1HXCOMM Use DEFHEADING() to define headings in both help text and rST. 2HXCOMM Text between SRST and ERST is copied to the rST version and 3HXCOMM discarded from C version. 4HXCOMM DEF(option, HAS_ARG/0, opt_enum, opt_help, arch_mask) is used to 5HXCOMM construct option structures, enums and help message for specified 6HXCOMM architectures. 7HXCOMM HXCOMM can be used for comments, discarded from both rST and C. 8 9DEFHEADING(Standard options:) 10 11DEF("help", 0, QEMU_OPTION_h, 12 "-h or -help display this help and exit\n", QEMU_ARCH_ALL) 13SRST 14``-h`` 15 Display help and exit 16ERST 17 18DEF("version", 0, QEMU_OPTION_version, 19 "-version display version information and exit\n", QEMU_ARCH_ALL) 20SRST 21``-version`` 22 Display version information and exit 23ERST 24 25DEF("machine", HAS_ARG, QEMU_OPTION_machine, \ 26 "-machine [type=]name[,prop[=value][,...]]\n" 27 " selects emulated machine ('-machine help' for list)\n" 28 " property accel=accel1[:accel2[:...]] selects accelerator\n" 29 " supported accelerators are kvm, xen, hvf, nvmm, whpx or tcg (default: tcg)\n" 30 " vmport=on|off|auto controls emulation of vmport (default: auto)\n" 31 " dump-guest-core=on|off include guest memory in a core dump (default=on)\n" 32 " mem-merge=on|off controls memory merge support (default: on)\n" 33 " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n" 34 " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n" 35 " suppress-vmdesc=on|off disables self-describing migration (default=off)\n" 36 " nvdimm=on|off controls NVDIMM support (default=off)\n" 37 " memory-encryption=@var{} memory encryption object to use (default=none)\n" 38 " hmat=on|off controls ACPI HMAT support (default=off)\n" 39 " memory-backend='backend-id' specifies explicitly provided backend for main RAM (default=none)\n" 40 " cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]\n", 41 QEMU_ARCH_ALL) 42SRST 43``-machine [type=]name[,prop=value[,...]]`` 44 Select the emulated machine by name. Use ``-machine help`` to list 45 available machines. 46 47 For architectures which aim to support live migration compatibility 48 across releases, each release will introduce a new versioned machine 49 type. For example, the 2.8.0 release introduced machine types 50 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures. 51 52 To allow live migration of guests from QEMU version 2.8.0, to QEMU 53 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8" 54 and "pc-q35-2.8" machines too. To allow users live migrating VMs to 55 skip multiple intermediate releases when upgrading, new releases of 56 QEMU will support machine types from many previous versions. 57 58 Supported machine properties are: 59 60 ``accel=accels1[:accels2[:...]]`` 61 This is used to enable an accelerator. Depending on the target 62 architecture, kvm, xen, hvf, nvmm, whpx or tcg can be available. 63 By default, tcg is used. If there is more than one accelerator 64 specified, the next one is used if the previous one fails to 65 initialize. 66 67 ``vmport=on|off|auto`` 68 Enables emulation of VMWare IO port, for vmmouse etc. auto says 69 to select the value based on accel. For accel=xen the default is 70 off otherwise the default is on. 71 72 ``dump-guest-core=on|off`` 73 Include guest memory in a core dump. The default is on. 74 75 ``mem-merge=on|off`` 76 Enables or disables memory merge support. This feature, when 77 supported by the host, de-duplicates identical memory pages 78 among VMs instances (enabled by default). 79 80 ``aes-key-wrap=on|off`` 81 Enables or disables AES key wrapping support on s390-ccw hosts. 82 This feature controls whether AES wrapping keys will be created 83 to allow execution of AES cryptographic functions. The default 84 is on. 85 86 ``dea-key-wrap=on|off`` 87 Enables or disables DEA key wrapping support on s390-ccw hosts. 88 This feature controls whether DEA wrapping keys will be created 89 to allow execution of DEA cryptographic functions. The default 90 is on. 91 92 ``nvdimm=on|off`` 93 Enables or disables NVDIMM support. The default is off. 94 95 ``memory-encryption=`` 96 Memory encryption object to use. The default is none. 97 98 ``hmat=on|off`` 99 Enables or disables ACPI Heterogeneous Memory Attribute Table 100 (HMAT) support. The default is off. 101 102 ``memory-backend='id'`` 103 An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options. 104 Allows to use a memory backend as main RAM. 105 106 For example: 107 :: 108 109 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on 110 -machine memory-backend=pc.ram 111 -m 512M 112 113 Migration compatibility note: 114 115 * as backend id one shall use value of 'default-ram-id', advertised by 116 machine type (available via ``query-machines`` QMP command), if migration 117 to/from old QEMU (<5.0) is expected. 118 * for machine types 4.0 and older, user shall 119 use ``x-use-canonical-path-for-ramblock-id=off`` backend option 120 if migration to/from old QEMU (<5.0) is expected. 121 122 For example: 123 :: 124 125 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off 126 -machine memory-backend=pc.ram 127 -m 512M 128 129 ``cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]`` 130 Define a CXL Fixed Memory Window (CFMW). 131 132 Described in the CXL 2.0 ECN: CEDT CFMWS & QTG _DSM. 133 134 They are regions of Host Physical Addresses (HPA) on a system which 135 may be interleaved across one or more CXL host bridges. The system 136 software will assign particular devices into these windows and 137 configure the downstream Host-managed Device Memory (HDM) decoders 138 in root ports, switch ports and devices appropriately to meet the 139 interleave requirements before enabling the memory devices. 140 141 ``targets.X=target`` provides the mapping to CXL host bridges 142 which may be identified by the id provided in the -device entry. 143 Multiple entries are needed to specify all the targets when 144 the fixed memory window represents interleaved memory. X is the 145 target index from 0. 146 147 ``size=size`` sets the size of the CFMW. This must be a multiple of 148 256MiB. The region will be aligned to 256MiB but the location is 149 platform and configuration dependent. 150 151 ``interleave-granularity=granularity`` sets the granularity of 152 interleave. Default 256KiB. Only 256KiB, 512KiB, 1024KiB, 2048KiB 153 4096KiB, 8192KiB and 16384KiB granularities supported. 154 155 Example: 156 157 :: 158 159 -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=512k 160ERST 161 162DEF("M", HAS_ARG, QEMU_OPTION_M, 163 " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n", 164 QEMU_ARCH_ALL) 165 166SRST 167``sgx-epc.0.memdev=@var{memid},sgx-epc.0.node=@var{numaid}`` 168 Define an SGX EPC section. 169ERST 170 171DEF("cpu", HAS_ARG, QEMU_OPTION_cpu, 172 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL) 173SRST 174``-cpu model`` 175 Select CPU model (``-cpu help`` for list and additional feature 176 selection) 177ERST 178 179DEF("accel", HAS_ARG, QEMU_OPTION_accel, 180 "-accel [accel=]accelerator[,prop[=value][,...]]\n" 181 " select accelerator (kvm, xen, hvf, nvmm, whpx or tcg; use 'help' for a list)\n" 182 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n" 183 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n" 184 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n" 185 " one-insn-per-tb=on|off (one guest instruction per TCG translation block)\n" 186 " split-wx=on|off (enable TCG split w^x mapping)\n" 187 " tb-size=n (TCG translation block cache size)\n" 188 " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n" 189 " eager-split-size=n (KVM Eager Page Split chunk size, default 0, disabled. ARM only)\n" 190 " notify-vmexit=run|internal-error|disable,notify-window=n (enable notify VM exit and set notify window, x86 only)\n" 191 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL) 192SRST 193``-accel name[,prop=value[,...]]`` 194 This is used to enable an accelerator. Depending on the target 195 architecture, kvm, xen, hvf, nvmm, whpx or tcg can be available. By 196 default, tcg is used. If there is more than one accelerator 197 specified, the next one is used if the previous one fails to 198 initialize. 199 200 ``igd-passthru=on|off`` 201 When Xen is in use, this option controls whether Intel 202 integrated graphics devices can be passed through to the guest 203 (default=off) 204 205 ``kernel-irqchip=on|off|split`` 206 Controls KVM in-kernel irqchip support. The default is full 207 acceleration of the interrupt controllers. On x86, split irqchip 208 reduces the kernel attack surface, at a performance cost for 209 non-MSI interrupts. Disabling the in-kernel irqchip completely 210 is not recommended except for debugging purposes. 211 212 ``kvm-shadow-mem=size`` 213 Defines the size of the KVM shadow MMU. 214 215 ``one-insn-per-tb=on|off`` 216 Makes the TCG accelerator put only one guest instruction into 217 each translation block. This slows down emulation a lot, but 218 can be useful in some situations, such as when trying to analyse 219 the logs produced by the ``-d`` option. 220 221 ``split-wx=on|off`` 222 Controls the use of split w^x mapping for the TCG code generation 223 buffer. Some operating systems require this to be enabled, and in 224 such a case this will default on. On other operating systems, this 225 will default off, but one may enable this for testing or debugging. 226 227 ``tb-size=n`` 228 Controls the size (in MiB) of the TCG translation block cache. 229 230 ``thread=single|multi`` 231 Controls number of TCG threads. When the TCG is multi-threaded 232 there will be one thread per vCPU therefore taking advantage of 233 additional host cores. The default is to enable multi-threading 234 where both the back-end and front-ends support it and no 235 incompatible TCG features have been enabled (e.g. 236 icount/replay). 237 238 ``dirty-ring-size=n`` 239 When the KVM accelerator is used, it controls the size of the per-vCPU 240 dirty page ring buffer (number of entries for each vCPU). It should 241 be a value that is power of two, and it should be 1024 or bigger (but 242 still less than the maximum value that the kernel supports). 4096 243 could be a good initial value if you have no idea which is the best. 244 Set this value to 0 to disable the feature. By default, this feature 245 is disabled (dirty-ring-size=0). When enabled, KVM will instead 246 record dirty pages in a bitmap. 247 248 ``eager-split-size=n`` 249 KVM implements dirty page logging at the PAGE_SIZE granularity and 250 enabling dirty-logging on a huge-page requires breaking it into 251 PAGE_SIZE pages in the first place. KVM on ARM does this splitting 252 lazily by default. There are performance benefits in doing huge-page 253 split eagerly, especially in situations where TLBI costs associated 254 with break-before-make sequences are considerable and also if guest 255 workloads are read intensive. The size here specifies how many pages 256 to break at a time and needs to be a valid block size which is 257 1GB/2MB/4KB, 32MB/16KB and 512MB/64KB for 4KB/16KB/64KB PAGE_SIZE 258 respectively. Be wary of specifying a higher size as it will have an 259 impact on the memory. By default, this feature is disabled 260 (eager-split-size=0). 261 262 ``notify-vmexit=run|internal-error|disable,notify-window=n`` 263 Enables or disables notify VM exit support on x86 host and specify 264 the corresponding notify window to trigger the VM exit if enabled. 265 ``run`` option enables the feature. It does nothing and continue 266 if the exit happens. ``internal-error`` option enables the feature. 267 It raises a internal error. ``disable`` option doesn't enable the feature. 268 This feature can mitigate the CPU stuck issue due to event windows don't 269 open up for a specified of time (i.e. notify-window). 270 Default: notify-vmexit=run,notify-window=0. 271 272ERST 273 274DEF("smp", HAS_ARG, QEMU_OPTION_smp, 275 "-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]\n" 276 " set the number of initial CPUs to 'n' [default=1]\n" 277 " maxcpus= maximum number of total CPUs, including\n" 278 " offline CPUs for hotplug, etc\n" 279 " sockets= number of sockets on the machine board\n" 280 " dies= number of dies in one socket\n" 281 " clusters= number of clusters in one die\n" 282 " cores= number of cores in one cluster\n" 283 " threads= number of threads in one core\n" 284 "Note: Different machines may have different subsets of the CPU topology\n" 285 " parameters supported, so the actual meaning of the supported parameters\n" 286 " will vary accordingly. For example, for a machine type that supports a\n" 287 " three-level CPU hierarchy of sockets/cores/threads, the parameters will\n" 288 " sequentially mean as below:\n" 289 " sockets means the number of sockets on the machine board\n" 290 " cores means the number of cores in one socket\n" 291 " threads means the number of threads in one core\n" 292 " For a particular machine type board, an expected CPU topology hierarchy\n" 293 " can be defined through the supported sub-option. Unsupported parameters\n" 294 " can also be provided in addition to the sub-option, but their values\n" 295 " must be set as 1 in the purpose of correct parsing.\n", 296 QEMU_ARCH_ALL) 297SRST 298``-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]`` 299 Simulate a SMP system with '\ ``n``\ ' CPUs initially present on 300 the machine type board. On boards supporting CPU hotplug, the optional 301 '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be 302 added at runtime. When both parameters are omitted, the maximum number 303 of CPUs will be calculated from the provided topology members and the 304 initial CPU count will match the maximum number. When only one of them 305 is given then the omitted one will be set to its counterpart's value. 306 Both parameters may be specified, but the maximum number of CPUs must 307 be equal to or greater than the initial CPU count. Product of the 308 CPU topology hierarchy must be equal to the maximum number of CPUs. 309 Both parameters are subject to an upper limit that is determined by 310 the specific machine type chosen. 311 312 To control reporting of CPU topology information, values of the topology 313 parameters can be specified. Machines may only support a subset of the 314 parameters and different machines may have different subsets supported 315 which vary depending on capacity of the corresponding CPU targets. So 316 for a particular machine type board, an expected topology hierarchy can 317 be defined through the supported sub-option. Unsupported parameters can 318 also be provided in addition to the sub-option, but their values must be 319 set as 1 in the purpose of correct parsing. 320 321 Either the initial CPU count, or at least one of the topology parameters 322 must be specified. The specified parameters must be greater than zero, 323 explicit configuration like "cpus=0" is not allowed. Values for any 324 omitted parameters will be computed from those which are given. 325 326 For example, the following sub-option defines a CPU topology hierarchy 327 (2 sockets totally on the machine, 2 cores per socket, 2 threads per 328 core) for a machine that only supports sockets/cores/threads. 329 Some members of the option can be omitted but their values will be 330 automatically computed: 331 332 :: 333 334 -smp 8,sockets=2,cores=2,threads=2,maxcpus=8 335 336 The following sub-option defines a CPU topology hierarchy (2 sockets 337 totally on the machine, 2 dies per socket, 2 cores per die, 2 threads 338 per core) for PC machines which support sockets/dies/cores/threads. 339 Some members of the option can be omitted but their values will be 340 automatically computed: 341 342 :: 343 344 -smp 16,sockets=2,dies=2,cores=2,threads=2,maxcpus=16 345 346 The following sub-option defines a CPU topology hierarchy (2 sockets 347 totally on the machine, 2 clusters per socket, 2 cores per cluster, 348 2 threads per core) for ARM virt machines which support sockets/clusters 349 /cores/threads. Some members of the option can be omitted but their values 350 will be automatically computed: 351 352 :: 353 354 -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16 355 356 Historically preference was given to the coarsest topology parameters 357 when computing missing values (ie sockets preferred over cores, which 358 were preferred over threads), however, this behaviour is considered 359 liable to change. Prior to 6.2 the preference was sockets over cores 360 over threads. Since 6.2 the preference is cores over sockets over threads. 361 362 For example, the following option defines a machine board with 2 sockets 363 of 1 core before 6.2 and 1 socket of 2 cores after 6.2: 364 365 :: 366 367 -smp 2 368 369 Note: The cluster topology will only be generated in ACPI and exposed 370 to guest if it's explicitly specified in -smp. 371ERST 372 373DEF("numa", HAS_ARG, QEMU_OPTION_numa, 374 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 375 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 376 "-numa dist,src=source,dst=destination,val=distance\n" 377 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n" 378 "-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" 379 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n", 380 QEMU_ARCH_ALL) 381SRST 382``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 383 \ 384``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 385 \ 386``-numa dist,src=source,dst=destination,val=distance`` 387 \ 388``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]`` 389 \ 390``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=type[,latency=lat][,bandwidth=bw]`` 391 \ 392``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]`` 393 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA 394 distance from a source node to a destination node. Set the ACPI 395 Heterogeneous Memory Attributes for the given nodes. 396 397 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and 398 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a 399 contiguous range of CPU indexes (or a single VCPU if lastcpu is 400 omitted). A non-contiguous set of VCPUs can be represented by 401 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is 402 omitted on all nodes, VCPUs are automatically split between them. 403 404 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a 405 NUMA node: 406 407 :: 408 409 -numa node,cpus=0-2,cpus=5 410 411 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option 412 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to 413 assign CPU objects to a node using topology layout properties of 414 CPU. The set of properties is machine specific, and depends on used 415 machine type/'\ ``smp``\ ' options. It could be queried with 416 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ ' 417 property specifies node to which CPU object will be assigned, it's 418 required for node to be declared with '\ ``node``\ ' option before 419 it's used with '\ ``cpu``\ ' option. 420 421 For example: 422 423 :: 424 425 -M pc \ 426 -smp 1,sockets=2,maxcpus=2 \ 427 -numa node,nodeid=0 -numa node,nodeid=1 \ 428 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 429 430 '\ ``memdev``\ ' option assigns RAM from a given memory backend 431 device to a node. It is recommended to use '\ ``memdev``\ ' option 432 over legacy '\ ``mem``\ ' option. This is because '\ ``memdev``\ ' 433 option provides better performance and more control over the 434 backend's RAM (e.g. '\ ``prealloc``\ ' parameter of 435 '\ ``-memory-backend-ram``\ ' allows memory preallocation). 436 437 For compatibility reasons, legacy '\ ``mem``\ ' option is 438 supported in 5.0 and older machine types. Note that '\ ``mem``\ ' 439 and '\ ``memdev``\ ' are mutually exclusive. If one node uses 440 '\ ``memdev``\ ', the rest nodes have to use '\ ``memdev``\ ' 441 option, and vice versa. 442 443 Users must specify memory for all NUMA nodes by '\ ``memdev``\ ' 444 (or legacy '\ ``mem``\ ' if available). In QEMU 5.2, the support 445 for '\ ``-numa node``\ ' without memory specified was removed. 446 447 '\ ``initiator``\ ' is an additional option that points to an 448 initiator NUMA node that has best performance (the lowest latency or 449 largest bandwidth) to this NUMA node. Note that this option can be 450 set only when the machine property 'hmat' is set to 'on'. 451 452 Following example creates a machine with 2 NUMA nodes, node 0 has 453 CPU. node 1 has only memory, and its initiator is node 0. Note that 454 because node 0 has CPU, by default the initiator of node 0 is itself 455 and must be itself. 456 457 :: 458 459 -machine hmat=on \ 460 -m 2G,slots=2,maxmem=4G \ 461 -object memory-backend-ram,size=1G,id=m0 \ 462 -object memory-backend-ram,size=1G,id=m1 \ 463 -numa node,nodeid=0,memdev=m0 \ 464 -numa node,nodeid=1,memdev=m1,initiator=0 \ 465 -smp 2,sockets=2,maxcpus=2 \ 466 -numa cpu,node-id=0,socket-id=0 \ 467 -numa cpu,node-id=0,socket-id=1 468 469 source and destination are NUMA node IDs. distance is the NUMA 470 distance from source to destination. The distance from a node to 471 itself is always 10. If any pair of nodes is given a distance, then 472 all pairs must be given distances. Although, when distances are only 473 given in one direction for each pair of nodes, then the distances in 474 the opposite directions are assumed to be the same. If, however, an 475 asymmetrical pair of distances is given for even one node pair, then 476 all node pairs must be provided distance values for both directions, 477 even when they are symmetrical. When a node is unreachable from 478 another node, set the pair's distance to 255. 479 480 Note that the -``numa`` option doesn't allocate any of the specified 481 resources, it just assigns existing resources to NUMA nodes. This 482 means that one still has to use the ``-m``, ``-smp`` options to 483 allocate RAM and VCPUs respectively. 484 485 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth 486 Information between initiator and target NUMA nodes in ACPI 487 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can 488 create memory requests, usually it has one or more processors. 489 Target NUMA node contains addressable memory. 490 491 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is 492 the memory hierarchy of the target NUMA node: if hierarchy is 493 'memory', the structure represents the memory performance; if 494 hierarchy is 'first-level\|second-level\|third-level', this 495 structure represents aggregated performance of memory side caches 496 for each domain. type of 'data-type' is type of data represented by 497 this structure instance: if 'hierarchy' is 'memory', 'data-type' is 498 'access\|read\|write' latency or 'access\|read\|write' bandwidth of 499 the target memory; if 'hierarchy' is 500 'first-level\|second-level\|third-level', 'data-type' is 501 'access\|read\|write' hit latency or 'access\|read\|write' hit 502 bandwidth of the target memory side cache. 503 504 lat is latency value in nanoseconds. bw is bandwidth value, the 505 possible value and units are NUM[M\|G\|T], mean that the bandwidth 506 value are NUM byte per second (or MB/s, GB/s or TB/s depending on 507 used suffix). Note that if latency or bandwidth value is 0, means 508 the corresponding latency or bandwidth information is not provided. 509 510 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory 511 belongs. size is the size of memory side cache in bytes. level is 512 the cache level described in this structure, note that the cache 513 level 0 should not be used with '\ ``hmat-cache``\ ' option. 514 associativity is the cache associativity, the possible value is 515 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy 516 is the write policy. line is the cache Line size in bytes. 517 518 For example, the following options describe 2 NUMA nodes. Node 0 has 519 2 cpus and a ram, node 1 has only a ram. The processors in node 0 520 access memory in node 0 with access-latency 5 nanoseconds, 521 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access 522 memory in NUMA node 1 with access-latency 10 nanoseconds, 523 access-bandwidth is 100 MB/s. And for memory side cache information, 524 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB, 525 policy is write-back, the cache Line size is 8 bytes: 526 527 :: 528 529 -machine hmat=on \ 530 -m 2G \ 531 -object memory-backend-ram,size=1G,id=m0 \ 532 -object memory-backend-ram,size=1G,id=m1 \ 533 -smp 2,sockets=2,maxcpus=2 \ 534 -numa node,nodeid=0,memdev=m0 \ 535 -numa node,nodeid=1,memdev=m1,initiator=0 \ 536 -numa cpu,node-id=0,socket-id=0 \ 537 -numa cpu,node-id=0,socket-id=1 \ 538 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \ 539 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \ 540 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \ 541 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \ 542 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \ 543 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8 544ERST 545 546DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd, 547 "-add-fd fd=fd,set=set[,opaque=opaque]\n" 548 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL) 549SRST 550``-add-fd fd=fd,set=set[,opaque=opaque]`` 551 Add a file descriptor to an fd set. Valid options are: 552 553 ``fd=fd`` 554 This option defines the file descriptor of which a duplicate is 555 added to fd set. The file descriptor cannot be stdin, stdout, or 556 stderr. 557 558 ``set=set`` 559 This option defines the ID of the fd set to add the file 560 descriptor to. 561 562 ``opaque=opaque`` 563 This option defines a free-form string that can be used to 564 describe fd. 565 566 You can open an image using pre-opened file descriptors from an fd 567 set: 568 569 .. parsed-literal:: 570 571 |qemu_system| \\ 572 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 573 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 574 -drive file=/dev/fdset/2,index=0,media=disk 575ERST 576 577DEF("set", HAS_ARG, QEMU_OPTION_set, 578 "-set group.id.arg=value\n" 579 " set <arg> parameter for item <id> of type <group>\n" 580 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL) 581SRST 582``-set group.id.arg=value`` 583 Set parameter arg for item id of type group 584ERST 585 586DEF("global", HAS_ARG, QEMU_OPTION_global, 587 "-global driver.property=value\n" 588 "-global driver=driver,property=property,value=value\n" 589 " set a global default for a driver property\n", 590 QEMU_ARCH_ALL) 591SRST 592``-global driver.prop=value`` 593 \ 594``-global driver=driver,property=property,value=value`` 595 Set default value of driver's property prop to value, e.g.: 596 597 .. parsed-literal:: 598 599 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img 600 601 In particular, you can use this to set driver properties for devices 602 which are created automatically by the machine model. To create a 603 device which is not created automatically and set properties on it, 604 use -``device``. 605 606 -global driver.prop=value is shorthand for -global 607 driver=driver,property=prop,value=value. The longhand syntax works 608 even when driver contains a dot. 609ERST 610 611DEF("boot", HAS_ARG, QEMU_OPTION_boot, 612 "-boot [order=drives][,once=drives][,menu=on|off]\n" 613 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n" 614 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n" 615 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n" 616 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n" 617 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n", 618 QEMU_ARCH_ALL) 619SRST 620``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]`` 621 Specify boot order drives as a string of drive letters. Valid drive 622 letters depend on the target architecture. The x86 PC uses: a, b 623 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p 624 (Etherboot from network adapter 1-4), hard disk boot is the default. 625 To apply a particular boot order only on the first startup, specify 626 it via ``once``. Note that the ``order`` or ``once`` parameter 627 should not be used together with the ``bootindex`` property of 628 devices, since the firmware implementations normally do not support 629 both at the same time. 630 631 Interactive boot menus/prompts can be enabled via ``menu=on`` as far 632 as firmware/BIOS supports them. The default is non-interactive boot. 633 634 A splash picture could be passed to bios, enabling user to show it 635 as logo, when option splash=sp\_name is given and menu=on, If 636 firmware/BIOS supports them. Currently Seabios for X86 system 637 support it. limitation: The splash file could be a jpeg file or a 638 BMP file in 24 BPP format(true color). The resolution should be 639 supported by the SVGA mode, so the recommended is 320x240, 640x480, 640 800x640. 641 642 A timeout could be passed to bios, guest will pause for rb\_timeout 643 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will 644 not reboot, qemu passes '-1' to bios by default. Currently Seabios 645 for X86 system support it. 646 647 Do strict boot via ``strict=on`` as far as firmware/BIOS supports 648 it. This only effects when boot priority is changed by bootindex 649 options. The default is non-strict boot. 650 651 .. parsed-literal:: 652 653 # try to boot from network first, then from hard disk 654 |qemu_system_x86| -boot order=nc 655 # boot from CD-ROM first, switch back to default order after reboot 656 |qemu_system_x86| -boot once=d 657 # boot with a splash picture for 5 seconds. 658 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000 659 660 Note: The legacy format '-boot drives' is still supported but its 661 use is discouraged as it may be removed from future versions. 662ERST 663 664DEF("m", HAS_ARG, QEMU_OPTION_m, 665 "-m [size=]megs[,slots=n,maxmem=size]\n" 666 " configure guest RAM\n" 667 " size: initial amount of guest memory\n" 668 " slots: number of hotplug slots (default: none)\n" 669 " maxmem: maximum amount of guest memory (default: none)\n" 670 " Note: Some architectures might enforce a specific granularity\n", 671 QEMU_ARCH_ALL) 672SRST 673``-m [size=]megs[,slots=n,maxmem=size]`` 674 Sets guest startup RAM size to megs megabytes. Default is 128 MiB. 675 Optionally, a suffix of "M" or "G" can be used to signify a value in 676 megabytes or gigabytes respectively. Optional pair slots, maxmem 677 could be used to set amount of hotpluggable memory slots and maximum 678 amount of memory. Note that maxmem must be aligned to the page size. 679 680 For example, the following command-line sets the guest startup RAM 681 size to 1GB, creates 3 slots to hotplug additional memory and sets 682 the maximum memory the guest can reach to 4GB: 683 684 .. parsed-literal:: 685 686 |qemu_system| -m 1G,slots=3,maxmem=4G 687 688 If slots and maxmem are not specified, memory hotplug won't be 689 enabled and the guest startup RAM will never increase. 690ERST 691 692DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath, 693 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL) 694SRST 695``-mem-path path`` 696 Allocate guest RAM from a temporarily created file in path. 697ERST 698 699DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc, 700 "-mem-prealloc preallocate guest memory (use with -mem-path)\n", 701 QEMU_ARCH_ALL) 702SRST 703``-mem-prealloc`` 704 Preallocate memory when using -mem-path. 705ERST 706 707DEF("k", HAS_ARG, QEMU_OPTION_k, 708 "-k language use keyboard layout (for example 'fr' for French)\n", 709 QEMU_ARCH_ALL) 710SRST 711``-k language`` 712 Use keyboard layout language (for example ``fr`` for French). This 713 option is only needed where it is not easy to get raw PC keycodes 714 (e.g. on Macs, with some X11 servers or with a VNC or curses 715 display). You don't normally need to use it on PC/Linux or 716 PC/Windows hosts. 717 718 The available layouts are: 719 720 :: 721 722 ar de-ch es fo fr-ca hu ja mk no pt-br sv 723 da en-gb et fr fr-ch is lt nl pl ru th 724 de en-us fi fr-be hr it lv nl-be pt sl tr 725 726 The default is ``en-us``. 727ERST 728 729 730DEF("audio", HAS_ARG, QEMU_OPTION_audio, 731 "-audio [driver=]driver[,prop[=value][,...]]\n" 732 " specifies default audio backend when `audiodev` is not\n" 733 " used to create a machine or sound device;" 734 " options are the same as for -audiodev\n" 735 "-audio [driver=]driver,model=value[,prop[=value][,...]]\n" 736 " specifies the audio backend and device to use;\n" 737 " apart from 'model', options are the same as for -audiodev.\n" 738 " use '-audio model=help' to show possible devices.\n", 739 QEMU_ARCH_ALL) 740SRST 741``-audio [driver=]driver[,model=value][,prop[=value][,...]]`` 742 If the ``model`` option is specified, ``-audio`` is a shortcut 743 for configuring both the guest audio hardware and the host audio 744 backend in one go. The guest hardware model can be set with 745 ``model=modelname``. Use ``model=help`` to list the available 746 device types. 747 748 The following two example do exactly the same, to show how ``-audio`` 749 can be used to shorten the command line length: 750 751 .. parsed-literal:: 752 753 |qemu_system| -audiodev pa,id=pa -device sb16,audiodev=pa 754 |qemu_system| -audio pa,model=sb16 755 756 If the ``model`` option is not specified, ``-audio`` is used to 757 configure a default audio backend that will be used whenever the 758 ``audiodev`` property is not set on a device or machine. In 759 particular, ``-audio none`` ensures that no audio is produced even 760 for machines that have embedded sound hardware. 761 762 In both cases, the driver option is the same as with the corresponding 763 ``-audiodev`` option below. Use ``driver=help`` to list the available 764 drivers. 765 766ERST 767 768DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, 769 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n" 770 " specifies the audio backend to use\n" 771 " Use ``-audiodev help`` to list the available drivers\n" 772 " id= identifier of the backend\n" 773 " timer-period= timer period in microseconds\n" 774 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n" 775 " in|out.fixed-settings= use fixed settings for host audio\n" 776 " in|out.frequency= frequency to use with fixed settings\n" 777 " in|out.channels= number of channels to use with fixed settings\n" 778 " in|out.format= sample format to use with fixed settings\n" 779 " valid values: s8, s16, s32, u8, u16, u32, f32\n" 780 " in|out.voices= number of voices to use\n" 781 " in|out.buffer-length= length of buffer in microseconds\n" 782 "-audiodev none,id=id,[,prop[=value][,...]]\n" 783 " dummy driver that discards all output\n" 784#ifdef CONFIG_AUDIO_ALSA 785 "-audiodev alsa,id=id[,prop[=value][,...]]\n" 786 " in|out.dev= name of the audio device to use\n" 787 " in|out.period-length= length of period in microseconds\n" 788 " in|out.try-poll= attempt to use poll mode\n" 789 " threshold= threshold (in microseconds) when playback starts\n" 790#endif 791#ifdef CONFIG_AUDIO_COREAUDIO 792 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n" 793 " in|out.buffer-count= number of buffers\n" 794#endif 795#ifdef CONFIG_AUDIO_DSOUND 796 "-audiodev dsound,id=id[,prop[=value][,...]]\n" 797 " latency= add extra latency to playback in microseconds\n" 798#endif 799#ifdef CONFIG_AUDIO_OSS 800 "-audiodev oss,id=id[,prop[=value][,...]]\n" 801 " in|out.dev= path of the audio device to use\n" 802 " in|out.buffer-count= number of buffers\n" 803 " in|out.try-poll= attempt to use poll mode\n" 804 " try-mmap= try using memory mapped access\n" 805 " exclusive= open device in exclusive mode\n" 806 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n" 807#endif 808#ifdef CONFIG_AUDIO_PA 809 "-audiodev pa,id=id[,prop[=value][,...]]\n" 810 " server= PulseAudio server address\n" 811 " in|out.name= source/sink device name\n" 812 " in|out.latency= desired latency in microseconds\n" 813#endif 814#ifdef CONFIG_AUDIO_PIPEWIRE 815 "-audiodev pipewire,id=id[,prop[=value][,...]]\n" 816 " in|out.name= source/sink device name\n" 817 " in|out.stream-name= name of pipewire stream\n" 818 " in|out.latency= desired latency in microseconds\n" 819#endif 820#ifdef CONFIG_AUDIO_SDL 821 "-audiodev sdl,id=id[,prop[=value][,...]]\n" 822 " in|out.buffer-count= number of buffers\n" 823#endif 824#ifdef CONFIG_AUDIO_SNDIO 825 "-audiodev sndio,id=id[,prop[=value][,...]]\n" 826#endif 827#ifdef CONFIG_SPICE 828 "-audiodev spice,id=id[,prop[=value][,...]]\n" 829#endif 830#ifdef CONFIG_DBUS_DISPLAY 831 "-audiodev dbus,id=id[,prop[=value][,...]]\n" 832#endif 833 "-audiodev wav,id=id[,prop[=value][,...]]\n" 834 " path= path of wav file to record\n", 835 QEMU_ARCH_ALL) 836SRST 837``-audiodev [driver=]driver,id=id[,prop[=value][,...]]`` 838 Adds a new audio backend driver identified by id. There are global 839 and driver specific properties. Some values can be set differently 840 for input and output, they're marked with ``in|out.``. You can set 841 the input's property with ``in.prop`` and the output's property with 842 ``out.prop``. For example: 843 844 :: 845 846 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000 847 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified 848 849 NOTE: parameter validation is known to be incomplete, in many cases 850 specifying an invalid option causes QEMU to print an error message 851 and continue emulation without sound. 852 853 Valid global options are: 854 855 ``id=identifier`` 856 Identifies the audio backend. 857 858 ``timer-period=period`` 859 Sets the timer period used by the audio subsystem in 860 microseconds. Default is 10000 (10 ms). 861 862 ``in|out.mixing-engine=on|off`` 863 Use QEMU's mixing engine to mix all streams inside QEMU and 864 convert audio formats when not supported by the backend. When 865 off, fixed-settings must be off too. Note that disabling this 866 option means that the selected backend must support multiple 867 streams and the audio formats used by the virtual cards, 868 otherwise you'll get no sound. It's not recommended to disable 869 this option unless you want to use 5.1 or 7.1 audio, as mixing 870 engine only supports mono and stereo audio. Default is on. 871 872 ``in|out.fixed-settings=on|off`` 873 Use fixed settings for host audio. When off, it will change 874 based on how the guest opens the sound card. In this case you 875 must not specify frequency, channels or format. Default is on. 876 877 ``in|out.frequency=frequency`` 878 Specify the frequency to use when using fixed-settings. Default 879 is 44100Hz. 880 881 ``in|out.channels=channels`` 882 Specify the number of channels to use when using fixed-settings. 883 Default is 2 (stereo). 884 885 ``in|out.format=format`` 886 Specify the sample format to use when using fixed-settings. 887 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``, 888 ``u32``, ``f32``. Default is ``s16``. 889 890 ``in|out.voices=voices`` 891 Specify the number of voices to use. Default is 1. 892 893 ``in|out.buffer-length=usecs`` 894 Sets the size of the buffer in microseconds. 895 896``-audiodev none,id=id[,prop[=value][,...]]`` 897 Creates a dummy backend that discards all outputs. This backend has 898 no backend specific properties. 899 900``-audiodev alsa,id=id[,prop[=value][,...]]`` 901 Creates backend using the ALSA. This backend is only available on 902 Linux. 903 904 ALSA specific options are: 905 906 ``in|out.dev=device`` 907 Specify the ALSA device to use for input and/or output. Default 908 is ``default``. 909 910 ``in|out.period-length=usecs`` 911 Sets the period length in microseconds. 912 913 ``in|out.try-poll=on|off`` 914 Attempt to use poll mode with the device. Default is on. 915 916 ``threshold=threshold`` 917 Threshold (in microseconds) when playback starts. Default is 0. 918 919``-audiodev coreaudio,id=id[,prop[=value][,...]]`` 920 Creates a backend using Apple's Core Audio. This backend is only 921 available on Mac OS and only supports playback. 922 923 Core Audio specific options are: 924 925 ``in|out.buffer-count=count`` 926 Sets the count of the buffers. 927 928``-audiodev dsound,id=id[,prop[=value][,...]]`` 929 Creates a backend using Microsoft's DirectSound. This backend is 930 only available on Windows and only supports playback. 931 932 DirectSound specific options are: 933 934 ``latency=usecs`` 935 Add extra usecs microseconds latency to playback. Default is 936 10000 (10 ms). 937 938``-audiodev oss,id=id[,prop[=value][,...]]`` 939 Creates a backend using OSS. This backend is available on most 940 Unix-like systems. 941 942 OSS specific options are: 943 944 ``in|out.dev=device`` 945 Specify the file name of the OSS device to use. Default is 946 ``/dev/dsp``. 947 948 ``in|out.buffer-count=count`` 949 Sets the count of the buffers. 950 951 ``in|out.try-poll=on|of`` 952 Attempt to use poll mode with the device. Default is on. 953 954 ``try-mmap=on|off`` 955 Try using memory mapped device access. Default is off. 956 957 ``exclusive=on|off`` 958 Open the device in exclusive mode (vmix won't work in this 959 case). Default is off. 960 961 ``dsp-policy=policy`` 962 Sets the timing policy (between 0 and 10, where smaller number 963 means smaller latency but higher CPU usage). Use -1 to use 964 buffer sizes specified by ``buffer`` and ``buffer-count``. This 965 option is ignored if you do not have OSS 4. Default is 5. 966 967``-audiodev pa,id=id[,prop[=value][,...]]`` 968 Creates a backend using PulseAudio. This backend is available on 969 most systems. 970 971 PulseAudio specific options are: 972 973 ``server=server`` 974 Sets the PulseAudio server to connect to. 975 976 ``in|out.name=sink`` 977 Use the specified source/sink for recording/playback. 978 979 ``in|out.latency=usecs`` 980 Desired latency in microseconds. The PulseAudio server will try 981 to honor this value but actual latencies may be lower or higher. 982 983``-audiodev pipewire,id=id[,prop[=value][,...]]`` 984 Creates a backend using PipeWire. This backend is available on 985 most systems. 986 987 PipeWire specific options are: 988 989 ``in|out.latency=usecs`` 990 Desired latency in microseconds. 991 992 ``in|out.name=sink`` 993 Use the specified source/sink for recording/playback. 994 995 ``in|out.stream-name`` 996 Specify the name of pipewire stream. 997 998``-audiodev sdl,id=id[,prop[=value][,...]]`` 999 Creates a backend using SDL. This backend is available on most 1000 systems, but you should use your platform's native backend if 1001 possible. 1002 1003 SDL specific options are: 1004 1005 ``in|out.buffer-count=count`` 1006 Sets the count of the buffers. 1007 1008``-audiodev sndio,id=id[,prop[=value][,...]]`` 1009 Creates a backend using SNDIO. This backend is available on 1010 OpenBSD and most other Unix-like systems. 1011 1012 Sndio specific options are: 1013 1014 ``in|out.dev=device`` 1015 Specify the sndio device to use for input and/or output. Default 1016 is ``default``. 1017 1018 ``in|out.latency=usecs`` 1019 Sets the desired period length in microseconds. 1020 1021``-audiodev spice,id=id[,prop[=value][,...]]`` 1022 Creates a backend that sends audio through SPICE. This backend 1023 requires ``-spice`` and automatically selected in that case, so 1024 usually you can ignore this option. This backend has no backend 1025 specific properties. 1026 1027``-audiodev wav,id=id[,prop[=value][,...]]`` 1028 Creates a backend that writes audio to a WAV file. 1029 1030 Backend specific options are: 1031 1032 ``path=path`` 1033 Write recorded audio into the specified file. Default is 1034 ``qemu.wav``. 1035ERST 1036 1037DEF("device", HAS_ARG, QEMU_OPTION_device, 1038 "-device driver[,prop[=value][,...]]\n" 1039 " add device (based on driver)\n" 1040 " prop=value,... sets driver properties\n" 1041 " use '-device help' to print all possible drivers\n" 1042 " use '-device driver,help' to print all possible properties\n", 1043 QEMU_ARCH_ALL) 1044SRST 1045``-device driver[,prop[=value][,...]]`` 1046 Add device driver. prop=value sets driver properties. Valid 1047 properties depend on the driver. To get help on possible drivers and 1048 properties, use ``-device help`` and ``-device driver,help``. 1049 1050 Some drivers are: 1051 1052``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]`` 1053 Add an IPMI BMC. This is a simulation of a hardware management 1054 interface processor that normally sits on a system. It provides a 1055 watchdog and the ability to reset and power control the system. You 1056 need to connect this to an IPMI interface to make it useful 1057 1058 The IPMI slave address to use for the BMC. The default is 0x20. This 1059 address is the BMC's address on the I2C network of management 1060 controllers. If you don't know what this means, it is safe to ignore 1061 it. 1062 1063 ``id=id`` 1064 The BMC id for interfaces to use this device. 1065 1066 ``slave_addr=val`` 1067 Define slave address to use for the BMC. The default is 0x20. 1068 1069 ``sdrfile=file`` 1070 file containing raw Sensor Data Records (SDR) data. The default 1071 is none. 1072 1073 ``fruareasize=val`` 1074 size of a Field Replaceable Unit (FRU) area. The default is 1075 1024. 1076 1077 ``frudatafile=file`` 1078 file containing raw Field Replaceable Unit (FRU) inventory data. 1079 The default is none. 1080 1081 ``guid=uuid`` 1082 value for the GUID for the BMC, in standard UUID format. If this 1083 is set, get "Get GUID" command to the BMC will return it. 1084 Otherwise "Get GUID" will return an error. 1085 1086``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]`` 1087 Add a connection to an external IPMI BMC simulator. Instead of 1088 locally emulating the BMC like the above item, instead connect to an 1089 external entity that provides the IPMI services. 1090 1091 A connection is made to an external BMC simulator. If you do this, 1092 it is strongly recommended that you use the "reconnect=" chardev 1093 option to reconnect to the simulator if the connection is lost. Note 1094 that if this is not used carefully, it can be a security issue, as 1095 the interface has the ability to send resets, NMIs, and power off 1096 the VM. It's best if QEMU makes a connection to an external 1097 simulator running on a secure port on localhost, so neither the 1098 simulator nor QEMU is exposed to any outside network. 1099 1100 See the "lanserv/README.vm" file in the OpenIPMI library for more 1101 details on the external interface. 1102 1103``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]`` 1104 Add a KCS IPMI interface on the ISA bus. This also adds a 1105 corresponding ACPI and SMBIOS entries, if appropriate. 1106 1107 ``bmc=id`` 1108 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern 1109 above. 1110 1111 ``ioport=val`` 1112 Define the I/O address of the interface. The default is 0xca0 1113 for KCS. 1114 1115 ``irq=val`` 1116 Define the interrupt to use. The default is 5. To disable 1117 interrupts, set this to 0. 1118 1119``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]`` 1120 Like the KCS interface, but defines a BT interface. The default port 1121 is 0xe4 and the default interrupt is 5. 1122 1123``-device pci-ipmi-kcs,bmc=id`` 1124 Add a KCS IPMI interface on the PCI bus. 1125 1126 ``bmc=id`` 1127 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above. 1128 1129``-device pci-ipmi-bt,bmc=id`` 1130 Like the KCS interface, but defines a BT interface on the PCI bus. 1131 1132``-device intel-iommu[,option=...]`` 1133 This is only supported by ``-machine q35``, which will enable Intel VT-d 1134 emulation within the guest. It supports below options: 1135 1136 ``intremap=on|off`` (default: auto) 1137 This enables interrupt remapping feature. It's required to enable 1138 complete x2apic. Currently it only supports kvm kernel-irqchip modes 1139 ``off`` or ``split``, while full kernel-irqchip is not yet supported. 1140 The default value is "auto", which will be decided by the mode of 1141 kernel-irqchip. 1142 1143 ``caching-mode=on|off`` (default: off) 1144 This enables caching mode for the VT-d emulated device. When 1145 caching-mode is enabled, each guest DMA buffer mapping will generate an 1146 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in 1147 a synchronous way. It is required for ``-device vfio-pci`` to work 1148 with the VT-d device, because host assigned devices requires to setup 1149 the DMA mapping on the host before guest DMA starts. 1150 1151 ``device-iotlb=on|off`` (default: off) 1152 This enables device-iotlb capability for the emulated VT-d device. So 1153 far virtio/vhost should be the only real user for this parameter, 1154 paired with ats=on configured for the device. 1155 1156 ``aw-bits=39|48`` (default: 39) 1157 This decides the address width of IOVA address space. The address 1158 space has 39 bits width for 3-level IOMMU page tables, and 48 bits for 1159 4-level IOMMU page tables. 1160 1161 Please also refer to the wiki page for general scenarios of VT-d 1162 emulation in QEMU: https://wiki.qemu.org/Features/VT-d. 1163 1164ERST 1165 1166DEF("name", HAS_ARG, QEMU_OPTION_name, 1167 "-name string1[,process=string2][,debug-threads=on|off]\n" 1168 " set the name of the guest\n" 1169 " string1 sets the window title and string2 the process name\n" 1170 " When debug-threads is enabled, individual threads are given a separate name\n" 1171 " NOTE: The thread names are for debugging and not a stable API.\n", 1172 QEMU_ARCH_ALL) 1173SRST 1174``-name name`` 1175 Sets the name of the guest. This name will be displayed in the SDL 1176 window caption. The name will also be used for the VNC server. Also 1177 optionally set the top visible process name in Linux. Naming of 1178 individual threads can also be enabled on Linux to aid debugging. 1179ERST 1180 1181DEF("uuid", HAS_ARG, QEMU_OPTION_uuid, 1182 "-uuid %08x-%04x-%04x-%04x-%012x\n" 1183 " specify machine UUID\n", QEMU_ARCH_ALL) 1184SRST 1185``-uuid uuid`` 1186 Set system UUID. 1187ERST 1188 1189DEFHEADING() 1190 1191DEFHEADING(Block device options:) 1192 1193SRST 1194The QEMU block device handling options have a long history and 1195have gone through several iterations as the feature set and complexity 1196of the block layer have grown. Many online guides to QEMU often 1197reference older and deprecated options, which can lead to confusion. 1198 1199The most explicit way to describe disks is to use a combination of 1200``-device`` to specify the hardware device and ``-blockdev`` to 1201describe the backend. The device defines what the guest sees and the 1202backend describes how QEMU handles the data. It is the only guaranteed 1203stable interface for describing block devices and as such is 1204recommended for management tools and scripting. 1205 1206The ``-drive`` option combines the device and backend into a single 1207command line option which is a more human friendly. There is however no 1208interface stability guarantee although some older board models still 1209need updating to work with the modern blockdev forms. 1210 1211Older options like ``-hda`` are essentially macros which expand into 1212``-drive`` options for various drive interfaces. The original forms 1213bake in a lot of assumptions from the days when QEMU was emulating a 1214legacy PC, they are not recommended for modern configurations. 1215 1216ERST 1217 1218DEF("fda", HAS_ARG, QEMU_OPTION_fda, 1219 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL) 1220DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL) 1221SRST 1222``-fda file`` 1223 \ 1224``-fdb file`` 1225 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in 1226 the System Emulation Users Guide). 1227ERST 1228 1229DEF("hda", HAS_ARG, QEMU_OPTION_hda, 1230 "-hda/-hdb file use 'file' as hard disk 0/1 image\n", QEMU_ARCH_ALL) 1231DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL) 1232DEF("hdc", HAS_ARG, QEMU_OPTION_hdc, 1233 "-hdc/-hdd file use 'file' as hard disk 2/3 image\n", QEMU_ARCH_ALL) 1234DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL) 1235SRST 1236``-hda file`` 1237 \ 1238``-hdb file`` 1239 \ 1240``-hdc file`` 1241 \ 1242``-hdd file`` 1243 Use file as hard disk 0, 1, 2 or 3 image on the default bus of the 1244 emulated machine (this is for example the IDE bus on most x86 machines, 1245 but it can also be SCSI, virtio or something else on other target 1246 architectures). See also the :ref:`disk images` chapter in the System 1247 Emulation Users Guide. 1248ERST 1249 1250DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom, 1251 "-cdrom file use 'file' as CD-ROM image\n", 1252 QEMU_ARCH_ALL) 1253SRST 1254``-cdrom file`` 1255 Use file as CD-ROM image on the default bus of the emulated machine 1256 (which is IDE1 master on x86, so you cannot use ``-hdc`` and ``-cdrom`` 1257 at the same time there). On systems that support it, you can use the 1258 host CD-ROM by using ``/dev/cdrom`` as filename. 1259ERST 1260 1261DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev, 1262 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n" 1263 " [,cache.direct=on|off][,cache.no-flush=on|off]\n" 1264 " [,read-only=on|off][,auto-read-only=on|off]\n" 1265 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n" 1266 " [,driver specific parameters...]\n" 1267 " configure a block backend\n", QEMU_ARCH_ALL) 1268SRST 1269``-blockdev option[,option[,option[,...]]]`` 1270 Define a new block driver node. Some of the options apply to all 1271 block drivers, other options are only accepted for a specific block 1272 driver. See below for a list of generic options and options for the 1273 most common block drivers. 1274 1275 Options that expect a reference to another node (e.g. ``file``) can 1276 be given in two ways. Either you specify the node name of an already 1277 existing node (file=node-name), or you define a new node inline, 1278 adding options for the referenced node after a dot 1279 (file.filename=path,file.aio=native). 1280 1281 A block driver node created with ``-blockdev`` can be used for a 1282 guest device by specifying its node name for the ``drive`` property 1283 in a ``-device`` argument that defines a block device. 1284 1285 ``Valid options for any block driver node:`` 1286 ``driver`` 1287 Specifies the block driver to use for the given node. 1288 1289 ``node-name`` 1290 This defines the name of the block driver node by which it 1291 will be referenced later. The name must be unique, i.e. it 1292 must not match the name of a different block driver node, or 1293 (if you use ``-drive`` as well) the ID of a drive. 1294 1295 If no node name is specified, it is automatically generated. 1296 The generated node name is not intended to be predictable 1297 and changes between QEMU invocations. For the top level, an 1298 explicit node name must be specified. 1299 1300 ``read-only`` 1301 Open the node read-only. Guest write attempts will fail. 1302 1303 Note that some block drivers support only read-only access, 1304 either generally or in certain configurations. In this case, 1305 the default value ``read-only=off`` does not work and the 1306 option must be specified explicitly. 1307 1308 ``auto-read-only`` 1309 If ``auto-read-only=on`` is set, QEMU may fall back to 1310 read-only usage even when ``read-only=off`` is requested, or 1311 even switch between modes as needed, e.g. depending on 1312 whether the image file is writable or whether a writing user 1313 is attached to the node. 1314 1315 ``force-share`` 1316 Override the image locking system of QEMU by forcing the 1317 node to utilize weaker shared access for permissions where 1318 it would normally request exclusive access. When there is 1319 the potential for multiple instances to have the same file 1320 open (whether this invocation of QEMU is the first or the 1321 second instance), both instances must permit shared access 1322 for the second instance to succeed at opening the file. 1323 1324 Enabling ``force-share=on`` requires ``read-only=on``. 1325 1326 ``cache.direct`` 1327 The host page cache can be avoided with ``cache.direct=on``. 1328 This will attempt to do disk IO directly to the guest's 1329 memory. QEMU may still perform an internal copy of the data. 1330 1331 ``cache.no-flush`` 1332 In case you don't care about data integrity over host 1333 failures, you can use ``cache.no-flush=on``. This option 1334 tells QEMU that it never needs to write any data to the disk 1335 but can instead keep things in cache. If anything goes 1336 wrong, like your host losing power, the disk storage getting 1337 disconnected accidentally, etc. your image will most 1338 probably be rendered unusable. 1339 1340 ``discard=discard`` 1341 discard is one of "ignore" (or "off") or "unmap" (or "on") 1342 and controls whether ``discard`` (also known as ``trim`` or 1343 ``unmap``) requests are ignored or passed to the filesystem. 1344 Some machine types may not support discard requests. 1345 1346 ``detect-zeroes=detect-zeroes`` 1347 detect-zeroes is "off", "on" or "unmap" and enables the 1348 automatic conversion of plain zero writes by the OS to 1349 driver specific optimized zero write commands. You may even 1350 choose "unmap" if discard is set to "unmap" to allow a zero 1351 write to be converted to an ``unmap`` operation. 1352 1353 ``Driver-specific options for file`` 1354 This is the protocol-level block driver for accessing regular 1355 files. 1356 1357 ``filename`` 1358 The path to the image file in the local filesystem 1359 1360 ``aio`` 1361 Specifies the AIO backend (threads/native/io_uring, 1362 default: threads) 1363 1364 ``locking`` 1365 Specifies whether the image file is protected with Linux OFD 1366 / POSIX locks. The default is to use the Linux Open File 1367 Descriptor API if available, otherwise no lock is applied. 1368 (auto/on/off, default: auto) 1369 1370 Example: 1371 1372 :: 1373 1374 -blockdev driver=file,node-name=disk,filename=disk.img 1375 1376 ``Driver-specific options for raw`` 1377 This is the image format block driver for raw images. It is 1378 usually stacked on top of a protocol level block driver such as 1379 ``file``. 1380 1381 ``file`` 1382 Reference to or definition of the data source block driver 1383 node (e.g. a ``file`` driver node) 1384 1385 Example 1: 1386 1387 :: 1388 1389 -blockdev driver=file,node-name=disk_file,filename=disk.img 1390 -blockdev driver=raw,node-name=disk,file=disk_file 1391 1392 Example 2: 1393 1394 :: 1395 1396 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img 1397 1398 ``Driver-specific options for qcow2`` 1399 This is the image format block driver for qcow2 images. It is 1400 usually stacked on top of a protocol level block driver such as 1401 ``file``. 1402 1403 ``file`` 1404 Reference to or definition of the data source block driver 1405 node (e.g. a ``file`` driver node) 1406 1407 ``backing`` 1408 Reference to or definition of the backing file block device 1409 (default is taken from the image file). It is allowed to 1410 pass ``null`` here in order to disable the default backing 1411 file. 1412 1413 ``lazy-refcounts`` 1414 Whether to enable the lazy refcounts feature (on/off; 1415 default is taken from the image file) 1416 1417 ``cache-size`` 1418 The maximum total size of the L2 table and refcount block 1419 caches in bytes (default: the sum of l2-cache-size and 1420 refcount-cache-size) 1421 1422 ``l2-cache-size`` 1423 The maximum size of the L2 table cache in bytes (default: if 1424 cache-size is not specified - 32M on Linux platforms, and 8M 1425 on non-Linux platforms; otherwise, as large as possible 1426 within the cache-size, while permitting the requested or the 1427 minimal refcount cache size) 1428 1429 ``refcount-cache-size`` 1430 The maximum size of the refcount block cache in bytes 1431 (default: 4 times the cluster size; or if cache-size is 1432 specified, the part of it which is not used for the L2 1433 cache) 1434 1435 ``cache-clean-interval`` 1436 Clean unused entries in the L2 and refcount caches. The 1437 interval is in seconds. The default value is 600 on 1438 supporting platforms, and 0 on other platforms. Setting it 1439 to 0 disables this feature. 1440 1441 ``pass-discard-request`` 1442 Whether discard requests to the qcow2 device should be 1443 forwarded to the data source (on/off; default: on if 1444 discard=unmap is specified, off otherwise) 1445 1446 ``pass-discard-snapshot`` 1447 Whether discard requests for the data source should be 1448 issued when a snapshot operation (e.g. deleting a snapshot) 1449 frees clusters in the qcow2 file (on/off; default: on) 1450 1451 ``pass-discard-other`` 1452 Whether discard requests for the data source should be 1453 issued on other occasions where a cluster gets freed 1454 (on/off; default: off) 1455 1456 ``discard-no-unref`` 1457 When enabled, discards from the guest will not cause cluster 1458 allocations to be relinquished. This prevents qcow2 fragmentation 1459 that would be caused by such discards. Besides potential 1460 performance degradation, such fragmentation can lead to increased 1461 allocation of clusters past the end of the image file, 1462 resulting in image files whose file length can grow much larger 1463 than their guest disk size would suggest. 1464 If image file length is of concern (e.g. when storing qcow2 1465 images directly on block devices), you should consider enabling 1466 this option. 1467 1468 ``overlap-check`` 1469 Which overlap checks to perform for writes to the image 1470 (none/constant/cached/all; default: cached). For details or 1471 finer granularity control refer to the QAPI documentation of 1472 ``blockdev-add``. 1473 1474 Example 1: 1475 1476 :: 1477 1478 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2 1479 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216 1480 1481 Example 2: 1482 1483 :: 1484 1485 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2 1486 1487 ``Driver-specific options for other drivers`` 1488 Please refer to the QAPI documentation of the ``blockdev-add`` 1489 QMP command. 1490ERST 1491 1492DEF("drive", HAS_ARG, QEMU_OPTION_drive, 1493 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n" 1494 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n" 1495 " [,snapshot=on|off][,rerror=ignore|stop|report]\n" 1496 " [,werror=ignore|stop|report|enospc][,id=name]\n" 1497 " [,aio=threads|native|io_uring]\n" 1498 " [,readonly=on|off][,copy-on-read=on|off]\n" 1499 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n" 1500 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n" 1501 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n" 1502 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n" 1503 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n" 1504 " [[,iops_size=is]]\n" 1505 " [[,group=g]]\n" 1506 " use 'file' as a drive image\n", QEMU_ARCH_ALL) 1507SRST 1508``-drive option[,option[,option[,...]]]`` 1509 Define a new drive. This includes creating a block driver node (the 1510 backend) as well as a guest device, and is mostly a shortcut for 1511 defining the corresponding ``-blockdev`` and ``-device`` options. 1512 1513 ``-drive`` accepts all options that are accepted by ``-blockdev``. 1514 In addition, it knows the following options: 1515 1516 ``file=file`` 1517 This option defines which disk image (see the :ref:`disk images` 1518 chapter in the System Emulation Users Guide) to use with this drive. 1519 If the filename contains comma, you must double it (for instance, 1520 "file=my,,file" to use file "my,file"). 1521 1522 Special files such as iSCSI devices can be specified using 1523 protocol specific URLs. See the section for "Device URL Syntax" 1524 for more information. 1525 1526 ``if=interface`` 1527 This option defines on which type on interface the drive is 1528 connected. Available types are: ide, scsi, sd, mtd, floppy, 1529 pflash, virtio, none. 1530 1531 ``bus=bus,unit=unit`` 1532 These options define where is connected the drive by defining 1533 the bus number and the unit id. 1534 1535 ``index=index`` 1536 This option defines where the drive is connected by using an 1537 index in the list of available connectors of a given interface 1538 type. 1539 1540 ``media=media`` 1541 This option defines the type of the media: disk or cdrom. 1542 1543 ``snapshot=snapshot`` 1544 snapshot is "on" or "off" and controls snapshot mode for the 1545 given drive (see ``-snapshot``). 1546 1547 ``cache=cache`` 1548 cache is "none", "writeback", "unsafe", "directsync" or 1549 "writethrough" and controls how the host cache is used to access 1550 block data. This is a shortcut that sets the ``cache.direct`` 1551 and ``cache.no-flush`` options (as in ``-blockdev``), and 1552 additionally ``cache.writeback``, which provides a default for 1553 the ``write-cache`` option of block guest devices (as in 1554 ``-device``). The modes correspond to the following settings: 1555 1556 ============= =============== ============ ============== 1557 \ cache.writeback cache.direct cache.no-flush 1558 ============= =============== ============ ============== 1559 writeback on off off 1560 none on on off 1561 writethrough off off off 1562 directsync off on off 1563 unsafe on off on 1564 ============= =============== ============ ============== 1565 1566 The default mode is ``cache=writeback``. 1567 1568 ``aio=aio`` 1569 aio is "threads", "native", or "io_uring" and selects between pthread 1570 based disk I/O, native Linux AIO, or Linux io_uring API. 1571 1572 ``format=format`` 1573 Specify which disk format will be used rather than detecting the 1574 format. Can be used to specify format=raw to avoid interpreting 1575 an untrusted format header. 1576 1577 ``werror=action,rerror=action`` 1578 Specify which action to take on write and read errors. Valid 1579 actions are: "ignore" (ignore the error and try to continue), 1580 "stop" (pause QEMU), "report" (report the error to the guest), 1581 "enospc" (pause QEMU only if the host disk is full; report the 1582 error to the guest otherwise). The default setting is 1583 ``werror=enospc`` and ``rerror=report``. 1584 1585 ``copy-on-read=copy-on-read`` 1586 copy-on-read is "on" or "off" and enables whether to copy read 1587 backing file sectors into the image file. 1588 1589 ``bps=b,bps_rd=r,bps_wr=w`` 1590 Specify bandwidth throttling limits in bytes per second, either 1591 for all request types or for reads or writes only. Small values 1592 can lead to timeouts or hangs inside the guest. A safe minimum 1593 for disks is 2 MB/s. 1594 1595 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm`` 1596 Specify bursts in bytes per second, either for all request types 1597 or for reads or writes only. Bursts allow the guest I/O to spike 1598 above the limit temporarily. 1599 1600 ``iops=i,iops_rd=r,iops_wr=w`` 1601 Specify request rate limits in requests per second, either for 1602 all request types or for reads or writes only. 1603 1604 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm`` 1605 Specify bursts in requests per second, either for all request 1606 types or for reads or writes only. Bursts allow the guest I/O to 1607 spike above the limit temporarily. 1608 1609 ``iops_size=is`` 1610 Let every is bytes of a request count as a new request for iops 1611 throttling purposes. Use this option to prevent guests from 1612 circumventing iops limits by sending fewer but larger requests. 1613 1614 ``group=g`` 1615 Join a throttling quota group with given name g. All drives that 1616 are members of the same group are accounted for together. Use 1617 this option to prevent guests from circumventing throttling 1618 limits by using many small disks instead of a single larger 1619 disk. 1620 1621 By default, the ``cache.writeback=on`` mode is used. It will report 1622 data writes as completed as soon as the data is present in the host 1623 page cache. This is safe as long as your guest OS makes sure to 1624 correctly flush disk caches where needed. If your guest OS does not 1625 handle volatile disk write caches correctly and your host crashes or 1626 loses power, then the guest may experience data corruption. 1627 1628 For such guests, you should consider using ``cache.writeback=off``. 1629 This means that the host page cache will be used to read and write 1630 data, but write notification will be sent to the guest only after 1631 QEMU has made sure to flush each write to the disk. Be aware that 1632 this has a major impact on performance. 1633 1634 When using the ``-snapshot`` option, unsafe caching is always used. 1635 1636 Copy-on-read avoids accessing the same backing file sectors 1637 repeatedly and is useful when the backing file is over a slow 1638 network. By default copy-on-read is off. 1639 1640 Instead of ``-cdrom`` you can use: 1641 1642 .. parsed-literal:: 1643 1644 |qemu_system| -drive file=file,index=2,media=cdrom 1645 1646 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use: 1647 1648 .. parsed-literal:: 1649 1650 |qemu_system| -drive file=file,index=0,media=disk 1651 |qemu_system| -drive file=file,index=1,media=disk 1652 |qemu_system| -drive file=file,index=2,media=disk 1653 |qemu_system| -drive file=file,index=3,media=disk 1654 1655 You can open an image using pre-opened file descriptors from an fd 1656 set: 1657 1658 .. parsed-literal:: 1659 1660 |qemu_system| \\ 1661 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 1662 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 1663 -drive file=/dev/fdset/2,index=0,media=disk 1664 1665 You can connect a CDROM to the slave of ide0: 1666 1667 .. parsed-literal:: 1668 1669 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom 1670 1671 If you don't specify the "file=" argument, you define an empty 1672 drive: 1673 1674 .. parsed-literal:: 1675 1676 |qemu_system_x86| -drive if=ide,index=1,media=cdrom 1677 1678 Instead of ``-fda``, ``-fdb``, you can use: 1679 1680 .. parsed-literal:: 1681 1682 |qemu_system_x86| -drive file=file,index=0,if=floppy 1683 |qemu_system_x86| -drive file=file,index=1,if=floppy 1684 1685 By default, interface is "ide" and index is automatically 1686 incremented: 1687 1688 .. parsed-literal:: 1689 1690 |qemu_system_x86| -drive file=a -drive file=b 1691 1692 is interpreted like: 1693 1694 .. parsed-literal:: 1695 1696 |qemu_system_x86| -hda a -hdb b 1697ERST 1698 1699DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock, 1700 "-mtdblock file use 'file' as on-board Flash memory image\n", 1701 QEMU_ARCH_ALL) 1702SRST 1703``-mtdblock file`` 1704 Use file as on-board Flash memory image. 1705ERST 1706 1707DEF("sd", HAS_ARG, QEMU_OPTION_sd, 1708 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL) 1709SRST 1710``-sd file`` 1711 Use file as SecureDigital card image. 1712ERST 1713 1714DEF("snapshot", 0, QEMU_OPTION_snapshot, 1715 "-snapshot write to temporary files instead of disk image files\n", 1716 QEMU_ARCH_ALL) 1717SRST 1718``-snapshot`` 1719 Write to temporary files instead of disk image files. In this case, 1720 the raw disk image you use is not written back. You can however 1721 force the write back by pressing C-a s (see the :ref:`disk images` 1722 chapter in the System Emulation Users Guide). 1723 1724 .. warning:: 1725 snapshot is incompatible with ``-blockdev`` (instead use qemu-img 1726 to manually create snapshot images to attach to your blockdev). 1727 If you have mixed ``-blockdev`` and ``-drive`` declarations you 1728 can use the 'snapshot' property on your drive declarations 1729 instead of this global option. 1730 1731ERST 1732 1733DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev, 1734 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1735 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n" 1736 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n" 1737 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n" 1738 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n" 1739 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n" 1740 " [[,throttling.iops-size=is]]\n" 1741 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n" 1742 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n" 1743 "-fsdev synth,id=id\n", 1744 QEMU_ARCH_ALL) 1745 1746SRST 1747``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]`` 1748 \ 1749``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]`` 1750 \ 1751``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]`` 1752 \ 1753``-fsdev synth,id=id[,readonly=on]`` 1754 Define a new file system device. Valid options are: 1755 1756 ``local`` 1757 Accesses to the filesystem are done by QEMU. 1758 1759 ``proxy`` 1760 Accesses to the filesystem are done by virtfs-proxy-helper(1). This 1761 option is deprecated (since QEMU 8.1) and will be removed in a future 1762 version of QEMU. Use ``local`` instead. 1763 1764 ``synth`` 1765 Synthetic filesystem, only used by QTests. 1766 1767 ``id=id`` 1768 Specifies identifier for this device. 1769 1770 ``path=path`` 1771 Specifies the export path for the file system device. Files 1772 under this path will be available to the 9p client on the guest. 1773 1774 ``security_model=security_model`` 1775 Specifies the security model to be used for this export path. 1776 Supported security models are "passthrough", "mapped-xattr", 1777 "mapped-file" and "none". In "passthrough" security model, files 1778 are stored using the same credentials as they are created on the 1779 guest. This requires QEMU to run as root. In "mapped-xattr" 1780 security model, some of the file attributes like uid, gid, mode 1781 bits and link target are stored as file attributes. For 1782 "mapped-file" these attributes are stored in the hidden 1783 .virtfs\_metadata directory. Directories exported by this 1784 security model cannot interact with other unix tools. "none" 1785 security model is same as passthrough except the sever won't 1786 report failures if it fails to set file attributes like 1787 ownership. Security model is mandatory only for local fsdriver. 1788 Other fsdrivers (like proxy) don't take security model as a 1789 parameter. 1790 1791 ``writeout=writeout`` 1792 This is an optional argument. The only supported value is 1793 "immediate". This means that host page cache will be used to 1794 read and write data but write notification will be sent to the 1795 guest only when the data has been reported as written by the 1796 storage subsystem. 1797 1798 ``readonly=on`` 1799 Enables exporting 9p share as a readonly mount for guests. By 1800 default read-write access is given. 1801 1802 ``socket=socket`` 1803 Enables proxy filesystem driver to use passed socket file for 1804 communicating with virtfs-proxy-helper(1). 1805 1806 ``sock_fd=sock_fd`` 1807 Enables proxy filesystem driver to use passed socket descriptor 1808 for communicating with virtfs-proxy-helper(1). Usually a helper 1809 like libvirt will create socketpair and pass one of the fds as 1810 sock\_fd. 1811 1812 ``fmode=fmode`` 1813 Specifies the default mode for newly created files on the host. 1814 Works only with security models "mapped-xattr" and 1815 "mapped-file". 1816 1817 ``dmode=dmode`` 1818 Specifies the default mode for newly created directories on the 1819 host. Works only with security models "mapped-xattr" and 1820 "mapped-file". 1821 1822 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w`` 1823 Specify bandwidth throttling limits in bytes per second, either 1824 for all request types or for reads or writes only. 1825 1826 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm`` 1827 Specify bursts in bytes per second, either for all request types 1828 or for reads or writes only. Bursts allow the guest I/O to spike 1829 above the limit temporarily. 1830 1831 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w`` 1832 Specify request rate limits in requests per second, either for 1833 all request types or for reads or writes only. 1834 1835 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm`` 1836 Specify bursts in requests per second, either for all request 1837 types or for reads or writes only. Bursts allow the guest I/O to 1838 spike above the limit temporarily. 1839 1840 ``throttling.iops-size=is`` 1841 Let every is bytes of a request count as a new request for iops 1842 throttling purposes. 1843 1844 -fsdev option is used along with -device driver "virtio-9p-...". 1845 1846``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag`` 1847 Options for virtio-9p-... driver are: 1848 1849 ``type`` 1850 Specifies the variant to be used. Supported values are "pci", 1851 "ccw" or "device", depending on the machine type. 1852 1853 ``fsdev=id`` 1854 Specifies the id value specified along with -fsdev option. 1855 1856 ``mount_tag=mount_tag`` 1857 Specifies the tag name to be used by the guest to mount this 1858 export point. 1859ERST 1860 1861DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs, 1862 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1863 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n" 1864 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n" 1865 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n" 1866 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n", 1867 QEMU_ARCH_ALL) 1868 1869SRST 1870``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]`` 1871 \ 1872``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1873 \ 1874``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1875 \ 1876``-virtfs synth,mount_tag=mount_tag`` 1877 Define a new virtual filesystem device and expose it to the guest using 1878 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain 1879 directory on host is made directly accessible by guest as a pass-through 1880 file system by using the 9P network protocol for communication between 1881 host and guests, if desired even accessible, shared by several guests 1882 simultaneously. 1883 1884 Note that ``-virtfs`` is actually just a convenience shortcut for its 1885 generalized form ``-fsdev -device virtio-9p-pci``. 1886 1887 The general form of pass-through file system options are: 1888 1889 ``local`` 1890 Accesses to the filesystem are done by QEMU. 1891 1892 ``proxy`` 1893 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1894 This option is deprecated (since QEMU 8.1) and will be removed in a 1895 future version of QEMU. Use ``local`` instead. 1896 1897 ``synth`` 1898 Synthetic filesystem, only used by QTests. 1899 1900 ``id=id`` 1901 Specifies identifier for the filesystem device 1902 1903 ``path=path`` 1904 Specifies the export path for the file system device. Files 1905 under this path will be available to the 9p client on the guest. 1906 1907 ``security_model=security_model`` 1908 Specifies the security model to be used for this export path. 1909 Supported security models are "passthrough", "mapped-xattr", 1910 "mapped-file" and "none". In "passthrough" security model, files 1911 are stored using the same credentials as they are created on the 1912 guest. This requires QEMU to run as root. In "mapped-xattr" 1913 security model, some of the file attributes like uid, gid, mode 1914 bits and link target are stored as file attributes. For 1915 "mapped-file" these attributes are stored in the hidden 1916 .virtfs\_metadata directory. Directories exported by this 1917 security model cannot interact with other unix tools. "none" 1918 security model is same as passthrough except the sever won't 1919 report failures if it fails to set file attributes like 1920 ownership. Security model is mandatory only for local fsdriver. 1921 Other fsdrivers (like proxy) don't take security model as a 1922 parameter. 1923 1924 ``writeout=writeout`` 1925 This is an optional argument. The only supported value is 1926 "immediate". This means that host page cache will be used to 1927 read and write data but write notification will be sent to the 1928 guest only when the data has been reported as written by the 1929 storage subsystem. 1930 1931 ``readonly=on`` 1932 Enables exporting 9p share as a readonly mount for guests. By 1933 default read-write access is given. 1934 1935 ``socket=socket`` 1936 Enables proxy filesystem driver to use passed socket file for 1937 communicating with virtfs-proxy-helper(1). Usually a helper like 1938 libvirt will create socketpair and pass one of the fds as 1939 sock\_fd. 1940 1941 ``sock_fd`` 1942 Enables proxy filesystem driver to use passed 'sock\_fd' as the 1943 socket descriptor for interfacing with virtfs-proxy-helper(1). 1944 1945 ``fmode=fmode`` 1946 Specifies the default mode for newly created files on the host. 1947 Works only with security models "mapped-xattr" and 1948 "mapped-file". 1949 1950 ``dmode=dmode`` 1951 Specifies the default mode for newly created directories on the 1952 host. Works only with security models "mapped-xattr" and 1953 "mapped-file". 1954 1955 ``mount_tag=mount_tag`` 1956 Specifies the tag name to be used by the guest to mount this 1957 export point. 1958 1959 ``multidevs=multidevs`` 1960 Specifies how to deal with multiple devices being shared with a 1961 9p export. Supported behaviours are either "remap", "forbid" or 1962 "warn". The latter is the default behaviour on which virtfs 9p 1963 expects only one device to be shared with the same export, and 1964 if more than one device is shared and accessed via the same 9p 1965 export then only a warning message is logged (once) by qemu on 1966 host side. In order to avoid file ID collisions on guest you 1967 should either create a separate virtfs export for each device to 1968 be shared with guests (recommended way) or you might use "remap" 1969 instead which allows you to share multiple devices with only one 1970 export instead, which is achieved by remapping the original 1971 inode numbers from host to guest in a way that would prevent 1972 such collisions. Remapping inodes in such use cases is required 1973 because the original device IDs from host are never passed and 1974 exposed on guest. Instead all files of an export shared with 1975 virtfs always share the same device id on guest. So two files 1976 with identical inode numbers but from actually different devices 1977 on host would otherwise cause a file ID collision and hence 1978 potential misbehaviours on guest. "forbid" on the other hand 1979 assumes like "warn" that only one device is shared by the same 1980 export, however it will not only log a warning message but also 1981 deny access to additional devices on guest. Note though that 1982 "forbid" does currently not block all possible file access 1983 operations (e.g. readdir() would still return entries from other 1984 devices). 1985ERST 1986 1987DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi, 1988 "-iscsi [user=user][,password=password][,password-secret=secret-id]\n" 1989 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE]\n" 1990 " [,initiator-name=initiator-iqn][,id=target-iqn]\n" 1991 " [,timeout=timeout]\n" 1992 " iSCSI session parameters\n", QEMU_ARCH_ALL) 1993 1994SRST 1995``-iscsi`` 1996 Configure iSCSI session parameters. 1997ERST 1998 1999DEFHEADING() 2000 2001DEFHEADING(USB convenience options:) 2002 2003DEF("usb", 0, QEMU_OPTION_usb, 2004 "-usb enable on-board USB host controller (if not enabled by default)\n", 2005 QEMU_ARCH_ALL) 2006SRST 2007``-usb`` 2008 Enable USB emulation on machine types with an on-board USB host 2009 controller (if not enabled by default). Note that on-board USB host 2010 controllers may not support USB 3.0. In this case 2011 ``-device qemu-xhci`` can be used instead on machines with PCI. 2012ERST 2013 2014DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice, 2015 "-usbdevice name add the host or guest USB device 'name'\n", 2016 QEMU_ARCH_ALL) 2017SRST 2018``-usbdevice devname`` 2019 Add the USB device devname, and enable an on-board USB controller 2020 if possible and necessary (just like it can be done via 2021 ``-machine usb=on``). Note that this option is mainly intended for 2022 the user's convenience only. More fine-grained control can be 2023 achieved by selecting a USB host controller (if necessary) and the 2024 desired USB device via the ``-device`` option instead. For example, 2025 instead of using ``-usbdevice mouse`` it is possible to use 2026 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse 2027 to a USB 3.0 controller instead (at least on machines that support 2028 PCI and do not have an USB controller enabled by default yet). 2029 For more details, see the chapter about 2030 :ref:`Connecting USB devices` in the System Emulation Users Guide. 2031 Possible devices for devname are: 2032 2033 ``braille`` 2034 Braille device. This will use BrlAPI to display the braille 2035 output on a real or fake device (i.e. it also creates a 2036 corresponding ``braille`` chardev automatically beside the 2037 ``usb-braille`` USB device). 2038 2039 ``keyboard`` 2040 Standard USB keyboard. Will override the PS/2 keyboard (if present). 2041 2042 ``mouse`` 2043 Virtual Mouse. This will override the PS/2 mouse emulation when 2044 activated. 2045 2046 ``tablet`` 2047 Pointer device that uses absolute coordinates (like a 2048 touchscreen). This means QEMU is able to report the mouse 2049 position without having to grab the mouse. Also overrides the 2050 PS/2 mouse emulation when activated. 2051 2052 ``wacom-tablet`` 2053 Wacom PenPartner USB tablet. 2054 2055 2056ERST 2057 2058DEFHEADING() 2059 2060DEFHEADING(Display options:) 2061 2062DEF("display", HAS_ARG, QEMU_OPTION_display, 2063#if defined(CONFIG_SPICE) 2064 "-display spice-app[,gl=on|off]\n" 2065#endif 2066#if defined(CONFIG_SDL) 2067 "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n" 2068 " [,window-close=on|off]\n" 2069#endif 2070#if defined(CONFIG_GTK) 2071 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n" 2072 " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n" 2073 " [,show-menubar=on|off]\n" 2074#endif 2075#if defined(CONFIG_VNC) 2076 "-display vnc=<display>[,<optargs>]\n" 2077#endif 2078#if defined(CONFIG_CURSES) 2079 "-display curses[,charset=<encoding>]\n" 2080#endif 2081#if defined(CONFIG_COCOA) 2082 "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n" 2083#endif 2084#if defined(CONFIG_OPENGL) 2085 "-display egl-headless[,rendernode=<file>]\n" 2086#endif 2087#if defined(CONFIG_DBUS_DISPLAY) 2088 "-display dbus[,addr=<dbusaddr>]\n" 2089 " [,gl=on|core|es|off][,rendernode=<file>]\n" 2090#endif 2091#if defined(CONFIG_COCOA) 2092 "-display cocoa[,show-cursor=on|off][,left-command-key=on|off]\n" 2093#endif 2094 "-display none\n" 2095 " select display backend type\n" 2096 " The default display is equivalent to\n " 2097#if defined(CONFIG_GTK) 2098 "\"-display gtk\"\n" 2099#elif defined(CONFIG_SDL) 2100 "\"-display sdl\"\n" 2101#elif defined(CONFIG_COCOA) 2102 "\"-display cocoa\"\n" 2103#elif defined(CONFIG_VNC) 2104 "\"-vnc localhost:0,to=99,id=default\"\n" 2105#else 2106 "\"-display none\"\n" 2107#endif 2108 , QEMU_ARCH_ALL) 2109SRST 2110``-display type`` 2111 Select type of display to use. Use ``-display help`` to list the available 2112 display types. Valid values for type are 2113 2114 ``spice-app[,gl=on|off]`` 2115 Start QEMU as a Spice server and launch the default Spice client 2116 application. The Spice server will redirect the serial consoles 2117 and QEMU monitors. (Since 4.0) 2118 2119 ``dbus`` 2120 Export the display over D-Bus interfaces. (Since 7.0) 2121 2122 The connection is registered with the "org.qemu" name (and queued when 2123 already owned). 2124 2125 ``addr=<dbusaddr>`` : D-Bus bus address to connect to. 2126 2127 ``p2p=yes|no`` : Use peer-to-peer connection, accepted via QMP ``add_client``. 2128 2129 ``gl=on|off|core|es`` : Use OpenGL for rendering (the D-Bus interface 2130 will share framebuffers with DMABUF file descriptors). 2131 2132 ``sdl`` 2133 Display video output via SDL (usually in a separate graphics 2134 window; see the SDL documentation for other possibilities). 2135 Valid parameters are: 2136 2137 ``grab-mod=<mods>`` : Used to select the modifier keys for toggling 2138 the mouse grabbing in conjunction with the "g" key. ``<mods>`` can be 2139 either ``lshift-lctrl-lalt`` or ``rctrl``. 2140 2141 ``gl=on|off|core|es`` : Use OpenGL for displaying 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 ``gtk`` 2148 Display video output in a GTK window. This interface provides 2149 drop-down menus and other UI elements to configure and control 2150 the VM during runtime. Valid parameters are: 2151 2152 ``full-screen=on|off`` : Start in fullscreen mode 2153 2154 ``gl=on|off`` : Use OpenGL for displaying 2155 2156 ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover 2157 2158 ``show-tabs=on|off`` : Display the tab bar for switching between the 2159 various graphical interfaces (e.g. VGA and 2160 virtual console character devices) by default. 2161 2162 ``show-cursor=on|off`` : Force showing the mouse cursor 2163 2164 ``window-close=on|off`` : Allow to quit qemu with window close button 2165 2166 ``show-menubar=on|off`` : Display the main window menubar, defaults to "on" 2167 2168 ``zoom-to-fit=on|off`` : Expand video output to the window size, 2169 defaults to "off" 2170 2171 ``curses[,charset=<encoding>]`` 2172 Display video output via curses. For graphics device models 2173 which support a text mode, QEMU can display this output using a 2174 curses/ncurses interface. Nothing is displayed when the graphics 2175 device is in graphical mode or if the graphics device does not 2176 support a text mode. Generally only the VGA device models 2177 support text mode. The font charset used by the guest can be 2178 specified with the ``charset`` option, for example 2179 ``charset=CP850`` for IBM CP850 encoding. The default is 2180 ``CP437``. 2181 2182 ``cocoa`` 2183 Display video output in a Cocoa window. Mac only. This interface 2184 provides drop-down menus and other UI elements to configure and 2185 control the VM during runtime. Valid parameters are: 2186 2187 ``show-cursor=on|off`` : Force showing the mouse cursor 2188 2189 ``left-command-key=on|off`` : Disable forwarding left command key to host 2190 2191 ``egl-headless[,rendernode=<file>]`` 2192 Offload all OpenGL operations to a local DRI device. For any 2193 graphical display, this display needs to be paired with either 2194 VNC or SPICE displays. 2195 2196 ``vnc=<display>`` 2197 Start a VNC server on display <display> 2198 2199 ``none`` 2200 Do not display video output. The guest will still see an 2201 emulated graphics card, but its output will not be displayed to 2202 the QEMU user. This option differs from the -nographic option in 2203 that it only affects what is done with video output; -nographic 2204 also changes the destination of the serial and parallel port 2205 data. 2206ERST 2207 2208DEF("nographic", 0, QEMU_OPTION_nographic, 2209 "-nographic disable graphical output and redirect serial I/Os to console\n", 2210 QEMU_ARCH_ALL) 2211SRST 2212``-nographic`` 2213 Normally, if QEMU is compiled with graphical window support, it 2214 displays output such as guest graphics, guest console, and the QEMU 2215 monitor in a window. With this option, you can totally disable 2216 graphical output so that QEMU is a simple command line application. 2217 The emulated serial port is redirected on the console and muxed with 2218 the monitor (unless redirected elsewhere explicitly). Therefore, you 2219 can still use QEMU to debug a Linux kernel with a serial console. 2220 Use C-a h for help on switching between the console and monitor. 2221ERST 2222 2223#ifdef CONFIG_SPICE 2224DEF("spice", HAS_ARG, QEMU_OPTION_spice, 2225 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n" 2226 " [,x509-key-file=<file>][,x509-key-password=<file>]\n" 2227 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n" 2228 " [,x509-dh-key-file=<file>][,addr=addr]\n" 2229 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n" 2230 " [,tls-ciphers=<list>]\n" 2231 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n" 2232 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n" 2233 " [,sasl=on|off][,disable-ticketing=on|off]\n" 2234 " [,password-secret=<secret-id>]\n" 2235 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n" 2236 " [,jpeg-wan-compression=[auto|never|always]]\n" 2237 " [,zlib-glz-wan-compression=[auto|never|always]]\n" 2238 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n" 2239 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n" 2240 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n" 2241 " [,gl=[on|off]][,rendernode=<file>]\n" 2242 " enable spice\n" 2243 " at least one of {port, tls-port} is mandatory\n", 2244 QEMU_ARCH_ALL) 2245#endif 2246SRST 2247``-spice option[,option[,...]]`` 2248 Enable the spice remote desktop protocol. Valid options are 2249 2250 ``port=<nr>`` 2251 Set the TCP port spice is listening on for plaintext channels. 2252 2253 ``addr=<addr>`` 2254 Set the IP address spice is listening on. Default is any 2255 address. 2256 2257 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off`` 2258 Force using the specified IP version. 2259 2260 ``password-secret=<secret-id>`` 2261 Set the ID of the ``secret`` object containing the password 2262 you need to authenticate. 2263 2264 ``sasl=on|off`` 2265 Require that the client use SASL to authenticate with the spice. 2266 The exact choice of authentication method used is controlled 2267 from the system / user's SASL configuration file for the 'qemu' 2268 service. This is typically found in /etc/sasl2/qemu.conf. If 2269 running QEMU as an unprivileged user, an environment variable 2270 SASL\_CONF\_PATH can be used to make it search alternate 2271 locations for the service config. While some SASL auth methods 2272 can also provide data encryption (eg GSSAPI), it is recommended 2273 that SASL always be combined with the 'tls' and 'x509' settings 2274 to enable use of SSL and server certificates. This ensures a 2275 data encryption preventing compromise of authentication 2276 credentials. 2277 2278 ``disable-ticketing=on|off`` 2279 Allow client connects without authentication. 2280 2281 ``disable-copy-paste=on|off`` 2282 Disable copy paste between the client and the guest. 2283 2284 ``disable-agent-file-xfer=on|off`` 2285 Disable spice-vdagent based file-xfer between the client and the 2286 guest. 2287 2288 ``tls-port=<nr>`` 2289 Set the TCP port spice is listening on for encrypted channels. 2290 2291 ``x509-dir=<dir>`` 2292 Set the x509 file directory. Expects same filenames as -vnc 2293 $display,x509=$dir 2294 2295 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>`` 2296 The x509 file names can also be configured individually. 2297 2298 ``tls-ciphers=<list>`` 2299 Specify which ciphers to use. 2300 2301 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]`` 2302 Force specific channel to be used with or without TLS 2303 encryption. The options can be specified multiple times to 2304 configure multiple channels. The special name "default" can be 2305 used to set the default mode. For channels which are not 2306 explicitly forced into one mode the spice client is allowed to 2307 pick tls/plaintext as he pleases. 2308 2309 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]`` 2310 Configure image compression (lossless). Default is auto\_glz. 2311 2312 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]`` 2313 Configure wan image compression (lossy for slow links). Default 2314 is auto. 2315 2316 ``streaming-video=[off|all|filter]`` 2317 Configure video stream detection. Default is off. 2318 2319 ``agent-mouse=[on|off]`` 2320 Enable/disable passing mouse events via vdagent. Default is on. 2321 2322 ``playback-compression=[on|off]`` 2323 Enable/disable audio stream compression (using celt 0.5.1). 2324 Default is on. 2325 2326 ``seamless-migration=[on|off]`` 2327 Enable/disable spice seamless migration. Default is off. 2328 2329 ``gl=[on|off]`` 2330 Enable/disable OpenGL context. Default is off. 2331 2332 ``rendernode=<file>`` 2333 DRM render node for OpenGL rendering. If not specified, it will 2334 pick the first available. (Since 2.9) 2335ERST 2336 2337DEF("portrait", 0, QEMU_OPTION_portrait, 2338 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n", 2339 QEMU_ARCH_ALL) 2340SRST 2341``-portrait`` 2342 Rotate graphical output 90 deg left (only PXA LCD). 2343ERST 2344 2345DEF("rotate", HAS_ARG, QEMU_OPTION_rotate, 2346 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n", 2347 QEMU_ARCH_ALL) 2348SRST 2349``-rotate deg`` 2350 Rotate graphical output some deg left (only PXA LCD). 2351ERST 2352 2353DEF("vga", HAS_ARG, QEMU_OPTION_vga, 2354 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n" 2355 " select video card type\n", QEMU_ARCH_ALL) 2356SRST 2357``-vga type`` 2358 Select type of VGA card to emulate. Valid values for type are 2359 2360 ``cirrus`` 2361 Cirrus Logic GD5446 Video card. All Windows versions starting 2362 from Windows 95 should recognize and use this graphic card. For 2363 optimal performances, use 16 bit color depth in the guest and 2364 the host OS. (This card was the default before QEMU 2.2) 2365 2366 ``std`` 2367 Standard VGA card with Bochs VBE extensions. If your guest OS 2368 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if 2369 you want to use high resolution modes (>= 1280x1024x16) then you 2370 should use this option. (This card is the default since QEMU 2371 2.2) 2372 2373 ``vmware`` 2374 VMWare SVGA-II compatible adapter. Use it if you have 2375 sufficiently recent XFree86/XOrg server or Windows guest with a 2376 driver for this card. 2377 2378 ``qxl`` 2379 QXL paravirtual graphic card. It is VGA compatible (including 2380 VESA 2.0 VBE support). Works best with qxl guest drivers 2381 installed though. Recommended choice when using the spice 2382 protocol. 2383 2384 ``tcx`` 2385 (sun4m only) Sun TCX framebuffer. This is the default 2386 framebuffer for sun4m machines and offers both 8-bit and 24-bit 2387 colour depths at a fixed resolution of 1024x768. 2388 2389 ``cg3`` 2390 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit 2391 framebuffer for sun4m machines available in both 1024x768 2392 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people 2393 wishing to run older Solaris versions. 2394 2395 ``virtio`` 2396 Virtio VGA card. 2397 2398 ``none`` 2399 Disable VGA card. 2400ERST 2401 2402DEF("full-screen", 0, QEMU_OPTION_full_screen, 2403 "-full-screen start in full screen\n", QEMU_ARCH_ALL) 2404SRST 2405``-full-screen`` 2406 Start in full screen. 2407ERST 2408 2409DEF("g", HAS_ARG, QEMU_OPTION_g , 2410 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n", 2411 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K) 2412SRST 2413``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]`` 2414 Set the initial graphical resolution and depth (PPC, SPARC only). 2415 2416 For PPC the default is 800x600x32. 2417 2418 For SPARC with the TCX graphics device, the default is 1024x768x8 2419 with the option of 1024x768x24. For cgthree, the default is 2420 1024x768x8 with the option of 1152x900x8 for people who wish to use 2421 OBP. 2422ERST 2423 2424DEF("vnc", HAS_ARG, QEMU_OPTION_vnc , 2425 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL) 2426SRST 2427``-vnc display[,option[,option[,...]]]`` 2428 Normally, if QEMU is compiled with graphical window support, it 2429 displays output such as guest graphics, guest console, and the QEMU 2430 monitor in a window. With this option, you can have QEMU listen on 2431 VNC display display and redirect the VGA display over the VNC 2432 session. It is very useful to enable the usb tablet device when 2433 using this option (option ``-device usb-tablet``). When using the 2434 VNC display, you must use the ``-k`` parameter to set the keyboard 2435 layout if you are not using en-us. Valid syntax for the display is 2436 2437 ``to=L`` 2438 With this option, QEMU will try next available VNC displays, 2439 until the number L, if the origianlly defined "-vnc display" is 2440 not available, e.g. port 5900+display is already used by another 2441 application. By default, to=0. 2442 2443 ``host:d`` 2444 TCP connections will only be allowed from host on display d. By 2445 convention the TCP port is 5900+d. Optionally, host can be 2446 omitted in which case the server will accept connections from 2447 any host. 2448 2449 ``unix:path`` 2450 Connections will be allowed over UNIX domain sockets where path 2451 is the location of a unix socket to listen for connections on. 2452 2453 ``none`` 2454 VNC is initialized but not started. The monitor ``change`` 2455 command can be used to later start the VNC server. 2456 2457 Following the display value there may be one or more option flags 2458 separated by commas. Valid options are 2459 2460 ``reverse=on|off`` 2461 Connect to a listening VNC client via a "reverse" connection. 2462 The client is specified by the display. For reverse network 2463 connections (host:d,``reverse``), the d argument is a TCP port 2464 number, not a display number. 2465 2466 ``websocket=on|off`` 2467 Opens an additional TCP listening port dedicated to VNC 2468 Websocket connections. If a bare websocket option is given, the 2469 Websocket port is 5700+display. An alternative port can be 2470 specified with the syntax ``websocket``\ =port. 2471 2472 If host is specified connections will only be allowed from this 2473 host. It is possible to control the websocket listen address 2474 independently, using the syntax ``websocket``\ =host:port. 2475 2476 If no TLS credentials are provided, the websocket connection 2477 runs in unencrypted mode. If TLS credentials are provided, the 2478 websocket connection requires encrypted client connections. 2479 2480 ``password=on|off`` 2481 Require that password based authentication is used for client 2482 connections. 2483 2484 The password must be set separately using the ``set_password`` 2485 command in the :ref:`QEMU monitor`. The 2486 syntax to change your password is: 2487 ``set_password <protocol> <password>`` where <protocol> could be 2488 either "vnc" or "spice". 2489 2490 If you would like to change <protocol> password expiration, you 2491 should use ``expire_password <protocol> <expiration-time>`` 2492 where expiration time could be one of the following options: 2493 now, never, +seconds or UNIX time of expiration, e.g. +60 to 2494 make password expire in 60 seconds, or 1335196800 to make 2495 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for 2496 this date and time). 2497 2498 You can also use keywords "now" or "never" for the expiration 2499 time to allow <protocol> password to expire immediately or never 2500 expire. 2501 2502 ``password-secret=<secret-id>`` 2503 Require that password based authentication is used for client 2504 connections, using the password provided by the ``secret`` 2505 object identified by ``secret-id``. 2506 2507 ``tls-creds=ID`` 2508 Provides the ID of a set of TLS credentials to use to secure the 2509 VNC server. They will apply to both the normal VNC server socket 2510 and the websocket socket (if enabled). Setting TLS credentials 2511 will cause the VNC server socket to enable the VeNCrypt auth 2512 mechanism. The credentials should have been previously created 2513 using the ``-object tls-creds`` argument. 2514 2515 ``tls-authz=ID`` 2516 Provides the ID of the QAuthZ authorization object against which 2517 the client's x509 distinguished name will validated. This object 2518 is only resolved at time of use, so can be deleted and recreated 2519 on the fly while the VNC server is active. If missing, it will 2520 default to denying access. 2521 2522 ``sasl=on|off`` 2523 Require that the client use SASL to authenticate with the VNC 2524 server. The exact choice of authentication method used is 2525 controlled from the system / user's SASL configuration file for 2526 the 'qemu' service. This is typically found in 2527 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user, 2528 an environment variable SASL\_CONF\_PATH can be used to make it 2529 search alternate locations for the service config. While some 2530 SASL auth methods can also provide data encryption (eg GSSAPI), 2531 it is recommended that SASL always be combined with the 'tls' 2532 and 'x509' settings to enable use of SSL and server 2533 certificates. This ensures a data encryption preventing 2534 compromise of authentication credentials. See the 2535 :ref:`VNC security` section in the System Emulation Users Guide 2536 for details on using SASL authentication. 2537 2538 ``sasl-authz=ID`` 2539 Provides the ID of the QAuthZ authorization object against which 2540 the client's SASL username will validated. This object is only 2541 resolved at time of use, so can be deleted and recreated on the 2542 fly while the VNC server is active. If missing, it will default 2543 to denying access. 2544 2545 ``acl=on|off`` 2546 Legacy method for enabling authorization of clients against the 2547 x509 distinguished name and SASL username. It results in the 2548 creation of two ``authz-list`` objects with IDs of 2549 ``vnc.username`` and ``vnc.x509dname``. The rules for these 2550 objects must be configured with the HMP ACL commands. 2551 2552 This option is deprecated and should no longer be used. The new 2553 ``sasl-authz`` and ``tls-authz`` options are a replacement. 2554 2555 ``lossy=on|off`` 2556 Enable lossy compression methods (gradient, JPEG, ...). If this 2557 option is set, VNC client may receive lossy framebuffer updates 2558 depending on its encoding settings. Enabling this option can 2559 save a lot of bandwidth at the expense of quality. 2560 2561 ``non-adaptive=on|off`` 2562 Disable adaptive encodings. Adaptive encodings are enabled by 2563 default. An adaptive encoding will try to detect frequently 2564 updated screen regions, and send updates in these regions using 2565 a lossy encoding (like JPEG). This can be really helpful to save 2566 bandwidth when playing videos. Disabling adaptive encodings 2567 restores the original static behavior of encodings like Tight. 2568 2569 ``share=[allow-exclusive|force-shared|ignore]`` 2570 Set display sharing policy. 'allow-exclusive' allows clients to 2571 ask for exclusive access. As suggested by the rfb spec this is 2572 implemented by dropping other connections. Connecting multiple 2573 clients in parallel requires all clients asking for a shared 2574 session (vncviewer: -shared switch). This is the default. 2575 'force-shared' disables exclusive client access. Useful for 2576 shared desktop sessions, where you don't want someone forgetting 2577 specify -shared disconnect everybody else. 'ignore' completely 2578 ignores the shared flag and allows everybody connect 2579 unconditionally. Doesn't conform to the rfb spec but is 2580 traditional QEMU behavior. 2581 2582 ``key-delay-ms`` 2583 Set keyboard delay, for key down and key up events, in 2584 milliseconds. Default is 10. Keyboards are low-bandwidth 2585 devices, so this slowdown can help the device and guest to keep 2586 up and not lose events in case events are arriving in bulk. 2587 Possible causes for the latter are flaky network connections, or 2588 scripts for automated testing. 2589 2590 ``audiodev=audiodev`` 2591 Use the specified audiodev when the VNC client requests audio 2592 transmission. When not using an -audiodev argument, this option 2593 must be omitted, otherwise is must be present and specify a 2594 valid audiodev. 2595 2596 ``power-control=on|off`` 2597 Permit the remote client to issue shutdown, reboot or reset power 2598 control requests. 2599ERST 2600 2601ARCHHEADING(, QEMU_ARCH_I386) 2602 2603ARCHHEADING(i386 target only:, QEMU_ARCH_I386) 2604 2605DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack, 2606 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n", 2607 QEMU_ARCH_I386) 2608SRST 2609``-win2k-hack`` 2610 Use it when installing Windows 2000 to avoid a disk full bug. After 2611 Windows 2000 is installed, you no longer need this option (this 2612 option slows down the IDE transfers). 2613ERST 2614 2615DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk, 2616 "-no-fd-bootchk disable boot signature checking for floppy disks\n", 2617 QEMU_ARCH_I386) 2618SRST 2619``-no-fd-bootchk`` 2620 Disable boot signature checking for floppy disks in BIOS. May be 2621 needed to boot from old floppy disks. 2622ERST 2623 2624DEF("no-acpi", 0, QEMU_OPTION_no_acpi, 2625 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2626SRST 2627``-no-acpi`` 2628 Disable ACPI (Advanced Configuration and Power Interface) support. 2629 Use it if your guest OS complains about ACPI problems (PC target 2630 machine only). 2631ERST 2632 2633DEF("no-hpet", 0, QEMU_OPTION_no_hpet, 2634 "-no-hpet disable HPET\n", QEMU_ARCH_I386) 2635SRST 2636``-no-hpet`` 2637 Disable HPET support. Deprecated, use '-machine hpet=off' instead. 2638ERST 2639 2640DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable, 2641 "-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" 2642 " ACPI table description\n", QEMU_ARCH_I386) 2643SRST 2644``-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]...]`` 2645 Add ACPI table with specified header fields and context from 2646 specified files. For file=, take whole ACPI table from the specified 2647 files, including all ACPI headers (possible overridden by other 2648 options). For data=, only data portion of the table is used, all 2649 header information is specified in the command line. If a SLIC table 2650 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id 2651 fields will override the same in the RSDT and the FADT (a.k.a. 2652 FACP), in order to ensure the field matches required by the 2653 Microsoft SLIC spec and the ACPI spec. 2654ERST 2655 2656DEF("smbios", HAS_ARG, QEMU_OPTION_smbios, 2657 "-smbios file=binary\n" 2658 " load SMBIOS entry from binary file\n" 2659 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n" 2660 " [,uefi=on|off]\n" 2661 " specify SMBIOS type 0 fields\n" 2662 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2663 " [,uuid=uuid][,sku=str][,family=str]\n" 2664 " specify SMBIOS type 1 fields\n" 2665 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2666 " [,asset=str][,location=str]\n" 2667 " specify SMBIOS type 2 fields\n" 2668 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n" 2669 " [,sku=str]\n" 2670 " specify SMBIOS type 3 fields\n" 2671 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n" 2672 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n" 2673 " [,processor-id=%d]\n" 2674 " specify SMBIOS type 4 fields\n" 2675 "-smbios type=8[,external_reference=str][,internal_reference=str][,connector_type=%d][,port_type=%d]\n" 2676 " specify SMBIOS type 8 fields\n" 2677 "-smbios type=11[,value=str][,path=filename]\n" 2678 " specify SMBIOS type 11 fields\n" 2679 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n" 2680 " [,asset=str][,part=str][,speed=%d]\n" 2681 " specify SMBIOS type 17 fields\n" 2682 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n" 2683 " specify SMBIOS type 41 fields\n", 2684 QEMU_ARCH_I386 | QEMU_ARCH_ARM | QEMU_ARCH_LOONGARCH) 2685SRST 2686``-smbios file=binary`` 2687 Load SMBIOS entry from binary file. 2688 2689``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]`` 2690 Specify SMBIOS type 0 fields 2691 2692``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]`` 2693 Specify SMBIOS type 1 fields 2694 2695``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]`` 2696 Specify SMBIOS type 2 fields 2697 2698``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]`` 2699 Specify SMBIOS type 3 fields 2700 2701``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-id=%d]`` 2702 Specify SMBIOS type 4 fields 2703 2704``-smbios type=11[,value=str][,path=filename]`` 2705 Specify SMBIOS type 11 fields 2706 2707 This argument can be repeated multiple times, and values are added in the order they are parsed. 2708 Applications intending to use OEM strings data are encouraged to use their application name as 2709 a prefix for the value string. This facilitates passing information for multiple applications 2710 concurrently. 2711 2712 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax 2713 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes. 2714 2715 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to 2716 the SMBIOS table in the order in which they appear. 2717 2718 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535 2719 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the 2720 guest. Instead it should be used as a indicator to inform the guest where to locate the real 2721 data set, for example, by specifying the serial ID of a block device. 2722 2723 An example passing three strings is 2724 2725 .. parsed-literal:: 2726 2727 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\ 2728 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\ 2729 path=/some/file/with/oemstringsdata.txt 2730 2731 In the guest OS this is visible with the ``dmidecode`` command 2732 2733 .. parsed-literal:: 2734 2735 $ dmidecode -t 11 2736 Handle 0x0E00, DMI type 11, 5 bytes 2737 OEM Strings 2738 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/ 2739 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os 2740 String 3: myapp:some extra data 2741 2742 2743``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]`` 2744 Specify SMBIOS type 17 fields 2745 2746``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]`` 2747 Specify SMBIOS type 41 fields 2748 2749 This argument can be repeated multiple times. Its main use is to allow network interfaces be created 2750 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface 2751 position on the PCI bus. 2752 2753 Here is an example of use: 2754 2755 .. parsed-literal:: 2756 2757 -netdev user,id=internet \\ 2758 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\ 2759 -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev 2760 2761 In the guest OS, the device should then appear as ``eno1``: 2762 2763 ..parsed-literal:: 2764 2765 $ ip -brief l 2766 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP> 2767 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP> 2768 2769 Currently, the PCI device has to be attached to the root bus. 2770 2771ERST 2772 2773DEFHEADING() 2774 2775DEFHEADING(Network options:) 2776 2777DEF("netdev", HAS_ARG, QEMU_OPTION_netdev, 2778#ifdef CONFIG_SLIRP 2779 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n" 2780 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n" 2781 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n" 2782 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n" 2783 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]" 2784#ifndef _WIN32 2785 "[,smb=dir[,smbserver=addr]]\n" 2786#endif 2787 " configure a user mode network backend with ID 'str',\n" 2788 " its DHCP server and optional services\n" 2789#endif 2790#ifdef _WIN32 2791 "-netdev tap,id=str,ifname=name\n" 2792 " configure a host TAP network backend with ID 'str'\n" 2793#else 2794 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n" 2795 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n" 2796 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n" 2797 " [,poll-us=n]\n" 2798 " configure a host TAP network backend with ID 'str'\n" 2799 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2800 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n" 2801 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n" 2802 " to deconfigure it\n" 2803 " use '[down]script=no' to disable script execution\n" 2804 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n" 2805 " configure it\n" 2806 " use 'fd=h' to connect to an already opened TAP interface\n" 2807 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n" 2808 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n" 2809 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n" 2810 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n" 2811 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n" 2812 " use vhost=on to enable experimental in kernel accelerator\n" 2813 " (only has effect for virtio guests which use MSIX)\n" 2814 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n" 2815 " use 'vhostfd=h' to connect to an already opened vhost net device\n" 2816 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n" 2817 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n" 2818 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n" 2819 " spent on busy polling for vhost net\n" 2820 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n" 2821 " configure a host TAP network backend with ID 'str' that is\n" 2822 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2823 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n" 2824#endif 2825#ifdef __linux__ 2826 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n" 2827 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n" 2828 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n" 2829 " [,rxcookie=rxcookie][,offset=offset]\n" 2830 " configure a network backend with ID 'str' connected to\n" 2831 " an Ethernet over L2TPv3 pseudowire.\n" 2832 " Linux kernel 3.3+ as well as most routers can talk\n" 2833 " L2TPv3. This transport allows connecting a VM to a VM,\n" 2834 " VM to a router and even VM to Host. It is a nearly-universal\n" 2835 " standard (RFC3931). Note - this implementation uses static\n" 2836 " pre-configured tunnels (same as the Linux kernel).\n" 2837 " use 'src=' to specify source address\n" 2838 " use 'dst=' to specify destination address\n" 2839 " use 'udp=on' to specify udp encapsulation\n" 2840 " use 'srcport=' to specify source udp port\n" 2841 " use 'dstport=' to specify destination udp port\n" 2842 " use 'ipv6=on' to force v6\n" 2843 " L2TPv3 uses cookies to prevent misconfiguration as\n" 2844 " well as a weak security measure\n" 2845 " use 'rxcookie=0x012345678' to specify a rxcookie\n" 2846 " use 'txcookie=0x012345678' to specify a txcookie\n" 2847 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n" 2848 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n" 2849 " use 'pincounter=on' to work around broken counter handling in peer\n" 2850 " use 'offset=X' to add an extra offset between header and data\n" 2851#endif 2852 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n" 2853 " configure a network backend to connect to another network\n" 2854 " using a socket connection\n" 2855 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n" 2856 " configure a network backend to connect to a multicast maddr and port\n" 2857 " use 'localaddr=addr' to specify the host address to send packets from\n" 2858 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n" 2859 " configure a network backend to connect to another network\n" 2860 " using an UDP tunnel\n" 2861 "-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" 2862 "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off][,reconnect=seconds]\n" 2863 "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor[,reconnect=seconds]\n" 2864 " configure a network backend to connect to another network\n" 2865 " using a socket connection in stream mode.\n" 2866 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n" 2867 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]\n" 2868 " configure a network backend to connect to a multicast maddr and port\n" 2869 " use ``local.host=addr`` to specify the host address to send packets from\n" 2870 "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n" 2871 "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n" 2872 "-netdev dgram,id=str,local.type=fd,local.str=file-descriptor\n" 2873 " configure a network backend to connect to another network\n" 2874 " using an UDP tunnel\n" 2875#ifdef CONFIG_VDE 2876 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n" 2877 " configure a network backend to connect to port 'n' of a vde switch\n" 2878 " running on host and listening for incoming connections on 'socketpath'.\n" 2879 " Use group 'groupname' and mode 'octalmode' to change default\n" 2880 " ownership and permissions for communication port.\n" 2881#endif 2882#ifdef CONFIG_NETMAP 2883 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n" 2884 " attach to the existing netmap-enabled network interface 'name', or to a\n" 2885 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n" 2886 " netmap device, defaults to '/dev/netmap')\n" 2887#endif 2888#ifdef CONFIG_AF_XDP 2889 "-netdev af-xdp,id=str,ifname=name[,mode=native|skb][,force-copy=on|off]\n" 2890 " [,queues=n][,start-queue=m][,inhibit=on|off][,sock-fds=x:y:...:z]\n" 2891 " attach to the existing network interface 'name' with AF_XDP socket\n" 2892 " use 'mode=MODE' to specify an XDP program attach mode\n" 2893 " use 'force-copy=on|off' to force XDP copy mode even if device supports zero-copy (default: off)\n" 2894 " use 'inhibit=on|off' to inhibit loading of a default XDP program (default: off)\n" 2895 " with inhibit=on,\n" 2896 " use 'sock-fds' to provide file descriptors for already open AF_XDP sockets\n" 2897 " added to a socket map in XDP program. One socket per queue.\n" 2898 " use 'queues=n' to specify how many queues of a multiqueue interface should be used\n" 2899 " use 'start-queue=m' to specify the first queue that should be used\n" 2900#endif 2901#ifdef CONFIG_POSIX 2902 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n" 2903 " configure a vhost-user network, backed by a chardev 'dev'\n" 2904#endif 2905#ifdef __linux__ 2906 "-netdev vhost-vdpa,id=str[,vhostdev=/path/to/dev][,vhostfd=h]\n" 2907 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n" 2908 " use 'vhostdev=/path/to/dev' to open a vhost vdpa device\n" 2909 " use 'vhostfd=h' to connect to an already opened vhost vdpa device\n" 2910#endif 2911#ifdef CONFIG_VMNET 2912 "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n" 2913 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" 2914 " configure a vmnet network backend in host mode with ID 'str',\n" 2915 " isolate this interface from others with 'isolated',\n" 2916 " configure the address range and choose a subnet mask,\n" 2917 " specify network UUID 'uuid' to disable DHCP and interact with\n" 2918 " vmnet-host interfaces within this isolated network\n" 2919 "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n" 2920 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" 2921 " configure a vmnet network backend in shared mode with ID 'str',\n" 2922 " configure the address range and choose a subnet mask,\n" 2923 " set IPv6 ULA prefix (of length 64) to use for internal network,\n" 2924 " isolate this interface from others with 'isolated'\n" 2925 "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n" 2926 " configure a vmnet network backend in bridged mode with ID 'str',\n" 2927 " use 'ifname=name' to select a physical network interface to be bridged,\n" 2928 " isolate this interface from others with 'isolated'\n" 2929#endif 2930 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n" 2931 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL) 2932DEF("nic", HAS_ARG, QEMU_OPTION_nic, 2933 "-nic [tap|bridge|" 2934#ifdef CONFIG_SLIRP 2935 "user|" 2936#endif 2937#ifdef __linux__ 2938 "l2tpv3|" 2939#endif 2940#ifdef CONFIG_VDE 2941 "vde|" 2942#endif 2943#ifdef CONFIG_NETMAP 2944 "netmap|" 2945#endif 2946#ifdef CONFIG_AF_XDP 2947 "af-xdp|" 2948#endif 2949#ifdef CONFIG_POSIX 2950 "vhost-user|" 2951#endif 2952#ifdef CONFIG_VMNET 2953 "vmnet-host|vmnet-shared|vmnet-bridged|" 2954#endif 2955 "socket][,option][,...][mac=macaddr]\n" 2956 " initialize an on-board / default host NIC (using MAC address\n" 2957 " macaddr) and connect it to the given host network backend\n" 2958 "-nic none use it alone to have zero network devices (the default is to\n" 2959 " provided a 'user' network connection)\n", 2960 QEMU_ARCH_ALL) 2961DEF("net", HAS_ARG, QEMU_OPTION_net, 2962 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n" 2963 " configure or create an on-board (or machine default) NIC and\n" 2964 " connect it to hub 0 (please use -nic unless you need a hub)\n" 2965 "-net [" 2966#ifdef CONFIG_SLIRP 2967 "user|" 2968#endif 2969 "tap|" 2970 "bridge|" 2971#ifdef CONFIG_VDE 2972 "vde|" 2973#endif 2974#ifdef CONFIG_NETMAP 2975 "netmap|" 2976#endif 2977#ifdef CONFIG_AF_XDP 2978 "af-xdp|" 2979#endif 2980#ifdef CONFIG_VMNET 2981 "vmnet-host|vmnet-shared|vmnet-bridged|" 2982#endif 2983 "socket][,option][,option][,...]\n" 2984 " old way to initialize a host network interface\n" 2985 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL) 2986SRST 2987``-nic [tap|bridge|user|l2tpv3|vde|netmap|af-xdp|vhost-user|socket][,...][,mac=macaddr][,model=mn]`` 2988 This option is a shortcut for configuring both the on-board 2989 (default) guest NIC hardware and the host network backend in one go. 2990 The host backend options are the same as with the corresponding 2991 ``-netdev`` options below. The guest NIC model can be set with 2992 ``model=modelname``. Use ``model=help`` to list the available device 2993 types. The hardware MAC address can be set with ``mac=macaddr``. 2994 2995 The following two example do exactly the same, to show how ``-nic`` 2996 can be used to shorten the command line length: 2997 2998 .. parsed-literal:: 2999 3000 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32 3001 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32 3002 3003``-nic none`` 3004 Indicate that no network devices should be configured. It is used to 3005 override the default configuration (default NIC with "user" host 3006 network backend) which is activated if no other networking options 3007 are provided. 3008 3009``-netdev user,id=id[,option][,option][,...]`` 3010 Configure user mode host network backend which requires no 3011 administrator privilege to run. Valid options are: 3012 3013 ``id=id`` 3014 Assign symbolic name for use in monitor commands. 3015 3016 ``ipv4=on|off and ipv6=on|off`` 3017 Specify that either IPv4 or IPv6 must be enabled. If neither is 3018 specified both protocols are enabled. 3019 3020 ``net=addr[/mask]`` 3021 Set IP network address the guest will see. Optionally specify 3022 the netmask, either in the form a.b.c.d or as number of valid 3023 top-most bits. Default is 10.0.2.0/24. 3024 3025 ``host=addr`` 3026 Specify the guest-visible address of the host. Default is the 3027 2nd IP in the guest network, i.e. x.x.x.2. 3028 3029 ``ipv6-net=addr[/int]`` 3030 Set IPv6 network address the guest will see (default is 3031 fec0::/64). The network prefix is given in the usual hexadecimal 3032 IPv6 address notation. The prefix size is optional, and is given 3033 as the number of valid top-most bits (default is 64). 3034 3035 ``ipv6-host=addr`` 3036 Specify the guest-visible IPv6 address of the host. Default is 3037 the 2nd IPv6 in the guest network, i.e. xxxx::2. 3038 3039 ``restrict=on|off`` 3040 If this option is enabled, the guest will be isolated, i.e. it 3041 will not be able to contact the host and no guest IP packets 3042 will be routed over the host to the outside. This option does 3043 not affect any explicitly set forwarding rules. 3044 3045 ``hostname=name`` 3046 Specifies the client hostname reported by the built-in DHCP 3047 server. 3048 3049 ``dhcpstart=addr`` 3050 Specify the first of the 16 IPs the built-in DHCP server can 3051 assign. Default is the 15th to 31st IP in the guest network, 3052 i.e. x.x.x.15 to x.x.x.31. 3053 3054 ``dns=addr`` 3055 Specify the guest-visible address of the virtual nameserver. The 3056 address must be different from the host address. Default is the 3057 3rd IP in the guest network, i.e. x.x.x.3. 3058 3059 ``ipv6-dns=addr`` 3060 Specify the guest-visible address of the IPv6 virtual 3061 nameserver. The address must be different from the host address. 3062 Default is the 3rd IP in the guest network, i.e. xxxx::3. 3063 3064 ``dnssearch=domain`` 3065 Provides an entry for the domain-search list sent by the 3066 built-in DHCP server. More than one domain suffix can be 3067 transmitted by specifying this option multiple times. If 3068 supported, this will cause the guest to automatically try to 3069 append the given domain suffix(es) in case a domain name can not 3070 be resolved. 3071 3072 Example: 3073 3074 .. parsed-literal:: 3075 3076 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org 3077 3078 ``domainname=domain`` 3079 Specifies the client domain name reported by the built-in DHCP 3080 server. 3081 3082 ``tftp=dir`` 3083 When using the user mode network stack, activate a built-in TFTP 3084 server. The files in dir will be exposed as the root of a TFTP 3085 server. The TFTP client on the guest must be configured in 3086 binary mode (use the command ``bin`` of the Unix TFTP client). 3087 3088 ``tftp-server-name=name`` 3089 In BOOTP reply, broadcast name as the "TFTP server name" 3090 (RFC2132 option 66). This can be used to advise the guest to 3091 load boot files or configurations from a different server than 3092 the host address. 3093 3094 ``bootfile=file`` 3095 When using the user mode network stack, broadcast file as the 3096 BOOTP filename. In conjunction with ``tftp``, this can be used 3097 to network boot a guest from a local directory. 3098 3099 Example (using pxelinux): 3100 3101 .. parsed-literal:: 3102 3103 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\ 3104 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0 3105 3106 ``smb=dir[,smbserver=addr]`` 3107 When using the user mode network stack, activate a built-in SMB 3108 server so that Windows OSes can access to the host files in 3109 ``dir`` transparently. The IP address of the SMB server can be 3110 set to addr. By default the 4th IP in the guest network is used, 3111 i.e. x.x.x.4. 3112 3113 In the guest Windows OS, the line: 3114 3115 :: 3116 3117 10.0.2.4 smbserver 3118 3119 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows 3120 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows 3121 NT/2000). 3122 3123 Then ``dir`` can be accessed in ``\\smbserver\qemu``. 3124 3125 Note that a SAMBA server must be installed on the host OS. 3126 3127 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport`` 3128 Redirect incoming TCP or UDP connections to the host port 3129 hostport to the guest IP address guestaddr on guest port 3130 guestport. If guestaddr is not specified, its value is x.x.x.15 3131 (default first address given by the built-in DHCP server). By 3132 specifying hostaddr, the rule can be bound to a specific host 3133 interface. If no connection type is set, TCP is used. This 3134 option can be given multiple times. 3135 3136 For example, to redirect host X11 connection from screen 1 to 3137 guest screen 0, use the following: 3138 3139 .. parsed-literal:: 3140 3141 # on the host 3142 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000 3143 # this host xterm should open in the guest X11 server 3144 xterm -display :1 3145 3146 To redirect telnet connections from host port 5555 to telnet 3147 port on the guest, use the following: 3148 3149 .. parsed-literal:: 3150 3151 # on the host 3152 |qemu_system| -nic user,hostfwd=tcp::5555-:23 3153 telnet localhost 5555 3154 3155 Then when you use on the host ``telnet localhost 5555``, you 3156 connect to the guest telnet server. 3157 3158 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command`` 3159 Forward guest TCP connections to the IP address server on port 3160 port to the character device dev or to a program executed by 3161 cmd:command which gets spawned for each connection. This option 3162 can be given multiple times. 3163 3164 You can either use a chardev directly and have that one used 3165 throughout QEMU's lifetime, like in the following example: 3166 3167 .. parsed-literal:: 3168 3169 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever 3170 # the guest accesses it 3171 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321 3172 3173 Or you can execute a command on every TCP connection established 3174 by the guest, so that QEMU behaves similar to an inetd process 3175 for that virtual server: 3176 3177 .. parsed-literal:: 3178 3179 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234 3180 # and connect the TCP stream to its stdin/stdout 3181 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321' 3182 3183``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]`` 3184 Configure a host TAP network backend with ID id. 3185 3186 Use the network script file to configure it and the network script 3187 dfile to deconfigure it. If name is not provided, the OS 3188 automatically provides one. The default network configure script is 3189 ``/etc/qemu-ifup`` and the default network deconfigure script is 3190 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to 3191 disable script execution. 3192 3193 If running QEMU as an unprivileged user, use the network helper 3194 to configure the TAP interface and attach it to the bridge. 3195 The default network helper executable is 3196 ``/path/to/qemu-bridge-helper`` and the default bridge device is 3197 ``br0``. 3198 3199 ``fd``\ =h can be used to specify the handle of an already opened 3200 host TAP interface. 3201 3202 Examples: 3203 3204 .. parsed-literal:: 3205 3206 #launch a QEMU instance with the default network script 3207 |qemu_system| linux.img -nic tap 3208 3209 .. parsed-literal:: 3210 3211 #launch a QEMU instance with two NICs, each one connected 3212 #to a TAP device 3213 |qemu_system| linux.img \\ 3214 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\ 3215 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1 3216 3217 .. parsed-literal:: 3218 3219 #launch a QEMU instance with the default network helper to 3220 #connect a TAP device to bridge br0 3221 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3222 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper" 3223 3224``-netdev bridge,id=id[,br=bridge][,helper=helper]`` 3225 Connect a host TAP network interface to a host bridge device. 3226 3227 Use the network helper helper to configure the TAP interface and 3228 attach it to the bridge. The default network helper executable is 3229 ``/path/to/qemu-bridge-helper`` and the default bridge device is 3230 ``br0``. 3231 3232 Examples: 3233 3234 .. parsed-literal:: 3235 3236 #launch a QEMU instance with the default network helper to 3237 #connect a TAP device to bridge br0 3238 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1 3239 3240 .. parsed-literal:: 3241 3242 #launch a QEMU instance with the default network helper to 3243 #connect a TAP device to bridge qemubr0 3244 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1 3245 3246``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]`` 3247 This host network backend can be used to connect the guest's network 3248 to another QEMU virtual machine using a TCP socket connection. If 3249 ``listen`` is specified, QEMU waits for incoming connections on port 3250 (host is optional). ``connect`` is used to connect to another QEMU 3251 instance using the ``listen`` option. ``fd``\ =h specifies an 3252 already opened TCP socket. 3253 3254 Example: 3255 3256 .. parsed-literal:: 3257 3258 # launch a first QEMU instance 3259 |qemu_system| linux.img \\ 3260 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3261 -netdev socket,id=n1,listen=:1234 3262 # connect the network of this instance to the network of the first instance 3263 |qemu_system| linux.img \\ 3264 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3265 -netdev socket,id=n2,connect=127.0.0.1:1234 3266 3267``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]`` 3268 Configure a socket host network backend to share the guest's network 3269 traffic with another QEMU virtual machines using a UDP multicast 3270 socket, effectively making a bus for every QEMU with same multicast 3271 address maddr and port. NOTES: 3272 3273 1. Several QEMU can be running on different hosts and share same bus 3274 (assuming correct multicast setup for these hosts). 3275 3276 2. mcast support is compatible with User Mode Linux (argument 3277 ``ethN=mcast``), see http://user-mode-linux.sf.net. 3278 3279 3. Use ``fd=h`` to specify an already opened UDP multicast socket. 3280 3281 Example: 3282 3283 .. parsed-literal:: 3284 3285 # launch one QEMU instance 3286 |qemu_system| linux.img \\ 3287 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3288 -netdev socket,id=n1,mcast=230.0.0.1:1234 3289 # launch another QEMU instance on same "bus" 3290 |qemu_system| linux.img \\ 3291 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3292 -netdev socket,id=n2,mcast=230.0.0.1:1234 3293 # launch yet another QEMU instance on same "bus" 3294 |qemu_system| linux.img \\ 3295 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\ 3296 -netdev socket,id=n3,mcast=230.0.0.1:1234 3297 3298 Example (User Mode Linux compat.): 3299 3300 .. parsed-literal:: 3301 3302 # launch QEMU instance (note mcast address selected is UML's default) 3303 |qemu_system| linux.img \\ 3304 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3305 -netdev socket,id=n1,mcast=239.192.168.1:1102 3306 # launch UML 3307 /path/to/linux ubd0=/path/to/root_fs eth0=mcast 3308 3309 Example (send packets from host's 1.2.3.4): 3310 3311 .. parsed-literal:: 3312 3313 |qemu_system| linux.img \\ 3314 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3315 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4 3316 3317``-netdev l2tpv3,id=id,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport],txsession=txsession[,rxsession=rxsession][,ipv6=on|off][,udp=on|off][,cookie64][,counter][,pincounter][,txcookie=txcookie][,rxcookie=rxcookie][,offset=offset]`` 3318 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931) 3319 is a popular protocol to transport Ethernet (and other Layer 2) data 3320 frames between two systems. It is present in routers, firewalls and 3321 the Linux kernel (from version 3.3 onwards). 3322 3323 This transport allows a VM to communicate to another VM, router or 3324 firewall directly. 3325 3326 ``src=srcaddr`` 3327 source address (mandatory) 3328 3329 ``dst=dstaddr`` 3330 destination address (mandatory) 3331 3332 ``udp`` 3333 select udp encapsulation (default is ip). 3334 3335 ``srcport=srcport`` 3336 source udp port. 3337 3338 ``dstport=dstport`` 3339 destination udp port. 3340 3341 ``ipv6`` 3342 force v6, otherwise defaults to v4. 3343 3344 ``rxcookie=rxcookie``; \ ``txcookie=txcookie`` 3345 Cookies are a weak form of security in the l2tpv3 specification. 3346 Their function is mostly to prevent misconfiguration. By default 3347 they are 32 bit. 3348 3349 ``cookie64`` 3350 Set cookie size to 64 bit instead of the default 32 3351 3352 ``counter=off`` 3353 Force a 'cut-down' L2TPv3 with no counter as in 3354 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00 3355 3356 ``pincounter=on`` 3357 Work around broken counter handling in peer. This may also help 3358 on networks which have packet reorder. 3359 3360 ``offset=offset`` 3361 Add an extra offset between header and data 3362 3363 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to 3364 the bridge br-lan on the remote Linux host 1.2.3.4: 3365 3366 .. parsed-literal:: 3367 3368 # Setup tunnel on linux host using raw ip as encapsulation 3369 # on 1.2.3.4 3370 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\ 3371 encap udp udp_sport 16384 udp_dport 16384 3372 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\ 3373 0xFFFFFFFF peer_session_id 0xFFFFFFFF 3374 ifconfig vmtunnel0 mtu 1500 3375 ifconfig vmtunnel0 up 3376 brctl addif br-lan vmtunnel0 3377 3378 3379 # on 4.3.2.1 3380 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter 3381 3382 |qemu_system| linux.img -device e1000,netdev=n1 \\ 3383 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter 3384 3385``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]`` 3386 Configure VDE backend to connect to PORT n of a vde switch running 3387 on host and listening for incoming connections on socketpath. Use 3388 GROUP groupname and MODE octalmode to change default ownership and 3389 permissions for communication port. This option is only available if 3390 QEMU has been compiled with vde support enabled. 3391 3392 Example: 3393 3394 .. parsed-literal:: 3395 3396 # launch vde switch 3397 vde_switch -F -sock /tmp/myswitch 3398 # launch QEMU instance 3399 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch 3400 3401``-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]`` 3402 Configure AF_XDP backend to connect to a network interface 'name' 3403 using AF_XDP socket. A specific program attach mode for a default 3404 XDP program can be forced with 'mode', defaults to best-effort, 3405 where the likely most performant mode will be in use. Number of queues 3406 'n' should generally match the number or queues in the interface, 3407 defaults to 1. Traffic arriving on non-configured device queues will 3408 not be delivered to the network backend. 3409 3410 .. parsed-literal:: 3411 3412 # set number of queues to 4 3413 ethtool -L eth0 combined 4 3414 # launch QEMU instance 3415 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3416 -netdev af-xdp,id=n1,ifname=eth0,queues=4 3417 3418 'start-queue' option can be specified if a particular range of queues 3419 [m, m + n] should be in use. For example, this is may be necessary in 3420 order to use certain NICs in native mode. Kernel allows the driver to 3421 create a separate set of XDP queues on top of regular ones, and only 3422 these queues can be used for AF_XDP sockets. NICs that work this way 3423 may also require an additional traffic redirection with ethtool to these 3424 special queues. 3425 3426 .. parsed-literal:: 3427 3428 # set number of queues to 1 3429 ethtool -L eth0 combined 1 3430 # redirect all the traffic to the second queue (id: 1) 3431 # note: drivers may require non-empty key/mask pair. 3432 ethtool -N eth0 flow-type ether \\ 3433 dst 00:00:00:00:00:00 m FF:FF:FF:FF:FF:FE action 1 3434 ethtool -N eth0 flow-type ether \\ 3435 dst 00:00:00:00:00:01 m FF:FF:FF:FF:FF:FE action 1 3436 # launch QEMU instance 3437 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3438 -netdev af-xdp,id=n1,ifname=eth0,queues=1,start-queue=1 3439 3440 XDP program can also be loaded externally. In this case 'inhibit' option 3441 should be set to 'on' and 'sock-fds' provided with file descriptors for 3442 already open but not bound XDP sockets already added to a socket map for 3443 corresponding queues. One socket per queue. 3444 3445 .. parsed-literal:: 3446 3447 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3448 -netdev af-xdp,id=n1,ifname=eth0,queues=3,inhibit=on,sock-fds=15:16:17 3449 3450``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]`` 3451 Establish a vhost-user netdev, backed by a chardev id. The chardev 3452 should be a unix domain socket backed one. The vhost-user uses a 3453 specifically defined protocol to pass vhost ioctl replacement 3454 messages to an application on the other end of the socket. On 3455 non-MSIX guests, the feature can be forced with vhostforce. Use 3456 'queues=n' to specify the number of queues to be created for 3457 multiqueue vhost-user. 3458 3459 Example: 3460 3461 :: 3462 3463 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \ 3464 -numa node,memdev=mem \ 3465 -chardev socket,id=chr0,path=/path/to/socket \ 3466 -netdev type=vhost-user,id=net0,chardev=chr0 \ 3467 -device virtio-net-pci,netdev=net0 3468 3469``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]`` 3470 Establish a vhost-vdpa netdev. 3471 3472 vDPA device is a device that uses a datapath which complies with 3473 the virtio specifications with a vendor specific control path. 3474 vDPA devices can be both physically located on the hardware or 3475 emulated by software. 3476 3477``-netdev hubport,id=id,hubid=hubid[,netdev=nd]`` 3478 Create a hub port on the emulated hub with ID hubid. 3479 3480 The hubport netdev lets you connect a NIC to a QEMU emulated hub 3481 instead of a single netdev. Alternatively, you can also connect the 3482 hubport to another netdev with ID nd by using the ``netdev=nd`` 3483 option. 3484 3485``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]`` 3486 Legacy option to configure or create an on-board (or machine 3487 default) Network Interface Card(NIC) and connect it either to the 3488 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd. 3489 If model is omitted, then the default NIC model associated with the 3490 machine type is used. Note that the default NIC model may change in 3491 future QEMU releases, so it is highly recommended to always specify 3492 a model. Optionally, the MAC address can be changed to mac, the 3493 device address set to addr (PCI cards only), and a name can be 3494 assigned for use in monitor commands. Optionally, for PCI cards, you 3495 can specify the number v of MSI-X vectors that the card should have; 3496 this option currently only affects virtio cards; set v = 0 to 3497 disable MSI-X. If no ``-net`` option is specified, a single NIC is 3498 created. QEMU can emulate several different models of network card. 3499 Use ``-net nic,model=help`` for a list of available devices for your 3500 target. 3501 3502``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]`` 3503 Configure a host network backend (with the options corresponding to 3504 the same ``-netdev`` option) and connect it to the emulated hub 0 3505 (the default hub). Use name to specify the name of the hub port. 3506ERST 3507 3508DEFHEADING() 3509 3510DEFHEADING(Character device options:) 3511 3512DEF("chardev", HAS_ARG, QEMU_OPTION_chardev, 3513 "-chardev help\n" 3514 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3515 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n" 3516 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n" 3517 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n" 3518 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n" 3519 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n" 3520 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n" 3521 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n" 3522 " [,logfile=PATH][,logappend=on|off]\n" 3523 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3524 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n" 3525 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3526 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n" 3527 "-chardev file,id=id,path=path[,input-path=input-file][,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3528 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3529#ifdef _WIN32 3530 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3531 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3532#else 3533 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3534 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n" 3535#endif 3536#ifdef CONFIG_BRLAPI 3537 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3538#endif 3539#if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \ 3540 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) 3541 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3542#endif 3543#if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__) 3544 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3545#endif 3546#if defined(CONFIG_SPICE) 3547 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3548 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3549#endif 3550 , QEMU_ARCH_ALL 3551) 3552 3553SRST 3554The general form of a character device option is: 3555 3556``-chardev backend,id=id[,mux=on|off][,options]`` 3557 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``, 3558 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``, 3559 ``pty``, ``stdio``, ``braille``, ``parallel``, 3560 ``spicevmc``, ``spiceport``. The specific backend will determine the 3561 applicable options. 3562 3563 Use ``-chardev help`` to print all available chardev backend types. 3564 3565 All devices must have an id, which can be any string up to 127 3566 characters long. It is used to uniquely identify this device in 3567 other command line directives. 3568 3569 A character device may be used in multiplexing mode by multiple 3570 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is 3571 a "1:N" device, and here the "1" end is your specified chardev 3572 backend, and the "N" end is the various parts of QEMU that can talk 3573 to a chardev. If you create a chardev with ``id=myid`` and 3574 ``mux=on``, QEMU will create a multiplexer with your specified ID, 3575 and you can then configure multiple front ends to use that chardev 3576 ID for their input/output. Up to four different front ends can be 3577 connected to a single multiplexed chardev. (Without multiplexing 3578 enabled, a chardev can only be used by a single front end.) For 3579 instance you could use this to allow a single stdio chardev to be 3580 used by two serial ports and the QEMU monitor: 3581 3582 :: 3583 3584 -chardev stdio,mux=on,id=char0 \ 3585 -mon chardev=char0,mode=readline \ 3586 -serial chardev:char0 \ 3587 -serial chardev:char0 3588 3589 You can have more than one multiplexer in a system configuration; 3590 for instance you could have a TCP port multiplexed between UART 0 3591 and UART 1, and stdio multiplexed between the QEMU monitor and a 3592 parallel port: 3593 3594 :: 3595 3596 -chardev stdio,mux=on,id=char0 \ 3597 -mon chardev=char0,mode=readline \ 3598 -parallel chardev:char0 \ 3599 -chardev tcp,...,mux=on,id=char1 \ 3600 -serial chardev:char1 \ 3601 -serial chardev:char1 3602 3603 When you're using a multiplexed character device, some escape 3604 sequences are interpreted in the input. See the chapter about 3605 :ref:`keys in the character backend multiplexer` in the 3606 System Emulation Users Guide for more details. 3607 3608 Note that some other command line options may implicitly create 3609 multiplexed character backends; for instance ``-serial mon:stdio`` 3610 creates a multiplexed stdio backend connected to the serial port and 3611 the QEMU monitor, and ``-nographic`` also multiplexes the console 3612 and the monitor to stdio. 3613 3614 There is currently no support for multiplexing in the other 3615 direction (where a single QEMU front end takes input and output from 3616 multiple chardevs). 3617 3618 Every backend supports the ``logfile`` option, which supplies the 3619 path to a file to record all data transmitted via the backend. The 3620 ``logappend`` option controls whether the log file will be truncated 3621 or appended to when opened. 3622 3623The available backends are: 3624 3625``-chardev null,id=id`` 3626 A void device. This device will not emit any data, and will drop any 3627 data it receives. The null backend does not take any options. 3628 3629``-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]`` 3630 Create a two-way stream socket, which can be either a TCP or a unix 3631 socket. A unix socket will be created if ``path`` is specified. 3632 Behaviour is undefined if TCP options are specified for a unix 3633 socket. 3634 3635 ``server=on|off`` specifies that the socket shall be a listening socket. 3636 3637 ``wait=on|off`` specifies that QEMU should not block waiting for a client 3638 to connect to a listening socket. 3639 3640 ``telnet=on|off`` specifies that traffic on the socket should interpret 3641 telnet escape sequences. 3642 3643 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for 3644 communication. 3645 3646 ``reconnect`` sets the timeout for reconnecting on non-server 3647 sockets when the remote end goes away. qemu will delay this many 3648 seconds and then attempt to reconnect. Zero disables reconnecting, 3649 and is the default. 3650 3651 ``tls-creds`` requests enablement of the TLS protocol for 3652 encryption, and specifies the id of the TLS credentials to use for 3653 the handshake. The credentials must be previously created with the 3654 ``-object tls-creds`` argument. 3655 3656 ``tls-auth`` provides the ID of the QAuthZ authorization object 3657 against which the client's x509 distinguished name will be 3658 validated. This object is only resolved at time of use, so can be 3659 deleted and recreated on the fly while the chardev server is active. 3660 If missing, it will default to denying access. 3661 3662 TCP and unix socket options are given below: 3663 3664 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]`` 3665 ``host`` for a listening socket specifies the local address to 3666 be bound. For a connecting socket species the remote host to 3667 connect to. ``host`` is optional for listening sockets. If not 3668 specified it defaults to ``0.0.0.0``. 3669 3670 ``port`` for a listening socket specifies the local port to be 3671 bound. For a connecting socket specifies the port on the remote 3672 host to connect to. ``port`` can be given as either a port 3673 number or a service name. ``port`` is required. 3674 3675 ``to`` is only relevant to listening sockets. If it is 3676 specified, and ``port`` cannot be bound, QEMU will attempt to 3677 bind to subsequent ports up to and including ``to`` until it 3678 succeeds. ``to`` must be specified as a port number. 3679 3680 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 3681 or IPv6 must be used. If neither is specified the socket may 3682 use either protocol. 3683 3684 ``nodelay=on|off`` disables the Nagle algorithm. 3685 3686 ``unix options: path=path[,abstract=on|off][,tight=on|off]`` 3687 ``path`` specifies the local path of the unix socket. ``path`` 3688 is required. 3689 ``abstract=on|off`` specifies the use of the abstract socket namespace, 3690 rather than the filesystem. Optional, defaults to false. 3691 ``tight=on|off`` sets the socket length of abstract sockets to their minimum, 3692 rather than the full sun_path length. Optional, defaults to true. 3693 3694``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]`` 3695 Sends all traffic from the guest to a remote host over UDP. 3696 3697 ``host`` specifies the remote host to connect to. If not specified 3698 it defaults to ``localhost``. 3699 3700 ``port`` specifies the port on the remote host to connect to. 3701 ``port`` is required. 3702 3703 ``localaddr`` specifies the local address to bind to. If not 3704 specified it defaults to ``0.0.0.0``. 3705 3706 ``localport`` specifies the local port to bind to. If not specified 3707 any available local port will be used. 3708 3709 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used. 3710 If neither is specified the device may use either protocol. 3711 3712``-chardev msmouse,id=id`` 3713 Forward QEMU's emulated msmouse events to the guest. ``msmouse`` 3714 does not take any options. 3715 3716``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]`` 3717 Connect to a QEMU text console. ``vc`` may optionally be given a 3718 specific size. 3719 3720 ``width`` and ``height`` specify the width and height respectively 3721 of the console, in pixels. 3722 3723 ``cols`` and ``rows`` specify that the console be sized to fit a 3724 text console with the given dimensions. 3725 3726``-chardev ringbuf,id=id[,size=size]`` 3727 Create a ring buffer with fixed size ``size``. size must be a power 3728 of two and defaults to ``64K``. 3729 3730``-chardev file,id=id,path=path[,input-path=input-path]`` 3731 Log all traffic received from the guest to a file. 3732 3733 ``path`` specifies the path of the file to be opened. This file will 3734 be created if it does not already exist, and overwritten if it does. 3735 ``path`` is required. 3736 3737 If ``input-path`` is specified, this is the path of a second file 3738 which will be used for input. If ``input-path`` is not specified, 3739 no input will be available from the chardev. 3740 3741 Note that ``input-path`` is not supported on Windows hosts. 3742 3743``-chardev pipe,id=id,path=path`` 3744 Create a two-way connection to the guest. The behaviour differs 3745 slightly between Windows hosts and other hosts: 3746 3747 On Windows, a single duplex pipe will be created at 3748 ``\\.pipe\path``. 3749 3750 On other hosts, 2 pipes will be created called ``path.in`` and 3751 ``path.out``. Data written to ``path.in`` will be received by the 3752 guest. Data written by the guest can be read from ``path.out``. QEMU 3753 will not create these fifos, and requires them to be present. 3754 3755 ``path`` forms part of the pipe path as described above. ``path`` is 3756 required. 3757 3758``-chardev console,id=id`` 3759 Send traffic from the guest to QEMU's standard output. ``console`` 3760 does not take any options. 3761 3762 ``console`` is only available on Windows hosts. 3763 3764``-chardev serial,id=id,path=path`` 3765 Send traffic from the guest to a serial device on the host. 3766 3767 On Unix hosts serial will actually accept any tty device, not only 3768 serial lines. 3769 3770 ``path`` specifies the name of the serial device to open. 3771 3772``-chardev pty,id=id`` 3773 Create a new pseudo-terminal on the host and connect to it. ``pty`` 3774 does not take any options. 3775 3776 ``pty`` is not available on Windows hosts. 3777 3778``-chardev stdio,id=id[,signal=on|off]`` 3779 Connect to standard input and standard output of the QEMU process. 3780 3781 ``signal`` controls if signals are enabled on the terminal, that 3782 includes exiting QEMU with the key sequence Control-c. This option 3783 is enabled by default, use ``signal=off`` to disable it. 3784 3785``-chardev braille,id=id`` 3786 Connect to a local BrlAPI server. ``braille`` does not take any 3787 options. 3788 3789``-chardev parallel,id=id,path=path`` 3790 \ 3791 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD 3792 hosts. 3793 3794 Connect to a local parallel port. 3795 3796 ``path`` specifies the path to the parallel port device. ``path`` is 3797 required. 3798 3799``-chardev spicevmc,id=id,debug=debug,name=name`` 3800 ``spicevmc`` is only available when spice support is built in. 3801 3802 ``debug`` debug level for spicevmc 3803 3804 ``name`` name of spice channel to connect to 3805 3806 Connect to a spice virtual machine channel, such as vdiport. 3807 3808``-chardev spiceport,id=id,debug=debug,name=name`` 3809 ``spiceport`` is only available when spice support is built in. 3810 3811 ``debug`` debug level for spicevmc 3812 3813 ``name`` name of spice port to connect to 3814 3815 Connect to a spice port, allowing a Spice client to handle the 3816 traffic identified by a name (preferably a fqdn). 3817ERST 3818 3819DEFHEADING() 3820 3821#ifdef CONFIG_TPM 3822DEFHEADING(TPM device options:) 3823 3824DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \ 3825 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n" 3826 " use path to provide path to a character device; default is /dev/tpm0\n" 3827 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n" 3828 " not provided it will be searched for in /sys/class/misc/tpm?/device\n" 3829 "-tpmdev emulator,id=id,chardev=dev\n" 3830 " configure the TPM device using chardev backend\n", 3831 QEMU_ARCH_ALL) 3832SRST 3833The general form of a TPM device option is: 3834 3835``-tpmdev backend,id=id[,options]`` 3836 The specific backend type will determine the applicable options. The 3837 ``-tpmdev`` option creates the TPM backend and requires a 3838 ``-device`` option that specifies the TPM frontend interface model. 3839 3840 Use ``-tpmdev help`` to print all available TPM backend types. 3841 3842The available backends are: 3843 3844``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path`` 3845 (Linux-host only) Enable access to the host's TPM using the 3846 passthrough driver. 3847 3848 ``path`` specifies the path to the host's TPM device, i.e., on a 3849 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by 3850 default ``/dev/tpm0`` is used. 3851 3852 ``cancel-path`` specifies the path to the host TPM device's sysfs 3853 entry allowing for cancellation of an ongoing TPM command. 3854 ``cancel-path`` is optional and by default QEMU will search for the 3855 sysfs entry to use. 3856 3857 Some notes about using the host's TPM with the passthrough driver: 3858 3859 The TPM device accessed by the passthrough driver must not be used 3860 by any other application on the host. 3861 3862 Since the host's firmware (BIOS/UEFI) has already initialized the 3863 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize 3864 the TPM again and may therefore not show a TPM-specific menu that 3865 would otherwise allow the user to configure the TPM, e.g., allow the 3866 user to enable/disable or activate/deactivate the TPM. Further, if 3867 TPM ownership is released from within a VM then the host's TPM will 3868 get disabled and deactivated. To enable and activate the TPM again 3869 afterwards, the host has to be rebooted and the user is required to 3870 enter the firmware's menu to enable and activate the TPM. If the TPM 3871 is left disabled and/or deactivated most TPM commands will fail. 3872 3873 To create a passthrough TPM use the following two options: 3874 3875 :: 3876 3877 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0 3878 3879 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by 3880 ``tpmdev=tpm0`` in the device option. 3881 3882``-tpmdev emulator,id=id,chardev=dev`` 3883 (Linux-host only) Enable access to a TPM emulator using Unix domain 3884 socket based chardev backend. 3885 3886 ``chardev`` specifies the unique ID of a character device backend 3887 that provides connection to the software TPM server. 3888 3889 To create a TPM emulator backend device with chardev socket backend: 3890 3891 :: 3892 3893 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0 3894ERST 3895 3896DEFHEADING() 3897 3898#endif 3899 3900DEFHEADING(Boot Image or Kernel specific:) 3901SRST 3902There are broadly 4 ways you can boot a system with QEMU. 3903 3904 - specify a firmware and let it control finding a kernel 3905 - specify a firmware and pass a hint to the kernel to boot 3906 - direct kernel image boot 3907 - manually load files into the guest's address space 3908 3909The third method is useful for quickly testing kernels but as there is 3910no firmware to pass configuration information to the kernel the 3911hardware must either be probeable, the kernel built for the exact 3912configuration or passed some configuration data (e.g. a DTB blob) 3913which tells the kernel what drivers it needs. This exact details are 3914often hardware specific. 3915 3916The final method is the most generic way of loading images into the 3917guest address space and used mostly for ``bare metal`` type 3918development where the reset vectors of the processor are taken into 3919account. 3920 3921ERST 3922 3923SRST 3924 3925For x86 machines and some other architectures ``-bios`` will generally 3926do the right thing with whatever it is given. For other machines the 3927more strict ``-pflash`` option needs an image that is sized for the 3928flash device for the given machine type. 3929 3930Please see the :ref:`system-targets-ref` section of the manual for 3931more detailed documentation. 3932 3933ERST 3934 3935DEF("bios", HAS_ARG, QEMU_OPTION_bios, \ 3936 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL) 3937SRST 3938``-bios file`` 3939 Set the filename for the BIOS. 3940ERST 3941 3942DEF("pflash", HAS_ARG, QEMU_OPTION_pflash, 3943 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL) 3944SRST 3945``-pflash file`` 3946 Use file as a parallel flash image. 3947ERST 3948 3949SRST 3950 3951The kernel options were designed to work with Linux kernels although 3952other things (like hypervisors) can be packaged up as a kernel 3953executable image. The exact format of a executable image is usually 3954architecture specific. 3955 3956The way in which the kernel is started (what address it is loaded at, 3957what if any information is passed to it via CPU registers, the state 3958of the hardware when it is started, and so on) is also architecture 3959specific. Typically it follows the specification laid down by the 3960Linux kernel for how kernels for that architecture must be started. 3961 3962ERST 3963 3964DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \ 3965 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL) 3966SRST 3967``-kernel bzImage`` 3968 Use bzImage as kernel image. The kernel can be either a Linux kernel 3969 or in multiboot format. 3970ERST 3971 3972DEF("append", HAS_ARG, QEMU_OPTION_append, \ 3973 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL) 3974SRST 3975``-append cmdline`` 3976 Use cmdline as kernel command line 3977ERST 3978 3979DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \ 3980 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL) 3981SRST 3982``-initrd file`` 3983 Use file as initial ram disk. 3984 3985``-initrd "file1 arg=foo,file2"`` 3986 This syntax is only available with multiboot. 3987 3988 Use file1 and file2 as modules and pass arg=foo as parameter to the 3989 first module. 3990ERST 3991 3992DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \ 3993 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL) 3994SRST 3995``-dtb file`` 3996 Use file as a device tree binary (dtb) image and pass it to the 3997 kernel on boot. 3998ERST 3999 4000SRST 4001 4002Finally you can also manually load images directly into the address 4003space of the guest. This is most useful for developers who already 4004know the layout of their guest and take care to ensure something sane 4005will happen when the reset vector executes. 4006 4007The generic loader can be invoked by using the loader device: 4008 4009``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]`` 4010 4011there is also the guest loader which operates in a similar way but 4012tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where 4013the guest image is: 4014 4015``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]`` 4016 4017ERST 4018 4019DEFHEADING() 4020 4021DEFHEADING(Debug/Expert options:) 4022 4023DEF("compat", HAS_ARG, QEMU_OPTION_compat, 4024 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n" 4025 " Policy for handling deprecated management interfaces\n" 4026 "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n" 4027 " Policy for handling unstable management interfaces\n", 4028 QEMU_ARCH_ALL) 4029SRST 4030``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]`` 4031 Set policy for handling deprecated management interfaces (experimental): 4032 4033 ``deprecated-input=accept`` (default) 4034 Accept deprecated commands and arguments 4035 ``deprecated-input=reject`` 4036 Reject deprecated commands and arguments 4037 ``deprecated-input=crash`` 4038 Crash on deprecated commands and arguments 4039 ``deprecated-output=accept`` (default) 4040 Emit deprecated command results and events 4041 ``deprecated-output=hide`` 4042 Suppress deprecated command results and events 4043 4044 Limitation: covers only syntactic aspects of QMP. 4045 4046``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]`` 4047 Set policy for handling unstable management interfaces (experimental): 4048 4049 ``unstable-input=accept`` (default) 4050 Accept unstable commands and arguments 4051 ``unstable-input=reject`` 4052 Reject unstable commands and arguments 4053 ``unstable-input=crash`` 4054 Crash on unstable commands and arguments 4055 ``unstable-output=accept`` (default) 4056 Emit unstable command results and events 4057 ``unstable-output=hide`` 4058 Suppress unstable command results and events 4059 4060 Limitation: covers only syntactic aspects of QMP. 4061ERST 4062 4063DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg, 4064 "-fw_cfg [name=]<name>,file=<file>\n" 4065 " add named fw_cfg entry with contents from file\n" 4066 "-fw_cfg [name=]<name>,string=<str>\n" 4067 " add named fw_cfg entry with contents from string\n", 4068 QEMU_ARCH_ALL) 4069SRST 4070``-fw_cfg [name=]name,file=file`` 4071 Add named fw\_cfg entry with contents from file file. 4072 4073``-fw_cfg [name=]name,string=str`` 4074 Add named fw\_cfg entry with contents from string str. 4075 4076 The terminating NUL character of the contents of str will not be 4077 included as part of the fw\_cfg item data. To insert contents with 4078 embedded NUL characters, you have to use the file parameter. 4079 4080 The fw\_cfg entries are passed by QEMU through to the guest. 4081 4082 Example: 4083 4084 :: 4085 4086 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin 4087 4088 creates an fw\_cfg entry named opt/com.mycompany/blob with contents 4089 from ./my\_blob.bin. 4090ERST 4091 4092DEF("serial", HAS_ARG, QEMU_OPTION_serial, \ 4093 "-serial dev redirect the serial port to char device 'dev'\n", 4094 QEMU_ARCH_ALL) 4095SRST 4096``-serial dev`` 4097 Redirect the virtual serial port to host character device dev. The 4098 default device is ``vc`` in graphical mode and ``stdio`` in non 4099 graphical mode. 4100 4101 This option can be used several times to simulate up to 4 serial 4102 ports. 4103 4104 Use ``-serial none`` to disable all serial ports. 4105 4106 Available character devices are: 4107 4108 ``vc[:WxH]`` 4109 Virtual console. Optionally, a width and height can be given in 4110 pixel with 4111 4112 :: 4113 4114 vc:800x600 4115 4116 It is also possible to specify width or height in characters: 4117 4118 :: 4119 4120 vc:80Cx24C 4121 4122 ``pty`` 4123 [Linux only] Pseudo TTY (a new PTY is automatically allocated) 4124 4125 ``none`` 4126 No device is allocated. 4127 4128 ``null`` 4129 void device 4130 4131 ``chardev:id`` 4132 Use a named character device defined with the ``-chardev`` 4133 option. 4134 4135 ``/dev/XXX`` 4136 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial 4137 port parameters are set according to the emulated ones. 4138 4139 ``/dev/parportN`` 4140 [Linux only, parallel port only] Use host parallel port N. 4141 Currently SPP and EPP parallel port features can be used. 4142 4143 ``file:filename`` 4144 Write output to filename. No character can be read. 4145 4146 ``stdio`` 4147 [Unix only] standard input/output 4148 4149 ``pipe:filename`` 4150 name pipe filename 4151 4152 ``COMn`` 4153 [Windows only] Use host serial port n 4154 4155 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]`` 4156 This implements UDP Net Console. When remote\_host or src\_ip 4157 are not specified they default to ``0.0.0.0``. When not using a 4158 specified src\_port a random port is automatically chosen. 4159 4160 If you just want a simple readonly console you can use 4161 ``netcat`` or ``nc``, by starting QEMU with: 4162 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time 4163 QEMU writes something to that port it will appear in the 4164 netconsole session. 4165 4166 If you plan to send characters back via netconsole or you want 4167 to stop and start QEMU a lot of times, you should have QEMU use 4168 the same source port each time by using something like ``-serial 4169 udp::4555@:4556`` to QEMU. Another approach is to use a patched 4170 version of netcat which can listen to a TCP port and send and 4171 receive characters via udp. If you have a patched version of 4172 netcat which activates telnet remote echo and single char 4173 transfer, then you can use the following options to set up a 4174 netcat redirector to allow telnet on port 5555 to access the 4175 QEMU port. 4176 4177 ``QEMU Options:`` 4178 -serial udp::4555@:4556 4179 4180 ``netcat options:`` 4181 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T 4182 4183 ``telnet options:`` 4184 localhost 5555 4185 4186 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]`` 4187 The TCP Net Console has two modes of operation. It can send the 4188 serial I/O to a location or wait for a connection from a 4189 location. By default the TCP Net Console is sent to host at the 4190 port. If you use the ``server=on`` option QEMU will wait for a client 4191 socket application to connect to the port before continuing, 4192 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off`` 4193 option disables the Nagle buffering algorithm. The ``reconnect=on`` 4194 option only applies if ``server=no`` is set, if the connection goes 4195 down it will attempt to reconnect at the given interval. If host 4196 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a 4197 time is accepted. You can use ``telnet=on`` to connect to the 4198 corresponding character device. 4199 4200 ``Example to send tcp console to 192.168.0.2 port 4444`` 4201 -serial tcp:192.168.0.2:4444 4202 4203 ``Example to listen and wait on port 4444 for connection`` 4204 -serial tcp::4444,server=on 4205 4206 ``Example to not wait and listen on ip 192.168.0.100 port 4444`` 4207 -serial tcp:192.168.0.100:4444,server=on,wait=off 4208 4209 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]`` 4210 The telnet protocol is used instead of raw tcp sockets. The 4211 options work the same as if you had specified ``-serial tcp``. 4212 The difference is that the port acts like a telnet server or 4213 client using telnet option negotiation. This will also allow you 4214 to send the MAGIC\_SYSRQ sequence if you use a telnet that 4215 supports sending the break sequence. Typically in unix telnet 4216 you do it with Control-] and then type "send break" followed by 4217 pressing the enter key. 4218 4219 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]`` 4220 The WebSocket protocol is used instead of raw tcp socket. The 4221 port acts as a WebSocket server. Client mode is not supported. 4222 4223 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]`` 4224 A unix domain socket is used instead of a tcp socket. The option 4225 works the same as if you had specified ``-serial tcp`` except 4226 the unix domain socket path is used for connections. 4227 4228 ``mon:dev_string`` 4229 This is a special option to allow the monitor to be multiplexed 4230 onto another serial port. The monitor is accessed with key 4231 sequence of Control-a and then pressing c. dev\_string should be 4232 any one of the serial devices specified above. An example to 4233 multiplex the monitor onto a telnet server listening on port 4234 4444 would be: 4235 4236 ``-serial mon:telnet::4444,server=on,wait=off`` 4237 4238 When the monitor is multiplexed to stdio in this way, Ctrl+C 4239 will not terminate QEMU any more but will be passed to the guest 4240 instead. 4241 4242 ``braille`` 4243 Braille device. This will use BrlAPI to display the braille 4244 output on a real or fake device. 4245 4246 ``msmouse`` 4247 Three button serial mouse. Configure the guest to use Microsoft 4248 protocol. 4249ERST 4250 4251DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \ 4252 "-parallel dev redirect the parallel port to char device 'dev'\n", 4253 QEMU_ARCH_ALL) 4254SRST 4255``-parallel dev`` 4256 Redirect the virtual parallel port to host device dev (same devices 4257 as the serial port). On Linux hosts, ``/dev/parportN`` can be used 4258 to use hardware devices connected on the corresponding host parallel 4259 port. 4260 4261 This option can be used several times to simulate up to 3 parallel 4262 ports. 4263 4264 Use ``-parallel none`` to disable all parallel ports. 4265ERST 4266 4267DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \ 4268 "-monitor dev redirect the monitor to char device 'dev'\n", 4269 QEMU_ARCH_ALL) 4270SRST 4271``-monitor dev`` 4272 Redirect the monitor to host device dev (same devices as the serial 4273 port). The default device is ``vc`` in graphical mode and ``stdio`` 4274 in non graphical mode. Use ``-monitor none`` to disable the default 4275 monitor. 4276ERST 4277DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \ 4278 "-qmp dev like -monitor but opens in 'control' mode\n", 4279 QEMU_ARCH_ALL) 4280SRST 4281``-qmp dev`` 4282 Like ``-monitor`` but opens in 'control' mode. For example, to make 4283 QMP available on localhost port 4444:: 4284 4285 -qmp tcp:localhost:4444,server=on,wait=off 4286 4287 Not all options are configurable via this syntax; for maximum 4288 flexibility use the ``-mon`` option and an accompanying ``-chardev``. 4289 4290ERST 4291DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \ 4292 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n", 4293 QEMU_ARCH_ALL) 4294SRST 4295``-qmp-pretty dev`` 4296 Like ``-qmp`` but uses pretty JSON formatting. 4297ERST 4298 4299DEF("mon", HAS_ARG, QEMU_OPTION_mon, \ 4300 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL) 4301SRST 4302``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]`` 4303 Set up a monitor connected to the chardev ``name``. 4304 QEMU supports two monitors: the Human Monitor Protocol 4305 (HMP; for human interaction), and the QEMU Monitor Protocol 4306 (QMP; a JSON RPC-style protocol). 4307 The default is HMP; ``mode=control`` selects QMP instead. 4308 ``pretty`` is only valid when ``mode=control``, 4309 turning on JSON pretty printing to ease 4310 human reading and debugging. 4311 4312 For example:: 4313 4314 -chardev socket,id=mon1,host=localhost,port=4444,server=on,wait=off \ 4315 -mon chardev=mon1,mode=control,pretty=on 4316 4317 enables the QMP monitor on localhost port 4444 with pretty-printing. 4318ERST 4319 4320DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \ 4321 "-debugcon dev redirect the debug console to char device 'dev'\n", 4322 QEMU_ARCH_ALL) 4323SRST 4324``-debugcon dev`` 4325 Redirect the debug console to host device dev (same devices as the 4326 serial port). The debug console is an I/O port which is typically 4327 port 0xe9; writing to that I/O port sends output to this device. The 4328 default device is ``vc`` in graphical mode and ``stdio`` in non 4329 graphical mode. 4330ERST 4331 4332DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \ 4333 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL) 4334SRST 4335``-pidfile file`` 4336 Store the QEMU process PID in file. It is useful if you launch QEMU 4337 from a script. 4338ERST 4339 4340DEF("singlestep", 0, QEMU_OPTION_singlestep, \ 4341 "-singlestep deprecated synonym for -accel tcg,one-insn-per-tb=on\n", QEMU_ARCH_ALL) 4342SRST 4343``-singlestep`` 4344 This is a deprecated synonym for the TCG accelerator property 4345 ``one-insn-per-tb``. 4346ERST 4347 4348DEF("preconfig", 0, QEMU_OPTION_preconfig, \ 4349 "--preconfig pause QEMU before machine is initialized (experimental)\n", 4350 QEMU_ARCH_ALL) 4351SRST 4352``--preconfig`` 4353 Pause QEMU for interactive configuration before the machine is 4354 created, which allows querying and configuring properties that will 4355 affect machine initialization. Use QMP command 'x-exit-preconfig' to 4356 exit the preconfig state and move to the next state (i.e. run guest 4357 if -S isn't used or pause the second time if -S is used). This 4358 option is experimental. 4359ERST 4360 4361DEF("S", 0, QEMU_OPTION_S, \ 4362 "-S freeze CPU at startup (use 'c' to start execution)\n", 4363 QEMU_ARCH_ALL) 4364SRST 4365``-S`` 4366 Do not start CPU at startup (you must type 'c' in the monitor). 4367ERST 4368 4369DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit, 4370 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n" 4371 " run qemu with overcommit hints\n" 4372 " mem-lock=on|off controls memory lock support (default: off)\n" 4373 " cpu-pm=on|off controls cpu power management (default: off)\n", 4374 QEMU_ARCH_ALL) 4375SRST 4376``-overcommit mem-lock=on|off`` 4377 \ 4378``-overcommit cpu-pm=on|off`` 4379 Run qemu with hints about host resource overcommit. The default is 4380 to assume that host overcommits all resources. 4381 4382 Locking qemu and guest memory can be enabled via ``mem-lock=on`` 4383 (disabled by default). This works when host memory is not 4384 overcommitted and reduces the worst-case latency for guest. 4385 4386 Guest ability to manage power state of host cpus (increasing latency 4387 for other processes on the same host cpu, but decreasing latency for 4388 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This 4389 works best when host CPU is not overcommitted. When used, host 4390 estimates of CPU cycle and power utilization will be incorrect, not 4391 taking into account guest idle time. 4392ERST 4393 4394DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \ 4395 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n" 4396 " the guest without waiting for gdb to connect; use -S too\n" 4397 " if you want it to not start execution.)\n", 4398 QEMU_ARCH_ALL) 4399SRST 4400``-gdb dev`` 4401 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter 4402 in the System Emulation Users Guide). Note that this option does not pause QEMU 4403 execution -- if you want QEMU to not start the guest until you 4404 connect with gdb and issue a ``continue`` command, you will need to 4405 also pass the ``-S`` option to QEMU. 4406 4407 The most usual configuration is to listen on a local TCP socket:: 4408 4409 -gdb tcp::3117 4410 4411 but you can specify other backends; UDP, pseudo TTY, or even stdio 4412 are all reasonable use cases. For example, a stdio connection 4413 allows you to start QEMU from within gdb and establish the 4414 connection via a pipe: 4415 4416 .. parsed-literal:: 4417 4418 (gdb) target remote | exec |qemu_system| -gdb stdio ... 4419ERST 4420 4421DEF("s", 0, QEMU_OPTION_s, \ 4422 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n", 4423 QEMU_ARCH_ALL) 4424SRST 4425``-s`` 4426 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234 4427 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide). 4428ERST 4429 4430DEF("d", HAS_ARG, QEMU_OPTION_d, \ 4431 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n", 4432 QEMU_ARCH_ALL) 4433SRST 4434``-d item1[,...]`` 4435 Enable logging of specified items. Use '-d help' for a list of log 4436 items. 4437ERST 4438 4439DEF("D", HAS_ARG, QEMU_OPTION_D, \ 4440 "-D logfile output log to logfile (default stderr)\n", 4441 QEMU_ARCH_ALL) 4442SRST 4443``-D logfile`` 4444 Output log in logfile instead of to stderr 4445ERST 4446 4447DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \ 4448 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n", 4449 QEMU_ARCH_ALL) 4450SRST 4451``-dfilter range1[,...]`` 4452 Filter debug output to that relevant to a range of target addresses. 4453 The filter spec can be either start+size, start-size or start..end 4454 where start end and size are the addresses and sizes required. For 4455 example: 4456 4457 :: 4458 4459 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000 4460 4461 Will dump output for any code in the 0x1000 sized block starting at 4462 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and 4463 another 0x1000 sized block starting at 0xffffffc00005f000. 4464ERST 4465 4466DEF("seed", HAS_ARG, QEMU_OPTION_seed, \ 4467 "-seed number seed the pseudo-random number generator\n", 4468 QEMU_ARCH_ALL) 4469SRST 4470``-seed number`` 4471 Force the guest to use a deterministic pseudo-random number 4472 generator, seeded with number. This does not affect crypto routines 4473 within the host. 4474ERST 4475 4476DEF("L", HAS_ARG, QEMU_OPTION_L, \ 4477 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n", 4478 QEMU_ARCH_ALL) 4479SRST 4480``-L path`` 4481 Set the directory for the BIOS, VGA BIOS and keymaps. 4482 4483 To list all the data directories, use ``-L help``. 4484ERST 4485 4486DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \ 4487 "-enable-kvm enable KVM full virtualization support\n", 4488 QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC | 4489 QEMU_ARCH_RISCV | QEMU_ARCH_S390X) 4490SRST 4491``-enable-kvm`` 4492 Enable KVM full virtualization support. This option is only 4493 available if KVM support is enabled when compiling. 4494ERST 4495 4496DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, 4497 "-xen-domid id specify xen guest domain id\n", 4498 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4499DEF("xen-attach", 0, QEMU_OPTION_xen_attach, 4500 "-xen-attach attach to existing xen domain\n" 4501 " libxl will use this when starting QEMU\n", 4502 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4503DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, 4504 "-xen-domid-restrict restrict set of available xen operations\n" 4505 " to specified domain id. (Does not affect\n" 4506 " xenpv machine type).\n", 4507 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4508SRST 4509``-xen-domid id`` 4510 Specify xen guest domain id (XEN only). 4511 4512``-xen-attach`` 4513 Attach to existing xen domain. libxl will use this when starting 4514 QEMU (XEN only). Restrict set of available xen operations to 4515 specified domain id (XEN only). 4516ERST 4517 4518DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \ 4519 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL) 4520SRST 4521``-no-reboot`` 4522 Exit instead of rebooting. 4523ERST 4524 4525DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \ 4526 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL) 4527SRST 4528``-no-shutdown`` 4529 Don't exit QEMU on guest shutdown, but instead only stop the 4530 emulation. This allows for instance switching to monitor to commit 4531 changes to the disk image. 4532ERST 4533 4534DEF("action", HAS_ARG, QEMU_OPTION_action, 4535 "-action reboot=reset|shutdown\n" 4536 " action when guest reboots [default=reset]\n" 4537 "-action shutdown=poweroff|pause\n" 4538 " action when guest shuts down [default=poweroff]\n" 4539 "-action panic=pause|shutdown|exit-failure|none\n" 4540 " action when guest panics [default=shutdown]\n" 4541 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" 4542 " action when watchdog fires [default=reset]\n", 4543 QEMU_ARCH_ALL) 4544SRST 4545``-action event=action`` 4546 The action parameter serves to modify QEMU's default behavior when 4547 certain guest events occur. It provides a generic method for specifying the 4548 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown`` 4549 parameters. 4550 4551 Examples: 4552 4553 ``-action panic=none`` 4554 ``-action reboot=shutdown,shutdown=pause`` 4555 ``-device i6300esb -action watchdog=pause`` 4556 4557ERST 4558 4559DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \ 4560 "-loadvm [tag|id]\n" \ 4561 " start right away with a saved state (loadvm in monitor)\n", 4562 QEMU_ARCH_ALL) 4563SRST 4564``-loadvm file`` 4565 Start right away with a saved state (``loadvm`` in monitor) 4566ERST 4567 4568#ifndef _WIN32 4569DEF("daemonize", 0, QEMU_OPTION_daemonize, \ 4570 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL) 4571#endif 4572SRST 4573``-daemonize`` 4574 Daemonize the QEMU process after initialization. QEMU will not 4575 detach from standard IO until it is ready to receive connections on 4576 any of its devices. This option is a useful way for external 4577 programs to launch QEMU without having to cope with initialization 4578 race conditions. 4579ERST 4580 4581DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \ 4582 "-option-rom rom load a file, rom, into the option ROM space\n", 4583 QEMU_ARCH_ALL) 4584SRST 4585``-option-rom file`` 4586 Load the contents of file as an option ROM. This option is useful to 4587 load things like EtherBoot. 4588ERST 4589 4590DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \ 4591 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \ 4592 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n", 4593 QEMU_ARCH_ALL) 4594 4595SRST 4596``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]`` 4597 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at 4598 the current UTC or local time, respectively. ``localtime`` is 4599 required for correct date in MS-DOS or Windows. To start at a 4600 specific point in time, provide datetime in the format 4601 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC. 4602 4603 By default the RTC is driven by the host system time. This allows 4604 using of the RTC as accurate reference clock inside the guest, 4605 specifically if the host time is smoothly following an accurate 4606 external reference clock, e.g. via NTP. If you want to isolate the 4607 guest time from the host, you can set ``clock`` to ``rt`` instead, 4608 which provides a host monotonic clock if host support it. To even 4609 prevent the RTC from progressing during suspension, you can set 4610 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is 4611 recommended especially in icount mode in order to preserve 4612 determinism; however, note that in icount mode the speed of the 4613 virtual clock is variable and can in general differ from the host 4614 clock. 4615 4616 Enable ``driftfix`` (i386 targets only) if you experience time drift 4617 problems, specifically with Windows' ACPI HAL. This option will try 4618 to figure out how many timer interrupts were not processed by the 4619 Windows guest and will re-inject them. 4620ERST 4621 4622DEF("icount", HAS_ARG, QEMU_OPTION_icount, \ 4623 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \ 4624 " enable virtual instruction counter with 2^N clock ticks per\n" \ 4625 " instruction, enable aligning the host and virtual clocks\n" \ 4626 " or disable real time cpu sleeping, and optionally enable\n" \ 4627 " record-and-replay mode\n", QEMU_ARCH_ALL) 4628SRST 4629``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]`` 4630 Enable virtual instruction counter. The virtual cpu will execute one 4631 instruction every 2^N ns of virtual time. If ``auto`` is specified 4632 then the virtual cpu speed will be automatically adjusted to keep 4633 virtual time within a few seconds of real time. 4634 4635 Note that while this option can give deterministic behavior, it does 4636 not provide cycle accurate emulation. Modern CPUs contain 4637 superscalar out of order cores with complex cache hierarchies. The 4638 number of instructions executed often has little or no correlation 4639 with actual performance. 4640 4641 When the virtual cpu is sleeping, the virtual time will advance at 4642 default speed unless ``sleep=on`` is specified. With 4643 ``sleep=on``, the virtual time will jump to the next timer 4644 deadline instantly whenever the virtual cpu goes to sleep mode and 4645 will not advance if no timer is enabled. This behavior gives 4646 deterministic execution times from the guest point of view. 4647 The default if icount is enabled is ``sleep=off``. 4648 ``sleep=on`` cannot be used together with either ``shift=auto`` 4649 or ``align=on``. 4650 4651 ``align=on`` will activate the delay algorithm which will try to 4652 synchronise the host clock and the virtual clock. The goal is to 4653 have a guest running at the real frequency imposed by the shift 4654 option. Whenever the guest clock is behind the host clock and if 4655 ``align=on`` is specified then we print a message to the user to 4656 inform about the delay. Currently this option does not work when 4657 ``shift`` is ``auto``. Note: The sync algorithm will work for those 4658 shift values for which the guest clock runs ahead of the host clock. 4659 Typically this happens when the shift value is high (how high 4660 depends on the host machine). The default if icount is enabled 4661 is ``align=off``. 4662 4663 When the ``rr`` option is specified deterministic record/replay is 4664 enabled. The ``rrfile=`` option must also be provided to 4665 specify the path to the replay log. In record mode data is written 4666 to this file, and in replay mode it is read back. 4667 If the ``rrsnapshot`` option is given then it specifies a VM snapshot 4668 name. In record mode, a new VM snapshot with the given name is created 4669 at the start of execution recording. In replay mode this option 4670 specifies the snapshot name used to load the initial VM state. 4671ERST 4672 4673DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \ 4674 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \ 4675 " action when watchdog fires [default=reset]\n", 4676 QEMU_ARCH_ALL) 4677SRST 4678``-watchdog-action action`` 4679 The action controls what QEMU will do when the watchdog timer 4680 expires. The default is ``reset`` (forcefully reset the guest). 4681 Other possible actions are: ``shutdown`` (attempt to gracefully 4682 shutdown the guest), ``poweroff`` (forcefully poweroff the guest), 4683 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the 4684 guest), ``debug`` (print a debug message and continue), or ``none`` 4685 (do nothing). 4686 4687 Note that the ``shutdown`` action requires that the guest responds 4688 to ACPI signals, which it may not be able to do in the sort of 4689 situations where the watchdog would have expired, and thus 4690 ``-watchdog-action shutdown`` is not recommended for production use. 4691 4692 Examples: 4693 4694 ``-device i6300esb -watchdog-action pause`` 4695 4696ERST 4697 4698DEF("echr", HAS_ARG, QEMU_OPTION_echr, \ 4699 "-echr chr set terminal escape character instead of ctrl-a\n", 4700 QEMU_ARCH_ALL) 4701SRST 4702``-echr numeric_ascii_value`` 4703 Change the escape character used for switching to the monitor when 4704 using monitor and serial sharing. The default is ``0x01`` when using 4705 the ``-nographic`` option. ``0x01`` is equal to pressing 4706 ``Control-a``. You can select a different character from the ascii 4707 control keys where 1 through 26 map to Control-a through Control-z. 4708 For instance you could use the either of the following to change the 4709 escape character to Control-t. 4710 4711 ``-echr 0x14``; \ ``-echr 20`` 4712 4713ERST 4714 4715DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \ 4716 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \ 4717 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \ 4718 "-incoming unix:socketpath\n" \ 4719 " prepare for incoming migration, listen on\n" \ 4720 " specified protocol and socket address\n" \ 4721 "-incoming fd:fd\n" \ 4722 "-incoming file:filename[,offset=offset]\n" \ 4723 "-incoming exec:cmdline\n" \ 4724 " accept incoming migration on given file descriptor\n" \ 4725 " or from given external command\n" \ 4726 "-incoming defer\n" \ 4727 " wait for the URI to be specified via migrate_incoming\n", 4728 QEMU_ARCH_ALL) 4729SRST 4730``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]`` 4731 \ 4732``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]`` 4733 Prepare for incoming migration, listen on a given tcp port. 4734 4735``-incoming unix:socketpath`` 4736 Prepare for incoming migration, listen on a given unix socket. 4737 4738``-incoming fd:fd`` 4739 Accept incoming migration from a given file descriptor. 4740 4741``-incoming file:filename[,offset=offset]`` 4742 Accept incoming migration from a given file starting at offset. 4743 offset allows the common size suffixes, or a 0x prefix, but not both. 4744 4745``-incoming exec:cmdline`` 4746 Accept incoming migration as an output from specified external 4747 command. 4748 4749``-incoming defer`` 4750 Wait for the URI to be specified via migrate\_incoming. The monitor 4751 can be used to change settings (such as migration parameters) prior 4752 to issuing the migrate\_incoming to allow the migration to begin. 4753ERST 4754 4755DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \ 4756 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL) 4757SRST 4758``-only-migratable`` 4759 Only allow migratable devices. Devices will not be allowed to enter 4760 an unmigratable state. 4761ERST 4762 4763DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \ 4764 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL) 4765SRST 4766``-nodefaults`` 4767 Don't create default devices. Normally, QEMU sets the default 4768 devices like serial port, parallel port, virtual console, monitor 4769 device, VGA adapter, floppy and CD-ROM drive and others. The 4770 ``-nodefaults`` option will disable all those default devices. 4771ERST 4772 4773#ifndef _WIN32 4774DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \ 4775 "-chroot dir chroot to dir just before starting the VM (deprecated)\n", 4776 QEMU_ARCH_ALL) 4777#endif 4778SRST 4779``-chroot dir`` 4780 Deprecated, use '-run-with chroot=...' instead. 4781 Immediately before starting guest execution, chroot to the specified 4782 directory. Especially useful in combination with -runas. 4783ERST 4784 4785#ifndef _WIN32 4786DEF("runas", HAS_ARG, QEMU_OPTION_runas, \ 4787 "-runas user change to user id user just before starting the VM\n" \ 4788 " user can be numeric uid:gid instead\n", 4789 QEMU_ARCH_ALL) 4790#endif 4791SRST 4792``-runas user`` 4793 Immediately before starting guest execution, drop root privileges, 4794 switching to the specified user. 4795ERST 4796 4797DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env, 4798 "-prom-env variable=value\n" 4799 " set OpenBIOS nvram variables\n", 4800 QEMU_ARCH_PPC | QEMU_ARCH_SPARC) 4801SRST 4802``-prom-env variable=value`` 4803 Set OpenBIOS nvram variable to given value (PPC, SPARC only). 4804 4805 :: 4806 4807 qemu-system-sparc -prom-env 'auto-boot?=false' \ 4808 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single' 4809 4810 :: 4811 4812 qemu-system-ppc -prom-env 'auto-boot?=false' \ 4813 -prom-env 'boot-device=hd:2,\yaboot' \ 4814 -prom-env 'boot-args=conf=hd:2,\yaboot.conf' 4815ERST 4816DEF("semihosting", 0, QEMU_OPTION_semihosting, 4817 "-semihosting semihosting mode\n", 4818 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 4819 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) 4820SRST 4821``-semihosting`` 4822 Enable :ref:`Semihosting` mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only). 4823 4824 .. warning:: 4825 Note that this allows guest direct access to the host filesystem, so 4826 should only be used with a trusted guest OS. 4827 4828 See the -semihosting-config option documentation for further 4829 information about the facilities this enables. 4830ERST 4831DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config, 4832 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]\n" \ 4833 " semihosting configuration\n", 4834QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 4835QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) 4836SRST 4837``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]`` 4838 Enable and configure :ref:`Semihosting` (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V 4839 only). 4840 4841 .. warning:: 4842 Note that this allows guest direct access to the host filesystem, so 4843 should only be used with a trusted guest OS. 4844 4845 ``target=native|gdb|auto`` 4846 Defines where the semihosting calls will be addressed, to QEMU 4847 (``native``) or to GDB (``gdb``). The default is ``auto``, which 4848 means ``gdb`` during debug sessions and ``native`` otherwise. 4849 4850 ``chardev=str1`` 4851 Send the output to a chardev backend output for native or auto 4852 output when not in gdb 4853 4854 ``userspace=on|off`` 4855 Allows code running in guest userspace to access the semihosting 4856 interface. The default is that only privileged guest code can 4857 make semihosting calls. Note that setting ``userspace=on`` should 4858 only be used if all guest code is trusted (for example, in 4859 bare-metal test case code). 4860 4861 ``arg=str1,arg=str2,...`` 4862 Allows the user to pass input arguments, and can be used 4863 multiple times to build up a list. The old-style 4864 ``-kernel``/``-append`` method of passing a command line is 4865 still supported for backward compatibility. If both the 4866 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are 4867 specified, the former is passed to semihosting as it always 4868 takes precedence. 4869ERST 4870DEF("old-param", 0, QEMU_OPTION_old_param, 4871 "-old-param old param mode\n", QEMU_ARCH_ARM) 4872SRST 4873``-old-param`` 4874 Old param mode (ARM only). 4875ERST 4876 4877DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \ 4878 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \ 4879 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \ 4880 " Enable seccomp mode 2 system call filter (default 'off').\n" \ 4881 " use 'obsolete' to allow obsolete system calls that are provided\n" \ 4882 " by the kernel, but typically no longer used by modern\n" \ 4883 " C library implementations.\n" \ 4884 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \ 4885 " to elevate privileges using set*uid|gid system calls.\n" \ 4886 " The value 'children' will deny set*uid|gid system calls for\n" \ 4887 " main QEMU process but will allow forks and execves to run unprivileged\n" \ 4888 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \ 4889 " blocking *fork and execve\n" \ 4890 " use 'resourcecontrol' to disable process affinity and schedular priority\n", 4891 QEMU_ARCH_ALL) 4892SRST 4893``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]`` 4894 Enable Seccomp mode 2 system call filter. 'on' will enable syscall 4895 filtering and 'off' will disable it. The default is 'off'. 4896 4897 ``obsolete=string`` 4898 Enable Obsolete system calls 4899 4900 ``elevateprivileges=string`` 4901 Disable set\*uid\|gid system calls 4902 4903 ``spawn=string`` 4904 Disable \*fork and execve 4905 4906 ``resourcecontrol=string`` 4907 Disable process affinity and schedular priority 4908ERST 4909 4910DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig, 4911 "-readconfig <file>\n" 4912 " read config file\n", QEMU_ARCH_ALL) 4913SRST 4914``-readconfig file`` 4915 Read device configuration from file. This approach is useful when 4916 you want to spawn QEMU process with many command line options but 4917 you don't want to exceed the command line character limit. 4918ERST 4919 4920DEF("no-user-config", 0, QEMU_OPTION_nouserconfig, 4921 "-no-user-config\n" 4922 " do not load default user-provided config files at startup\n", 4923 QEMU_ARCH_ALL) 4924SRST 4925``-no-user-config`` 4926 The ``-no-user-config`` option makes QEMU not load any of the 4927 user-provided config files on sysconfdir. 4928ERST 4929 4930DEF("trace", HAS_ARG, QEMU_OPTION_trace, 4931 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n" 4932 " specify tracing options\n", 4933 QEMU_ARCH_ALL) 4934SRST 4935``-trace [[enable=]pattern][,events=file][,file=file]`` 4936 .. include:: ../qemu-option-trace.rst.inc 4937 4938ERST 4939DEF("plugin", HAS_ARG, QEMU_OPTION_plugin, 4940 "-plugin [file=]<file>[,<argname>=<argvalue>]\n" 4941 " load a plugin\n", 4942 QEMU_ARCH_ALL) 4943SRST 4944``-plugin file=file[,argname=argvalue]`` 4945 Load a plugin. 4946 4947 ``file=file`` 4948 Load the given plugin from a shared library file. 4949 4950 ``argname=argvalue`` 4951 Argument passed to the plugin. (Can be given multiple times.) 4952ERST 4953 4954HXCOMM Internal use 4955DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL) 4956DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL) 4957 4958#ifdef __linux__ 4959DEF("async-teardown", 0, QEMU_OPTION_asyncteardown, 4960 "-async-teardown enable asynchronous teardown\n", 4961 QEMU_ARCH_ALL) 4962SRST 4963``-async-teardown`` 4964 This option is deprecated and should no longer be used. The new option 4965 ``-run-with async-teardown=on`` is a replacement. 4966ERST 4967#endif 4968#ifdef CONFIG_POSIX 4969DEF("run-with", HAS_ARG, QEMU_OPTION_run_with, 4970 "-run-with [async-teardown=on|off][,chroot=dir]\n" 4971 " Set miscellaneous QEMU process lifecycle options:\n" 4972 " async-teardown=on enables asynchronous teardown (Linux only)\n" 4973 " chroot=dir chroot to dir just before starting the VM\n", 4974 QEMU_ARCH_ALL) 4975SRST 4976``-run-with [async-teardown=on|off][,chroot=dir]`` 4977 Set QEMU process lifecycle options. 4978 4979 ``async-teardown=on`` enables asynchronous teardown. A new process called 4980 "cleanup/<QEMU_PID>" will be created at startup sharing the address 4981 space with the main QEMU process, using clone. It will wait for the 4982 main QEMU process to terminate completely, and then exit. This allows 4983 QEMU to terminate very quickly even if the guest was huge, leaving the 4984 teardown of the address space to the cleanup process. Since the cleanup 4985 process shares the same cgroups as the main QEMU process, accounting is 4986 performed correctly. This only works if the cleanup process is not 4987 forcefully killed with SIGKILL before the main QEMU process has 4988 terminated completely. 4989 4990 ``chroot=dir`` can be used for doing a chroot to the specified directory 4991 immediately before starting the guest execution. This is especially useful 4992 in combination with -runas. 4993ERST 4994#endif 4995 4996DEF("msg", HAS_ARG, QEMU_OPTION_msg, 4997 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n" 4998 " control error message format\n" 4999 " timestamp=on enables timestamps (default: off)\n" 5000 " guest-name=on enables guest name prefix but only if\n" 5001 " -name guest option is set (default: off)\n", 5002 QEMU_ARCH_ALL) 5003SRST 5004``-msg [timestamp[=on|off]][,guest-name[=on|off]]`` 5005 Control error message format. 5006 5007 ``timestamp=on|off`` 5008 Prefix messages with a timestamp. Default is off. 5009 5010 ``guest-name=on|off`` 5011 Prefix messages with guest name but only if -name guest option is set 5012 otherwise the option is ignored. Default is off. 5013ERST 5014 5015DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate, 5016 "-dump-vmstate <file>\n" 5017 " Output vmstate information in JSON format to file.\n" 5018 " Use the scripts/vmstate-static-checker.py file to\n" 5019 " check for possible regressions in migration code\n" 5020 " by comparing two such vmstate dumps.\n", 5021 QEMU_ARCH_ALL) 5022SRST 5023``-dump-vmstate file`` 5024 Dump json-encoded vmstate information for current machine type to 5025 file in file 5026ERST 5027 5028DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile, 5029 "-enable-sync-profile\n" 5030 " enable synchronization profiling\n", 5031 QEMU_ARCH_ALL) 5032SRST 5033``-enable-sync-profile`` 5034 Enable synchronization profiling. 5035ERST 5036 5037#if defined(CONFIG_TCG) && defined(CONFIG_LINUX) 5038DEF("perfmap", 0, QEMU_OPTION_perfmap, 5039 "-perfmap generate a /tmp/perf-${pid}.map file for perf\n", 5040 QEMU_ARCH_ALL) 5041SRST 5042``-perfmap`` 5043 Generate a map file for Linux perf tools that will allow basic profiling 5044 information to be broken down into basic blocks. 5045ERST 5046 5047DEF("jitdump", 0, QEMU_OPTION_jitdump, 5048 "-jitdump generate a jit-${pid}.dump file for perf\n", 5049 QEMU_ARCH_ALL) 5050SRST 5051``-jitdump`` 5052 Generate a dump file for Linux perf tools that maps basic blocks to symbol 5053 names, line numbers and JITted code. 5054ERST 5055#endif 5056 5057DEFHEADING() 5058 5059DEFHEADING(Generic object creation:) 5060 5061DEF("object", HAS_ARG, QEMU_OPTION_object, 5062 "-object TYPENAME[,PROP1=VALUE1,...]\n" 5063 " create a new object of type TYPENAME setting properties\n" 5064 " in the order they are specified. Note that the 'id'\n" 5065 " property must be set. These objects are placed in the\n" 5066 " '/objects' path.\n", 5067 QEMU_ARCH_ALL) 5068SRST 5069``-object typename[,prop1=value1,...]`` 5070 Create a new object of type typename setting properties in the order 5071 they are specified. Note that the 'id' property must be set. These 5072 objects are placed in the '/objects' path. 5073 5074 ``-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`` 5075 Creates a memory file backend object, which can be used to back 5076 the guest RAM with huge pages. 5077 5078 The ``id`` parameter is a unique ID that will be used to 5079 reference this memory region in other parameters, e.g. ``-numa``, 5080 ``-device nvdimm``, etc. 5081 5082 The ``size`` option provides the size of the memory region, and 5083 accepts common suffixes, e.g. ``500M``. 5084 5085 The ``mem-path`` provides the path to either a shared memory or 5086 huge page filesystem mount. 5087 5088 The ``share`` boolean option determines whether the memory 5089 region is marked as private to QEMU, or shared. The latter 5090 allows a co-operating external process to access the QEMU memory 5091 region. 5092 5093 The ``share`` is also required for pvrdma devices due to 5094 limitations in the RDMA API provided by Linux. 5095 5096 Setting share=on might affect the ability to configure NUMA 5097 bindings for the memory backend under some circumstances, see 5098 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel 5099 source tree for additional details. 5100 5101 Setting the ``discard-data`` boolean option to on indicates that 5102 file contents can be destroyed when QEMU exits, to avoid 5103 unnecessarily flushing data to the backing file. Note that 5104 ``discard-data`` is only an optimization, and QEMU might not 5105 discard file contents if it aborts unexpectedly or is terminated 5106 using SIGKILL. 5107 5108 The ``merge`` boolean option enables memory merge, also known as 5109 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider 5110 the pages for memory deduplication. 5111 5112 Setting the ``dump`` boolean option to off excludes the memory 5113 from core dumps. This feature is also known as MADV\_DONTDUMP. 5114 5115 The ``prealloc`` boolean option enables memory preallocation. 5116 5117 The ``host-nodes`` option binds the memory range to a list of 5118 NUMA host nodes. 5119 5120 The ``policy`` option sets the NUMA policy to one of the 5121 following values: 5122 5123 ``default`` 5124 default host policy 5125 5126 ``preferred`` 5127 prefer the given host node list for allocation 5128 5129 ``bind`` 5130 restrict memory allocation to the given host node list 5131 5132 ``interleave`` 5133 interleave memory allocations across the given host node 5134 list 5135 5136 The ``align`` option specifies the base address alignment when 5137 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg 5138 ``2M``. Some backend store specified by ``mem-path`` requires an 5139 alignment different than the default one used by QEMU, eg the 5140 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In 5141 such cases, users can specify the required alignment via this 5142 option. 5143 5144 The ``offset`` option specifies the offset into the target file 5145 that the region starts at. You can use this parameter to back 5146 multiple regions with a single file. 5147 5148 The ``pmem`` option specifies whether the backing file specified 5149 by ``mem-path`` is in host persistent memory that can be 5150 accessed using the SNIA NVM programming model (e.g. Intel 5151 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary 5152 operations to guarantee the persistence of its own writes to 5153 ``mem-path`` (e.g. in vNVDIMM label emulation and live 5154 migration). Also, we will map the backend-file with MAP\_SYNC 5155 flag, which ensures the file metadata is in sync for 5156 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC 5157 requires support from both the host kernel (since Linux kernel 5158 4.15) and the filesystem of ``mem-path`` mounted with DAX 5159 option. 5160 5161 The ``readonly`` option specifies whether the backing file is opened 5162 read-only or read-write (default). 5163 5164 The ``rom`` option specifies whether to create Read Only Memory 5165 (ROM) that cannot be modified by the VM. Any write attempts to such 5166 ROM will be denied. Most use cases want proper RAM instead of ROM. 5167 However, selected use cases, like R/O NVDIMMs, can benefit from 5168 ROM. If set to ``on``, create ROM; if set to ``off``, create 5169 writable RAM; if set to ``auto`` (default), the value of the 5170 ``readonly`` option is used. This option is primarily helpful when 5171 we want to have writable RAM in configurations that would 5172 traditionally create ROM before the ``rom`` option was introduced: 5173 VM templating, where we want to open a file readonly 5174 (``readonly=on``) and mark the memory to be private for QEMU 5175 (``share=off``). For this use case, we need writable RAM instead 5176 of ROM, and want to also set ``rom=off``. 5177 5178 ``-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`` 5179 Creates a memory backend object, which can be used to back the 5180 guest RAM. Memory backend objects offer more control than the 5181 ``-m`` option that is traditionally used to define guest RAM. 5182 Please refer to ``memory-backend-file`` for a description of the 5183 options. 5184 5185 ``-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`` 5186 Creates an anonymous memory file backend object, which allows 5187 QEMU to share the memory with an external process (e.g. when 5188 using vhost-user). The memory is allocated with memfd and 5189 optional sealing. (Linux only) 5190 5191 The ``seal`` option creates a sealed-file, that will block 5192 further resizing the memory ('on' by default). 5193 5194 The ``hugetlb`` option specify the file to be created resides in 5195 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction 5196 with the ``hugetlb`` option, the ``hugetlbsize`` option specify 5197 the hugetlb page size on systems that support multiple hugetlb 5198 page sizes (it must be a power of 2 value supported by the 5199 system). 5200 5201 In some versions of Linux, the ``hugetlb`` option is 5202 incompatible with the ``seal`` option (requires at least Linux 5203 4.16). 5204 5205 Please refer to ``memory-backend-file`` for a description of the 5206 other options. 5207 5208 The ``share`` boolean option is on by default with memfd. 5209 5210 ``-object rng-builtin,id=id`` 5211 Creates a random number generator backend which obtains entropy 5212 from QEMU builtin functions. The ``id`` parameter is a unique ID 5213 that will be used to reference this entropy backend from the 5214 ``virtio-rng`` device. By default, the ``virtio-rng`` device 5215 uses this RNG backend. 5216 5217 ``-object rng-random,id=id,filename=/dev/random`` 5218 Creates a random number generator backend which obtains entropy 5219 from a device on the host. The ``id`` parameter is a unique ID 5220 that will be used to reference this entropy backend from the 5221 ``virtio-rng`` device. The ``filename`` parameter specifies 5222 which file to obtain entropy from and if omitted defaults to 5223 ``/dev/urandom``. 5224 5225 ``-object rng-egd,id=id,chardev=chardevid`` 5226 Creates a random number generator backend which obtains entropy 5227 from an external daemon running on the host. The ``id`` 5228 parameter is a unique ID that will be used to reference this 5229 entropy backend from the ``virtio-rng`` device. The ``chardev`` 5230 parameter is the unique ID of a character device backend that 5231 provides the connection to the RNG daemon. 5232 5233 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off`` 5234 Creates a TLS anonymous credentials object, which can be used to 5235 provide TLS support on network backends. The ``id`` parameter is 5236 a unique ID which network backends will use to access the 5237 credentials. The ``endpoint`` is either ``server`` or ``client`` 5238 depending on whether the QEMU network backend that uses the 5239 credentials will be acting as a client or as a server. If 5240 ``verify-peer`` is enabled (the default) then once the handshake 5241 is completed, the peer credentials will be verified, though this 5242 is a no-op for anonymous credentials. 5243 5244 The dir parameter tells QEMU where to find the credential files. 5245 For server endpoints, this directory may contain a file 5246 dh-params.pem providing diffie-hellman parameters to use for the 5247 TLS server. If the file is missing, QEMU will generate a set of 5248 DH parameters at startup. This is a computationally expensive 5249 operation that consumes random pool entropy, so it is 5250 recommended that a persistent set of parameters be generated 5251 upfront and saved. 5252 5253 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]`` 5254 Creates a TLS Pre-Shared Keys (PSK) credentials object, which 5255 can be used to provide TLS support on network backends. The 5256 ``id`` parameter is a unique ID which network backends will use 5257 to access the credentials. The ``endpoint`` is either ``server`` 5258 or ``client`` depending on whether the QEMU network backend that 5259 uses the credentials will be acting as a client or as a server. 5260 For clients only, ``username`` is the username which will be 5261 sent to the server. If omitted it defaults to "qemu". 5262 5263 The dir parameter tells QEMU where to find the keys file. It is 5264 called "dir/keys.psk" and contains "username:key" pairs. This 5265 file can most easily be created using the GnuTLS ``psktool`` 5266 program. 5267 5268 For server endpoints, dir may also contain a file dh-params.pem 5269 providing diffie-hellman parameters to use for the TLS server. 5270 If the file is missing, QEMU will generate a set of DH 5271 parameters at startup. This is a computationally expensive 5272 operation that consumes random pool entropy, so it is 5273 recommended that a persistent set of parameters be generated up 5274 front and saved. 5275 5276 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id`` 5277 Creates a TLS anonymous credentials object, which can be used to 5278 provide TLS support on network backends. The ``id`` parameter is 5279 a unique ID which network backends will use to access the 5280 credentials. The ``endpoint`` is either ``server`` or ``client`` 5281 depending on whether the QEMU network backend that uses the 5282 credentials will be acting as a client or as a server. If 5283 ``verify-peer`` is enabled (the default) then once the handshake 5284 is completed, the peer credentials will be verified. With x509 5285 certificates, this implies that the clients must be provided 5286 with valid client certificates too. 5287 5288 The dir parameter tells QEMU where to find the credential files. 5289 For server endpoints, this directory may contain a file 5290 dh-params.pem providing diffie-hellman parameters to use for the 5291 TLS server. If the file is missing, QEMU will generate a set of 5292 DH parameters at startup. This is a computationally expensive 5293 operation that consumes random pool entropy, so it is 5294 recommended that a persistent set of parameters be generated 5295 upfront and saved. 5296 5297 For x509 certificate credentials the directory will contain 5298 further files providing the x509 certificates. The certificates 5299 must be stored in PEM format, in filenames ca-cert.pem, 5300 ca-crl.pem (optional), server-cert.pem (only servers), 5301 server-key.pem (only servers), client-cert.pem (only clients), 5302 and client-key.pem (only clients). 5303 5304 For the server-key.pem and client-key.pem files which contain 5305 sensitive private keys, it is possible to use an encrypted 5306 version by providing the passwordid parameter. This provides the 5307 ID of a previously created ``secret`` object containing the 5308 password for decryption. 5309 5310 The priority parameter allows to override the global default 5311 priority used by gnutls. This can be useful if the system 5312 administrator needs to use a weaker set of crypto priorities for 5313 QEMU without potentially forcing the weakness onto all 5314 applications. Or conversely if one wants wants a stronger 5315 default for QEMU than for all other applications, they can do 5316 this through this parameter. Its format is a gnutls priority 5317 string as described at 5318 https://gnutls.org/manual/html_node/Priority-Strings.html. 5319 5320 ``-object tls-cipher-suites,id=id,priority=priority`` 5321 Creates a TLS cipher suites object, which can be used to control 5322 the TLS cipher/protocol algorithms that applications are permitted 5323 to use. 5324 5325 The ``id`` parameter is a unique ID which frontends will use to 5326 access the ordered list of permitted TLS cipher suites from the 5327 host. 5328 5329 The ``priority`` parameter allows to override the global default 5330 priority used by gnutls. This can be useful if the system 5331 administrator needs to use a weaker set of crypto priorities for 5332 QEMU without potentially forcing the weakness onto all 5333 applications. Or conversely if one wants wants a stronger 5334 default for QEMU than for all other applications, they can do 5335 this through this parameter. Its format is a gnutls priority 5336 string as described at 5337 https://gnutls.org/manual/html_node/Priority-Strings.html. 5338 5339 An example of use of this object is to control UEFI HTTPS Boot. 5340 The tls-cipher-suites object exposes the ordered list of permitted 5341 TLS cipher suites from the host side to the guest firmware, via 5342 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER 5343 objects. The firmware uses the IANA_TLS_CIPHER array for configuring 5344 guest-side TLS. 5345 5346 In the following example, the priority at which the host-side policy 5347 is retrieved is given by the ``priority`` property. 5348 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to 5349 refer to /etc/crypto-policies/back-ends/gnutls.config. 5350 5351 .. parsed-literal:: 5352 5353 # |qemu_system| \\ 5354 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\ 5355 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0 5356 5357 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]`` 5358 Interval t can't be 0, this filter batches the packet delivery: 5359 all packets arriving in a given interval on netdev netdevid are 5360 delayed until the end of the interval. Interval is in 5361 microseconds. ``status`` is optional that indicate whether the 5362 netfilter is on (enabled) or off (disabled), the default status 5363 for netfilter will be 'on'. 5364 5365 queue all\|rx\|tx is an option that can be applied to any 5366 netfilter. 5367 5368 ``all``: the filter is attached both to the receive and the 5369 transmit queue of the netdev (default). 5370 5371 ``rx``: the filter is attached to the receive queue of the 5372 netdev, where it will receive packets sent to the netdev. 5373 5374 ``tx``: the filter is attached to the transmit queue of the 5375 netdev, where it will receive packets sent by the netdev. 5376 5377 position head\|tail\|id=<id> is an option to specify where the 5378 filter should be inserted in the filter list. It can be applied 5379 to any netfilter. 5380 5381 ``head``: the filter is inserted at the head of the filter list, 5382 before any existing filters. 5383 5384 ``tail``: the filter is inserted at the tail of the filter list, 5385 behind any existing filters (default). 5386 5387 ``id=<id>``: the filter is inserted before or behind the filter 5388 specified by <id>, see the insert option below. 5389 5390 insert behind\|before is an option to specify where to insert 5391 the new filter relative to the one specified with 5392 position=id=<id>. It can be applied to any netfilter. 5393 5394 ``before``: insert before the specified filter. 5395 5396 ``behind``: insert behind the specified filter (default). 5397 5398 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5399 filter-mirror on netdev netdevid,mirror net packet to 5400 chardevchardevid, if it has the vnet\_hdr\_support flag, 5401 filter-mirror will mirror packet with vnet\_hdr\_len. 5402 5403 ``-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]`` 5404 filter-redirector on netdev netdevid,redirect filter's net 5405 packet to chardev chardevid,and redirect indev's packet to 5406 filter.if it has the vnet\_hdr\_support flag, filter-redirector 5407 will redirect packet with vnet\_hdr\_len. Create a 5408 filter-redirector we need to differ outdev id from indev id, id 5409 can not be the same. we can just use indev or outdev, but at 5410 least one of indev or outdev need to be specified. 5411 5412 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5413 Filter-rewriter is a part of COLO project.It will rewrite tcp 5414 packet to secondary from primary to keep secondary tcp 5415 connection,and rewrite tcp packet to primary from secondary make 5416 tcp packet can be handled by client.if it has the 5417 vnet\_hdr\_support flag, we can parse packet with vnet header. 5418 5419 usage: colo secondary: -object 5420 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object 5421 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object 5422 filter-rewriter,id=rew0,netdev=hn0,queue=all 5423 5424 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]`` 5425 Dump the network traffic on netdev dev to the file specified by 5426 filename. At most len bytes (64k by default) per packet are 5427 stored. The file format is libpcap, so it can be analyzed with 5428 tools such as tcpdump or Wireshark. 5429 5430 ``-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}]`` 5431 Colo-compare gets packet from primary\_in chardevid and 5432 secondary\_in, then compare whether the payload of primary packet 5433 and secondary packet are the same. If same, it will output 5434 primary packet to out\_dev, else it will notify COLO-framework to do 5435 checkpoint and send primary packet to out\_dev. In order to 5436 improve efficiency, we need to put the task of comparison in 5437 another iothread. If it has the vnet\_hdr\_support flag, 5438 colo compare will send/recv packet with vnet\_hdr\_len. 5439 The compare\_timeout=@var{ms} determines the maximum time of the 5440 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms} 5441 is to set the period of scanning expired primary node network packets. 5442 The max\_queue\_size=@var{size} is to set the max compare queue 5443 size depend on user environment. 5444 If user want to use Xen COLO, need to add the notify\_dev to 5445 notify Xen colo-frame to do checkpoint. 5446 5447 COLO-compare must be used with the help of filter-mirror, 5448 filter-redirector and filter-rewriter. 5449 5450 :: 5451 5452 KVM COLO 5453 5454 primary: 5455 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 5456 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5457 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5458 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5459 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5460 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5461 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5462 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5463 -object iothread,id=iothread1 5464 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5465 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5466 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5467 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1 5468 5469 secondary: 5470 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 5471 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5472 -chardev socket,id=red0,host=3.3.3.3,port=9003 5473 -chardev socket,id=red1,host=3.3.3.3,port=9004 5474 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5475 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5476 5477 5478 Xen COLO 5479 5480 primary: 5481 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 5482 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5483 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5484 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5485 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5486 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5487 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5488 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5489 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off 5490 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5491 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5492 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5493 -object iothread,id=iothread1 5494 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1 5495 5496 secondary: 5497 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 5498 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5499 -chardev socket,id=red0,host=3.3.3.3,port=9003 5500 -chardev socket,id=red1,host=3.3.3.3,port=9004 5501 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5502 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5503 5504 If you want to know the detail of above command line, you can 5505 read the colo-compare git log. 5506 5507 ``-object cryptodev-backend-builtin,id=id[,queues=queues]`` 5508 Creates a cryptodev backend which executes crypto operations from 5509 the QEMU cipher APIs. The id parameter is a unique ID that will 5510 be used to reference this cryptodev backend from the 5511 ``virtio-crypto`` device. The queues parameter is optional, 5512 which specify the queue number of cryptodev backend, the default 5513 of queues is 1. 5514 5515 .. parsed-literal:: 5516 5517 # |qemu_system| \\ 5518 [...] \\ 5519 -object cryptodev-backend-builtin,id=cryptodev0 \\ 5520 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5521 [...] 5522 5523 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]`` 5524 Creates a vhost-user cryptodev backend, backed by a chardev 5525 chardevid. The id parameter is a unique ID that will be used to 5526 reference this cryptodev backend from the ``virtio-crypto`` 5527 device. The chardev should be a unix domain socket backed one. 5528 The vhost-user uses a specifically defined protocol to pass 5529 vhost ioctl replacement messages to an application on the other 5530 end of the socket. The queues parameter is optional, which 5531 specify the queue number of cryptodev backend for multiqueue 5532 vhost-user, the default of queues is 1. 5533 5534 .. parsed-literal:: 5535 5536 # |qemu_system| \\ 5537 [...] \\ 5538 -chardev socket,id=chardev0,path=/path/to/socket \\ 5539 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\ 5540 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5541 [...] 5542 5543 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]`` 5544 \ 5545 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]`` 5546 Defines a secret to store a password, encryption key, or some 5547 other sensitive data. The sensitive data can either be passed 5548 directly via the data parameter, or indirectly via the file 5549 parameter. Using the data parameter is insecure unless the 5550 sensitive data is encrypted. 5551 5552 The sensitive data can be provided in raw format (the default), 5553 or base64. When encoded as JSON, the raw format only supports 5554 valid UTF-8 characters, so base64 is recommended for sending 5555 binary data. QEMU will convert from which ever format is 5556 provided to the format it needs internally. eg, an RBD password 5557 can be provided in raw format, even though it will be base64 5558 encoded when passed onto the RBD sever. 5559 5560 For added protection, it is possible to encrypt the data 5561 associated with a secret using the AES-256-CBC cipher. Use of 5562 encryption is indicated by providing the keyid and iv 5563 parameters. The keyid parameter provides the ID of a previously 5564 defined secret that contains the AES-256 decryption key. This 5565 key should be 32-bytes long and be base64 encoded. The iv 5566 parameter provides the random initialization vector used for 5567 encryption of this particular secret and should be a base64 5568 encrypted string of the 16-byte IV. 5569 5570 The simplest (insecure) usage is to provide the secret inline 5571 5572 .. parsed-literal:: 5573 5574 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw 5575 5576 The simplest secure usage is to provide the secret via a file 5577 5578 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object 5579 secret,id=sec0,file=mypasswd.txt,format=raw 5580 5581 For greater security, AES-256-CBC should be used. To illustrate 5582 usage, consider the openssl command line tool which can encrypt 5583 the data. Note that when encrypting, the plaintext must be 5584 padded to the cipher block size (32 bytes) using the standard 5585 PKCS#5/6 compatible padding algorithm. 5586 5587 First a master key needs to be created in base64 encoding: 5588 5589 :: 5590 5591 # openssl rand -base64 32 > key.b64 5592 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"') 5593 5594 Each secret to be encrypted needs to have a random 5595 initialization vector generated. These do not need to be kept 5596 secret 5597 5598 :: 5599 5600 # openssl rand -base64 16 > iv.b64 5601 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"') 5602 5603 The secret to be defined can now be encrypted, in this case 5604 we're telling openssl to base64 encode the result, but it could 5605 be left as raw bytes if desired. 5606 5607 :: 5608 5609 # SECRET=$(printf "letmein" | 5610 openssl enc -aes-256-cbc -a -K $KEY -iv $IV) 5611 5612 When launching QEMU, create a master secret pointing to 5613 ``key.b64`` and specify that to be used to decrypt the user 5614 password. Pass the contents of ``iv.b64`` to the second secret 5615 5616 .. parsed-literal:: 5617 5618 # |qemu_system| \\ 5619 -object secret,id=secmaster0,format=base64,file=key.b64 \\ 5620 -object secret,id=sec0,keyid=secmaster0,format=base64,\\ 5621 data=$SECRET,iv=$(<iv.b64) 5622 5623 ``-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]`` 5624 Create a Secure Encrypted Virtualization (SEV) guest object, 5625 which can be used to provide the guest memory encryption support 5626 on AMD processors. 5627 5628 When memory encryption is enabled, one of the physical address 5629 bit (aka the C-bit) is utilized to mark if a memory page is 5630 protected. The ``cbitpos`` is used to provide the C-bit 5631 position. The C-bit position is Host family dependent hence user 5632 must provide this value. On EPYC, the value should be 47. 5633 5634 When memory encryption is enabled, we loose certain bits in 5635 physical address space. The ``reduced-phys-bits`` is used to 5636 provide the number of bits we loose in physical address space. 5637 Similar to C-bit, the value is Host family dependent. On EPYC, 5638 a guest will lose a maximum of 1 bit, so the value should be 1. 5639 5640 The ``sev-device`` provides the device file to use for 5641 communicating with the SEV firmware running inside AMD Secure 5642 Processor. The default device is '/dev/sev'. If hardware 5643 supports memory encryption then /dev/sev devices are created by 5644 CCP driver. 5645 5646 The ``policy`` provides the guest policy to be enforced by the 5647 SEV firmware and restrict what configuration and operational 5648 commands can be performed on this guest by the hypervisor. The 5649 policy should be provided by the guest owner and is bound to the 5650 guest and cannot be changed throughout the lifetime of the 5651 guest. The default is 0. 5652 5653 If guest ``policy`` allows sharing the key with another SEV 5654 guest then ``handle`` can be use to provide handle of the guest 5655 from which to share the key. 5656 5657 The ``dh-cert-file`` and ``session-file`` provides the guest 5658 owner's Public Diffie-Hillman key defined in SEV spec. The PDH 5659 and session parameters are used for establishing a cryptographic 5660 session with the guest owner to negotiate keys used for 5661 attestation. The file must be encoded in base64. 5662 5663 The ``kernel-hashes`` adds the hashes of given kernel/initrd/ 5664 cmdline to a designated guest firmware page for measured Linux 5665 boot with -kernel. The default is off. (Since 6.2) 5666 5667 e.g to launch a SEV guest 5668 5669 .. parsed-literal:: 5670 5671 # |qemu_system_x86| \\ 5672 ...... \\ 5673 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=1 \\ 5674 -machine ...,memory-encryption=sev0 \\ 5675 ..... 5676 5677 ``-object authz-simple,id=id,identity=string`` 5678 Create an authorization object that will control access to 5679 network services. 5680 5681 The ``identity`` parameter is identifies the user and its format 5682 depends on the network service that authorization object is 5683 associated with. For authorizing based on TLS x509 certificates, 5684 the identity must be the x509 distinguished name. Note that care 5685 must be taken to escape any commas in the distinguished name. 5686 5687 An example authorization object to validate a x509 distinguished 5688 name would look like: 5689 5690 .. parsed-literal:: 5691 5692 # |qemu_system| \\ 5693 ... \\ 5694 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\ 5695 ... 5696 5697 Note the use of quotes due to the x509 distinguished name 5698 containing whitespace, and escaping of ','. 5699 5700 ``-object authz-listfile,id=id,filename=path,refresh=on|off`` 5701 Create an authorization object that will control access to 5702 network services. 5703 5704 The ``filename`` parameter is the fully qualified path to a file 5705 containing the access control list rules in JSON format. 5706 5707 An example set of rules that match against SASL usernames might 5708 look like: 5709 5710 :: 5711 5712 { 5713 "rules": [ 5714 { "match": "fred", "policy": "allow", "format": "exact" }, 5715 { "match": "bob", "policy": "allow", "format": "exact" }, 5716 { "match": "danb", "policy": "deny", "format": "glob" }, 5717 { "match": "dan*", "policy": "allow", "format": "exact" }, 5718 ], 5719 "policy": "deny" 5720 } 5721 5722 When checking access the object will iterate over all the rules 5723 and the first rule to match will have its ``policy`` value 5724 returned as the result. If no rules match, then the default 5725 ``policy`` value is returned. 5726 5727 The rules can either be an exact string match, or they can use 5728 the simple UNIX glob pattern matching to allow wildcards to be 5729 used. 5730 5731 If ``refresh`` is set to true the file will be monitored and 5732 automatically reloaded whenever its content changes. 5733 5734 As with the ``authz-simple`` object, the format of the identity 5735 strings being matched depends on the network service, but is 5736 usually a TLS x509 distinguished name, or a SASL username. 5737 5738 An example authorization object to validate a SASL username 5739 would look like: 5740 5741 .. parsed-literal:: 5742 5743 # |qemu_system| \\ 5744 ... \\ 5745 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\ 5746 ... 5747 5748 ``-object authz-pam,id=id,service=string`` 5749 Create an authorization object that will control access to 5750 network services. 5751 5752 The ``service`` parameter provides the name of a PAM service to 5753 use for authorization. It requires that a file 5754 ``/etc/pam.d/service`` exist to provide the configuration for 5755 the ``account`` subsystem. 5756 5757 An example authorization object to validate a TLS x509 5758 distinguished name would look like: 5759 5760 .. parsed-literal:: 5761 5762 # |qemu_system| \\ 5763 ... \\ 5764 -object authz-pam,id=auth0,service=qemu-vnc \\ 5765 ... 5766 5767 There would then be a corresponding config file for PAM at 5768 ``/etc/pam.d/qemu-vnc`` that contains: 5769 5770 :: 5771 5772 account requisite pam_listfile.so item=user sense=allow \ 5773 file=/etc/qemu/vnc.allow 5774 5775 Finally the ``/etc/qemu/vnc.allow`` file would contain the list 5776 of x509 distinguished names that are permitted access 5777 5778 :: 5779 5780 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB 5781 5782 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch`` 5783 Creates a dedicated event loop thread that devices can be 5784 assigned to. This is known as an IOThread. By default device 5785 emulation happens in vCPU threads or the main event loop thread. 5786 This can become a scalability bottleneck. IOThreads allow device 5787 emulation and I/O to run on other host CPUs. 5788 5789 The ``id`` parameter is a unique ID that will be used to 5790 reference this IOThread from ``-device ...,iothread=id``. 5791 Multiple devices can be assigned to an IOThread. Note that not 5792 all devices support an ``iothread`` parameter. 5793 5794 The ``query-iothreads`` QMP command lists IOThreads and reports 5795 their thread IDs so that the user can configure host CPU 5796 pinning/affinity. 5797 5798 IOThreads use an adaptive polling algorithm to reduce event loop 5799 latency. Instead of entering a blocking system call to monitor 5800 file descriptors and then pay the cost of being woken up when an 5801 event occurs, the polling algorithm spins waiting for events for 5802 a short time. The algorithm's default parameters are suitable 5803 for many cases but can be adjusted based on knowledge of the 5804 workload and/or host device latency. 5805 5806 The ``poll-max-ns`` parameter is the maximum number of 5807 nanoseconds to busy wait for events. Polling can be disabled by 5808 setting this value to 0. 5809 5810 The ``poll-grow`` parameter is the multiplier used to increase 5811 the polling time when the algorithm detects it is missing events 5812 due to not polling long enough. 5813 5814 The ``poll-shrink`` parameter is the divisor used to decrease 5815 the polling time when the algorithm detects it is spending too 5816 long polling without encountering events. 5817 5818 The ``aio-max-batch`` parameter is the maximum number of requests 5819 in a batch for the AIO engine, 0 means that the engine will use 5820 its default. 5821 5822 The IOThread parameters can be modified at run-time using the 5823 ``qom-set`` command (where ``iothread1`` is the IOThread's 5824 ``id``): 5825 5826 :: 5827 5828 (qemu) qom-set /objects/iothread1 poll-max-ns 100000 5829ERST 5830 5831 5832HXCOMM This is the last statement. Insert new options before this line! 5833 5834#undef DEF 5835#undef DEFHEADING 5836#undef ARCHHEADING 5837
