guid=uuid`` 1095 value for the GUID for the BMC, in standard UUID format. If this 1096 is set, get "Get GUID" command to the BMC will return it. 1097 Otherwise "Get GUID" will return an error. 1098 1099``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]`` 1100 Add a connection to an external IPMI BMC simulator. Instead of 1101 locally emulating the BMC like the above item, instead connect to an 1102 external entity that provides the IPMI services. 1103 1104 A connection is made to an external BMC simulator. If you do this, 1105 it is strongly recommended that you use the "reconnect=" chardev 1106 option to reconnect to the simulator if the connection is lost. Note 1107 that if this is not used carefully, it can be a security issue, as 1108 the interface has the ability to send resets, NMIs, and power off 1109 the VM. It's best if QEMU makes a connection to an external 1110 simulator running on a secure port on localhost, so neither the 1111 simulator nor QEMU is exposed to any outside network. 1112 1113 See the "lanserv/README.vm" file in the OpenIPMI library for more 1114 details on the external interface. 1115 1116``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]`` 1117 Add a KCS IPMI interface on the ISA bus. This also adds a 1118 corresponding ACPI and SMBIOS entries, if appropriate. 1119 1120 ``bmc=id`` 1121 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern 1122 above. 1123 1124 ``ioport=val`` 1125 Define the I/O address of the interface. The default is 0xca0 1126 for KCS. 1127 1128 ``irq=val`` 1129 Define the interrupt to use. The default is 5. To disable 1130 interrupts, set this to 0. 1131 1132``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]`` 1133 Like the KCS interface, but defines a BT interface. The default port 1134 is 0xe4 and the default interrupt is 5. 1135 1136``-device pci-ipmi-kcs,bmc=id`` 1137 Add a KCS IPMI interface on the PCI bus. 1138 1139 ``bmc=id`` 1140 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above. 1141 1142``-device pci-ipmi-bt,bmc=id`` 1143 Like the KCS interface, but defines a BT interface on the PCI bus. 1144 1145``-device intel-iommu[,option=...]`` 1146 This is only supported by ``-machine q35``, which will enable Intel VT-d 1147 emulation within the guest. It supports below options: 1148 1149 ``intremap=on|off`` (default: auto) 1150 This enables interrupt remapping feature. It's required to enable 1151 complete x2apic. Currently it only supports kvm kernel-irqchip modes 1152 ``off`` or ``split``, while full kernel-irqchip is not yet supported. 1153 The default value is "auto", which will be decided by the mode of 1154 kernel-irqchip. 1155 1156 ``caching-mode=on|off`` (default: off) 1157 This enables caching mode for the VT-d emulated device. When 1158 caching-mode is enabled, each guest DMA buffer mapping will generate an 1159 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in 1160 a synchronous way. It is required for ``-device vfio-pci`` to work 1161 with the VT-d device, because host assigned devices requires to setup 1162 the DMA mapping on the host before guest DMA starts. 1163 1164 ``device-iotlb=on|off`` (default: off) 1165 This enables device-iotlb capability for the emulated VT-d device. So 1166 far virtio/vhost should be the only real user for this parameter, 1167 paired with ats=on configured for the device. 1168 1169 ``aw-bits=39|48`` (default: 39) 1170 This decides the address width of IOVA address space. The address 1171 space has 39 bits width for 3-level IOMMU page tables, and 48 bits for 1172 4-level IOMMU page tables. 1173 1174 Please also refer to the wiki page for general scenarios of VT-d 1175 emulation in QEMU: https://wiki.qemu.org/Features/VT-d. 1176 1177``-device virtio-iommu-pci[,option=...]`` 1178 This is only supported by ``-machine q35`` (x86_64) and ``-machine virt`` (ARM). 1179 It supports below options: 1180 1181 ``granule=val`` (possible values are 4k, 8k, 16k, 64k and host; default: host) 1182 This decides the default granule to be be exposed by the 1183 virtio-iommu. If host, the granule matches the host page size. 1184 1185 ``aw-bits=val`` (val between 32 and 64, default depends on machine) 1186 This decides the address width of the IOVA address space. 1187 1188ERST 1189 1190DEF("name", HAS_ARG, QEMU_OPTION_name, 1191 "-name string1[,process=string2][,debug-threads=on|off]\n" 1192 " set the name of the guest\n" 1193 " string1 sets the window title and string2 the process name\n" 1194 " When debug-threads is enabled, individual threads are given a separate name\n" 1195 " NOTE: The thread names are for debugging and not a stable API.\n", 1196 QEMU_ARCH_ALL) 1197SRST 1198``-name name`` 1199 Sets the name of the guest. This name will be displayed in the SDL 1200 window caption. The name will also be used for the VNC server. Also 1201 optionally set the top visible process name in Linux. Naming of 1202 individual threads can also be enabled on Linux to aid debugging. 1203ERST 1204 1205DEF("uuid", HAS_ARG, QEMU_OPTION_uuid, 1206 "-uuid %08x-%04x-%04x-%04x-%012x\n" 1207 " specify machine UUID\n", QEMU_ARCH_ALL) 1208SRST 1209``-uuid uuid`` 1210 Set system UUID. 1211ERST 1212 1213DEFHEADING() 1214 1215DEFHEADING(Block device options:) 1216 1217SRST 1218The QEMU block device handling options have a long history and 1219have gone through several iterations as the feature set and complexity 1220of the block layer have grown. Many online guides to QEMU often 1221reference older and deprecated options, which can lead to confusion. 1222 1223The most explicit way to describe disks is to use a combination of 1224``-device`` to specify the hardware device and ``-blockdev`` to 1225describe the backend. The device defines what the guest sees and the 1226backend describes how QEMU handles the data. It is the only guaranteed 1227stable interface for describing block devices and as such is 1228recommended for management tools and scripting. 1229 1230The ``-drive`` option combines the device and backend into a single 1231command line option which is a more human friendly. There is however no 1232interface stability guarantee although some older board models still 1233need updating to work with the modern blockdev forms. 1234 1235Older options like ``-hda`` are essentially macros which expand into 1236``-drive`` options for various drive interfaces. The original forms 1237bake in a lot of assumptions from the days when QEMU was emulating a 1238legacy PC, they are not recommended for modern configurations. 1239 1240ERST 1241 1242DEF("fda", HAS_ARG, QEMU_OPTION_fda, 1243 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL) 1244DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL) 1245SRST 1246``-fda file`` 1247 \ 1248``-fdb file`` 1249 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in 1250 the System Emulation Users Guide). 1251ERST 1252 1253DEF("hda", HAS_ARG, QEMU_OPTION_hda, 1254 "-hda/-hdb file use 'file' as hard disk 0/1 image\n", QEMU_ARCH_ALL) 1255DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL) 1256DEF("hdc", HAS_ARG, QEMU_OPTION_hdc, 1257 "-hdc/-hdd file use 'file' as hard disk 2/3 image\n", QEMU_ARCH_ALL) 1258DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL) 1259SRST 1260``-hda file`` 1261 \ 1262``-hdb file`` 1263 \ 1264``-hdc file`` 1265 \ 1266``-hdd file`` 1267 Use file as hard disk 0, 1, 2 or 3 image on the default bus of the 1268 emulated machine (this is for example the IDE bus on most x86 machines, 1269 but it can also be SCSI, virtio or something else on other target 1270 architectures). See also the :ref:`disk images` chapter in the System 1271 Emulation Users Guide. 1272ERST 1273 1274DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom, 1275 "-cdrom file use 'file' as CD-ROM image\n", 1276 QEMU_ARCH_ALL) 1277SRST 1278``-cdrom file`` 1279 Use file as CD-ROM image on the default bus of the emulated machine 1280 (which is IDE1 master on x86, so you cannot use ``-hdc`` and ``-cdrom`` 1281 at the same time there). On systems that support it, you can use the 1282 host CD-ROM by using ``/dev/cdrom`` as filename. 1283ERST 1284 1285DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev, 1286 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n" 1287 " [,cache.direct=on|off][,cache.no-flush=on|off]\n" 1288 " [,read-only=on|off][,auto-read-only=on|off]\n" 1289 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n" 1290 " [,driver specific parameters...]\n" 1291 " configure a block backend\n", QEMU_ARCH_ALL) 1292SRST 1293``-blockdev option[,option[,option[,...]]]`` 1294 Define a new block driver node. Some of the options apply to all 1295 block drivers, other options are only accepted for a specific block 1296 driver. See below for a list of generic options and options for the 1297 most common block drivers. 1298 1299 Options that expect a reference to another node (e.g. ``file``) can 1300 be given in two ways. Either you specify the node name of an already 1301 existing node (file=node-name), or you define a new node inline, 1302 adding options for the referenced node after a dot 1303 (file.filename=path,file.aio=native). 1304 1305 A block driver node created with ``-blockdev`` can be used for a 1306 guest device by specifying its node name for the ``drive`` property 1307 in a ``-device`` argument that defines a block device. 1308 1309 ``Valid options for any block driver node:`` 1310 ``driver`` 1311 Specifies the block driver to use for the given node. 1312 1313 ``node-name`` 1314 This defines the name of the block driver node by which it 1315 will be referenced later. The name must be unique, i.e. it 1316 must not match the name of a different block driver node, or 1317 (if you use ``-drive`` as well) the ID of a drive. 1318 1319 If no node name is specified, it is automatically generated. 1320 The generated node name is not intended to be predictable 1321 and changes between QEMU invocations. For the top level, an 1322 explicit node name must be specified. 1323 1324 ``read-only`` 1325 Open the node read-only. Guest write attempts will fail. 1326 1327 Note that some block drivers support only read-only access, 1328 either generally or in certain configurations. In this case, 1329 the default value ``read-only=off`` does not work and the 1330 option must be specified explicitly. 1331 1332 ``auto-read-only`` 1333 If ``auto-read-only=on`` is set, QEMU may fall back to 1334 read-only usage even when ``read-only=off`` is requested, or 1335 even switch between modes as needed, e.g. depending on 1336 whether the image file is writable or whether a writing user 1337 is attached to the node. 1338 1339 ``force-share`` 1340 Override the image locking system of QEMU by forcing the 1341 node to utilize weaker shared access for permissions where 1342 it would normally request exclusive access. When there is 1343 the potential for multiple instances to have the same file 1344 open (whether this invocation of QEMU is the first or the 1345 second instance), both instances must permit shared access 1346 for the second instance to succeed at opening the file. 1347 1348 Enabling ``force-share=on`` requires ``read-only=on``. 1349 1350 ``cache.direct`` 1351 The host page cache can be avoided with ``cache.direct=on``. 1352 This will attempt to do disk IO directly to the guest's 1353 memory. QEMU may still perform an internal copy of the data. 1354 1355 ``cache.no-flush`` 1356 In case you don't care about data integrity over host 1357 failures, you can use ``cache.no-flush=on``. This option 1358 tells QEMU that it never needs to write any data to the disk 1359 but can instead keep things in cache. If anything goes 1360 wrong, like your host losing power, the disk storage getting 1361 disconnected accidentally, etc. your image will most 1362 probably be rendered unusable. 1363 1364 ``discard=discard`` 1365 discard is one of "ignore" (or "off") or "unmap" (or "on") 1366 and controls whether ``discard`` (also known as ``trim`` or 1367 ``unmap``) requests are ignored or passed to the filesystem. 1368 Some machine types may not support discard requests. 1369 1370 ``detect-zeroes=detect-zeroes`` 1371 detect-zeroes is "off", "on" or "unmap" and enables the 1372 automatic conversion of plain zero writes by the OS to 1373 driver specific optimized zero write commands. You may even 1374 choose "unmap" if discard is set to "unmap" to allow a zero 1375 write to be converted to an ``unmap`` operation. 1376 1377 ``Driver-specific options for file`` 1378 This is the protocol-level block driver for accessing regular 1379 files. 1380 1381 ``filename`` 1382 The path to the image file in the local filesystem 1383 1384 ``aio`` 1385 Specifies the AIO backend (threads/native/io_uring, 1386 default: threads) 1387 1388 ``locking`` 1389 Specifies whether the image file is protected with Linux OFD 1390 / POSIX locks. The default is to use the Linux Open File 1391 Descriptor API if available, otherwise no lock is applied. 1392 (auto/on/off, default: auto) 1393 1394 Example: 1395 1396 :: 1397 1398 -blockdev driver=file,node-name=disk,filename=disk.img 1399 1400 ``Driver-specific options for raw`` 1401 This is the image format block driver for raw images. It is 1402 usually stacked on top of a protocol level block driver such as 1403 ``file``. 1404 1405 ``file`` 1406 Reference to or definition of the data source block driver 1407 node (e.g. a ``file`` driver node) 1408 1409 Example 1: 1410 1411 :: 1412 1413 -blockdev driver=file,node-name=disk_file,filename=disk.img 1414 -blockdev driver=raw,node-name=disk,file=disk_file 1415 1416 Example 2: 1417 1418 :: 1419 1420 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img 1421 1422 ``Driver-specific options for qcow2`` 1423 This is the image format block driver for qcow2 images. It is 1424 usually stacked on top of a protocol level block driver such as 1425 ``file``. 1426 1427 ``file`` 1428 Reference to or definition of the data source block driver 1429 node (e.g. a ``file`` driver node) 1430 1431 ``backing`` 1432 Reference to or definition of the backing file block device 1433 (default is taken from the image file). It is allowed to 1434 pass ``null`` here in order to disable the default backing 1435 file. 1436 1437 ``lazy-refcounts`` 1438 Whether to enable the lazy refcounts feature (on/off; 1439 default is taken from the image file) 1440 1441 ``cache-size`` 1442 The maximum total size of the L2 table and refcount block 1443 caches in bytes (default: the sum of l2-cache-size and 1444 refcount-cache-size) 1445 1446 ``l2-cache-size`` 1447 The maximum size of the L2 table cache in bytes (default: if 1448 cache-size is not specified - 32M on Linux platforms, and 8M 1449 on non-Linux platforms; otherwise, as large as possible 1450 within the cache-size, while permitting the requested or the 1451 minimal refcount cache size) 1452 1453 ``refcount-cache-size`` 1454 The maximum size of the refcount block cache in bytes 1455 (default: 4 times the cluster size; or if cache-size is 1456 specified, the part of it which is not used for the L2 1457 cache) 1458 1459 ``cache-clean-interval`` 1460 Clean unused entries in the L2 and refcount caches. The 1461 interval is in seconds. The default value is 600 on 1462 supporting platforms, and 0 on other platforms. Setting it 1463 to 0 disables this feature. 1464 1465 ``pass-discard-request`` 1466 Whether discard requests to the qcow2 device should be 1467 forwarded to the data source (on/off; default: on if 1468 discard=unmap is specified, off otherwise) 1469 1470 ``pass-discard-snapshot`` 1471 Whether discard requests for the data source should be 1472 issued when a snapshot operation (e.g. deleting a snapshot) 1473 frees clusters in the qcow2 file (on/off; default: on) 1474 1475 ``pass-discard-other`` 1476 Whether discard requests for the data source should be 1477 issued on other occasions where a cluster gets freed 1478 (on/off; default: off) 1479 1480 ``discard-no-unref`` 1481 When enabled, data clusters will remain preallocated when they are 1482 no longer used, e.g. because they are discarded or converted to 1483 zero clusters. As usual, whether the old data is discarded or kept 1484 on the protocol level (i.e. in the image file) depends on the 1485 setting of the pass-discard-request option. Keeping the clusters 1486 preallocated prevents qcow2 fragmentation that would otherwise be 1487 caused by freeing and re-allocating them later. Besides potential 1488 performance degradation, such fragmentation can lead to increased 1489 allocation of clusters past the end of the image file, 1490 resulting in image files whose file length can grow much larger 1491 than their guest disk size would suggest. 1492 If image file length is of concern (e.g. when storing qcow2 1493 images directly on block devices), you should consider enabling 1494 this option. 1495 1496 ``overlap-check`` 1497 Which overlap checks to perform for writes to the image 1498 (none/constant/cached/all; default: cached). For details or 1499 finer granularity control refer to the QAPI documentation of 1500 ``blockdev-add``. 1501 1502 Example 1: 1503 1504 :: 1505 1506 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2 1507 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216 1508 1509 Example 2: 1510 1511 :: 1512 1513 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2 1514 1515 ``Driver-specific options for other drivers`` 1516 Please refer to the QAPI documentation of the ``blockdev-add`` 1517 QMP command. 1518ERST 1519 1520DEF("drive", HAS_ARG, QEMU_OPTION_drive, 1521 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n" 1522 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n" 1523 " [,snapshot=on|off][,rerror=ignore|stop|report]\n" 1524 " [,werror=ignore|stop|report|enospc][,id=name]\n" 1525 " [,aio=threads|native|io_uring]\n" 1526 " [,readonly=on|off][,copy-on-read=on|off]\n" 1527 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n" 1528 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n" 1529 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n" 1530 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n" 1531 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n" 1532 " [[,iops_size=is]]\n" 1533 " [[,group=g]]\n" 1534 " use 'file' as a drive image\n", QEMU_ARCH_ALL) 1535SRST 1536``-drive option[,option[,option[,...]]]`` 1537 Define a new drive. This includes creating a block driver node (the 1538 backend) as well as a guest device, and is mostly a shortcut for 1539 defining the corresponding ``-blockdev`` and ``-device`` options. 1540 1541 ``-drive`` accepts all options that are accepted by ``-blockdev``. 1542 In addition, it knows the following options: 1543 1544 ``file=file`` 1545 This option defines which disk image (see the :ref:`disk images` 1546 chapter in the System Emulation Users Guide) to use with this drive. 1547 If the filename contains comma, you must double it (for instance, 1548 "file=my,,file" to use file "my,file"). 1549 1550 Special files such as iSCSI devices can be specified using 1551 protocol specific URLs. See the section for "Device URL Syntax" 1552 for more information. 1553 1554 ``if=interface`` 1555 This option defines on which type on interface the drive is 1556 connected. Available types are: ide, scsi, sd, mtd, floppy, 1557 pflash, virtio, none. 1558 1559 ``bus=bus,unit=unit`` 1560 These options define where is connected the drive by defining 1561 the bus number and the unit id. 1562 1563 ``index=index`` 1564 This option defines where the drive is connected by using an 1565 index in the list of available connectors of a given interface 1566 type. 1567 1568 ``media=media`` 1569 This option defines the type of the media: disk or cdrom. 1570 1571 ``snapshot=snapshot`` 1572 snapshot is "on" or "off" and controls snapshot mode for the 1573 given drive (see ``-snapshot``). 1574 1575 ``cache=cache`` 1576 cache is "none", "writeback", "unsafe", "directsync" or 1577 "writethrough" and controls how the host cache is used to access 1578 block data. This is a shortcut that sets the ``cache.direct`` 1579 and ``cache.no-flush`` options (as in ``-blockdev``), and 1580 additionally ``cache.writeback``, which provides a default for 1581 the ``write-cache`` option of block guest devices (as in 1582 ``-device``). The modes correspond to the following settings: 1583 1584 ============= =============== ============ ============== 1585 \ cache.writeback cache.direct cache.no-flush 1586 ============= =============== ============ ============== 1587 writeback on off off 1588 none on on off 1589 writethrough off off off 1590 directsync off on off 1591 unsafe on off on 1592 ============= =============== ============ ============== 1593 1594 The default mode is ``cache=writeback``. 1595 1596 ``aio=aio`` 1597 aio is "threads", "native", or "io_uring" and selects between pthread 1598 based disk I/O, native Linux AIO, or Linux io_uring API. 1599 1600 ``format=format`` 1601 Specify which disk format will be used rather than detecting the 1602 format. Can be used to specify format=raw to avoid interpreting 1603 an untrusted format header. 1604 1605 ``werror=action,rerror=action`` 1606 Specify which action to take on write and read errors. Valid 1607 actions are: "ignore" (ignore the error and try to continue), 1608 "stop" (pause QEMU), "report" (report the error to the guest), 1609 "enospc" (pause QEMU only if the host disk is full; report the 1610 error to the guest otherwise). The default setting is 1611 ``werror=enospc`` and ``rerror=report``. 1612 1613 ``copy-on-read=copy-on-read`` 1614 copy-on-read is "on" or "off" and enables whether to copy read 1615 backing file sectors into the image file. 1616 1617 ``bps=b,bps_rd=r,bps_wr=w`` 1618 Specify bandwidth throttling limits in bytes per second, either 1619 for all request types or for reads or writes only. Small values 1620 can lead to timeouts or hangs inside the guest. A safe minimum 1621 for disks is 2 MB/s. 1622 1623 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm`` 1624 Specify bursts in bytes per second, either for all request types 1625 or for reads or writes only. Bursts allow the guest I/O to spike 1626 above the limit temporarily. 1627 1628 ``iops=i,iops_rd=r,iops_wr=w`` 1629 Specify request rate limits in requests per second, either for 1630 all request types or for reads or writes only. 1631 1632 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm`` 1633 Specify bursts in requests per second, either for all request 1634 types or for reads or writes only. Bursts allow the guest I/O to 1635 spike above the limit temporarily. 1636 1637 ``iops_size=is`` 1638 Let every is bytes of a request count as a new request for iops 1639 throttling purposes. Use this option to prevent guests from 1640 circumventing iops limits by sending fewer but larger requests. 1641 1642 ``group=g`` 1643 Join a throttling quota group with given name g. All drives that 1644 are members of the same group are accounted for together. Use 1645 this option to prevent guests from circumventing throttling 1646 limits by using many small disks instead of a single larger 1647 disk. 1648 1649 By default, the ``cache.writeback=on`` mode is used. It will report 1650 data writes as completed as soon as the data is present in the host 1651 page cache. This is safe as long as your guest OS makes sure to 1652 correctly flush disk caches where needed. If your guest OS does not 1653 handle volatile disk write caches correctly and your host crashes or 1654 loses power, then the guest may experience data corruption. 1655 1656 For such guests, you should consider using ``cache.writeback=off``. 1657 This means that the host page cache will be used to read and write 1658 data, but write notification will be sent to the guest only after 1659 QEMU has made sure to flush each write to the disk. Be aware that 1660 this has a major impact on performance. 1661 1662 When using the ``-snapshot`` option, unsafe caching is always used. 1663 1664 Copy-on-read avoids accessing the same backing file sectors 1665 repeatedly and is useful when the backing file is over a slow 1666 network. By default copy-on-read is off. 1667 1668 Instead of ``-cdrom`` you can use: 1669 1670 .. parsed-literal:: 1671 1672 |qemu_system| -drive file=file,index=2,media=cdrom 1673 1674 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use: 1675 1676 .. parsed-literal:: 1677 1678 |qemu_system| -drive file=file,index=0,media=disk 1679 |qemu_system| -drive file=file,index=1,media=disk 1680 |qemu_system| -drive file=file,index=2,media=disk 1681 |qemu_system| -drive file=file,index=3,media=disk 1682 1683 You can open an image using pre-opened file descriptors from an fd 1684 set: 1685 1686 .. parsed-literal:: 1687 1688 |qemu_system| \\ 1689 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 1690 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 1691 -drive file=/dev/fdset/2,index=0,media=disk 1692 1693 You can connect a CDROM to the slave of ide0: 1694 1695 .. parsed-literal:: 1696 1697 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom 1698 1699 If you don't specify the "file=" argument, you define an empty 1700 drive: 1701 1702 .. parsed-literal:: 1703 1704 |qemu_system_x86| -drive if=ide,index=1,media=cdrom 1705 1706 Instead of ``-fda``, ``-fdb``, you can use: 1707 1708 .. parsed-literal:: 1709 1710 |qemu_system_x86| -drive file=file,index=0,if=floppy 1711 |qemu_system_x86| -drive file=file,index=1,if=floppy 1712 1713 By default, interface is "ide" and index is automatically 1714 incremented: 1715 1716 .. parsed-literal:: 1717 1718 |qemu_system_x86| -drive file=a -drive file=b 1719 1720 is interpreted like: 1721 1722 .. parsed-literal:: 1723 1724 |qemu_system_x86| -hda a -hdb b 1725ERST 1726 1727DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock, 1728 "-mtdblock file use 'file' as on-board Flash memory image\n", 1729 QEMU_ARCH_ALL) 1730SRST 1731``-mtdblock file`` 1732 Use file as on-board Flash memory image. 1733ERST 1734 1735DEF("sd", HAS_ARG, QEMU_OPTION_sd, 1736 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL) 1737SRST 1738``-sd file`` 1739 Use file as SecureDigital card image. 1740ERST 1741 1742DEF("snapshot", 0, QEMU_OPTION_snapshot, 1743 "-snapshot write to temporary files instead of disk image files\n", 1744 QEMU_ARCH_ALL) 1745SRST 1746``-snapshot`` 1747 Write to temporary files instead of disk image files. In this case, 1748 the raw disk image you use is not written back. You can however 1749 force the write back by pressing C-a s (see the :ref:`disk images` 1750 chapter in the System Emulation Users Guide). 1751 1752 .. warning:: 1753 snapshot is incompatible with ``-blockdev`` (instead use qemu-img 1754 to manually create snapshot images to attach to your blockdev). 1755 If you have mixed ``-blockdev`` and ``-drive`` declarations you 1756 can use the 'snapshot' property on your drive declarations 1757 instead of this global option. 1758 1759ERST 1760 1761DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev, 1762 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1763 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n" 1764 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n" 1765 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n" 1766 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n" 1767 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n" 1768 " [[,throttling.iops-size=is]]\n" 1769 "-fsdev synth,id=id\n", 1770 QEMU_ARCH_ALL) 1771 1772SRST 1773``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]`` 1774 \ 1775``-fsdev synth,id=id[,readonly=on]`` 1776 Define a new file system device. Valid options are: 1777 1778 ``local`` 1779 Accesses to the filesystem are done by QEMU. 1780 1781 ``synth`` 1782 Synthetic filesystem, only used by QTests. 1783 1784 ``id=id`` 1785 Specifies identifier for this device. 1786 1787 ``path=path`` 1788 Specifies the export path for the file system device. Files 1789 under this path will be available to the 9p client on the guest. 1790 1791 ``security_model=security_model`` 1792 Specifies the security model to be used for this export path. 1793 Supported security models are "passthrough", "mapped-xattr", 1794 "mapped-file" and "none". In "passthrough" security model, files 1795 are stored using the same credentials as they are created on the 1796 guest. This requires QEMU to run as root. In "mapped-xattr" 1797 security model, some of the file attributes like uid, gid, mode 1798 bits and link target are stored as file attributes. For 1799 "mapped-file" these attributes are stored in the hidden 1800 .virtfs\_metadata directory. Directories exported by this 1801 security model cannot interact with other unix tools. "none" 1802 security model is same as passthrough except the sever won't 1803 report failures if it fails to set file attributes like 1804 ownership. Security model is mandatory only for local fsdriver. 1805 1806 ``writeout=writeout`` 1807 This is an optional argument. The only supported value is 1808 "immediate". This means that host page cache will be used to 1809 read and write data but write notification will be sent to the 1810 guest only when the data has been reported as written by the 1811 storage subsystem. 1812 1813 ``readonly=on`` 1814 Enables exporting 9p share as a readonly mount for guests. By 1815 default read-write access is given. 1816 1817 ``fmode=fmode`` 1818 Specifies the default mode for newly created files on the host. 1819 Works only with security models "mapped-xattr" and 1820 "mapped-file". 1821 1822 ``dmode=dmode`` 1823 Specifies the default mode for newly created directories on the 1824 host. Works only with security models "mapped-xattr" and 1825 "mapped-file". 1826 1827 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w`` 1828 Specify bandwidth throttling limits in bytes per second, either 1829 for all request types or for reads or writes only. 1830 1831 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm`` 1832 Specify bursts in bytes per second, either for all request types 1833 or for reads or writes only. Bursts allow the guest I/O to spike 1834 above the limit temporarily. 1835 1836 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w`` 1837 Specify request rate limits in requests per second, either for 1838 all request types or for reads or writes only. 1839 1840 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm`` 1841 Specify bursts in requests per second, either for all request 1842 types or for reads or writes only. Bursts allow the guest I/O to 1843 spike above the limit temporarily. 1844 1845 ``throttling.iops-size=is`` 1846 Let every is bytes of a request count as a new request for iops 1847 throttling purposes. 1848 1849 -fsdev option is used along with -device driver "virtio-9p-...". 1850 1851``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag`` 1852 Options for virtio-9p-... driver are: 1853 1854 ``type`` 1855 Specifies the variant to be used. Supported values are "pci", 1856 "ccw" or "device", depending on the machine type. 1857 1858 ``fsdev=id`` 1859 Specifies the id value specified along with -fsdev option. 1860 1861 ``mount_tag=mount_tag`` 1862 Specifies the tag name to be used by the guest to mount this 1863 export point. 1864ERST 1865 1866DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs, 1867 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1868 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n" 1869 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n", 1870 QEMU_ARCH_ALL) 1871 1872SRST 1873``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]`` 1874 \ 1875``-virtfs synth,mount_tag=mount_tag`` 1876 Define a new virtual filesystem device and expose it to the guest using 1877 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain 1878 directory on host is made directly accessible by guest as a pass-through 1879 file system by using the 9P network protocol for communication between 1880 host and guests, if desired even accessible, shared by several guests 1881 simultaneously. 1882 1883 Note that ``-virtfs`` is actually just a convenience shortcut for its 1884 generalized form ``-fsdev -device virtio-9p-pci``. 1885 1886 The general form of pass-through file system options are: 1887 1888 ``local`` 1889 Accesses to the filesystem are done by QEMU. 1890 1891 ``synth`` 1892 Synthetic filesystem, only used by QTests. 1893 1894 ``id=id`` 1895 Specifies identifier for the filesystem device 1896 1897 ``path=path`` 1898 Specifies the export path for the file system device. Files 1899 under this path will be available to the 9p client on the guest. 1900 1901 ``security_model=security_model`` 1902 Specifies the security model to be used for this export path. 1903 Supported security models are "passthrough", "mapped-xattr", 1904 "mapped-file" and "none". In "passthrough" security model, files 1905 are stored using the same credentials as they are created on the 1906 guest. This requires QEMU to run as root. In "mapped-xattr" 1907 security model, some of the file attributes like uid, gid, mode 1908 bits and link target are stored as file attributes. For 1909 "mapped-file" these attributes are stored in the hidden 1910 .virtfs\_metadata directory. Directories exported by this 1911 security model cannot interact with other unix tools. "none" 1912 security model is same as passthrough except the sever won't 1913 report failures if it fails to set file attributes like 1914 ownership. Security model is mandatory only for local fsdriver. 1915 1916 ``writeout=writeout`` 1917 This is an optional argument. The only supported value is 1918 "immediate". This means that host page cache will be used to 1919 read and write data but write notification will be sent to the 1920 guest only when the data has been reported as written by the 1921 storage subsystem. 1922 1923 ``readonly=on`` 1924 Enables exporting 9p share as a readonly mount for guests. By 1925 default read-write access is given. 1926 1927 ``fmode=fmode`` 1928 Specifies the default mode for newly created files on the host. 1929 Works only with security models "mapped-xattr" and 1930 "mapped-file". 1931 1932 ``dmode=dmode`` 1933 Specifies the default mode for newly created directories on the 1934 host. Works only with security models "mapped-xattr" and 1935 "mapped-file". 1936 1937 ``mount_tag=mount_tag`` 1938 Specifies the tag name to be used by the guest to mount this 1939 export point. 1940 1941 ``multidevs=multidevs`` 1942 Specifies how to deal with multiple devices being shared with a 1943 9p export. Supported behaviours are either "remap", "forbid" or 1944 "warn". The latter is the default behaviour on which virtfs 9p 1945 expects only one device to be shared with the same export, and 1946 if more than one device is shared and accessed via the same 9p 1947 export then only a warning message is logged (once) by qemu on 1948 host side. In order to avoid file ID collisions on guest you 1949 should either create a separate virtfs export for each device to 1950 be shared with guests (recommended way) or you might use "remap" 1951 instead which allows you to share multiple devices with only one 1952 export instead, which is achieved by remapping the original 1953 inode numbers from host to guest in a way that would prevent 1954 such collisions. Remapping inodes in such use cases is required 1955 because the original device IDs from host are never passed and 1956 exposed on guest. Instead all files of an export shared with 1957 virtfs always share the same device id on guest. So two files 1958 with identical inode numbers but from actually different devices 1959 on host would otherwise cause a file ID collision and hence 1960 potential misbehaviours on guest. "forbid" on the other hand 1961 assumes like "warn" that only one device is shared by the same 1962 export, however it will not only log a warning message but also 1963 deny access to additional devices on guest. Note though that 1964 "forbid" does currently not block all possible file access 1965 operations (e.g. readdir() would still return entries from other 1966 devices). 1967ERST 1968 1969DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi, 1970 "-iscsi [user=user][,password=password][,password-secret=secret-id]\n" 1971 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE]\n" 1972 " [,initiator-name=initiator-iqn][,id=target-iqn]\n" 1973 " [,timeout=timeout]\n" 1974 " iSCSI session parameters\n", QEMU_ARCH_ALL) 1975 1976SRST 1977``-iscsi`` 1978 Configure iSCSI session parameters. 1979ERST 1980 1981DEFHEADING() 1982 1983DEFHEADING(USB convenience options:) 1984 1985DEF("usb", 0, QEMU_OPTION_usb, 1986 "-usb enable on-board USB host controller (if not enabled by default)\n", 1987 QEMU_ARCH_ALL) 1988SRST 1989``-usb`` 1990 Enable USB emulation on machine types with an on-board USB host 1991 controller (if not enabled by default). Note that on-board USB host 1992 controllers may not support USB 3.0. In this case 1993 ``-device qemu-xhci`` can be used instead on machines with PCI. 1994ERST 1995 1996DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice, 1997 "-usbdevice name add the host or guest USB device 'name'\n", 1998 QEMU_ARCH_ALL) 1999SRST 2000``-usbdevice devname`` 2001 Add the USB device devname, and enable an on-board USB controller 2002 if possible and necessary (just like it can be done via 2003 ``-machine usb=on``). Note that this option is mainly intended for 2004 the user's convenience only. More fine-grained control can be 2005 achieved by selecting a USB host controller (if necessary) and the 2006 desired USB device via the ``-device`` option instead. For example, 2007 instead of using ``-usbdevice mouse`` it is possible to use 2008 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse 2009 to a USB 3.0 controller instead (at least on machines that support 2010 PCI and do not have an USB controller enabled by default yet). 2011 For more details, see the chapter about 2012 :ref:`Connecting USB devices` in the System Emulation Users Guide. 2013 Possible devices for devname are: 2014 2015 ``braille`` 2016 Braille device. This will use BrlAPI to display the braille 2017 output on a real or fake device (i.e. it also creates a 2018 corresponding ``braille`` chardev automatically beside the 2019 ``usb-braille`` USB device). 2020 2021 ``keyboard`` 2022 Standard USB keyboard. Will override the PS/2 keyboard (if present). 2023 2024 ``mouse`` 2025 Virtual Mouse. This will override the PS/2 mouse emulation when 2026 activated. 2027 2028 ``tablet`` 2029 Pointer device that uses absolute coordinates (like a 2030 touchscreen). This means QEMU is able to report the mouse 2031 position without having to grab the mouse. Also overrides the 2032 PS/2 mouse emulation when activated. 2033 2034 ``wacom-tablet`` 2035 Wacom PenPartner USB tablet. 2036 2037 2038ERST 2039 2040DEFHEADING() 2041 2042DEFHEADING(Display options:) 2043 2044DEF("display", HAS_ARG, QEMU_OPTION_display, 2045#if defined(CONFIG_SPICE) 2046 "-display spice-app[,gl=on|off]\n" 2047#endif 2048#if defined(CONFIG_SDL) 2049 "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n" 2050 " [,window-close=on|off]\n" 2051#endif 2052#if defined(CONFIG_GTK) 2053 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n" 2054 " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n" 2055 " [,show-menubar=on|off][,zoom-to-fit=on|off]\n" 2056#endif 2057#if defined(CONFIG_VNC) 2058 "-display vnc=<display>[,<optargs>]\n" 2059#endif 2060#if defined(CONFIG_CURSES) 2061 "-display curses[,charset=<encoding>]\n" 2062#endif 2063#if defined(CONFIG_COCOA) 2064 "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n" 2065 " [,show-cursor=on|off][,left-command-key=on|off]\n" 2066 " [,full-screen=on|off][,zoom-to-fit=on|off]\n" 2067#endif 2068#if defined(CONFIG_OPENGL) 2069 "-display egl-headless[,rendernode=<file>]\n" 2070#endif 2071#if defined(CONFIG_DBUS_DISPLAY) 2072 "-display dbus[,addr=<dbusaddr>]\n" 2073 " [,gl=on|core|es|off][,rendernode=<file>]\n" 2074#endif 2075 "-display none\n" 2076 " select display backend type\n" 2077 " The default display is equivalent to\n " 2078#if defined(CONFIG_GTK) 2079 "\"-display gtk\"\n" 2080#elif defined(CONFIG_SDL) 2081 "\"-display sdl\"\n" 2082#elif defined(CONFIG_COCOA) 2083 "\"-display cocoa\"\n" 2084#elif defined(CONFIG_VNC) 2085 "\"-vnc localhost:0,to=99,id=default\"\n" 2086#else 2087 "\"-display none\"\n" 2088#endif 2089 , QEMU_ARCH_ALL) 2090SRST 2091``-display type`` 2092 Select type of display to use. Use ``-display help`` to list the available 2093 display types. Valid values for type are 2094 2095 ``spice-app[,gl=on|off]`` 2096 Start QEMU as a Spice server and launch the default Spice client 2097 application. The Spice server will redirect the serial consoles 2098 and QEMU monitors. (Since 4.0) 2099 2100 ``dbus`` 2101 Export the display over D-Bus interfaces. (Since 7.0) 2102 2103 The connection is registered with the "org.qemu" name (and queued when 2104 already owned). 2105 2106 ``addr=<dbusaddr>`` : D-Bus bus address to connect to. 2107 2108 ``p2p=yes|no`` : Use peer-to-peer connection, accepted via QMP ``add_client``. 2109 2110 ``gl=on|off|core|es`` : Use OpenGL for rendering (the D-Bus interface 2111 will share framebuffers with DMABUF file descriptors). 2112 2113 ``sdl`` 2114 Display video output via SDL (usually in a separate graphics 2115 window; see the SDL documentation for other possibilities). 2116 Valid parameters are: 2117 2118 ``grab-mod=<mods>`` : Used to select the modifier keys for toggling 2119 the mouse grabbing in conjunction with the "g" key. ``<mods>`` can be 2120 either ``lshift-lctrl-lalt`` or ``rctrl``. 2121 2122 ``gl=on|off|core|es`` : Use OpenGL for displaying 2123 2124 ``show-cursor=on|off`` : Force showing the mouse cursor 2125 2126 ``window-close=on|off`` : Allow to quit qemu with window close button 2127 2128 ``gtk`` 2129 Display video output in a GTK window. This interface provides 2130 drop-down menus and other UI elements to configure and control 2131 the VM during runtime. Valid parameters are: 2132 2133 ``full-screen=on|off`` : Start in fullscreen mode 2134 2135 ``gl=on|off`` : Use OpenGL for displaying 2136 2137 ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover 2138 2139 ``show-tabs=on|off`` : Display the tab bar for switching between the 2140 various graphical interfaces (e.g. VGA and 2141 virtual console character devices) by default. 2142 2143 ``show-cursor=on|off`` : Force showing the mouse cursor 2144 2145 ``window-close=on|off`` : Allow to quit qemu with window close button 2146 2147 ``show-menubar=on|off`` : Display the main window menubar, defaults to "on" 2148 2149 ``zoom-to-fit=on|off`` : Expand video output to the window size, 2150 defaults to "off" 2151 2152 ``curses[,charset=<encoding>]`` 2153 Display video output via curses. For graphics device models 2154 which support a text mode, QEMU can display this output using a 2155 curses/ncurses interface. Nothing is displayed when the graphics 2156 device is in graphical mode or if the graphics device does not 2157 support a text mode. Generally only the VGA device models 2158 support text mode. The font charset used by the guest can be 2159 specified with the ``charset`` option, for example 2160 ``charset=CP850`` for IBM CP850 encoding. The default is 2161 ``CP437``. 2162 2163 ``cocoa`` 2164 Display video output in a Cocoa window. Mac only. This interface 2165 provides drop-down menus and other UI elements to configure and 2166 control the VM during runtime. Valid parameters are: 2167 2168 ``full-grab=on|off`` : Capture all key presses, including system combos. 2169 This requires accessibility permissions, since it 2170 performs a global grab on key events. 2171 (default: off) See 2172 https://support.apple.com/en-in/guide/mac-help/mh32356/mac 2173 2174 ``swap-opt-cmd=on|off`` : Swap the Option and Command keys so that their 2175 key codes match their position on non-Mac 2176 keyboards and you can use Meta/Super and Alt 2177 where you expect them. (default: off) 2178 2179 ``show-cursor=on|off`` : Force showing the mouse cursor 2180 2181 ``left-command-key=on|off`` : Disable forwarding left command key to host 2182 2183 ``full-screen=on|off`` : Start in fullscreen mode 2184 2185 ``zoom-to-fit=on|off`` : Expand video output to the window size, 2186 defaults to "off" 2187 2188 ``egl-headless[,rendernode=<file>]`` 2189 Offload all OpenGL operations to a local DRI device. For any 2190 graphical display, this display needs to be paired with either 2191 VNC or SPICE displays. 2192 2193 ``vnc=<display>`` 2194 Start a VNC server on display <display> 2195 2196 ``none`` 2197 Do not display video output. The guest will still see an 2198 emulated graphics card, but its output will not be displayed to 2199 the QEMU user. This option differs from the -nographic option in 2200 that it only affects what is done with video output; -nographic 2201 also changes the destination of the serial and parallel port 2202 data. 2203ERST 2204 2205DEF("nographic", 0, QEMU_OPTION_nographic, 2206 "-nographic disable graphical output and redirect serial I/Os to console\n", 2207 QEMU_ARCH_ALL) 2208SRST 2209``-nographic`` 2210 Normally, if QEMU is compiled with graphical window support, it 2211 displays output such as guest graphics, guest console, and the QEMU 2212 monitor in a window. With this option, you can totally disable 2213 graphical output so that QEMU is a simple command line application. 2214 The emulated serial port is redirected on the console and muxed with 2215 the monitor (unless redirected elsewhere explicitly). Therefore, you 2216 can still use QEMU to debug a Linux kernel with a serial console. 2217 Use C-a h for help on switching between the console and monitor. 2218ERST 2219 2220#ifdef CONFIG_SPICE 2221DEF("spice", HAS_ARG, QEMU_OPTION_spice, 2222 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n" 2223 " [,x509-key-file=<file>][,x509-key-password=<file>]\n" 2224 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n" 2225 " [,x509-dh-key-file=<file>][,addr=addr]\n" 2226 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n" 2227 " [,tls-ciphers=<list>]\n" 2228 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n" 2229 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n" 2230 " [,sasl=on|off][,disable-ticketing=on|off]\n" 2231 " [,password-secret=<secret-id>]\n" 2232 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n" 2233 " [,jpeg-wan-compression=[auto|never|always]]\n" 2234 " [,zlib-glz-wan-compression=[auto|never|always]]\n" 2235 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n" 2236 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n" 2237 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n" 2238 " [,gl=[on|off]][,rendernode=<file>]\n" 2239 " enable spice\n" 2240 " at least one of {port, tls-port} is mandatory\n", 2241 QEMU_ARCH_ALL) 2242#endif 2243SRST 2244``-spice option[,option[,...]]`` 2245 Enable the spice remote desktop protocol. Valid options are 2246 2247 ``port=<nr>`` 2248 Set the TCP port spice is listening on for plaintext channels. 2249 2250 ``addr=<addr>`` 2251 Set the IP address spice is listening on. Default is any 2252 address. 2253 2254 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off`` 2255 Force using the specified IP version. 2256 2257 ``password-secret=<secret-id>`` 2258 Set the ID of the ``secret`` object containing the password 2259 you need to authenticate. 2260 2261 ``sasl=on|off`` 2262 Require that the client use SASL to authenticate with the spice. 2263 The exact choice of authentication method used is controlled 2264 from the system / user's SASL configuration file for the 'qemu' 2265 service. This is typically found in /etc/sasl2/qemu.conf. If 2266 running QEMU as an unprivileged user, an environment variable 2267 SASL\_CONF\_PATH can be used to make it search alternate 2268 locations for the service config. While some SASL auth methods 2269 can also provide data encryption (eg GSSAPI), it is recommended 2270 that SASL always be combined with the 'tls' and 'x509' settings 2271 to enable use of SSL and server certificates. This ensures a 2272 data encryption preventing compromise of authentication 2273 credentials. 2274 2275 ``disable-ticketing=on|off`` 2276 Allow client connects without authentication. 2277 2278 ``disable-copy-paste=on|off`` 2279 Disable copy paste between the client and the guest. 2280 2281 ``disable-agent-file-xfer=on|off`` 2282 Disable spice-vdagent based file-xfer between the client and the 2283 guest. 2284 2285 ``tls-port=<nr>`` 2286 Set the TCP port spice is listening on for encrypted channels. 2287 2288 ``x509-dir=<dir>`` 2289 Set the x509 file directory. Expects same filenames as -vnc 2290 $display,x509=$dir 2291 2292 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>`` 2293 The x509 file names can also be configured individually. 2294 2295 ``tls-ciphers=<list>`` 2296 Specify which ciphers to use. 2297 2298 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]`` 2299 Force specific channel to be used with or without TLS 2300 encryption. The options can be specified multiple times to 2301 configure multiple channels. The special name "default" can be 2302 used to set the default mode. For channels which are not 2303 explicitly forced into one mode the spice client is allowed to 2304 pick tls/plaintext as he pleases. 2305 2306 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]`` 2307 Configure image compression (lossless). Default is auto\_glz. 2308 2309 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]`` 2310 Configure wan image compression (lossy for slow links). Default 2311 is auto. 2312 2313 ``streaming-video=[off|all|filter]`` 2314 Configure video stream detection. Default is off. 2315 2316 ``agent-mouse=[on|off]`` 2317 Enable/disable passing mouse events via vdagent. Default is on. 2318 2319 ``playback-compression=[on|off]`` 2320 Enable/disable audio stream compression (using celt 0.5.1). 2321 Default is on. 2322 2323 ``seamless-migration=[on|off]`` 2324 Enable/disable spice seamless migration. Default is off. 2325 2326 ``gl=[on|off]`` 2327 Enable/disable OpenGL context. Default is off. 2328 2329 ``rendernode=<file>`` 2330 DRM render node for OpenGL rendering. If not specified, it will 2331 pick the first available. (Since 2.9) 2332ERST 2333 2334DEF("portrait", 0, QEMU_OPTION_portrait, 2335 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n", 2336 QEMU_ARCH_ALL) 2337SRST 2338``-portrait`` 2339 Rotate graphical output 90 deg left (only PXA LCD). 2340ERST 2341 2342DEF("rotate", HAS_ARG, QEMU_OPTION_rotate, 2343 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n", 2344 QEMU_ARCH_ALL) 2345SRST 2346``-rotate deg`` 2347 Rotate graphical output some deg left (only PXA LCD). 2348ERST 2349 2350DEF("vga", HAS_ARG, QEMU_OPTION_vga, 2351 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n" 2352 " select video card type\n", QEMU_ARCH_ALL) 2353SRST 2354``-vga type`` 2355 Select type of VGA card to emulate. Valid values for type are 2356 2357 ``cirrus`` 2358 Cirrus Logic GD5446 Video card. All Windows versions starting 2359 from Windows 95 should recognize and use this graphic card. For 2360 optimal performances, use 16 bit color depth in the guest and 2361 the host OS. (This card was the default before QEMU 2.2) 2362 2363 ``std`` 2364 Standard VGA card with Bochs VBE extensions. If your guest OS 2365 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if 2366 you want to use high resolution modes (>= 1280x1024x16) then you 2367 should use this option. (This card is the default since QEMU 2368 2.2) 2369 2370 ``vmware`` 2371 VMWare SVGA-II compatible adapter. Use it if you have 2372 sufficiently recent XFree86/XOrg server or Windows guest with a 2373 driver for this card. 2374 2375 ``qxl`` 2376 QXL paravirtual graphic card. It is VGA compatible (including 2377 VESA 2.0 VBE support). Works best with qxl guest drivers 2378 installed though. Recommended choice when using the spice 2379 protocol. 2380 2381 ``tcx`` 2382 (sun4m only) Sun TCX framebuffer. This is the default 2383 framebuffer for sun4m machines and offers both 8-bit and 24-bit 2384 colour depths at a fixed resolution of 1024x768. 2385 2386 ``cg3`` 2387 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit 2388 framebuffer for sun4m machines available in both 1024x768 2389 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people 2390 wishing to run older Solaris versions. 2391 2392 ``virtio`` 2393 Virtio VGA card. 2394 2395 ``none`` 2396 Disable VGA card. 2397ERST 2398 2399DEF("full-screen", 0, QEMU_OPTION_full_screen, 2400 "-full-screen start in full screen\n", QEMU_ARCH_ALL) 2401SRST 2402``-full-screen`` 2403 Start in full screen. 2404ERST 2405 2406DEF("g", HAS_ARG, QEMU_OPTION_g , 2407 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n", 2408 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K) 2409SRST 2410``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]`` 2411 Set the initial graphical resolution and depth (PPC, SPARC only). 2412 2413 For PPC the default is 800x600x32. 2414 2415 For SPARC with the TCX graphics device, the default is 1024x768x8 2416 with the option of 1024x768x24. For cgthree, the default is 2417 1024x768x8 with the option of 1152x900x8 for people who wish to use 2418 OBP. 2419ERST 2420 2421#ifdef CONFIG_VNC 2422DEF("vnc", HAS_ARG, QEMU_OPTION_vnc , 2423 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL) 2424#endif 2425SRST 2426``-vnc display[,option[,option[,...]]]`` 2427 Normally, if QEMU is compiled with graphical window support, it 2428 displays output such as guest graphics, guest console, and the QEMU 2429 monitor in a window. With this option, you can have QEMU listen on 2430 VNC display display and redirect the VGA display over the VNC 2431 session. It is very useful to enable the usb tablet device when 2432 using this option (option ``-device usb-tablet``). When using the 2433 VNC display, you must use the ``-k`` parameter to set the keyboard 2434 layout if you are not using en-us. Valid syntax for the display is 2435 2436 ``to=L`` 2437 With this option, QEMU will try next available VNC displays, 2438 until the number L, if the originally defined "-vnc display" is 2439 not available, e.g. port 5900+display is already used by another 2440 application. By default, to=0. 2441 2442 ``host:d`` 2443 TCP connections will only be allowed from host on display d. By 2444 convention the TCP port is 5900+d. Optionally, host can be 2445 omitted in which case the server will accept connections from 2446 any host. 2447 2448 ``unix:path`` 2449 Connections will be allowed over UNIX domain sockets where path 2450 is the location of a unix socket to listen for connections on. 2451 2452 ``none`` 2453 VNC is initialized but not started. The monitor ``change`` 2454 command can be used to later start the VNC server. 2455 2456 Following the display value there may be one or more option flags 2457 separated by commas. Valid options are 2458 2459 ``reverse=on|off`` 2460 Connect to a listening VNC client via a "reverse" connection. 2461 The client is specified by the display. For reverse network 2462 connections (host:d,``reverse``), the d argument is a TCP port 2463 number, not a display number. 2464 2465 ``websocket=on|off`` 2466 Opens an additional TCP listening port dedicated to VNC 2467 Websocket connections. If a bare websocket option is given, the 2468 Websocket port is 5700+display. An alternative port can be 2469 specified with the syntax ``websocket``\ =port. 2470 2471 If host is specified connections will only be allowed from this 2472 host. It is possible to control the websocket listen address 2473 independently, using the syntax ``websocket``\ =host:port. 2474 2475 Websocket could be allowed over UNIX domain socket, using the syntax 2476 ``websocket``\ =unix:path, where path is the location of a unix socket 2477 to listen for connections on. 2478 2479 If no TLS credentials are provided, the websocket connection 2480 runs in unencrypted mode. If TLS credentials are provided, the 2481 websocket connection requires encrypted client connections. 2482 2483 ``password=on|off`` 2484 Require that password based authentication is used for client 2485 connections. 2486 2487 The password must be set separately using the ``set_password`` 2488 command in the :ref:`QEMU monitor`. The 2489 syntax to change your password is: 2490 ``set_password <protocol> <password>`` where <protocol> could be 2491 either "vnc" or "spice". 2492 2493 If you would like to change <protocol> password expiration, you 2494 should use ``expire_password <protocol> <expiration-time>`` 2495 where expiration time could be one of the following options: 2496 now, never, +seconds or UNIX time of expiration, e.g. +60 to 2497 make password expire in 60 seconds, or 1335196800 to make 2498 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for 2499 this date and time). 2500 2501 You can also use keywords "now" or "never" for the expiration 2502 time to allow <protocol> password to expire immediately or never 2503 expire. 2504 2505 ``password-secret=<secret-id>`` 2506 Require that password based authentication is used for client 2507 connections, using the password provided by the ``secret`` 2508 object identified by ``secret-id``. 2509 2510 ``tls-creds=ID`` 2511 Provides the ID of a set of TLS credentials to use to secure the 2512 VNC server. They will apply to both the normal VNC server socket 2513 and the websocket socket (if enabled). Setting TLS credentials 2514 will cause the VNC server socket to enable the VeNCrypt auth 2515 mechanism. The credentials should have been previously created 2516 using the ``-object tls-creds`` argument. 2517 2518 ``tls-authz=ID`` 2519 Provides the ID of the QAuthZ authorization object against which 2520 the client's x509 distinguished name will validated. This object 2521 is only resolved at time of use, so can be deleted and recreated 2522 on the fly while the VNC server is active. If missing, it will 2523 default to denying access. 2524 2525 ``sasl=on|off`` 2526 Require that the client use SASL to authenticate with the VNC 2527 server. The exact choice of authentication method used is 2528 controlled from the system / user's SASL configuration file for 2529 the 'qemu' service. This is typically found in 2530 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user, 2531 an environment variable SASL\_CONF\_PATH can be used to make it 2532 search alternate locations for the service config. While some 2533 SASL auth methods can also provide data encryption (eg GSSAPI), 2534 it is recommended that SASL always be combined with the 'tls' 2535 and 'x509' settings to enable use of SSL and server 2536 certificates. This ensures a data encryption preventing 2537 compromise of authentication credentials. See the 2538 :ref:`VNC security` section in the System Emulation Users Guide 2539 for details on using SASL authentication. 2540 2541 ``sasl-authz=ID`` 2542 Provides the ID of the QAuthZ authorization object against which 2543 the client's SASL username will validated. This object is only 2544 resolved at time of use, so can be deleted and recreated on the 2545 fly while the VNC server is active. If missing, it will default 2546 to denying access. 2547 2548 ``acl=on|off`` 2549 Legacy method for enabling authorization of clients against the 2550 x509 distinguished name and SASL username. It results in the 2551 creation of two ``authz-list`` objects with IDs of 2552 ``vnc.username`` and ``vnc.x509dname``. The rules for these 2553 objects must be configured with the HMP ACL commands. 2554 2555 This option is deprecated and should no longer be used. The new 2556 ``sasl-authz`` and ``tls-authz`` options are a replacement. 2557 2558 ``lossy=on|off`` 2559 Enable lossy compression methods (gradient, JPEG, ...). If this 2560 option is set, VNC client may receive lossy framebuffer updates 2561 depending on its encoding settings. Enabling this option can 2562 save a lot of bandwidth at the expense of quality. 2563 2564 ``non-adaptive=on|off`` 2565 Disable adaptive encodings. Adaptive encodings are enabled by 2566 default. An adaptive encoding will try to detect frequently 2567 updated screen regions, and send updates in these regions using 2568 a lossy encoding (like JPEG). This can be really helpful to save 2569 bandwidth when playing videos. Disabling adaptive encodings 2570 restores the original static behavior of encodings like Tight. 2571 2572 ``share=[allow-exclusive|force-shared|ignore]`` 2573 Set display sharing policy. 'allow-exclusive' allows clients to 2574 ask for exclusive access. As suggested by the rfb spec this is 2575 implemented by dropping other connections. Connecting multiple 2576 clients in parallel requires all clients asking for a shared 2577 session (vncviewer: -shared switch). This is the default. 2578 'force-shared' disables exclusive client access. Useful for 2579 shared desktop sessions, where you don't want someone forgetting 2580 specify -shared disconnect everybody else. 'ignore' completely 2581 ignores the shared flag and allows everybody connect 2582 unconditionally. Doesn't conform to the rfb spec but is 2583 traditional QEMU behavior. 2584 2585 ``key-delay-ms`` 2586 Set keyboard delay, for key down and key up events, in 2587 milliseconds. Default is 10. Keyboards are low-bandwidth 2588 devices, so this slowdown can help the device and guest to keep 2589 up and not lose events in case events are arriving in bulk. 2590 Possible causes for the latter are flaky network connections, or 2591 scripts for automated testing. 2592 2593 ``audiodev=audiodev`` 2594 Use the specified audiodev when the VNC client requests audio 2595 transmission. When not using an -audiodev argument, this option 2596 must be omitted, otherwise is must be present and specify a 2597 valid audiodev. 2598 2599 ``power-control=on|off`` 2600 Permit the remote client to issue shutdown, reboot or reset power 2601 control requests. 2602ERST 2603 2604ARCHHEADING(, QEMU_ARCH_I386) 2605 2606ARCHHEADING(i386 target only:, QEMU_ARCH_I386) 2607 2608DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack, 2609 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n", 2610 QEMU_ARCH_I386) 2611SRST 2612``-win2k-hack`` 2613 Use it when installing Windows 2000 to avoid a disk full bug. After 2614 Windows 2000 is installed, you no longer need this option (this 2615 option slows down the IDE transfers). Synonym of ``-global 2616 ide-device.win2k-install-hack=on``. 2617ERST 2618 2619DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk, 2620 "-no-fd-bootchk disable boot signature checking for floppy disks\n", 2621 QEMU_ARCH_I386) 2622SRST 2623``-no-fd-bootchk`` 2624 Disable boot signature checking for floppy disks in BIOS. May be 2625 needed to boot from old floppy disks. Synonym of ``-m fd-bootchk=off``. 2626ERST 2627 2628DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable, 2629 "-acpitable [sig=str][,rev=n][,oem_id=str][,oem_table_id=str][,oem_rev=n][,asl_compiler_id=str][,asl_compiler_rev=n][,{data|file}=file1[:file2]...]\n" 2630 " ACPI table description\n", QEMU_ARCH_I386) 2631SRST 2632``-acpitable [sig=str][,rev=n][,oem_id=str][,oem_table_id=str][,oem_rev=n] [,asl_compiler_id=str][,asl_compiler_rev=n][,data=file1[:file2]...]`` 2633 Add ACPI table with specified header fields and context from 2634 specified files. For file=, take whole ACPI table from the specified 2635 files, including all ACPI headers (possible overridden by other 2636 options). For data=, only data portion of the table is used, all 2637 header information is specified in the command line. If a SLIC table 2638 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id 2639 fields will override the same in the RSDT and the FADT (a.k.a. 2640 FACP), in order to ensure the field matches required by the 2641 Microsoft SLIC spec and the ACPI spec. 2642ERST 2643 2644DEF("smbios", HAS_ARG, QEMU_OPTION_smbios, 2645 "-smbios file=binary\n" 2646 " load SMBIOS entry from binary file\n" 2647 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n" 2648 " [,uefi=on|off]\n" 2649 " specify SMBIOS type 0 fields\n" 2650 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2651 " [,uuid=uuid][,sku=str][,family=str]\n" 2652 " specify SMBIOS type 1 fields\n" 2653 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2654 " [,asset=str][,location=str]\n" 2655 " specify SMBIOS type 2 fields\n" 2656 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n" 2657 " [,sku=str]\n" 2658 " specify SMBIOS type 3 fields\n" 2659 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n" 2660 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n" 2661 " [,processor-family=%d][,processor-id=%d]\n" 2662 " specify SMBIOS type 4 fields\n" 2663 "-smbios type=8[,external_reference=str][,internal_reference=str][,connector_type=%d][,port_type=%d]\n" 2664 " specify SMBIOS type 8 fields\n" 2665 "-smbios type=11[,value=str][,path=filename]\n" 2666 " specify SMBIOS type 11 fields\n" 2667 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n" 2668 " [,asset=str][,part=str][,speed=%d]\n" 2669 " specify SMBIOS type 17 fields\n" 2670 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n" 2671 " specify SMBIOS type 41 fields\n", 2672 QEMU_ARCH_I386 | QEMU_ARCH_ARM | QEMU_ARCH_LOONGARCH | QEMU_ARCH_RISCV) 2673SRST 2674``-smbios file=binary`` 2675 Load SMBIOS entry from binary file. 2676 2677``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]`` 2678 Specify SMBIOS type 0 fields 2679 2680``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]`` 2681 Specify SMBIOS type 1 fields 2682 2683``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]`` 2684 Specify SMBIOS type 2 fields 2685 2686``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]`` 2687 Specify SMBIOS type 3 fields 2688 2689``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-family=%d][,processor-id=%d]`` 2690 Specify SMBIOS type 4 fields 2691 2692``-smbios type=9[,slot_designation=str][,slot_type=%d][,slot_data_bus_width=%d][,current_usage=%d][,slot_length=%d][,slot_id=%d][,slot_characteristics1=%d][,slot_characteristics12=%d][,pci_device=str]`` 2693 Specify SMBIOS type 9 fields 2694 2695``-smbios type=11[,value=str][,path=filename]`` 2696 Specify SMBIOS type 11 fields 2697 2698 This argument can be repeated multiple times, and values are added in the order they are parsed. 2699 Applications intending to use OEM strings data are encouraged to use their application name as 2700 a prefix for the value string. This facilitates passing information for multiple applications 2701 concurrently. 2702 2703 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax 2704 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes. 2705 2706 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to 2707 the SMBIOS table in the order in which they appear. 2708 2709 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535 2710 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the 2711 guest. Instead it should be used as a indicator to inform the guest where to locate the real 2712 data set, for example, by specifying the serial ID of a block device. 2713 2714 An example passing three strings is 2715 2716 .. parsed-literal:: 2717 2718 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\ 2719 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\ 2720 path=/some/file/with/oemstringsdata.txt 2721 2722 In the guest OS this is visible with the ``dmidecode`` command 2723 2724 .. parsed-literal:: 2725 2726 $ dmidecode -t 11 2727 Handle 0x0E00, DMI type 11, 5 bytes 2728 OEM Strings 2729 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/ 2730 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os 2731 String 3: myapp:some extra data 2732 2733 2734``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]`` 2735 Specify SMBIOS type 17 fields 2736 2737``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]`` 2738 Specify SMBIOS type 41 fields 2739 2740 This argument can be repeated multiple times. Its main use is to allow network interfaces be created 2741 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface 2742 position on the PCI bus. 2743 2744 Here is an example of use: 2745 2746 .. parsed-literal:: 2747 2748 -netdev user,id=internet \\ 2749 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\ 2750 -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev 2751 2752 In the guest OS, the device should then appear as ``eno1``: 2753 2754 ..parsed-literal:: 2755 2756 $ ip -brief l 2757 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP> 2758 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP> 2759 2760 Currently, the PCI device has to be attached to the root bus. 2761 2762ERST 2763 2764DEFHEADING() 2765 2766DEFHEADING(Network options:) 2767 2768DEF("netdev", HAS_ARG, QEMU_OPTION_netdev, 2769#ifdef CONFIG_SLIRP 2770 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n" 2771 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n" 2772 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n" 2773 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n" 2774 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]" 2775#ifndef _WIN32 2776 "[,smb=dir[,smbserver=addr]]\n" 2777#endif 2778 " configure a user mode network backend with ID 'str',\n" 2779 " its DHCP server and optional services\n" 2780#endif 2781#ifdef _WIN32 2782 "-netdev tap,id=str,ifname=name\n" 2783 " configure a host TAP network backend with ID 'str'\n" 2784#else 2785 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n" 2786 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n" 2787 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n" 2788 " [,poll-us=n]\n" 2789 " configure a host TAP network backend with ID 'str'\n" 2790 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2791 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n" 2792 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n" 2793 " to deconfigure it\n" 2794 " use '[down]script=no' to disable script execution\n" 2795 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n" 2796 " configure it\n" 2797 " use 'fd=h' to connect to an already opened TAP interface\n" 2798 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n" 2799 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n" 2800 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n" 2801 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n" 2802 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n" 2803 " use vhost=on to enable experimental in kernel accelerator\n" 2804 " (only has effect for virtio guests which use MSIX)\n" 2805 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n" 2806 " use 'vhostfd=h' to connect to an already opened vhost net device\n" 2807 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n" 2808 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n" 2809 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n" 2810 " spent on busy polling for vhost net\n" 2811 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n" 2812 " configure a host TAP network backend with ID 'str' that is\n" 2813 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2814 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n" 2815#endif 2816#ifdef __linux__ 2817 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n" 2818 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n" 2819 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n" 2820 " [,rxcookie=rxcookie][,offset=offset]\n" 2821 " configure a network backend with ID 'str' connected to\n" 2822 " an Ethernet over L2TPv3 pseudowire.\n" 2823 " Linux kernel 3.3+ as well as most routers can talk\n" 2824 " L2TPv3. This transport allows connecting a VM to a VM,\n" 2825 " VM to a router and even VM to Host. It is a nearly-universal\n" 2826 " standard (RFC3931). Note - this implementation uses static\n" 2827 " pre-configured tunnels (same as the Linux kernel).\n" 2828 " use 'src=' to specify source address\n" 2829 " use 'dst=' to specify destination address\n" 2830 " use 'udp=on' to specify udp encapsulation\n" 2831 " use 'srcport=' to specify source udp port\n" 2832 " use 'dstport=' to specify destination udp port\n" 2833 " use 'ipv6=on' to force v6\n" 2834 " L2TPv3 uses cookies to prevent misconfiguration as\n" 2835 " well as a weak security measure\n" 2836 " use 'rxcookie=0x012345678' to specify a rxcookie\n" 2837 " use 'txcookie=0x012345678' to specify a txcookie\n" 2838 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n" 2839 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n" 2840 " use 'pincounter=on' to work around broken counter handling in peer\n" 2841 " use 'offset=X' to add an extra offset between header and data\n" 2842#endif 2843 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n" 2844 " configure a network backend to connect to another network\n" 2845 " using a socket connection\n" 2846 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n" 2847 " configure a network backend to connect to a multicast maddr and port\n" 2848 " use 'localaddr=addr' to specify the host address to send packets from\n" 2849 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n" 2850 " configure a network backend to connect to another network\n" 2851 " using an UDP tunnel\n" 2852 "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port[,to=maxport][,numeric=on|off][,keep-alive=on|off][,mptcp=on|off][,addr.ipv4=on|off][,addr.ipv6=on|off][,reconnect=seconds]\n" 2853 "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off][,reconnect=seconds]\n" 2854 "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor[,reconnect=seconds]\n" 2855 " configure a network backend to connect to another network\n" 2856 " using a socket connection in stream mode.\n" 2857 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n" 2858 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]\n" 2859 " configure a network backend to connect to a multicast maddr and port\n" 2860 " use ``local.host=addr`` to specify the host address to send packets from\n" 2861 "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n" 2862 "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n" 2863 "-netdev dgram,id=str,local.type=fd,local.str=file-descriptor\n" 2864 " configure a network backend to connect to another network\n" 2865 " using an UDP tunnel\n" 2866#ifdef CONFIG_VDE 2867 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n" 2868 " configure a network backend to connect to port 'n' of a vde switch\n" 2869 " running on host and listening for incoming connections on 'socketpath'.\n" 2870 " Use group 'groupname' and mode 'octalmode' to change default\n" 2871 " ownership and permissions for communication port.\n" 2872#endif 2873#ifdef CONFIG_NETMAP 2874 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n" 2875 " attach to the existing netmap-enabled network interface 'name', or to a\n" 2876 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n" 2877 " netmap device, defaults to '/dev/netmap')\n" 2878#endif 2879#ifdef CONFIG_AF_XDP 2880 "-netdev af-xdp,id=str,ifname=name[,mode=native|skb][,force-copy=on|off]\n" 2881 " [,queues=n][,start-queue=m][,inhibit=on|off][,sock-fds=x:y:...:z]\n" 2882 " attach to the existing network interface 'name' with AF_XDP socket\n" 2883 " use 'mode=MODE' to specify an XDP program attach mode\n" 2884 " use 'force-copy=on|off' to force XDP copy mode even if device supports zero-copy (default: off)\n" 2885 " use 'inhibit=on|off' to inhibit loading of a default XDP program (default: off)\n" 2886 " with inhibit=on,\n" 2887 " use 'sock-fds' to provide file descriptors for already open AF_XDP sockets\n" 2888 " added to a socket map in XDP program. One socket per queue.\n" 2889 " use 'queues=n' to specify how many queues of a multiqueue interface should be used\n" 2890 " use 'start-queue=m' to specify the first queue that should be used\n" 2891#endif 2892#ifdef CONFIG_POSIX 2893 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n" 2894 " configure a vhost-user network, backed by a chardev 'dev'\n" 2895#endif 2896#ifdef __linux__ 2897 "-netdev vhost-vdpa,id=str[,vhostdev=/path/to/dev][,vhostfd=h]\n" 2898 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n" 2899 " use 'vhostdev=/path/to/dev' to open a vhost vdpa device\n" 2900 " use 'vhostfd=h' to connect to an already opened vhost vdpa device\n" 2901#endif 2902#ifdef CONFIG_VMNET 2903 "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n" 2904 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" 2905 " configure a vmnet network backend in host mode with ID 'str',\n" 2906 " isolate this interface from others with 'isolated',\n" 2907 " configure the address range and choose a subnet mask,\n" 2908 " specify network UUID 'uuid' to disable DHCP and interact with\n" 2909 " vmnet-host interfaces within this isolated network\n" 2910 "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n" 2911 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" 2912 " configure a vmnet network backend in shared mode with ID 'str',\n" 2913 " configure the address range and choose a subnet mask,\n" 2914 " set IPv6 ULA prefix (of length 64) to use for internal network,\n" 2915 " isolate this interface from others with 'isolated'\n" 2916 "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n" 2917 " configure a vmnet network backend in bridged mode with ID 'str',\n" 2918 " use 'ifname=name' to select a physical network interface to be bridged,\n" 2919 " isolate this interface from others with 'isolated'\n" 2920#endif 2921 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n" 2922 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL) 2923DEF("nic", HAS_ARG, QEMU_OPTION_nic, 2924 "-nic [tap|bridge|" 2925#ifdef CONFIG_SLIRP 2926 "user|" 2927#endif 2928#ifdef __linux__ 2929 "l2tpv3|" 2930#endif 2931#ifdef CONFIG_VDE 2932 "vde|" 2933#endif 2934#ifdef CONFIG_NETMAP 2935 "netmap|" 2936#endif 2937#ifdef CONFIG_AF_XDP 2938 "af-xdp|" 2939#endif 2940#ifdef CONFIG_POSIX 2941 "vhost-user|" 2942#endif 2943#ifdef CONFIG_VMNET 2944 "vmnet-host|vmnet-shared|vmnet-bridged|" 2945#endif 2946 "socket][,option][,...][mac=macaddr]\n" 2947 " initialize an on-board / default host NIC (using MAC address\n" 2948 " macaddr) and connect it to the given host network backend\n" 2949 "-nic none use it alone to have zero network devices (the default is to\n" 2950 " provided a 'user' network connection)\n", 2951 QEMU_ARCH_ALL) 2952DEF("net", HAS_ARG, QEMU_OPTION_net, 2953 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n" 2954 " configure or create an on-board (or machine default) NIC and\n" 2955 " connect it to hub 0 (please use -nic unless you need a hub)\n" 2956 "-net [" 2957#ifdef CONFIG_SLIRP 2958 "user|" 2959#endif 2960 "tap|" 2961 "bridge|" 2962#ifdef CONFIG_VDE 2963 "vde|" 2964#endif 2965#ifdef CONFIG_NETMAP 2966 "netmap|" 2967#endif 2968#ifdef CONFIG_AF_XDP 2969 "af-xdp|" 2970#endif 2971#ifdef CONFIG_VMNET 2972 "vmnet-host|vmnet-shared|vmnet-bridged|" 2973#endif 2974 "socket][,option][,option][,...]\n" 2975 " old way to initialize a host network interface\n" 2976 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL) 2977SRST 2978``-nic [tap|bridge|user|l2tpv3|vde|netmap|af-xdp|vhost-user|socket][,...][,mac=macaddr][,model=mn]`` 2979 This option is a shortcut for configuring both the on-board 2980 (default) guest NIC hardware and the host network backend in one go. 2981 The host backend options are the same as with the corresponding 2982 ``-netdev`` options below. The guest NIC model can be set with 2983 ``model=modelname``. Use ``model=help`` to list the available device 2984 types. The hardware MAC address can be set with ``mac=macaddr``. 2985 2986 The following two example do exactly the same, to show how ``-nic`` 2987 can be used to shorten the command line length: 2988 2989 .. parsed-literal:: 2990 2991 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32 2992 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32 2993 2994``-nic none`` 2995 Indicate that no network devices should be configured. It is used to 2996 override the default configuration (default NIC with "user" host 2997 network backend) which is activated if no other networking options 2998 are provided. 2999 3000``-netdev user,id=id[,option][,option][,...]`` 3001 Configure user mode host network backend which requires no 3002 administrator privilege to run. Valid options are: 3003 3004 ``id=id`` 3005 Assign symbolic name for use in monitor commands. 3006 3007 ``ipv4=on|off and ipv6=on|off`` 3008 Specify that either IPv4 or IPv6 must be enabled. If neither is 3009 specified both protocols are enabled. 3010 3011 ``net=addr[/mask]`` 3012 Set IP network address the guest will see. Optionally specify 3013 the netmask, either in the form a.b.c.d or as number of valid 3014 top-most bits. Default is 10.0.2.0/24. 3015 3016 ``host=addr`` 3017 Specify the guest-visible address of the host. Default is the 3018 2nd IP in the guest network, i.e. x.x.x.2. 3019 3020 ``ipv6-net=addr[/int]`` 3021 Set IPv6 network address the guest will see (default is 3022 fec0::/64). The network prefix is given in the usual hexadecimal 3023 IPv6 address notation. The prefix size is optional, and is given 3024 as the number of valid top-most bits (default is 64). 3025 3026 ``ipv6-host=addr`` 3027 Specify the guest-visible IPv6 address of the host. Default is 3028 the 2nd IPv6 in the guest network, i.e. xxxx::2. 3029 3030 ``restrict=on|off`` 3031 If this option is enabled, the guest will be isolated, i.e. it 3032 will not be able to contact the host and no guest IP packets 3033 will be routed over the host to the outside. This option does 3034 not affect any explicitly set forwarding rules. 3035 3036 ``hostname=name`` 3037 Specifies the client hostname reported by the built-in DHCP 3038 server. 3039 3040 ``dhcpstart=addr`` 3041 Specify the first of the 16 IPs the built-in DHCP server can 3042 assign. Default is the 15th to 31st IP in the guest network, 3043 i.e. x.x.x.15 to x.x.x.31. 3044 3045 ``dns=addr`` 3046 Specify the guest-visible address of the virtual nameserver. The 3047 address must be different from the host address. Default is the 3048 3rd IP in the guest network, i.e. x.x.x.3. 3049 3050 ``ipv6-dns=addr`` 3051 Specify the guest-visible address of the IPv6 virtual 3052 nameserver. The address must be different from the host address. 3053 Default is the 3rd IP in the guest network, i.e. xxxx::3. 3054 3055 ``dnssearch=domain`` 3056 Provides an entry for the domain-search list sent by the 3057 built-in DHCP server. More than one domain suffix can be 3058 transmitted by specifying this option multiple times. If 3059 supported, this will cause the guest to automatically try to 3060 append the given domain suffix(es) in case a domain name can not 3061 be resolved. 3062 3063 Example: 3064 3065 .. parsed-literal:: 3066 3067 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org 3068 3069 ``domainname=domain`` 3070 Specifies the client domain name reported by the built-in DHCP 3071 server. 3072 3073 ``tftp=dir`` 3074 When using the user mode network stack, activate a built-in TFTP 3075 server. The files in dir will be exposed as the root of a TFTP 3076 server. The TFTP client on the guest must be configured in 3077 binary mode (use the command ``bin`` of the Unix TFTP client). 3078 The built-in TFTP server is read-only; it does not implement any 3079 command for writing files. QEMU will not write to this directory. 3080 3081 ``tftp-server-name=name`` 3082 In BOOTP reply, broadcast name as the "TFTP server name" 3083 (RFC2132 option 66). This can be used to advise the guest to 3084 load boot files or configurations from a different server than 3085 the host address. 3086 3087 ``bootfile=file`` 3088 When using the user mode network stack, broadcast file as the 3089 BOOTP filename. In conjunction with ``tftp``, this can be used 3090 to network boot a guest from a local directory. 3091 3092 Example (using pxelinux): 3093 3094 .. parsed-literal:: 3095 3096 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\ 3097 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0 3098 3099 ``smb=dir[,smbserver=addr]`` 3100 When using the user mode network stack, activate a built-in SMB 3101 server so that Windows OSes can access to the host files in 3102 ``dir`` transparently. The IP address of the SMB server can be 3103 set to addr. By default the 4th IP in the guest network is used, 3104 i.e. x.x.x.4. 3105 3106 In the guest Windows OS, the line: 3107 3108 :: 3109 3110 10.0.2.4 smbserver 3111 3112 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows 3113 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows 3114 NT/2000). 3115 3116 Then ``dir`` can be accessed in ``\\smbserver\qemu``. 3117 3118 Note that a SAMBA server must be installed on the host OS. 3119 3120 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport`` 3121 Redirect incoming TCP or UDP connections to the host port 3122 hostport to the guest IP address guestaddr on guest port 3123 guestport. If guestaddr is not specified, its value is x.x.x.15 3124 (default first address given by the built-in DHCP server). By 3125 specifying hostaddr, the rule can be bound to a specific host 3126 interface. If no connection type is set, TCP is used. This 3127 option can be given multiple times. 3128 3129 For example, to redirect host X11 connection from screen 1 to 3130 guest screen 0, use the following: 3131 3132 .. parsed-literal:: 3133 3134 # on the host 3135 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000 3136 # this host xterm should open in the guest X11 server 3137 xterm -display :1 3138 3139 To redirect telnet connections from host port 5555 to telnet 3140 port on the guest, use the following: 3141 3142 .. parsed-literal:: 3143 3144 # on the host 3145 |qemu_system| -nic user,hostfwd=tcp::5555-:23 3146 telnet localhost 5555 3147 3148 Then when you use on the host ``telnet localhost 5555``, you 3149 connect to the guest telnet server. 3150 3151 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command`` 3152 Forward guest TCP connections to the IP address server on port 3153 port to the character device dev or to a program executed by 3154 cmd:command which gets spawned for each connection. This option 3155 can be given multiple times. 3156 3157 You can either use a chardev directly and have that one used 3158 throughout QEMU's lifetime, like in the following example: 3159 3160 .. parsed-literal:: 3161 3162 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever 3163 # the guest accesses it 3164 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321 3165 3166 Or you can execute a command on every TCP connection established 3167 by the guest, so that QEMU behaves similar to an inetd process 3168 for that virtual server: 3169 3170 .. parsed-literal:: 3171 3172 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234 3173 # and connect the TCP stream to its stdin/stdout 3174 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321' 3175 3176``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]`` 3177 Configure a host TAP network backend with ID id. 3178 3179 Use the network script file to configure it and the network script 3180 dfile to deconfigure it. If name is not provided, the OS 3181 automatically provides one. The default network configure script is 3182 ``/etc/qemu-ifup`` and the default network deconfigure script is 3183 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to 3184 disable script execution. 3185 3186 If running QEMU as an unprivileged user, use the network helper 3187 to configure the TAP interface and attach it to the bridge. 3188 The default network helper executable is 3189 ``/path/to/qemu-bridge-helper`` and the default bridge device is 3190 ``br0``. 3191 3192 ``fd``\ =h can be used to specify the handle of an already opened 3193 host TAP interface. 3194 3195 Examples: 3196 3197 .. parsed-literal:: 3198 3199 #launch a QEMU instance with the default network script 3200 |qemu_system| linux.img -nic tap 3201 3202 .. parsed-literal:: 3203 3204 #launch a QEMU instance with two NICs, each one connected 3205 #to a TAP device 3206 |qemu_system| linux.img \\ 3207 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\ 3208 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1 3209 3210 .. parsed-literal:: 3211 3212 #launch a QEMU instance with the default network helper to 3213 #connect a TAP device to bridge br0 3214 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3215 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper" 3216 3217``-netdev bridge,id=id[,br=bridge][,helper=helper]`` 3218 Connect a host TAP network interface to a host bridge device. 3219 3220 Use the network helper helper to configure the TAP interface and 3221 attach it to the bridge. The default network helper executable is 3222 ``/path/to/qemu-bridge-helper`` and the default bridge device is 3223 ``br0``. 3224 3225 Examples: 3226 3227 .. parsed-literal:: 3228 3229 #launch a QEMU instance with the default network helper to 3230 #connect a TAP device to bridge br0 3231 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1 3232 3233 .. parsed-literal:: 3234 3235 #launch a QEMU instance with the default network helper to 3236 #connect a TAP device to bridge qemubr0 3237 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1 3238 3239``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]`` 3240 This host network backend can be used to connect the guest's network 3241 to another QEMU virtual machine using a TCP socket connection. If 3242 ``listen`` is specified, QEMU waits for incoming connections on port 3243 (host is optional). ``connect`` is used to connect to another QEMU 3244 instance using the ``listen`` option. ``fd``\ =h specifies an 3245 already opened TCP socket. 3246 3247 Example: 3248 3249 .. parsed-literal:: 3250 3251 # launch a first QEMU instance 3252 |qemu_system| linux.img \\ 3253 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3254 -netdev socket,id=n1,listen=:1234 3255 # connect the network of this instance to the network of the first instance 3256 |qemu_system| linux.img \\ 3257 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3258 -netdev socket,id=n2,connect=127.0.0.1:1234 3259 3260``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]`` 3261 Configure a socket host network backend to share the guest's network 3262 traffic with another QEMU virtual machines using a UDP multicast 3263 socket, effectively making a bus for every QEMU with same multicast 3264 address maddr and port. NOTES: 3265 3266 1. Several QEMU can be running on different hosts and share same bus 3267 (assuming correct multicast setup for these hosts). 3268 3269 2. mcast support is compatible with User Mode Linux (argument 3270 ``ethN=mcast``), see http://user-mode-linux.sf.net. 3271 3272 3. Use ``fd=h`` to specify an already opened UDP multicast socket. 3273 3274 Example: 3275 3276 .. parsed-literal:: 3277 3278 # launch one QEMU instance 3279 |qemu_system| linux.img \\ 3280 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3281 -netdev socket,id=n1,mcast=230.0.0.1:1234 3282 # launch another QEMU instance on same "bus" 3283 |qemu_system| linux.img \\ 3284 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3285 -netdev socket,id=n2,mcast=230.0.0.1:1234 3286 # launch yet another QEMU instance on same "bus" 3287 |qemu_system| linux.img \\ 3288 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\ 3289 -netdev socket,id=n3,mcast=230.0.0.1:1234 3290 3291 Example (User Mode Linux compat.): 3292 3293 .. parsed-literal:: 3294 3295 # launch QEMU instance (note mcast address selected is UML's default) 3296 |qemu_system| linux.img \\ 3297 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3298 -netdev socket,id=n1,mcast=239.192.168.1:1102 3299 # launch UML 3300 /path/to/linux ubd0=/path/to/root_fs eth0=mcast 3301 3302 Example (send packets from host's 1.2.3.4): 3303 3304 .. parsed-literal:: 3305 3306 |qemu_system| linux.img \\ 3307 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3308 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4 3309 3310``-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port[,to=maxport][,numeric=on|off][,keep-alive=on|off][,mptcp=on|off][,addr.ipv4=on|off][,addr.ipv6=on|off][,reconnect=seconds]`` 3311 Configure a network backend to connect to another QEMU virtual machine or a proxy using a TCP/IP socket. 3312 3313 ``server=on|off`` 3314 if ``on`` create a server socket 3315 3316 ``addr.host=host,addr.port=port`` 3317 socket address to listen on (server=on) or connect to (server=off) 3318 3319 ``to=maxport`` 3320 if present, this is range of possible addresses, with port between ``port`` and ``maxport``. 3321 3322 ``numeric=on|off`` 3323 if ``on`` ``host`` and ``port`` are guaranteed to be numeric, otherwise a name resolution should be attempted (default: ``off``) 3324 3325 ``keep-alive=on|off`` 3326 enable keep-alive when connecting to this socket. Not supported for passive sockets. 3327 3328 ``mptcp=on|off`` 3329 enable multipath TCP 3330 3331 ``ipv4=on|off`` 3332 whether to accept IPv4 addresses, default to try both IPv4 and IPv6 3333 3334 ``ipv6=on|off`` 3335 whether to accept IPv6 addresses, default to try both IPv4 and IPv6 3336 3337 ``reconnect=seconds`` 3338 for a client socket, if a socket is disconnected, then attempt a reconnect after the given number of seconds. 3339 Setting this to zero disables this function. (default: 0) 3340 3341 Example (two guests connected using a TCP/IP socket): 3342 3343 .. parsed-literal:: 3344 3345 # first VM 3346 |qemu_system| linux.img \\ 3347 -device virtio-net,netdev=net0,mac=52:54:00:12:34:56 \\ 3348 -netdev stream,id=net0,server=on,addr.type=inet,addr.host=localhost,addr.port=1234 3349 # second VM 3350 |qemu_system| linux.img \\ 3351 -device virtio-net,netdev=net0,mac=52:54:00:12:34:57 \\ 3352 -netdev stream,id=net0,server=off,addr.type=inet,addr.host=localhost,addr.port=1234,reconnect=5 3353 3354``-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off][,reconnect=seconds]`` 3355 Configure a network backend to connect to another QEMU virtual machine or a proxy using a stream oriented unix domain socket. 3356 3357 ``server=on|off`` 3358 if ``on`` create a server socket 3359 3360 ``addr.path=path`` 3361 filesystem path to use 3362 3363 ``abstract=on|off`` 3364 if ``on``, this is a Linux abstract socket address. 3365 3366 ``tight=on|off`` 3367 if false, pad an abstract socket address with enough null bytes to make it fill struct sockaddr_un member sun_path. 3368 3369 ``reconnect=seconds`` 3370 for a client socket, if a socket is disconnected, then attempt a reconnect after the given number of seconds. 3371 Setting this to zero disables this function. (default: 0) 3372 3373 Example (using passt as a replacement of -netdev user): 3374 3375 .. parsed-literal:: 3376 3377 # start passt server as a non privileged user 3378 passt 3379 UNIX domain socket bound at /tmp/passt_1.socket 3380 # start QEMU to connect to passt 3381 |qemu_system| linux.img \\ 3382 -device virtio-net,netdev=net0 \\ 3383 -netdev stream,id=net0,server=off,addr.type=unix,addr.path=/tmp/passt_1.socket 3384 3385 Example (two guests connected using a stream oriented unix domain socket): 3386 3387 .. parsed-literal:: 3388 3389 # first VM 3390 |qemu_system| linux.img \\ 3391 -device virtio-net,netdev=net0,mac=52:54:00:12:34:56 \\ 3392 netdev stream,id=net0,server=on,addr.type=unix,addr.path=/tmp/qemu0 3393 # second VM 3394 |qemu_system| linux.img \\ 3395 -device virtio-net,netdev=net0,mac=52:54:00:12:34:57 \\ 3396 -netdev stream,id=net0,server=off,addr.type=unix,addr.path=/tmp/qemu0,reconnect=5 3397 3398``-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor[,reconnect=seconds]`` 3399 Configure a network backend to connect to another QEMU virtual machine or a proxy using a stream oriented socket file descriptor. 3400 3401 ``server=on|off`` 3402 if ``on`` create a server socket 3403 3404 ``addr.str=file-descriptor`` 3405 file descriptor number to use as a socket 3406 3407 ``reconnect=seconds`` 3408 for a client socket, if a socket is disconnected, then attempt a reconnect after the given number of seconds. 3409 Setting this to zero disables this function. (default: 0) 3410 3411``-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]`` 3412 Configure a network backend to connect to a multicast address. 3413 3414 ``remote.host=maddr,remote.port=port`` 3415 multicast address 3416 3417 ``local.host=addr`` 3418 specify the host address to send packets from 3419 3420 Example: 3421 3422 .. parsed-literal:: 3423 3424 # launch one QEMU instance 3425 |qemu_system| linux.img \\ 3426 -device virtio-net,netdev=net0,mac=52:54:00:12:34:56 \\ 3427 -netdev dgram,id=net0,remote.type=inet,remote.host=224.0.0.1,remote.port=1234 3428 # launch another QEMU instance on same "bus" 3429 |qemu_system| linux.img \\ 3430 -device virtio-net,netdev=net0,mac=52:54:00:12:34:57 \\ 3431 -netdev dgram,id=net0,remote.type=inet,remote.host=224.0.0.1,remote.port=1234 3432 # launch yet another QEMU instance on same "bus" 3433 |qemu_system| linux.img \\ 3434 -device virtio-net,netdev=net0,mac=52:54:00:12:34:58 \\ 3435 -netdev dgram,id=net0,remote.type=inet,remote.host=224.0.0.1,remote.port=1234 3436 3437``-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]`` 3438 Configure a network backend to connect to a multicast address using a UDP socket file descriptor. 3439 3440 ``remote.host=maddr,remote.port=port`` 3441 multicast address 3442 3443 ``local.str=file-descriptor`` 3444 File descriptor to use to send packets 3445 3446``-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]`` 3447 Configure a network backend to connect to another QEMU virtual 3448 machine or a proxy using a datagram oriented unix domain socket. 3449 3450 ``local.host=addr,local.port=port`` 3451 IP address to use to send the packets from 3452 3453 ``remote.host=addr,remote.port=port`` 3454 Destination IP address 3455 3456 Example (two guests connected using an UDP/IP socket): 3457 3458 .. parsed-literal:: 3459 3460 # first VM 3461 |qemu_system| linux.img \\ 3462 -device virtio-net,netdev=net0,mac=52:54:00:12:34:56 \\ 3463 -netdev dgram,id=net0,local.type=inet,local.host=localhost,local.port=1234,remote.type=inet,remote.host=localhost,remote.port=1235 3464 # second VM 3465 |qemu_system| linux.img \\ 3466 -device virtio-net,netdev=net0,mac=52:54:00:12:34:56 \\ 3467 -netdev dgram,id=net0,local.type=inet,local.host=localhost,local.port=1235,remote.type=inet,remote.host=localhost,remote.port=1234 3468 3469``-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]`` 3470 Configure a network backend to connect to another QEMU virtual 3471 machine or a proxy using a datagram oriented unix socket. 3472 3473 ``local.path=path`` 3474 filesystem path to use to bind the socket 3475 3476 ``remote.path=path`` 3477 filesystem path to use as a destination (see sendto(2)) 3478 3479 Example (two guests connected using an UDP/UNIX socket): 3480 3481 .. parsed-literal:: 3482 3483 # first VM 3484 |qemu_system| linux.img \\ 3485 -device virtio-net,netdev=net0,mac=52:54:00:12:34:56 \\ 3486 -netdev dgram,id=net0,local.type=unix,local.path=/tmp/qemu0,remote.type=unix,remote.path=/tmp/qemu1 3487 # second VM 3488 |qemu_system| linux.img \\ 3489 -device virtio-net,netdev=net0,mac=52:54:00:12:34:57 \\ 3490 -netdev dgram,id=net0,local.type=unix,local.path=/tmp/qemu1,remote.type=unix,remote.path=/tmp/qemu0 3491 3492``-netdev dgram,id=str,local.type=fd,local.str=file-descriptor`` 3493 Configure a network backend to connect to another QEMU virtual 3494 machine or a proxy using a datagram oriented socket file descriptor. 3495 3496 ``local.str=file-descriptor`` 3497 File descriptor to use to send packets 3498 3499``-netdev l2tpv3,id=id,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport],txsession=txsession[,rxsession=rxsession][,ipv6=on|off][,udp=on|off][,cookie64=on|off][,counter=on|off][,pincounter=on|off][,txcookie=txcookie][,rxcookie=rxcookie][,offset=offset]`` 3500 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931) 3501 is a popular protocol to transport Ethernet (and other Layer 2) data 3502 frames between two systems. It is present in routers, firewalls and 3503 the Linux kernel (from version 3.3 onwards). 3504 3505 This transport allows a VM to communicate to another VM, router or 3506 firewall directly. 3507 3508 ``src=srcaddr`` 3509 source address (mandatory) 3510 3511 ``dst=dstaddr`` 3512 destination address (mandatory) 3513 3514 ``udp=on`` 3515 select udp encapsulation (default is ip). 3516 3517 ``srcport=srcport`` 3518 source udp port. 3519 3520 ``dstport=dstport`` 3521 destination udp port. 3522 3523 ``ipv6=on`` 3524 force v6, otherwise defaults to v4. 3525 3526 ``rxcookie=rxcookie``; \ ``txcookie=txcookie`` 3527 Cookies are a weak form of security in the l2tpv3 specification. 3528 Their function is mostly to prevent misconfiguration. By default 3529 they are 32 bit. 3530 3531 ``cookie64=on`` 3532 Set cookie size to 64 bit instead of the default 32 3533 3534 ``counter=off`` 3535 Force a 'cut-down' L2TPv3 with no counter as in 3536 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00 3537 3538 ``pincounter=on`` 3539 Work around broken counter handling in peer. This may also help 3540 on networks which have packet reorder. 3541 3542 ``offset=offset`` 3543 Add an extra offset between header and data 3544 3545 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to 3546 the bridge br-lan on the remote Linux host 1.2.3.4: 3547 3548 .. parsed-literal:: 3549 3550 # Setup tunnel on linux host using raw ip as encapsulation 3551 # on 1.2.3.4 3552 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\ 3553 encap udp udp_sport 16384 udp_dport 16384 3554 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\ 3555 0xFFFFFFFF peer_session_id 0xFFFFFFFF 3556 ifconfig vmtunnel0 mtu 1500 3557 ifconfig vmtunnel0 up 3558 brctl addif br-lan vmtunnel0 3559 3560 3561 # on 4.3.2.1 3562 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter 3563 3564 |qemu_system| linux.img -device e1000,netdev=n1 \\ 3565 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp=on,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter=on 3566 3567``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]`` 3568 Configure VDE backend to connect to PORT n of a vde switch running 3569 on host and listening for incoming connections on socketpath. Use 3570 GROUP groupname and MODE octalmode to change default ownership and 3571 permissions for communication port. This option is only available if 3572 QEMU has been compiled with vde support enabled. 3573 3574 Example: 3575 3576 .. parsed-literal:: 3577 3578 # launch vde switch 3579 vde_switch -F -sock /tmp/myswitch 3580 # launch QEMU instance 3581 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch 3582 3583``-netdev af-xdp,id=str,ifname=name[,mode=native|skb][,force-copy=on|off][,queues=n][,start-queue=m][,inhibit=on|off][,sock-fds=x:y:...:z]`` 3584 Configure AF_XDP backend to connect to a network interface 'name' 3585 using AF_XDP socket. A specific program attach mode for a default 3586 XDP program can be forced with 'mode', defaults to best-effort, 3587 where the likely most performant mode will be in use. Number of queues 3588 'n' should generally match the number or queues in the interface, 3589 defaults to 1. Traffic arriving on non-configured device queues will 3590 not be delivered to the network backend. 3591 3592 .. parsed-literal:: 3593 3594 # set number of queues to 4 3595 ethtool -L eth0 combined 4 3596 # launch QEMU instance 3597 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3598 -netdev af-xdp,id=n1,ifname=eth0,queues=4 3599 3600 'start-queue' option can be specified if a particular range of queues 3601 [m, m + n] should be in use. For example, this is may be necessary in 3602 order to use certain NICs in native mode. Kernel allows the driver to 3603 create a separate set of XDP queues on top of regular ones, and only 3604 these queues can be used for AF_XDP sockets. NICs that work this way 3605 may also require an additional traffic redirection with ethtool to these 3606 special queues. 3607 3608 .. parsed-literal:: 3609 3610 # set number of queues to 1 3611 ethtool -L eth0 combined 1 3612 # redirect all the traffic to the second queue (id: 1) 3613 # note: drivers may require non-empty key/mask pair. 3614 ethtool -N eth0 flow-type ether \\ 3615 dst 00:00:00:00:00:00 m FF:FF:FF:FF:FF:FE action 1 3616 ethtool -N eth0 flow-type ether \\ 3617 dst 00:00:00:00:00:01 m FF:FF:FF:FF:FF:FE action 1 3618 # launch QEMU instance 3619 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3620 -netdev af-xdp,id=n1,ifname=eth0,queues=1,start-queue=1 3621 3622 XDP program can also be loaded externally. In this case 'inhibit' option 3623 should be set to 'on' and 'sock-fds' provided with file descriptors for 3624 already open but not bound XDP sockets already added to a socket map for 3625 corresponding queues. One socket per queue. 3626 3627 .. parsed-literal:: 3628 3629 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3630 -netdev af-xdp,id=n1,ifname=eth0,queues=3,inhibit=on,sock-fds=15:16:17 3631 3632``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]`` 3633 Establish a vhost-user netdev, backed by a chardev id. The chardev 3634 should be a unix domain socket backed one. The vhost-user uses a 3635 specifically defined protocol to pass vhost ioctl replacement 3636 messages to an application on the other end of the socket. On 3637 non-MSIX guests, the feature can be forced with vhostforce. Use 3638 'queues=n' to specify the number of queues to be created for 3639 multiqueue vhost-user. 3640 3641 Example: 3642 3643 :: 3644 3645 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \ 3646 -numa node,memdev=mem \ 3647 -chardev socket,id=chr0,path=/path/to/socket \ 3648 -netdev type=vhost-user,id=net0,chardev=chr0 \ 3649 -device virtio-net-pci,netdev=net0 3650 3651``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]`` 3652 Establish a vhost-vdpa netdev. 3653 3654 vDPA device is a device that uses a datapath which complies with 3655 the virtio specifications with a vendor specific control path. 3656 vDPA devices can be both physically located on the hardware or 3657 emulated by software. 3658 3659``-netdev hubport,id=id,hubid=hubid[,netdev=nd]`` 3660 Create a hub port on the emulated hub with ID hubid. 3661 3662 The hubport netdev lets you connect a NIC to a QEMU emulated hub 3663 instead of a single netdev. Alternatively, you can also connect the 3664 hubport to another netdev with ID nd by using the ``netdev=nd`` 3665 option. 3666 3667``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]`` 3668 Legacy option to configure or create an on-board (or machine 3669 default) Network Interface Card(NIC) and connect it either to the 3670 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd. 3671 If model is omitted, then the default NIC model associated with the 3672 machine type is used. Note that the default NIC model may change in 3673 future QEMU releases, so it is highly recommended to always specify 3674 a model. Optionally, the MAC address can be changed to mac, the 3675 device address set to addr (PCI cards only), and a name can be 3676 assigned for use in monitor commands. Optionally, for PCI cards, you 3677 can specify the number v of MSI-X vectors that the card should have; 3678 this option currently only affects virtio cards; set v = 0 to 3679 disable MSI-X. If no ``-net`` option is specified, a single NIC is 3680 created. QEMU can emulate several different models of network card. 3681 Use ``-net nic,model=help`` for a list of available devices for your 3682 target. 3683 3684``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]`` 3685 Configure a host network backend (with the options corresponding to 3686 the same ``-netdev`` option) and connect it to the emulated hub 0 3687 (the default hub). Use name to specify the name of the hub port. 3688ERST 3689 3690DEFHEADING() 3691 3692DEFHEADING(Character device options:) 3693 3694DEF("chardev", HAS_ARG, QEMU_OPTION_chardev, 3695 "-chardev help\n" 3696 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3697 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n" 3698 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n" 3699 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n" 3700 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n" 3701 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n" 3702 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n" 3703 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n" 3704 " [,logfile=PATH][,logappend=on|off]\n" 3705 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3706 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n" 3707 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3708 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n" 3709 "-chardev file,id=id,path=path[,input-path=input-file][,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3710 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3711#ifdef _WIN32 3712 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3713 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3714#else 3715 "-chardev pty,id=id[,path=path][,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3716 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n" 3717#endif 3718#ifdef CONFIG_BRLAPI 3719 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3720#endif 3721#if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \ 3722 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) 3723 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3724#endif 3725#if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__) 3726 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3727#endif 3728#if defined(CONFIG_SPICE) 3729 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3730 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3731#endif 3732 , QEMU_ARCH_ALL 3733) 3734 3735SRST 3736The general form of a character device option is: 3737 3738``-chardev backend,id=id[,mux=on|off][,options]`` 3739 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``, 3740 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``, 3741 ``pty``, ``stdio``, ``braille``, ``parallel``, 3742 ``spicevmc``, ``spiceport``. The specific backend will determine the 3743 applicable options. 3744 3745 Use ``-chardev help`` to print all available chardev backend types. 3746 3747 All devices must have an id, which can be any string up to 127 3748 characters long. It is used to uniquely identify this device in 3749 other command line directives. 3750 3751 A character device may be used in multiplexing mode by multiple 3752 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is 3753 a "1:N" device, and here the "1" end is your specified chardev 3754 backend, and the "N" end is the various parts of QEMU that can talk 3755 to a chardev. If you create a chardev with ``id=myid`` and 3756 ``mux=on``, QEMU will create a multiplexer with your specified ID, 3757 and you can then configure multiple front ends to use that chardev 3758 ID for their input/output. Up to four different front ends can be 3759 connected to a single multiplexed chardev. (Without multiplexing 3760 enabled, a chardev can only be used by a single front end.) For 3761 instance you could use this to allow a single stdio chardev to be 3762 used by two serial ports and the QEMU monitor: 3763 3764 :: 3765 3766 -chardev stdio,mux=on,id=char0 \ 3767 -mon chardev=char0,mode=readline \ 3768 -serial chardev:char0 \ 3769 -serial chardev:char0 3770 3771 You can have more than one multiplexer in a system configuration; 3772 for instance you could have a TCP port multiplexed between UART 0 3773 and UART 1, and stdio multiplexed between the QEMU monitor and a 3774 parallel port: 3775 3776 :: 3777 3778 -chardev stdio,mux=on,id=char0 \ 3779 -mon chardev=char0,mode=readline \ 3780 -parallel chardev:char0 \ 3781 -chardev tcp,...,mux=on,id=char1 \ 3782 -serial chardev:char1 \ 3783 -serial chardev:char1 3784 3785 When you're using a multiplexed character device, some escape 3786 sequences are interpreted in the input. See the chapter about 3787 :ref:`keys in the character backend multiplexer` in the 3788 System Emulation Users Guide for more details. 3789 3790 Note that some other command line options may implicitly create 3791 multiplexed character backends; for instance ``-serial mon:stdio`` 3792 creates a multiplexed stdio backend connected to the serial port and 3793 the QEMU monitor, and ``-nographic`` also multiplexes the console 3794 and the monitor to stdio. 3795 3796 There is currently no support for multiplexing in the other 3797 direction (where a single QEMU front end takes input and output from 3798 multiple chardevs). 3799 3800 Every backend supports the ``logfile`` option, which supplies the 3801 path to a file to record all data transmitted via the backend. The 3802 ``logappend`` option controls whether the log file will be truncated 3803 or appended to when opened. 3804 3805The available backends are: 3806 3807``-chardev null,id=id`` 3808 A void device. This device will not emit any data, and will drop any 3809 data it receives. The null backend does not take any options. 3810 3811``-chardev socket,id=id[,TCP options or unix options][,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,tls-creds=id][,tls-authz=id]`` 3812 Create a two-way stream socket, which can be either a TCP or a unix 3813 socket. A unix socket will be created if ``path`` is specified. 3814 Behaviour is undefined if TCP options are specified for a unix 3815 socket. 3816 3817 ``server=on|off`` specifies that the socket shall be a listening socket. 3818 3819 ``wait=on|off`` specifies that QEMU should not block waiting for a client 3820 to connect to a listening socket. 3821 3822 ``telnet=on|off`` specifies that traffic on the socket should interpret 3823 telnet escape sequences. 3824 3825 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for 3826 communication. 3827 3828 ``reconnect`` sets the timeout for reconnecting on non-server 3829 sockets when the remote end goes away. qemu will delay this many 3830 seconds and then attempt to reconnect. Zero disables reconnecting, 3831 and is the default. 3832 3833 ``tls-creds`` requests enablement of the TLS protocol for 3834 encryption, and specifies the id of the TLS credentials to use for 3835 the handshake. The credentials must be previously created with the 3836 ``-object tls-creds`` argument. 3837 3838 ``tls-auth`` provides the ID of the QAuthZ authorization object 3839 against which the client's x509 distinguished name will be 3840 validated. This object is only resolved at time of use, so can be 3841 deleted and recreated on the fly while the chardev server is active. 3842 If missing, it will default to denying access. 3843 3844 TCP and unix socket options are given below: 3845 3846 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]`` 3847 ``host`` for a listening socket specifies the local address to 3848 be bound. For a connecting socket species the remote host to 3849 connect to. ``host`` is optional for listening sockets. If not 3850 specified it defaults to ``0.0.0.0``. 3851 3852 ``port`` for a listening socket specifies the local port to be 3853 bound. For a connecting socket specifies the port on the remote 3854 host to connect to. ``port`` can be given as either a port 3855 number or a service name. ``port`` is required. 3856 3857 ``to`` is only relevant to listening sockets. If it is 3858 specified, and ``port`` cannot be bound, QEMU will attempt to 3859 bind to subsequent ports up to and including ``to`` until it 3860 succeeds. ``to`` must be specified as a port number. 3861 3862 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 3863 or IPv6 must be used. If neither is specified the socket may 3864 use either protocol. 3865 3866 ``nodelay=on|off`` disables the Nagle algorithm. 3867 3868 ``unix options: path=path[,abstract=on|off][,tight=on|off]`` 3869 ``path`` specifies the local path of the unix socket. ``path`` 3870 is required. 3871 ``abstract=on|off`` specifies the use of the abstract socket namespace, 3872 rather than the filesystem. Optional, defaults to false. 3873 ``tight=on|off`` sets the socket length of abstract sockets to their minimum, 3874 rather than the full sun_path length. Optional, defaults to true. 3875 3876``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]`` 3877 Sends all traffic from the guest to a remote host over UDP. 3878 3879 ``host`` specifies the remote host to connect to. If not specified 3880 it defaults to ``localhost``. 3881 3882 ``port`` specifies the port on the remote host to connect to. 3883 ``port`` is required. 3884 3885 ``localaddr`` specifies the local address to bind to. If not 3886 specified it defaults to ``0.0.0.0``. 3887 3888 ``localport`` specifies the local port to bind to. If not specified 3889 any available local port will be used. 3890 3891 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used. 3892 If neither is specified the device may use either protocol. 3893 3894``-chardev msmouse,id=id`` 3895 Forward QEMU's emulated msmouse events to the guest. ``msmouse`` 3896 does not take any options. 3897 3898``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]`` 3899 Connect to a QEMU text console. ``vc`` may optionally be given a 3900 specific size. 3901 3902 ``width`` and ``height`` specify the width and height respectively 3903 of the console, in pixels. 3904 3905 ``cols`` and ``rows`` specify that the console be sized to fit a 3906 text console with the given dimensions. 3907 3908``-chardev ringbuf,id=id[,size=size]`` 3909 Create a ring buffer with fixed size ``size``. size must be a power 3910 of two and defaults to ``64K``. 3911 3912``-chardev file,id=id,path=path[,input-path=input-path]`` 3913 Log all traffic received from the guest to a file. 3914 3915 ``path`` specifies the path of the file to be opened. This file will 3916 be created if it does not already exist, and overwritten if it does. 3917 ``path`` is required. 3918 3919 If ``input-path`` is specified, this is the path of a second file 3920 which will be used for input. If ``input-path`` is not specified, 3921 no input will be available from the chardev. 3922 3923 Note that ``input-path`` is not supported on Windows hosts. 3924 3925``-chardev pipe,id=id,path=path`` 3926 Create a two-way connection to the guest. The behaviour differs 3927 slightly between Windows hosts and other hosts: 3928 3929 On Windows, a single duplex pipe will be created at 3930 ``\\.pipe\path``. 3931 3932 On other hosts, 2 pipes will be created called ``path.in`` and 3933 ``path.out``. Data written to ``path.in`` will be received by the 3934 guest. Data written by the guest can be read from ``path.out``. QEMU 3935 will not create these fifos, and requires them to be present. 3936 3937 ``path`` forms part of the pipe path as described above. ``path`` is 3938 required. 3939 3940``-chardev console,id=id`` 3941 Send traffic from the guest to QEMU's standard output. ``console`` 3942 does not take any options. 3943 3944 ``console`` is only available on Windows hosts. 3945 3946``-chardev serial,id=id,path=path`` 3947 Send traffic from the guest to a serial device on the host. 3948 3949 On Unix hosts serial will actually accept any tty device, not only 3950 serial lines. 3951 3952 ``path`` specifies the name of the serial device to open. 3953 3954``-chardev pty,id=id[,path=path]`` 3955 Create a new pseudo-terminal on the host and connect to it. 3956 3957 ``pty`` is not available on Windows hosts. 3958 3959 If ``path`` is specified, QEMU will create a symbolic link at 3960 that location which points to the new PTY device. 3961 3962 This avoids having to make QMP or HMP monitor queries to find out 3963 what the new PTY device path is. 3964 3965 Note that while QEMU will remove the symlink when it exits 3966 gracefully, it will not do so in case of crashes or on certain 3967 startup errors. It is recommended that the user checks and removes 3968 the symlink after QEMU terminates to account for this. 3969 3970``-chardev stdio,id=id[,signal=on|off]`` 3971 Connect to standard input and standard output of the QEMU process. 3972 3973 ``signal`` controls if signals are enabled on the terminal, that 3974 includes exiting QEMU with the key sequence Control-c. This option 3975 is enabled by default, use ``signal=off`` to disable it. 3976 3977``-chardev braille,id=id`` 3978 Connect to a local BrlAPI server. ``braille`` does not take any 3979 options. 3980 3981``-chardev parallel,id=id,path=path`` 3982 \ 3983 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD 3984 hosts. 3985 3986 Connect to a local parallel port. 3987 3988 ``path`` specifies the path to the parallel port device. ``path`` is 3989 required. 3990 3991``-chardev spicevmc,id=id,debug=debug,name=name`` 3992 ``spicevmc`` is only available when spice support is built in. 3993 3994 ``debug`` debug level for spicevmc 3995 3996 ``name`` name of spice channel to connect to 3997 3998 Connect to a spice virtual machine channel, such as vdiport. 3999 4000``-chardev spiceport,id=id,debug=debug,name=name`` 4001 ``spiceport`` is only available when spice support is built in. 4002 4003 ``debug`` debug level for spicevmc 4004 4005 ``name`` name of spice port to connect to 4006 4007 Connect to a spice port, allowing a Spice client to handle the 4008 traffic identified by a name (preferably a fqdn). 4009ERST 4010 4011DEFHEADING() 4012 4013#ifdef CONFIG_TPM 4014DEFHEADING(TPM device options:) 4015 4016DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \ 4017 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n" 4018 " use path to provide path to a character device; default is /dev/tpm0\n" 4019 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n" 4020 " not provided it will be searched for in /sys/class/misc/tpm?/device\n" 4021 "-tpmdev emulator,id=id,chardev=dev\n" 4022 " configure the TPM device using chardev backend\n", 4023 QEMU_ARCH_ALL) 4024SRST 4025The general form of a TPM device option is: 4026 4027``-tpmdev backend,id=id[,options]`` 4028 The specific backend type will determine the applicable options. The 4029 ``-tpmdev`` option creates the TPM backend and requires a 4030 ``-device`` option that specifies the TPM frontend interface model. 4031 4032 Use ``-tpmdev help`` to print all available TPM backend types. 4033 4034The available backends are: 4035 4036``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path`` 4037 (Linux-host only) Enable access to the host's TPM using the 4038 passthrough driver. 4039 4040 ``path`` specifies the path to the host's TPM device, i.e., on a 4041 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by 4042 default ``/dev/tpm0`` is used. 4043 4044 ``cancel-path`` specifies the path to the host TPM device's sysfs 4045 entry allowing for cancellation of an ongoing TPM command. 4046 ``cancel-path`` is optional and by default QEMU will search for the 4047 sysfs entry to use. 4048 4049 Some notes about using the host's TPM with the passthrough driver: 4050 4051 The TPM device accessed by the passthrough driver must not be used 4052 by any other application on the host. 4053 4054 Since the host's firmware (BIOS/UEFI) has already initialized the 4055 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize 4056 the TPM again and may therefore not show a TPM-specific menu that 4057 would otherwise allow the user to configure the TPM, e.g., allow the 4058 user to enable/disable or activate/deactivate the TPM. Further, if 4059 TPM ownership is released from within a VM then the host's TPM will 4060 get disabled and deactivated. To enable and activate the TPM again 4061 afterwards, the host has to be rebooted and the user is required to 4062 enter the firmware's menu to enable and activate the TPM. If the TPM 4063 is left disabled and/or deactivated most TPM commands will fail. 4064 4065 To create a passthrough TPM use the following two options: 4066 4067 :: 4068 4069 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0 4070 4071 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by 4072 ``tpmdev=tpm0`` in the device option. 4073 4074``-tpmdev emulator,id=id,chardev=dev`` 4075 (Linux-host only) Enable access to a TPM emulator using Unix domain 4076 socket based chardev backend. 4077 4078 ``chardev`` specifies the unique ID of a character device backend 4079 that provides connection to the software TPM server. 4080 4081 To create a TPM emulator backend device with chardev socket backend: 4082 4083 :: 4084 4085 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0 4086ERST 4087 4088DEFHEADING() 4089 4090#endif 4091 4092DEFHEADING(Boot Image or Kernel specific:) 4093SRST 4094There are broadly 4 ways you can boot a system with QEMU. 4095 4096 - specify a firmware and let it control finding a kernel 4097 - specify a firmware and pass a hint to the kernel to boot 4098 - direct kernel image boot 4099 - manually load files into the guest's address space 4100 4101The third method is useful for quickly testing kernels but as there is 4102no firmware to pass configuration information to the kernel the 4103hardware must either be probeable, the kernel built for the exact 4104configuration or passed some configuration data (e.g. a DTB blob) 4105which tells the kernel what drivers it needs. This exact details are 4106often hardware specific. 4107 4108The final method is the most generic way of loading images into the 4109guest address space and used mostly for ``bare metal`` type 4110development where the reset vectors of the processor are taken into 4111account. 4112 4113ERST 4114 4115SRST 4116 4117For x86 machines and some other architectures ``-bios`` will generally 4118do the right thing with whatever it is given. For other machines the 4119more strict ``-pflash`` option needs an image that is sized for the 4120flash device for the given machine type. 4121 4122Please see the :ref:`system-targets-ref` section of the manual for 4123more detailed documentation. 4124 4125ERST 4126 4127DEF("bios", HAS_ARG, QEMU_OPTION_bios, \ 4128 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL) 4129SRST 4130``-bios file`` 4131 Set the filename for the BIOS. 4132ERST 4133 4134DEF("pflash", HAS_ARG, QEMU_OPTION_pflash, 4135 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL) 4136SRST 4137``-pflash file`` 4138 Use file as a parallel flash image. 4139ERST 4140 4141SRST 4142 4143The kernel options were designed to work with Linux kernels although 4144other things (like hypervisors) can be packaged up as a kernel 4145executable image. The exact format of a executable image is usually 4146architecture specific. 4147 4148The way in which the kernel is started (what address it is loaded at, 4149what if any information is passed to it via CPU registers, the state 4150of the hardware when it is started, and so on) is also architecture 4151specific. Typically it follows the specification laid down by the 4152Linux kernel for how kernels for that architecture must be started. 4153 4154ERST 4155 4156DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \ 4157 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL) 4158SRST 4159``-kernel bzImage`` 4160 Use bzImage as kernel image. The kernel can be either a Linux kernel 4161 or in multiboot format. 4162ERST 4163 4164DEF("append", HAS_ARG, QEMU_OPTION_append, \ 4165 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL) 4166SRST 4167``-append cmdline`` 4168 Use cmdline as kernel command line 4169ERST 4170 4171DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \ 4172 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL) 4173SRST(initrd) 4174 4175``-initrd file`` 4176 Use file as initial ram disk. 4177 4178``-initrd "file1 arg=foo,file2"`` 4179 This syntax is only available with multiboot. 4180 4181 Use file1 and file2 as modules and pass ``arg=foo`` as parameter to the 4182 first module. Commas can be provided in module parameters by doubling 4183 them on the command line to escape them: 4184 4185``-initrd "bzImage earlyprintk=xen,,keep root=/dev/xvda1,initrd.img"`` 4186 Multiboot only. Use bzImage as the first module with 4187 "``earlyprintk=xen,keep root=/dev/xvda1``" as its command line, 4188 and initrd.img as the second module. 4189 4190ERST 4191 4192DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \ 4193 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL) 4194SRST 4195``-dtb file`` 4196 Use file as a device tree binary (dtb) image and pass it to the 4197 kernel on boot. 4198ERST 4199 4200SRST 4201 4202Finally you can also manually load images directly into the address 4203space of the guest. This is most useful for developers who already 4204know the layout of their guest and take care to ensure something sane 4205will happen when the reset vector executes. 4206 4207The generic loader can be invoked by using the loader device: 4208 4209``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]`` 4210 4211there is also the guest loader which operates in a similar way but 4212tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where 4213the guest image is: 4214 4215``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]`` 4216 4217ERST 4218 4219DEFHEADING() 4220 4221DEFHEADING(Debug/Expert options:) 4222 4223DEF("compat", HAS_ARG, QEMU_OPTION_compat, 4224 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n" 4225 " Policy for handling deprecated management interfaces\n" 4226 "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n" 4227 " Policy for handling unstable management interfaces\n", 4228 QEMU_ARCH_ALL) 4229SRST 4230``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]`` 4231 Set policy for handling deprecated management interfaces (experimental): 4232 4233 ``deprecated-input=accept`` (default) 4234 Accept deprecated commands and arguments 4235 ``deprecated-input=reject`` 4236 Reject deprecated commands and arguments 4237 ``deprecated-input=crash`` 4238 Crash on deprecated commands and arguments 4239 ``deprecated-output=accept`` (default) 4240 Emit deprecated command results and events 4241 ``deprecated-output=hide`` 4242 Suppress deprecated command results and events 4243 4244 Limitation: covers only syntactic aspects of QMP. 4245 4246``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]`` 4247 Set policy for handling unstable management interfaces (experimental): 4248 4249 ``unstable-input=accept`` (default) 4250 Accept unstable commands and arguments 4251 ``unstable-input=reject`` 4252 Reject unstable commands and arguments 4253 ``unstable-input=crash`` 4254 Crash on unstable commands and arguments 4255 ``unstable-output=accept`` (default) 4256 Emit unstable command results and events 4257 ``unstable-output=hide`` 4258 Suppress unstable command results and events 4259 4260 Limitation: covers only syntactic aspects of QMP. 4261ERST 4262 4263DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg, 4264 "-fw_cfg [name=]<name>,file=<file>\n" 4265 " add named fw_cfg entry with contents from file\n" 4266 "-fw_cfg [name=]<name>,string=<str>\n" 4267 " add named fw_cfg entry with contents from string\n", 4268 QEMU_ARCH_ALL) 4269SRST 4270``-fw_cfg [name=]name,file=file`` 4271 Add named fw\_cfg entry with contents from file file. 4272 If the filename contains comma, you must double it (for instance, 4273 "file=my,,file" to use file "my,file"). 4274 4275``-fw_cfg [name=]name,string=str`` 4276 Add named fw\_cfg entry with contents from string str. 4277 If the string contains comma, you must double it (for instance, 4278 "string=my,,string" to use file "my,string"). 4279 4280 The terminating NUL character of the contents of str will not be 4281 included as part of the fw\_cfg item data. To insert contents with 4282 embedded NUL characters, you have to use the file parameter. 4283 4284 The fw\_cfg entries are passed by QEMU through to the guest. 4285 4286 Example: 4287 4288 :: 4289 4290 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin 4291 4292 creates an fw\_cfg entry named opt/com.mycompany/blob with contents 4293 from ./my\_blob.bin. 4294ERST 4295 4296DEF("serial", HAS_ARG, QEMU_OPTION_serial, \ 4297 "-serial dev redirect the serial port to char device 'dev'\n", 4298 QEMU_ARCH_ALL) 4299SRST 4300``-serial dev`` 4301 Redirect the virtual serial port to host character device dev. The 4302 default device is ``vc`` in graphical mode and ``stdio`` in non 4303 graphical mode. 4304 4305 This option can be used several times to simulate multiple serial 4306 ports. 4307 4308 You can use ``-serial none`` to suppress the creation of default 4309 serial devices. 4310 4311 Available character devices are: 4312 4313 ``vc[:WxH]`` 4314 Virtual console. Optionally, a width and height can be given in 4315 pixel with 4316 4317 :: 4318 4319 vc:800x600 4320 4321 It is also possible to specify width or height in characters: 4322 4323 :: 4324 4325 vc:80Cx24C 4326 4327 ``pty[:path]`` 4328 [Linux only] Pseudo TTY (a new PTY is automatically allocated). 4329 4330 If ``path`` is specified, QEMU will create a symbolic link at 4331 that location which points to the new PTY device. 4332 4333 This avoids having to make QMP or HMP monitor queries to find 4334 out what the new PTY device path is. 4335 4336 Note that while QEMU will remove the symlink when it exits 4337 gracefully, it will not do so in case of crashes or on certain 4338 startup errors. It is recommended that the user checks and 4339 removes the symlink after QEMU terminates to account for this. 4340 4341 ``none`` 4342 No device is allocated. Note that for machine types which 4343 emulate systems where a serial device is always present in 4344 real hardware, this may be equivalent to the ``null`` option, 4345 in that the serial device is still present but all output 4346 is discarded. For boards where the number of serial ports is 4347 truly variable, this suppresses the creation of the device. 4348 4349 ``null`` 4350 A guest will see the UART or serial device as present in the 4351 machine, but all output is discarded, and there is no input. 4352 Conceptually equivalent to redirecting the output to ``/dev/null``. 4353 4354 ``chardev:id`` 4355 Use a named character device defined with the ``-chardev`` 4356 option. 4357 4358 ``/dev/XXX`` 4359 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial 4360 port parameters are set according to the emulated ones. 4361 4362 ``/dev/parportN`` 4363 [Linux only, parallel port only] Use host parallel port N. 4364 Currently SPP and EPP parallel port features can be used. 4365 4366 ``file:filename`` 4367 Write output to filename. No character can be read. 4368 4369 ``stdio`` 4370 [Unix only] standard input/output 4371 4372 ``pipe:filename`` 4373 name pipe filename 4374 4375 ``COMn`` 4376 [Windows only] Use host serial port n 4377 4378 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]`` 4379 This implements UDP Net Console. When remote\_host or src\_ip 4380 are not specified they default to ``0.0.0.0``. When not using a 4381 specified src\_port a random port is automatically chosen. 4382 4383 If you just want a simple readonly console you can use 4384 ``netcat`` or ``nc``, by starting QEMU with: 4385 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time 4386 QEMU writes something to that port it will appear in the 4387 netconsole session. 4388 4389 If you plan to send characters back via netconsole or you want 4390 to stop and start QEMU a lot of times, you should have QEMU use 4391 the same source port each time by using something like ``-serial 4392 udp::4555@:4556`` to QEMU. Another approach is to use a patched 4393 version of netcat which can listen to a TCP port and send and 4394 receive characters via udp. If you have a patched version of 4395 netcat which activates telnet remote echo and single char 4396 transfer, then you can use the following options to set up a 4397 netcat redirector to allow telnet on port 5555 to access the 4398 QEMU port. 4399 4400 ``QEMU Options:`` 4401 -serial udp::4555@:4556 4402 4403 ``netcat options:`` 4404 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T 4405 4406 ``telnet options:`` 4407 localhost 5555 4408 4409 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]`` 4410 The TCP Net Console has two modes of operation. It can send the 4411 serial I/O to a location or wait for a connection from a 4412 location. By default the TCP Net Console is sent to host at the 4413 port. If you use the ``server=on`` option QEMU will wait for a client 4414 socket application to connect to the port before continuing, 4415 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off`` 4416 option disables the Nagle buffering algorithm. The ``reconnect=on`` 4417 option only applies if ``server=no`` is set, if the connection goes 4418 down it will attempt to reconnect at the given interval. If host 4419 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a 4420 time is accepted. You can use ``telnet=on`` to connect to the 4421 corresponding character device. 4422 4423 ``Example to send tcp console to 192.168.0.2 port 4444`` 4424 -serial tcp:192.168.0.2:4444 4425 4426 ``Example to listen and wait on port 4444 for connection`` 4427 -serial tcp::4444,server=on 4428 4429 ``Example to not wait and listen on ip 192.168.0.100 port 4444`` 4430 -serial tcp:192.168.0.100:4444,server=on,wait=off 4431 4432 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]`` 4433 The telnet protocol is used instead of raw tcp sockets. The 4434 options work the same as if you had specified ``-serial tcp``. 4435 The difference is that the port acts like a telnet server or 4436 client using telnet option negotiation. This will also allow you 4437 to send the MAGIC\_SYSRQ sequence if you use a telnet that 4438 supports sending the break sequence. Typically in unix telnet 4439 you do it with Control-] and then type "send break" followed by 4440 pressing the enter key. 4441 4442 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]`` 4443 The WebSocket protocol is used instead of raw tcp socket. The 4444 port acts as a WebSocket server. Client mode is not supported. 4445 4446 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]`` 4447 A unix domain socket is used instead of a tcp socket. The option 4448 works the same as if you had specified ``-serial tcp`` except 4449 the unix domain socket path is used for connections. 4450 4451 ``mon:dev_string`` 4452 This is a special option to allow the monitor to be multiplexed 4453 onto another serial port. The monitor is accessed with key 4454 sequence of Control-a and then pressing c. dev\_string should be 4455 any one of the serial devices specified above. An example to 4456 multiplex the monitor onto a telnet server listening on port 4457 4444 would be: 4458 4459 ``-serial mon:telnet::4444,server=on,wait=off`` 4460 4461 When the monitor is multiplexed to stdio in this way, Ctrl+C 4462 will not terminate QEMU any more but will be passed to the guest 4463 instead. 4464 4465 ``braille`` 4466 Braille device. This will use BrlAPI to display the braille 4467 output on a real or fake device. 4468 4469 ``msmouse`` 4470 Three button serial mouse. Configure the guest to use Microsoft 4471 protocol. 4472ERST 4473 4474DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \ 4475 "-parallel dev redirect the parallel port to char device 'dev'\n", 4476 QEMU_ARCH_ALL) 4477SRST 4478``-parallel dev`` 4479 Redirect the virtual parallel port to host device dev (same devices 4480 as the serial port). On Linux hosts, ``/dev/parportN`` can be used 4481 to use hardware devices connected on the corresponding host parallel 4482 port. 4483 4484 This option can be used several times to simulate up to 3 parallel 4485 ports. 4486 4487 Use ``-parallel none`` to disable all parallel ports. 4488ERST 4489 4490DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \ 4491 "-monitor dev redirect the monitor to char device 'dev'\n", 4492 QEMU_ARCH_ALL) 4493SRST 4494``-monitor dev`` 4495 Redirect the monitor to host device dev (same devices as the serial 4496 port). The default device is ``vc`` in graphical mode and ``stdio`` 4497 in non graphical mode. Use ``-monitor none`` to disable the default 4498 monitor. 4499ERST 4500DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \ 4501 "-qmp dev like -monitor but opens in 'control' mode\n", 4502 QEMU_ARCH_ALL) 4503SRST 4504``-qmp dev`` 4505 Like ``-monitor`` but opens in 'control' mode. For example, to make 4506 QMP available on localhost port 4444:: 4507 4508 -qmp tcp:localhost:4444,server=on,wait=off 4509 4510 Not all options are configurable via this syntax; for maximum 4511 flexibility use the ``-mon`` option and an accompanying ``-chardev``. 4512 4513ERST 4514DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \ 4515 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n", 4516 QEMU_ARCH_ALL) 4517SRST 4518``-qmp-pretty dev`` 4519 Like ``-qmp`` but uses pretty JSON formatting. 4520ERST 4521 4522DEF("mon", HAS_ARG, QEMU_OPTION_mon, \ 4523 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL) 4524SRST 4525``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]`` 4526 Set up a monitor connected to the chardev ``name``. 4527 QEMU supports two monitors: the Human Monitor Protocol 4528 (HMP; for human interaction), and the QEMU Monitor Protocol 4529 (QMP; a JSON RPC-style protocol). 4530 The default is HMP; ``mode=control`` selects QMP instead. 4531 ``pretty`` is only valid when ``mode=control``, 4532 turning on JSON pretty printing to ease 4533 human reading and debugging. 4534 4535 For example:: 4536 4537 -chardev socket,id=mon1,host=localhost,port=4444,server=on,wait=off \ 4538 -mon chardev=mon1,mode=control,pretty=on 4539 4540 enables the QMP monitor on localhost port 4444 with pretty-printing. 4541ERST 4542 4543DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \ 4544 "-debugcon dev redirect the debug console to char device 'dev'\n", 4545 QEMU_ARCH_ALL) 4546SRST 4547``-debugcon dev`` 4548 Redirect the debug console to host device dev (same devices as the 4549 serial port). The debug console is an I/O port which is typically 4550 port 0xe9; writing to that I/O port sends output to this device. The 4551 default device is ``vc`` in graphical mode and ``stdio`` in non 4552 graphical mode. 4553ERST 4554 4555DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \ 4556 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL) 4557SRST 4558``-pidfile file`` 4559 Store the QEMU process PID in file. It is useful if you launch QEMU 4560 from a script. 4561ERST 4562 4563DEF("preconfig", 0, QEMU_OPTION_preconfig, \ 4564 "--preconfig pause QEMU before machine is initialized (experimental)\n", 4565 QEMU_ARCH_ALL) 4566SRST 4567``--preconfig`` 4568 Pause QEMU for interactive configuration before the machine is 4569 created, which allows querying and configuring properties that will 4570 affect machine initialization. Use QMP command 'x-exit-preconfig' to 4571 exit the preconfig state and move to the next state (i.e. run guest 4572 if -S isn't used or pause the second time if -S is used). This 4573 option is experimental. 4574ERST 4575 4576DEF("S", 0, QEMU_OPTION_S, \ 4577 "-S freeze CPU at startup (use 'c' to start execution)\n", 4578 QEMU_ARCH_ALL) 4579SRST 4580``-S`` 4581 Do not start CPU at startup (you must type 'c' in the monitor). 4582ERST 4583 4584DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit, 4585 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n" 4586 " run qemu with overcommit hints\n" 4587 " mem-lock=on|off controls memory lock support (default: off)\n" 4588 " cpu-pm=on|off controls cpu power management (default: off)\n", 4589 QEMU_ARCH_ALL) 4590SRST 4591``-overcommit mem-lock=on|off`` 4592 \ 4593``-overcommit cpu-pm=on|off`` 4594 Run qemu with hints about host resource overcommit. The default is 4595 to assume that host overcommits all resources. 4596 4597 Locking qemu and guest memory can be enabled via ``mem-lock=on`` 4598 (disabled by default). This works when host memory is not 4599 overcommitted and reduces the worst-case latency for guest. 4600 4601 Guest ability to manage power state of host cpus (increasing latency 4602 for other processes on the same host cpu, but decreasing latency for 4603 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This 4604 works best when host CPU is not overcommitted. When used, host 4605 estimates of CPU cycle and power utilization will be incorrect, not 4606 taking into account guest idle time. 4607ERST 4608 4609DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \ 4610 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n" 4611 " the guest without waiting for gdb to connect; use -S too\n" 4612 " if you want it to not start execution.)\n", 4613 QEMU_ARCH_ALL) 4614SRST 4615``-gdb dev`` 4616 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter 4617 in the System Emulation Users Guide). Note that this option does not pause QEMU 4618 execution -- if you want QEMU to not start the guest until you 4619 connect with gdb and issue a ``continue`` command, you will need to 4620 also pass the ``-S`` option to QEMU. 4621 4622 The most usual configuration is to listen on a local TCP socket:: 4623 4624 -gdb tcp::3117 4625 4626 but you can specify other backends; UDP, pseudo TTY, or even stdio 4627 are all reasonable use cases. For example, a stdio connection 4628 allows you to start QEMU from within gdb and establish the 4629 connection via a pipe: 4630 4631 .. parsed-literal:: 4632 4633 (gdb) target remote | exec |qemu_system| -gdb stdio ... 4634ERST 4635 4636DEF("s", 0, QEMU_OPTION_s, \ 4637 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n", 4638 QEMU_ARCH_ALL) 4639SRST 4640``-s`` 4641 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234 4642 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide). 4643ERST 4644 4645DEF("d", HAS_ARG, QEMU_OPTION_d, \ 4646 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n", 4647 QEMU_ARCH_ALL) 4648SRST 4649``-d item1[,...]`` 4650 Enable logging of specified items. Use '-d help' for a list of log 4651 items. 4652ERST 4653 4654DEF("D", HAS_ARG, QEMU_OPTION_D, \ 4655 "-D logfile output log to logfile (default stderr)\n", 4656 QEMU_ARCH_ALL) 4657SRST 4658``-D logfile`` 4659 Output log in logfile instead of to stderr 4660ERST 4661 4662DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \ 4663 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n", 4664 QEMU_ARCH_ALL) 4665SRST 4666``-dfilter range1[,...]`` 4667 Filter debug output to that relevant to a range of target addresses. 4668 The filter spec can be either start+size, start-size or start..end 4669 where start end and size are the addresses and sizes required. For 4670 example: 4671 4672 :: 4673 4674 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000 4675 4676 Will dump output for any code in the 0x1000 sized block starting at 4677 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and 4678 another 0x1000 sized block starting at 0xffffffc00005f000. 4679ERST 4680 4681DEF("seed", HAS_ARG, QEMU_OPTION_seed, \ 4682 "-seed number seed the pseudo-random number generator\n", 4683 QEMU_ARCH_ALL) 4684SRST 4685``-seed number`` 4686 Force the guest to use a deterministic pseudo-random number 4687 generator, seeded with number. This does not affect crypto routines 4688 within the host. 4689ERST 4690 4691DEF("L", HAS_ARG, QEMU_OPTION_L, \ 4692 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n", 4693 QEMU_ARCH_ALL) 4694SRST 4695``-L path`` 4696 Set the directory for the BIOS, VGA BIOS and keymaps. 4697 4698 To list all the data directories, use ``-L help``. 4699ERST 4700 4701DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \ 4702 "-enable-kvm enable KVM full virtualization support\n", 4703 QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC | 4704 QEMU_ARCH_RISCV | QEMU_ARCH_S390X) 4705SRST 4706``-enable-kvm`` 4707 Enable KVM full virtualization support. This option is only 4708 available if KVM support is enabled when compiling. 4709ERST 4710 4711DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, 4712 "-xen-domid id specify xen guest domain id\n", 4713 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4714DEF("xen-attach", 0, QEMU_OPTION_xen_attach, 4715 "-xen-attach attach to existing xen domain\n" 4716 " libxl will use this when starting QEMU\n", 4717 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4718DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, 4719 "-xen-domid-restrict restrict set of available xen operations\n" 4720 " to specified domain id. (Does not affect\n" 4721 " xenpv machine type).\n", 4722 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4723SRST 4724``-xen-domid id`` 4725 Specify xen guest domain id (XEN only). 4726 4727``-xen-attach`` 4728 Attach to existing xen domain. libxl will use this when starting 4729 QEMU (XEN only). Restrict set of available xen operations to 4730 specified domain id (XEN only). 4731ERST 4732 4733DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \ 4734 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL) 4735SRST 4736``-no-reboot`` 4737 Exit instead of rebooting. 4738ERST 4739 4740DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \ 4741 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL) 4742SRST 4743``-no-shutdown`` 4744 Don't exit QEMU on guest shutdown, but instead only stop the 4745 emulation. This allows for instance switching to monitor to commit 4746 changes to the disk image. 4747ERST 4748 4749DEF("action", HAS_ARG, QEMU_OPTION_action, 4750 "-action reboot=reset|shutdown\n" 4751 " action when guest reboots [default=reset]\n" 4752 "-action shutdown=poweroff|pause\n" 4753 " action when guest shuts down [default=poweroff]\n" 4754 "-action panic=pause|shutdown|exit-failure|none\n" 4755 " action when guest panics [default=shutdown]\n" 4756 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" 4757 " action when watchdog fires [default=reset]\n", 4758 QEMU_ARCH_ALL) 4759SRST 4760``-action event=action`` 4761 The action parameter serves to modify QEMU's default behavior when 4762 certain guest events occur. It provides a generic method for specifying the 4763 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown`` 4764 parameters. 4765 4766 Examples: 4767 4768 ``-action panic=none`` 4769 ``-action reboot=shutdown,shutdown=pause`` 4770 ``-device i6300esb -action watchdog=pause`` 4771 4772ERST 4773 4774DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \ 4775 "-loadvm [tag|id]\n" \ 4776 " start right away with a saved state (loadvm in monitor)\n", 4777 QEMU_ARCH_ALL) 4778SRST 4779``-loadvm file`` 4780 Start right away with a saved state (``loadvm`` in monitor) 4781ERST 4782 4783#ifndef _WIN32 4784DEF("daemonize", 0, QEMU_OPTION_daemonize, \ 4785 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL) 4786#endif 4787SRST 4788``-daemonize`` 4789 Daemonize the QEMU process after initialization. QEMU will not 4790 detach from standard IO until it is ready to receive connections on 4791 any of its devices. This option is a useful way for external 4792 programs to launch QEMU without having to cope with initialization 4793 race conditions. 4794ERST 4795 4796DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \ 4797 "-option-rom rom load a file, rom, into the option ROM space\n", 4798 QEMU_ARCH_ALL) 4799SRST 4800``-option-rom file`` 4801 Load the contents of file as an option ROM. This option is useful to 4802 load things like EtherBoot. 4803ERST 4804 4805DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \ 4806 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \ 4807 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n", 4808 QEMU_ARCH_ALL) 4809 4810SRST 4811``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]`` 4812 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at 4813 the current UTC or local time, respectively. ``localtime`` is 4814 required for correct date in MS-DOS or Windows. To start at a 4815 specific point in time, provide datetime in the format 4816 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC. 4817 4818 By default the RTC is driven by the host system time. This allows 4819 using of the RTC as accurate reference clock inside the guest, 4820 specifically if the host time is smoothly following an accurate 4821 external reference clock, e.g. via NTP. If you want to isolate the 4822 guest time from the host, you can set ``clock`` to ``rt`` instead, 4823 which provides a host monotonic clock if host support it. To even 4824 prevent the RTC from progressing during suspension, you can set 4825 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is 4826 recommended especially in icount mode in order to preserve 4827 determinism; however, note that in icount mode the speed of the 4828 virtual clock is variable and can in general differ from the host 4829 clock. 4830 4831 Enable ``driftfix`` (i386 targets only) if you experience time drift 4832 problems, specifically with Windows' ACPI HAL. This option will try 4833 to figure out how many timer interrupts were not processed by the 4834 Windows guest and will re-inject them. 4835ERST 4836 4837DEF("icount", HAS_ARG, QEMU_OPTION_icount, \ 4838 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \ 4839 " enable virtual instruction counter with 2^N clock ticks per\n" \ 4840 " instruction, enable aligning the host and virtual clocks\n" \ 4841 " or disable real time cpu sleeping, and optionally enable\n" \ 4842 " record-and-replay mode\n", QEMU_ARCH_ALL) 4843SRST 4844``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]`` 4845 Enable virtual instruction counter. The virtual cpu will execute one 4846 instruction every 2^N ns of virtual time. If ``auto`` is specified 4847 then the virtual cpu speed will be automatically adjusted to keep 4848 virtual time within a few seconds of real time. 4849 4850 Note that while this option can give deterministic behavior, it does 4851 not provide cycle accurate emulation. Modern CPUs contain 4852 superscalar out of order cores with complex cache hierarchies. The 4853 number of instructions executed often has little or no correlation 4854 with actual performance. 4855 4856 When the virtual cpu is sleeping, the virtual time will advance at 4857 default speed unless ``sleep=on`` is specified. With 4858 ``sleep=on``, the virtual time will jump to the next timer 4859 deadline instantly whenever the virtual cpu goes to sleep mode and 4860 will not advance if no timer is enabled. This behavior gives 4861 deterministic execution times from the guest point of view. 4862 The default if icount is enabled is ``sleep=off``. 4863 ``sleep=on`` cannot be used together with either ``shift=auto`` 4864 or ``align=on``. 4865 4866 ``align=on`` will activate the delay algorithm which will try to 4867 synchronise the host clock and the virtual clock. The goal is to 4868 have a guest running at the real frequency imposed by the shift 4869 option. Whenever the guest clock is behind the host clock and if 4870 ``align=on`` is specified then we print a message to the user to 4871 inform about the delay. Currently this option does not work when 4872 ``shift`` is ``auto``. Note: The sync algorithm will work for those 4873 shift values for which the guest clock runs ahead of the host clock. 4874 Typically this happens when the shift value is high (how high 4875 depends on the host machine). The default if icount is enabled 4876 is ``align=off``. 4877 4878 When the ``rr`` option is specified deterministic record/replay is 4879 enabled. The ``rrfile=`` option must also be provided to 4880 specify the path to the replay log. In record mode data is written 4881 to this file, and in replay mode it is read back. 4882 If the ``rrsnapshot`` option is given then it specifies a VM snapshot 4883 name. In record mode, a new VM snapshot with the given name is created 4884 at the start of execution recording. In replay mode this option 4885 specifies the snapshot name used to load the initial VM state. 4886ERST 4887 4888DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \ 4889 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \ 4890 " action when watchdog fires [default=reset]\n", 4891 QEMU_ARCH_ALL) 4892SRST 4893``-watchdog-action action`` 4894 The action controls what QEMU will do when the watchdog timer 4895 expires. The default is ``reset`` (forcefully reset the guest). 4896 Other possible actions are: ``shutdown`` (attempt to gracefully 4897 shutdown the guest), ``poweroff`` (forcefully poweroff the guest), 4898 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the 4899 guest), ``debug`` (print a debug message and continue), or ``none`` 4900 (do nothing). 4901 4902 Note that the ``shutdown`` action requires that the guest responds 4903 to ACPI signals, which it may not be able to do in the sort of 4904 situations where the watchdog would have expired, and thus 4905 ``-watchdog-action shutdown`` is not recommended for production use. 4906 4907 Examples: 4908 4909 ``-device i6300esb -watchdog-action pause`` 4910 4911ERST 4912 4913DEF("echr", HAS_ARG, QEMU_OPTION_echr, \ 4914 "-echr chr set terminal escape character instead of ctrl-a\n", 4915 QEMU_ARCH_ALL) 4916SRST 4917``-echr numeric_ascii_value`` 4918 Change the escape character used for switching to the monitor when 4919 using monitor and serial sharing. The default is ``0x01`` when using 4920 the ``-nographic`` option. ``0x01`` is equal to pressing 4921 ``Control-a``. You can select a different character from the ascii 4922 control keys where 1 through 26 map to Control-a through Control-z. 4923 For instance you could use the either of the following to change the 4924 escape character to Control-t. 4925 4926 ``-echr 0x14``; \ ``-echr 20`` 4927 4928ERST 4929 4930DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \ 4931 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \ 4932 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \ 4933 "-incoming unix:socketpath\n" \ 4934 " prepare for incoming migration, listen on\n" \ 4935 " specified protocol and socket address\n" \ 4936 "-incoming fd:fd\n" \ 4937 "-incoming file:filename[,offset=offset]\n" \ 4938 "-incoming exec:cmdline\n" \ 4939 " accept incoming migration on given file descriptor\n" \ 4940 " or from given external command\n" \ 4941 "-incoming defer\n" \ 4942 " wait for the URI to be specified via migrate_incoming\n", 4943 QEMU_ARCH_ALL) 4944SRST 4945``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]`` 4946 \ 4947``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]`` 4948 Prepare for incoming migration, listen on a given tcp port. 4949 4950``-incoming unix:socketpath`` 4951 Prepare for incoming migration, listen on a given unix socket. 4952 4953``-incoming fd:fd`` 4954 Accept incoming migration from a given file descriptor. 4955 4956``-incoming file:filename[,offset=offset]`` 4957 Accept incoming migration from a given file starting at offset. 4958 offset allows the common size suffixes, or a 0x prefix, but not both. 4959 4960``-incoming exec:cmdline`` 4961 Accept incoming migration as an output from specified external 4962 command. 4963 4964``-incoming defer`` 4965 Wait for the URI to be specified via migrate\_incoming. The monitor 4966 can be used to change settings (such as migration parameters) prior 4967 to issuing the migrate\_incoming to allow the migration to begin. 4968ERST 4969 4970DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \ 4971 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL) 4972SRST 4973``-only-migratable`` 4974 Only allow migratable devices. Devices will not be allowed to enter 4975 an unmigratable state. 4976ERST 4977 4978DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \ 4979 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL) 4980SRST 4981``-nodefaults`` 4982 Don't create default devices. Normally, QEMU sets the default 4983 devices like serial port, parallel port, virtual console, monitor 4984 device, VGA adapter, floppy and CD-ROM drive and others. The 4985 ``-nodefaults`` option will disable all those default devices. 4986ERST 4987 4988#ifndef _WIN32 4989DEF("runas", HAS_ARG, QEMU_OPTION_runas, \ 4990 "-runas user change to user id user just before starting the VM\n" \ 4991 " user can be numeric uid:gid instead\n", 4992 QEMU_ARCH_ALL) 4993#endif 4994SRST 4995``-runas user`` 4996 Immediately before starting guest execution, drop root privileges, 4997 switching to the specified user. This option is deprecated, use 4998 ``-run-with user=...`` instead. 4999ERST 5000 5001DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env, 5002 "-prom-env variable=value\n" 5003 " set OpenBIOS nvram variables\n", 5004 QEMU_ARCH_PPC | QEMU_ARCH_SPARC) 5005SRST 5006``-prom-env variable=value`` 5007 Set OpenBIOS nvram variable to given value (PPC, SPARC only). 5008 5009 :: 5010 5011 qemu-system-sparc -prom-env 'auto-boot?=false' \ 5012 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single' 5013 5014 :: 5015 5016 qemu-system-ppc -prom-env 'auto-boot?=false' \ 5017 -prom-env 'boot-device=hd:2,\yaboot' \ 5018 -prom-env 'boot-args=conf=hd:2,\yaboot.conf' 5019ERST 5020DEF("semihosting", 0, QEMU_OPTION_semihosting, 5021 "-semihosting semihosting mode\n", 5022 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 5023 QEMU_ARCH_MIPS | QEMU_ARCH_RISCV) 5024SRST 5025``-semihosting`` 5026 Enable :ref:`Semihosting` mode (ARM, M68K, Xtensa, MIPS, RISC-V only). 5027 5028 .. warning:: 5029 Note that this allows guest direct access to the host filesystem, so 5030 should only be used with a trusted guest OS. 5031 5032 See the -semihosting-config option documentation for further 5033 information about the facilities this enables. 5034ERST 5035DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config, 5036 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]\n" \ 5037 " semihosting configuration\n", 5038QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 5039QEMU_ARCH_MIPS | QEMU_ARCH_RISCV) 5040SRST 5041``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]`` 5042 Enable and configure :ref:`Semihosting` (ARM, M68K, Xtensa, MIPS, RISC-V 5043 only). 5044 5045 .. warning:: 5046 Note that this allows guest direct access to the host filesystem, so 5047 should only be used with a trusted guest OS. 5048 5049 ``target=native|gdb|auto`` 5050 Defines where the semihosting calls will be addressed, to QEMU 5051 (``native``) or to GDB (``gdb``). The default is ``auto``, which 5052 means ``gdb`` during debug sessions and ``native`` otherwise. 5053 5054 ``chardev=str1`` 5055 Send the output to a chardev backend output for native or auto 5056 output when not in gdb 5057 5058 ``userspace=on|off`` 5059 Allows code running in guest userspace to access the semihosting 5060 interface. The default is that only privileged guest code can 5061 make semihosting calls. Note that setting ``userspace=on`` should 5062 only be used if all guest code is trusted (for example, in 5063 bare-metal test case code). 5064 5065 ``arg=str1,arg=str2,...`` 5066 Allows the user to pass input arguments, and can be used 5067 multiple times to build up a list. The old-style 5068 ``-kernel``/``-append`` method of passing a command line is 5069 still supported for backward compatibility. If both the 5070 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are 5071 specified, the former is passed to semihosting as it always 5072 takes precedence. 5073ERST 5074DEF("old-param", 0, QEMU_OPTION_old_param, 5075 "-old-param old param mode\n", QEMU_ARCH_ARM) 5076SRST 5077``-old-param`` 5078 Old param mode (ARM only). 5079ERST 5080 5081DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \ 5082 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \ 5083 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \ 5084 " Enable seccomp mode 2 system call filter (default 'off').\n" \ 5085 " use 'obsolete' to allow obsolete system calls that are provided\n" \ 5086 " by the kernel, but typically no longer used by modern\n" \ 5087 " C library implementations.\n" \ 5088 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \ 5089 " to elevate privileges using set*uid|gid system calls.\n" \ 5090 " The value 'children' will deny set*uid|gid system calls for\n" \ 5091 " main QEMU process but will allow forks and execves to run unprivileged\n" \ 5092 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \ 5093 " blocking *fork and execve\n" \ 5094 " use 'resourcecontrol' to disable process affinity and schedular priority\n", 5095 QEMU_ARCH_ALL) 5096SRST 5097``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]`` 5098 Enable Seccomp mode 2 system call filter. 'on' will enable syscall 5099 filtering and 'off' will disable it. The default is 'off'. 5100 5101 ``obsolete=string`` 5102 Enable Obsolete system calls 5103 5104 ``elevateprivileges=string`` 5105 Disable set\*uid\|gid system calls 5106 5107 ``spawn=string`` 5108 Disable \*fork and execve 5109 5110 ``resourcecontrol=string`` 5111 Disable process affinity and schedular priority 5112ERST 5113 5114DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig, 5115 "-readconfig <file>\n" 5116 " read config file\n", QEMU_ARCH_ALL) 5117SRST 5118``-readconfig file`` 5119 Read device configuration from file. This approach is useful when 5120 you want to spawn QEMU process with many command line options but 5121 you don't want to exceed the command line character limit. 5122ERST 5123 5124DEF("no-user-config", 0, QEMU_OPTION_nouserconfig, 5125 "-no-user-config\n" 5126 " do not load default user-provided config files at startup\n", 5127 QEMU_ARCH_ALL) 5128SRST 5129``-no-user-config`` 5130 The ``-no-user-config`` option makes QEMU not load any of the 5131 user-provided config files on sysconfdir. 5132ERST 5133 5134DEF("trace", HAS_ARG, QEMU_OPTION_trace, 5135 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n" 5136 " specify tracing options\n", 5137 QEMU_ARCH_ALL) 5138SRST 5139``-trace [[enable=]pattern][,events=file][,file=file]`` 5140 .. include:: ../qemu-option-trace.rst.inc 5141 5142ERST 5143DEF("plugin", HAS_ARG, QEMU_OPTION_plugin, 5144 "-plugin [file=]<file>[,<argname>=<argvalue>]\n" 5145 " load a plugin\n", 5146 QEMU_ARCH_ALL) 5147SRST 5148``-plugin file=file[,argname=argvalue]`` 5149 Load a plugin. 5150 5151 ``file=file`` 5152 Load the given plugin from a shared library file. 5153 5154 ``argname=argvalue`` 5155 Argument passed to the plugin. (Can be given multiple times.) 5156ERST 5157 5158HXCOMM Internal use 5159DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL) 5160DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL) 5161 5162#ifdef CONFIG_POSIX 5163DEF("run-with", HAS_ARG, QEMU_OPTION_run_with, 5164 "-run-with [async-teardown=on|off][,chroot=dir][user=username|uid:gid]\n" 5165 " Set miscellaneous QEMU process lifecycle options:\n" 5166 " async-teardown=on enables asynchronous teardown (Linux only)\n" 5167 " chroot=dir chroot to dir just before starting the VM\n" 5168 " user=username switch to the specified user before starting the VM\n" 5169 " user=uid:gid ditto, but use specified user-ID and group-ID instead\n", 5170 QEMU_ARCH_ALL) 5171SRST 5172``-run-with [async-teardown=on|off][,chroot=dir][user=username|uid:gid]`` 5173 Set QEMU process lifecycle options. 5174 5175 ``async-teardown=on`` enables asynchronous teardown. A new process called 5176 "cleanup/<QEMU_PID>" will be created at startup sharing the address 5177 space with the main QEMU process, using clone. It will wait for the 5178 main QEMU process to terminate completely, and then exit. This allows 5179 QEMU to terminate very quickly even if the guest was huge, leaving the 5180 teardown of the address space to the cleanup process. Since the cleanup 5181 process shares the same cgroups as the main QEMU process, accounting is 5182 performed correctly. This only works if the cleanup process is not 5183 forcefully killed with SIGKILL before the main QEMU process has 5184 terminated completely. 5185 5186 ``chroot=dir`` can be used for doing a chroot to the specified directory 5187 immediately before starting the guest execution. This is especially useful 5188 in combination with -runas. 5189 5190 ``user=username`` or ``user=uid:gid`` can be used to drop root privileges 5191 before starting guest execution. QEMU will use the ``setuid`` and ``setgid`` 5192 system calls to switch to the specified identity. Note that the 5193 ``user=username`` syntax will also apply the full set of supplementary 5194 groups for the user, whereas the ``user=uid:gid`` will use only the 5195 ``gid`` group. 5196ERST 5197#endif 5198 5199DEF("msg", HAS_ARG, QEMU_OPTION_msg, 5200 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n" 5201 " control error message format\n" 5202 " timestamp=on enables timestamps (default: off)\n" 5203 " guest-name=on enables guest name prefix but only if\n" 5204 " -name guest option is set (default: off)\n", 5205 QEMU_ARCH_ALL) 5206SRST 5207``-msg [timestamp[=on|off]][,guest-name[=on|off]]`` 5208 Control error message format. 5209 5210 ``timestamp=on|off`` 5211 Prefix messages with a timestamp. Default is off. 5212 5213 ``guest-name=on|off`` 5214 Prefix messages with guest name but only if -name guest option is set 5215 otherwise the option is ignored. Default is off. 5216ERST 5217 5218DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate, 5219 "-dump-vmstate <file>\n" 5220 " Output vmstate information in JSON format to file.\n" 5221 " Use the scripts/vmstate-static-checker.py file to\n" 5222 " check for possible regressions in migration code\n" 5223 " by comparing two such vmstate dumps.\n", 5224 QEMU_ARCH_ALL) 5225SRST 5226``-dump-vmstate file`` 5227 Dump json-encoded vmstate information for current machine type to 5228 file in file 5229ERST 5230 5231DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile, 5232 "-enable-sync-profile\n" 5233 " enable synchronization profiling\n", 5234 QEMU_ARCH_ALL) 5235SRST 5236``-enable-sync-profile`` 5237 Enable synchronization profiling. 5238ERST 5239 5240#if defined(CONFIG_TCG) && defined(CONFIG_LINUX) 5241DEF("perfmap", 0, QEMU_OPTION_perfmap, 5242 "-perfmap generate a /tmp/perf-${pid}.map file for perf\n", 5243 QEMU_ARCH_ALL) 5244SRST 5245``-perfmap`` 5246 Generate a map file for Linux perf tools that will allow basic profiling 5247 information to be broken down into basic blocks. 5248ERST 5249 5250DEF("jitdump", 0, QEMU_OPTION_jitdump, 5251 "-jitdump generate a jit-${pid}.dump file for perf\n", 5252 QEMU_ARCH_ALL) 5253SRST 5254``-jitdump`` 5255 Generate a dump file for Linux perf tools that maps basic blocks to symbol 5256 names, line numbers and JITted code. 5257ERST 5258#endif 5259 5260DEFHEADING() 5261 5262DEFHEADING(Generic object creation:) 5263 5264DEF("object", HAS_ARG, QEMU_OPTION_object, 5265 "-object TYPENAME[,PROP1=VALUE1,...]\n" 5266 " create a new object of type TYPENAME setting properties\n" 5267 " in the order they are specified. Note that the 'id'\n" 5268 " property must be set. These objects are placed in the\n" 5269 " '/objects' path.\n", 5270 QEMU_ARCH_ALL) 5271SRST 5272``-object typename[,prop1=value1,...]`` 5273 Create a new object of type typename setting properties in the order 5274 they are specified. Note that the 'id' property must be set. These 5275 objects are placed in the '/objects' path. 5276 5277 ``-object memory-backend-file,id=id,size=size,mem-path=dir,share=on|off,discard-data=on|off,merge=on|off,dump=on|off,prealloc=on|off,host-nodes=host-nodes,policy=default|preferred|bind|interleave,align=align,offset=offset,readonly=on|off,rom=on|off|auto`` 5278 Creates a memory file backend object, which can be used to back 5279 the guest RAM with huge pages. 5280 5281 The ``id`` parameter is a unique ID that will be used to 5282 reference this memory region in other parameters, e.g. ``-numa``, 5283 ``-device nvdimm``, etc. 5284 5285 The ``size`` option provides the size of the memory region, and 5286 accepts common suffixes, e.g. ``500M``. 5287 5288 The ``mem-path`` provides the path to either a shared memory or 5289 huge page filesystem mount. 5290 5291 The ``share`` boolean option determines whether the memory 5292 region is marked as private to QEMU, or shared. The latter 5293 allows a co-operating external process to access the QEMU memory 5294 region. 5295 5296 Setting share=on might affect the ability to configure NUMA 5297 bindings for the memory backend under some circumstances, see 5298 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel 5299 source tree for additional details. 5300 5301 Setting the ``discard-data`` boolean option to on indicates that 5302 file contents can be destroyed when QEMU exits, to avoid 5303 unnecessarily flushing data to the backing file. Note that 5304 ``discard-data`` is only an optimization, and QEMU might not 5305 discard file contents if it aborts unexpectedly or is terminated 5306 using SIGKILL. 5307 5308 The ``merge`` boolean option enables memory merge, also known as 5309 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider 5310 the pages for memory deduplication. 5311 5312 Setting the ``dump`` boolean option to off excludes the memory 5313 from core dumps. This feature is also known as MADV\_DONTDUMP. 5314 5315 The ``prealloc`` boolean option enables memory preallocation. 5316 5317 The ``host-nodes`` option binds the memory range to a list of 5318 NUMA host nodes. 5319 5320 The ``policy`` option sets the NUMA policy to one of the 5321 following values: 5322 5323 ``default`` 5324 default host policy 5325 5326 ``preferred`` 5327 prefer the given host node list for allocation 5328 5329 ``bind`` 5330 restrict memory allocation to the given host node list 5331 5332 ``interleave`` 5333 interleave memory allocations across the given host node 5334 list 5335 5336 The ``align`` option specifies the base address alignment when 5337 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg 5338 ``2M``. Some backend store specified by ``mem-path`` requires an 5339 alignment different than the default one used by QEMU, eg the 5340 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In 5341 such cases, users can specify the required alignment via this 5342 option. 5343 5344 The ``offset`` option specifies the offset into the target file 5345 that the region starts at. You can use this parameter to back 5346 multiple regions with a single file. 5347 5348 The ``pmem`` option specifies whether the backing file specified 5349 by ``mem-path`` is in host persistent memory that can be 5350 accessed using the SNIA NVM programming model (e.g. Intel 5351 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary 5352 operations to guarantee the persistence of its own writes to 5353 ``mem-path`` (e.g. in vNVDIMM label emulation and live 5354 migration). Also, we will map the backend-file with MAP\_SYNC 5355 flag, which ensures the file metadata is in sync for 5356 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC 5357 requires support from both the host kernel (since Linux kernel 5358 4.15) and the filesystem of ``mem-path`` mounted with DAX 5359 option. 5360 5361 The ``readonly`` option specifies whether the backing file is opened 5362 read-only or read-write (default). 5363 5364 The ``rom`` option specifies whether to create Read Only Memory 5365 (ROM) that cannot be modified by the VM. Any write attempts to such 5366 ROM will be denied. Most use cases want proper RAM instead of ROM. 5367 However, selected use cases, like R/O NVDIMMs, can benefit from 5368 ROM. If set to ``on``, create ROM; if set to ``off``, create 5369 writable RAM; if set to ``auto`` (default), the value of the 5370 ``readonly`` option is used. This option is primarily helpful when 5371 we want to have writable RAM in configurations that would 5372 traditionally create ROM before the ``rom`` option was introduced: 5373 VM templating, where we want to open a file readonly 5374 (``readonly=on``) and mark the memory to be private for QEMU 5375 (``share=off``). For this use case, we need writable RAM instead 5376 of ROM, and want to also set ``rom=off``. 5377 5378 ``-object memory-backend-ram,id=id,merge=on|off,dump=on|off,share=on|off,prealloc=on|off,size=size,host-nodes=host-nodes,policy=default|preferred|bind|interleave`` 5379 Creates a memory backend object, which can be used to back the 5380 guest RAM. Memory backend objects offer more control than the 5381 ``-m`` option that is traditionally used to define guest RAM. 5382 Please refer to ``memory-backend-file`` for a description of the 5383 options. 5384 5385 ``-object memory-backend-memfd,id=id,merge=on|off,dump=on|off,share=on|off,prealloc=on|off,size=size,host-nodes=host-nodes,policy=default|preferred|bind|interleave,seal=on|off,hugetlb=on|off,hugetlbsize=size`` 5386 Creates an anonymous memory file backend object, which allows 5387 QEMU to share the memory with an external process (e.g. when 5388 using vhost-user). The memory is allocated with memfd and 5389 optional sealing. (Linux only) 5390 5391 The ``seal`` option creates a sealed-file, that will block 5392 further resizing the memory ('on' by default). 5393 5394 The ``hugetlb`` option specify the file to be created resides in 5395 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction 5396 with the ``hugetlb`` option, the ``hugetlbsize`` option specify 5397 the hugetlb page size on systems that support multiple hugetlb 5398 page sizes (it must be a power of 2 value supported by the 5399 system). 5400 5401 In some versions of Linux, the ``hugetlb`` option is 5402 incompatible with the ``seal`` option (requires at least Linux 5403 4.16). 5404 5405 Please refer to ``memory-backend-file`` for a description of the 5406 other options. 5407 5408 The ``share`` boolean option is on by default with memfd. 5409 5410 ``-object memory-backend-shm,id=id,merge=on|off,dump=on|off,share=on|off,prealloc=on|off,size=size,host-nodes=host-nodes,policy=default|preferred|bind|interleave`` 5411 Creates a POSIX shared memory backend object, which allows 5412 QEMU to share the memory with an external process (e.g. when 5413 using vhost-user). 5414 5415 ``memory-backend-shm`` is a more portable and less featureful version 5416 of ``memory-backend-memfd``. It can then be used in any POSIX system, 5417 especially when memfd is not supported. 5418 5419 Please refer to ``memory-backend-file`` for a description of the 5420 options. 5421 5422 The ``share`` boolean option is on by default with shm. Setting it to 5423 off will cause a failure during allocation because it is not supported 5424 by this backend. 5425 5426 ``-object iommufd,id=id[,fd=fd]`` 5427 Creates an iommufd backend which allows control of DMA mapping 5428 through the ``/dev/iommu`` device. 5429 5430 The ``id`` parameter is a unique ID which frontends (such as 5431 vfio-pci of vdpa) will use to connect with the iommufd backend. 5432 5433 The ``fd`` parameter is an optional pre-opened file descriptor 5434 resulting from ``/dev/iommu`` opening. Usually the iommufd is shared 5435 across all subsystems, bringing the benefit of centralized 5436 reference counting. 5437 5438 ``-object rng-builtin,id=id`` 5439 Creates a random number generator backend which obtains entropy 5440 from QEMU builtin functions. The ``id`` parameter is a unique ID 5441 that will be used to reference this entropy backend from the 5442 ``virtio-rng`` device. By default, the ``virtio-rng`` device 5443 uses this RNG backend. 5444 5445 ``-object rng-random,id=id,filename=/dev/random`` 5446 Creates a random number generator backend which obtains entropy 5447 from a device on the host. The ``id`` parameter is a unique ID 5448 that will be used to reference this entropy backend from the 5449 ``virtio-rng`` device. The ``filename`` parameter specifies 5450 which file to obtain entropy from and if omitted defaults to 5451 ``/dev/urandom``. 5452 5453 ``-object rng-egd,id=id,chardev=chardevid`` 5454 Creates a random number generator backend which obtains entropy 5455 from an external daemon running on the host. The ``id`` 5456 parameter is a unique ID that will be used to reference this 5457 entropy backend from the ``virtio-rng`` device. The ``chardev`` 5458 parameter is the unique ID of a character device backend that 5459 provides the connection to the RNG daemon. 5460 5461 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off`` 5462 Creates a TLS anonymous credentials object, which can be used to 5463 provide TLS support on network backends. The ``id`` parameter is 5464 a unique ID which network backends will use to access the 5465 credentials. The ``endpoint`` is either ``server`` or ``client`` 5466 depending on whether the QEMU network backend that uses the 5467 credentials will be acting as a client or as a server. If 5468 ``verify-peer`` is enabled (the default) then once the handshake 5469 is completed, the peer credentials will be verified, though this 5470 is a no-op for anonymous credentials. 5471 5472 The dir parameter tells QEMU where to find the credential files. 5473 For server endpoints, this directory may contain a file 5474 dh-params.pem providing diffie-hellman parameters to use for the 5475 TLS server. If the file is missing, QEMU will generate a set of 5476 DH parameters at startup. This is a computationally expensive 5477 operation that consumes random pool entropy, so it is 5478 recommended that a persistent set of parameters be generated 5479 upfront and saved. 5480 5481 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]`` 5482 Creates a TLS Pre-Shared Keys (PSK) credentials object, which 5483 can be used to provide TLS support on network backends. The 5484 ``id`` parameter is a unique ID which network backends will use 5485 to access the credentials. The ``endpoint`` is either ``server`` 5486 or ``client`` depending on whether the QEMU network backend that 5487 uses the credentials will be acting as a client or as a server. 5488 For clients only, ``username`` is the username which will be 5489 sent to the server. If omitted it defaults to "qemu". 5490 5491 The dir parameter tells QEMU where to find the keys file. It is 5492 called "dir/keys.psk" and contains "username:key" pairs. This 5493 file can most easily be created using the GnuTLS ``psktool`` 5494 program. 5495 5496 For server endpoints, dir may also contain a file dh-params.pem 5497 providing diffie-hellman parameters to use for the TLS server. 5498 If the file is missing, QEMU will generate a set of DH 5499 parameters at startup. This is a computationally expensive 5500 operation that consumes random pool entropy, so it is 5501 recommended that a persistent set of parameters be generated up 5502 front and saved. 5503 5504 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id`` 5505 Creates a TLS anonymous credentials object, which can be used to 5506 provide TLS support on network backends. The ``id`` parameter is 5507 a unique ID which network backends will use to access the 5508 credentials. The ``endpoint`` is either ``server`` or ``client`` 5509 depending on whether the QEMU network backend that uses the 5510 credentials will be acting as a client or as a server. If 5511 ``verify-peer`` is enabled (the default) then once the handshake 5512 is completed, the peer credentials will be verified. With x509 5513 certificates, this implies that the clients must be provided 5514 with valid client certificates too. 5515 5516 The dir parameter tells QEMU where to find the credential files. 5517 For server endpoints, this directory may contain a file 5518 dh-params.pem providing diffie-hellman parameters to use for the 5519 TLS server. If the file is missing, QEMU will generate a set of 5520 DH parameters at startup. This is a computationally expensive 5521 operation that consumes random pool entropy, so it is 5522 recommended that a persistent set of parameters be generated 5523 upfront and saved. 5524 5525 For x509 certificate credentials the directory will contain 5526 further files providing the x509 certificates. The certificates 5527 must be stored in PEM format, in filenames ca-cert.pem, 5528 ca-crl.pem (optional), server-cert.pem (only servers), 5529 server-key.pem (only servers), client-cert.pem (only clients), 5530 and client-key.pem (only clients). 5531 5532 For the server-key.pem and client-key.pem files which contain 5533 sensitive private keys, it is possible to use an encrypted 5534 version by providing the passwordid parameter. This provides the 5535 ID of a previously created ``secret`` object containing the 5536 password for decryption. 5537 5538 The priority parameter allows to override the global default 5539 priority used by gnutls. This can be useful if the system 5540 administrator needs to use a weaker set of crypto priorities for 5541 QEMU without potentially forcing the weakness onto all 5542 applications. Or conversely if one wants wants a stronger 5543 default for QEMU than for all other applications, they can do 5544 this through this parameter. Its format is a gnutls priority 5545 string as described at 5546 https://gnutls.org/manual/html_node/Priority-Strings.html. 5547 5548 ``-object tls-cipher-suites,id=id,priority=priority`` 5549 Creates a TLS cipher suites object, which can be used to control 5550 the TLS cipher/protocol algorithms that applications are permitted 5551 to use. 5552 5553 The ``id`` parameter is a unique ID which frontends will use to 5554 access the ordered list of permitted TLS cipher suites from the 5555 host. 5556 5557 The ``priority`` parameter allows to override the global default 5558 priority used by gnutls. This can be useful if the system 5559 administrator needs to use a weaker set of crypto priorities for 5560 QEMU without potentially forcing the weakness onto all 5561 applications. Or conversely if one wants wants a stronger 5562 default for QEMU than for all other applications, they can do 5563 this through this parameter. Its format is a gnutls priority 5564 string as described at 5565 https://gnutls.org/manual/html_node/Priority-Strings.html. 5566 5567 An example of use of this object is to control UEFI HTTPS Boot. 5568 The tls-cipher-suites object exposes the ordered list of permitted 5569 TLS cipher suites from the host side to the guest firmware, via 5570 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER 5571 objects. The firmware uses the IANA_TLS_CIPHER array for configuring 5572 guest-side TLS. 5573 5574 In the following example, the priority at which the host-side policy 5575 is retrieved is given by the ``priority`` property. 5576 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to 5577 refer to /etc/crypto-policies/back-ends/gnutls.config. 5578 5579 .. parsed-literal:: 5580 5581 # |qemu_system| \\ 5582 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\ 5583 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0 5584 5585 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]`` 5586 Interval t can't be 0, this filter batches the packet delivery: 5587 all packets arriving in a given interval on netdev netdevid are 5588 delayed until the end of the interval. Interval is in 5589 microseconds. ``status`` is optional that indicate whether the 5590 netfilter is on (enabled) or off (disabled), the default status 5591 for netfilter will be 'on'. 5592 5593 queue all\|rx\|tx is an option that can be applied to any 5594 netfilter. 5595 5596 ``all``: the filter is attached both to the receive and the 5597 transmit queue of the netdev (default). 5598 5599 ``rx``: the filter is attached to the receive queue of the 5600 netdev, where it will receive packets sent to the netdev. 5601 5602 ``tx``: the filter is attached to the transmit queue of the 5603 netdev, where it will receive packets sent by the netdev. 5604 5605 position head\|tail\|id=<id> is an option to specify where the 5606 filter should be inserted in the filter list. It can be applied 5607 to any netfilter. 5608 5609 ``head``: the filter is inserted at the head of the filter list, 5610 before any existing filters. 5611 5612 ``tail``: the filter is inserted at the tail of the filter list, 5613 behind any existing filters (default). 5614 5615 ``id=<id>``: the filter is inserted before or behind the filter 5616 specified by <id>, see the insert option below. 5617 5618 insert behind\|before is an option to specify where to insert 5619 the new filter relative to the one specified with 5620 position=id=<id>. It can be applied to any netfilter. 5621 5622 ``before``: insert before the specified filter. 5623 5624 ``behind``: insert behind the specified filter (default). 5625 5626 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5627 filter-mirror on netdev netdevid,mirror net packet to 5628 chardevchardevid, if it has the vnet\_hdr\_support flag, 5629 filter-mirror will mirror packet with vnet\_hdr\_len. 5630 5631 ``-object filter-redirector,id=id,netdev=netdevid,indev=chardevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5632 filter-redirector on netdev netdevid,redirect filter's net 5633 packet to chardev chardevid,and redirect indev's packet to 5634 filter.if it has the vnet\_hdr\_support flag, filter-redirector 5635 will redirect packet with vnet\_hdr\_len. Create a 5636 filter-redirector we need to differ outdev id from indev id, id 5637 can not be the same. we can just use indev or outdev, but at 5638 least one of indev or outdev need to be specified. 5639 5640 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5641 Filter-rewriter is a part of COLO project.It will rewrite tcp 5642 packet to secondary from primary to keep secondary tcp 5643 connection,and rewrite tcp packet to primary from secondary make 5644 tcp packet can be handled by client.if it has the 5645 vnet\_hdr\_support flag, we can parse packet with vnet header. 5646 5647 usage: colo secondary: -object 5648 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object 5649 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object 5650 filter-rewriter,id=rew0,netdev=hn0,queue=all 5651 5652 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]`` 5653 Dump the network traffic on netdev dev to the file specified by 5654 filename. At most len bytes (64k by default) per packet are 5655 stored. The file format is libpcap, so it can be analyzed with 5656 tools such as tcpdump or Wireshark. 5657 5658 ``-object colo-compare,id=id,primary_in=chardevid,secondary_in=chardevid,outdev=chardevid,iothread=id[,vnet_hdr_support][,notify_dev=id][,compare_timeout=@var{ms}][,expired_scan_cycle=@var{ms}][,max_queue_size=@var{size}]`` 5659 Colo-compare gets packet from primary\_in chardevid and 5660 secondary\_in, then compare whether the payload of primary packet 5661 and secondary packet are the same. If same, it will output 5662 primary packet to out\_dev, else it will notify COLO-framework to do 5663 checkpoint and send primary packet to out\_dev. In order to 5664 improve efficiency, we need to put the task of comparison in 5665 another iothread. If it has the vnet\_hdr\_support flag, 5666 colo compare will send/recv packet with vnet\_hdr\_len. 5667 The compare\_timeout=@var{ms} determines the maximum time of the 5668 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms} 5669 is to set the period of scanning expired primary node network packets. 5670 The max\_queue\_size=@var{size} is to set the max compare queue 5671 size depend on user environment. 5672 If user want to use Xen COLO, need to add the notify\_dev to 5673 notify Xen colo-frame to do checkpoint. 5674 5675 COLO-compare must be used with the help of filter-mirror, 5676 filter-redirector and filter-rewriter. 5677 5678 :: 5679 5680 KVM COLO 5681 5682 primary: 5683 -netdev tap,id=hn0,vhost=off 5684 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5685 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5686 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5687 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5688 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5689 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5690 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5691 -object iothread,id=iothread1 5692 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5693 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5694 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5695 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1 5696 5697 secondary: 5698 -netdev tap,id=hn0,vhost=off 5699 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5700 -chardev socket,id=red0,host=3.3.3.3,port=9003 5701 -chardev socket,id=red1,host=3.3.3.3,port=9004 5702 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5703 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5704 5705 5706 Xen COLO 5707 5708 primary: 5709 -netdev tap,id=hn0,vhost=off 5710 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5711 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5712 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5713 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5714 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5715 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5716 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5717 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off 5718 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5719 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5720 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5721 -object iothread,id=iothread1 5722 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=notify_way,iothread=iothread1 5723 5724 secondary: 5725 -netdev tap,id=hn0,vhost=off 5726 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5727 -chardev socket,id=red0,host=3.3.3.3,port=9003 5728 -chardev socket,id=red1,host=3.3.3.3,port=9004 5729 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5730 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5731 5732 If you want to know the detail of above command line, you can 5733 read the colo-compare git log. 5734 5735 ``-object cryptodev-backend-builtin,id=id[,queues=queues]`` 5736 Creates a cryptodev backend which executes crypto operations from 5737 the QEMU cipher APIs. The id parameter is a unique ID that will 5738 be used to reference this cryptodev backend from the 5739 ``virtio-crypto`` device. The queues parameter is optional, 5740 which specify the queue number of cryptodev backend, the default 5741 of queues is 1. 5742 5743 .. parsed-literal:: 5744 5745 # |qemu_system| \\ 5746 [...] \\ 5747 -object cryptodev-backend-builtin,id=cryptodev0 \\ 5748 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5749 [...] 5750 5751 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]`` 5752 Creates a vhost-user cryptodev backend, backed by a chardev 5753 chardevid. The id parameter is a unique ID that will be used to 5754 reference this cryptodev backend from the ``virtio-crypto`` 5755 device. The chardev should be a unix domain socket backed one. 5756 The vhost-user uses a specifically defined protocol to pass 5757 vhost ioctl replacement messages to an application on the other 5758 end of the socket. The queues parameter is optional, which 5759 specify the queue number of cryptodev backend for multiqueue 5760 vhost-user, the default of queues is 1. 5761 5762 .. parsed-literal:: 5763 5764 # |qemu_system| \\ 5765 [...] \\ 5766 -chardev socket,id=chardev0,path=/path/to/socket \\ 5767 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\ 5768 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5769 [...] 5770 5771 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]`` 5772 \ 5773 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]`` 5774 Defines a secret to store a password, encryption key, or some 5775 other sensitive data. The sensitive data can either be passed 5776 directly via the data parameter, or indirectly via the file 5777 parameter. Using the data parameter is insecure unless the 5778 sensitive data is encrypted. 5779 5780 The sensitive data can be provided in raw format (the default), 5781 or base64. When encoded as JSON, the raw format only supports 5782 valid UTF-8 characters, so base64 is recommended for sending 5783 binary data. QEMU will convert from which ever format is 5784 provided to the format it needs internally. eg, an RBD password 5785 can be provided in raw format, even though it will be base64 5786 encoded when passed onto the RBD sever. 5787 5788 For added protection, it is possible to encrypt the data 5789 associated with a secret using the AES-256-CBC cipher. Use of 5790 encryption is indicated by providing the keyid and iv 5791 parameters. The keyid parameter provides the ID of a previously 5792 defined secret that contains the AES-256 decryption key. This 5793 key should be 32-bytes long and be base64 encoded. The iv 5794 parameter provides the random initialization vector used for 5795 encryption of this particular secret and should be a base64 5796 encrypted string of the 16-byte IV. 5797 5798 The simplest (insecure) usage is to provide the secret inline 5799 5800 .. parsed-literal:: 5801 5802 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw 5803 5804 The simplest secure usage is to provide the secret via a file 5805 5806 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object 5807 secret,id=sec0,file=mypasswd.txt,format=raw 5808 5809 For greater security, AES-256-CBC should be used. To illustrate 5810 usage, consider the openssl command line tool which can encrypt 5811 the data. Note that when encrypting, the plaintext must be 5812 padded to the cipher block size (32 bytes) using the standard 5813 PKCS#5/6 compatible padding algorithm. 5814 5815 First a master key needs to be created in base64 encoding: 5816 5817 :: 5818 5819 # openssl rand -base64 32 > key.b64 5820 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"') 5821 5822 Each secret to be encrypted needs to have a random 5823 initialization vector generated. These do not need to be kept 5824 secret 5825 5826 :: 5827 5828 # openssl rand -base64 16 > iv.b64 5829 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"') 5830 5831 The secret to be defined can now be encrypted, in this case 5832 we're telling openssl to base64 encode the result, but it could 5833 be left as raw bytes if desired. 5834 5835 :: 5836 5837 # SECRET=$(printf "letmein" | 5838 openssl enc -aes-256-cbc -a -K $KEY -iv $IV) 5839 5840 When launching QEMU, create a master secret pointing to 5841 ``key.b64`` and specify that to be used to decrypt the user 5842 password. Pass the contents of ``iv.b64`` to the second secret 5843 5844 .. parsed-literal:: 5845 5846 # |qemu_system| \\ 5847 -object secret,id=secmaster0,format=base64,file=key.b64 \\ 5848 -object secret,id=sec0,keyid=secmaster0,format=base64,\\ 5849 data=$SECRET,iv=$(<iv.b64) 5850 5851 ``-object sev-guest,id=id,cbitpos=cbitpos,reduced-phys-bits=val,[sev-device=string,policy=policy,handle=handle,dh-cert-file=file,session-file=file,kernel-hashes=on|off]`` 5852 Create a Secure Encrypted Virtualization (SEV) guest object, 5853 which can be used to provide the guest memory encryption support 5854 on AMD processors. 5855 5856 When memory encryption is enabled, one of the physical address 5857 bit (aka the C-bit) is utilized to mark if a memory page is 5858 protected. The ``cbitpos`` is used to provide the C-bit 5859 position. The C-bit position is Host family dependent hence user 5860 must provide this value. On EPYC, the value should be 47. 5861 5862 When memory encryption is enabled, we loose certain bits in 5863 physical address space. The ``reduced-phys-bits`` is used to 5864 provide the number of bits we loose in physical address space. 5865 Similar to C-bit, the value is Host family dependent. On EPYC, 5866 a guest will lose a maximum of 1 bit, so the value should be 1. 5867 5868 The ``sev-device`` provides the device file to use for 5869 communicating with the SEV firmware running inside AMD Secure 5870 Processor. The default device is '/dev/sev'. If hardware 5871 supports memory encryption then /dev/sev devices are created by 5872 CCP driver. 5873 5874 The ``policy`` provides the guest policy to be enforced by the 5875 SEV firmware and restrict what configuration and operational 5876 commands can be performed on this guest by the hypervisor. The 5877 policy should be provided by the guest owner and is bound to the 5878 guest and cannot be changed throughout the lifetime of the 5879 guest. The default is 0. 5880 5881 If guest ``policy`` allows sharing the key with another SEV 5882 guest then ``handle`` can be use to provide handle of the guest 5883 from which to share the key. 5884 5885 The ``dh-cert-file`` and ``session-file`` provides the guest 5886 owner's Public Diffie-Hillman key defined in SEV spec. The PDH 5887 and session parameters are used for establishing a cryptographic 5888 session with the guest owner to negotiate keys used for 5889 attestation. The file must be encoded in base64. 5890 5891 The ``kernel-hashes`` adds the hashes of given kernel/initrd/ 5892 cmdline to a designated guest firmware page for measured Linux 5893 boot with -kernel. The default is off. (Since 6.2) 5894 5895 e.g to launch a SEV guest 5896 5897 .. parsed-literal:: 5898 5899 # |qemu_system_x86| \\ 5900 ...... \\ 5901 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=1 \\ 5902 -machine ...,memory-encryption=sev0 \\ 5903 ..... 5904 5905 ``-object authz-simple,id=id,identity=string`` 5906 Create an authorization object that will control access to 5907 network services. 5908 5909 The ``identity`` parameter is identifies the user and its format 5910 depends on the network service that authorization object is 5911 associated with. For authorizing based on TLS x509 certificates, 5912 the identity must be the x509 distinguished name. Note that care 5913 must be taken to escape any commas in the distinguished name. 5914 5915 An example authorization object to validate a x509 distinguished 5916 name would look like: 5917 5918 .. parsed-literal:: 5919 5920 # |qemu_system| \\ 5921 ... \\ 5922 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\ 5923 ... 5924 5925 Note the use of quotes due to the x509 distinguished name 5926 containing whitespace, and escaping of ','. 5927 5928 ``-object authz-listfile,id=id,filename=path,refresh=on|off`` 5929 Create an authorization object that will control access to 5930 network services. 5931 5932 The ``filename`` parameter is the fully qualified path to a file 5933 containing the access control list rules in JSON format. 5934 5935 An example set of rules that match against SASL usernames might 5936 look like: 5937 5938 :: 5939 5940 { 5941 "rules": [ 5942 { "match": "fred", "policy": "allow", "format": "exact" }, 5943 { "match": "bob", "policy": "allow", "format": "exact" }, 5944 { "match": "danb", "policy": "deny", "format": "glob" }, 5945 { "match": "dan*", "policy": "allow", "format": "exact" }, 5946 ], 5947 "policy": "deny" 5948 } 5949 5950 When checking access the object will iterate over all the rules 5951 and the first rule to match will have its ``policy`` value 5952 returned as the result. If no rules match, then the default 5953 ``policy`` value is returned. 5954 5955 The rules can either be an exact string match, or they can use 5956 the simple UNIX glob pattern matching to allow wildcards to be 5957 used. 5958 5959 If ``refresh`` is set to true the file will be monitored and 5960 automatically reloaded whenever its content changes. 5961 5962 As with the ``authz-simple`` object, the format of the identity 5963 strings being matched depends on the network service, but is 5964 usually a TLS x509 distinguished name, or a SASL username. 5965 5966 An example authorization object to validate a SASL username 5967 would look like: 5968 5969 .. parsed-literal:: 5970 5971 # |qemu_system| \\ 5972 ... \\ 5973 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\ 5974 ... 5975 5976 ``-object authz-pam,id=id,service=string`` 5977 Create an authorization object that will control access to 5978 network services. 5979 5980 The ``service`` parameter provides the name of a PAM service to 5981 use for authorization. It requires that a file 5982 ``/etc/pam.d/service`` exist to provide the configuration for 5983 the ``account`` subsystem. 5984 5985 An example authorization object to validate a TLS x509 5986 distinguished name would look like: 5987 5988 .. parsed-literal:: 5989 5990 # |qemu_system| \\ 5991 ... \\ 5992 -object authz-pam,id=auth0,service=qemu-vnc \\ 5993 ... 5994 5995 There would then be a corresponding config file for PAM at 5996 ``/etc/pam.d/qemu-vnc`` that contains: 5997 5998 :: 5999 6000 account requisite pam_listfile.so item=user sense=allow \ 6001 file=/etc/qemu/vnc.allow 6002 6003 Finally the ``/etc/qemu/vnc.allow`` file would contain the list 6004 of x509 distinguished names that are permitted access 6005 6006 :: 6007 6008 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB 6009 6010 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch`` 6011 Creates a dedicated event loop thread that devices can be 6012 assigned to. This is known as an IOThread. By default device 6013 emulation happens in vCPU threads or the main event loop thread. 6014 This can become a scalability bottleneck. IOThreads allow device 6015 emulation and I/O to run on other host CPUs. 6016 6017 The ``id`` parameter is a unique ID that will be used to 6018 reference this IOThread from ``-device ...,iothread=id``. 6019 Multiple devices can be assigned to an IOThread. Note that not 6020 all devices support an ``iothread`` parameter. 6021 6022 The ``query-iothreads`` QMP command lists IOThreads and reports 6023 their thread IDs so that the user can configure host CPU 6024 pinning/affinity. 6025 6026 IOThreads use an adaptive polling algorithm to reduce event loop 6027 latency. Instead of entering a blocking system call to monitor 6028 file descriptors and then pay the cost of being woken up when an 6029 event occurs, the polling algorithm spins waiting for events for 6030 a short time. The algorithm's default parameters are suitable 6031 for many cases but can be adjusted based on knowledge of the 6032 workload and/or host device latency. 6033 6034 The ``poll-max-ns`` parameter is the maximum number of 6035 nanoseconds to busy wait for events. Polling can be disabled by 6036 setting this value to 0. 6037 6038 The ``poll-grow`` parameter is the multiplier used to increase 6039 the polling time when the algorithm detects it is missing events 6040 due to not polling long enough. 6041 6042 The ``poll-shrink`` parameter is the divisor used to decrease 6043 the polling time when the algorithm detects it is spending too 6044 long polling without encountering events. 6045 6046 The ``aio-max-batch`` parameter is the maximum number of requests 6047 in a batch for the AIO engine, 0 means that the engine will use 6048 its default. 6049 6050 The IOThread parameters can be modified at run-time using the 6051 ``qom-set`` command (where ``iothread1`` is the IOThread's 6052 ``id``): 6053 6054 :: 6055 6056 (qemu) qom-set /objects/iothread1 poll-max-ns 100000 6057ERST 6058 6059 6060HXCOMM This is the last statement. Insert new options before this line! 6061 6062#undef DEF 6063#undef DEFHEADING 6064#undef ARCHHEADING 6065
