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, hax, 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 " zpcii-disable=on|off disables zPCI interpretation facilities (default=off)\n", 42 QEMU_ARCH_ALL) 43SRST 44``-machine [type=]name[,prop=value[,...]]`` 45 Select the emulated machine by name. Use ``-machine help`` to list 46 available machines. 47 48 For architectures which aim to support live migration compatibility 49 across releases, each release will introduce a new versioned machine 50 type. For example, the 2.8.0 release introduced machine types 51 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures. 52 53 To allow live migration of guests from QEMU version 2.8.0, to QEMU 54 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8" 55 and "pc-q35-2.8" machines too. To allow users live migrating VMs to 56 skip multiple intermediate releases when upgrading, new releases of 57 QEMU will support machine types from many previous versions. 58 59 Supported machine properties are: 60 61 ``accel=accels1[:accels2[:...]]`` 62 This is used to enable an accelerator. Depending on the target 63 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available. 64 By default, tcg is used. If there is more than one accelerator 65 specified, the next one is used if the previous one fails to 66 initialize. 67 68 ``vmport=on|off|auto`` 69 Enables emulation of VMWare IO port, for vmmouse etc. auto says 70 to select the value based on accel. For accel=xen the default is 71 off otherwise the default is on. 72 73 ``dump-guest-core=on|off`` 74 Include guest memory in a core dump. The default is on. 75 76 ``mem-merge=on|off`` 77 Enables or disables memory merge support. This feature, when 78 supported by the host, de-duplicates identical memory pages 79 among VMs instances (enabled by default). 80 81 ``aes-key-wrap=on|off`` 82 Enables or disables AES key wrapping support on s390-ccw hosts. 83 This feature controls whether AES wrapping keys will be created 84 to allow execution of AES cryptographic functions. The default 85 is on. 86 87 ``dea-key-wrap=on|off`` 88 Enables or disables DEA key wrapping support on s390-ccw hosts. 89 This feature controls whether DEA wrapping keys will be created 90 to allow execution of DEA cryptographic functions. The default 91 is on. 92 93 ``nvdimm=on|off`` 94 Enables or disables NVDIMM support. The default is off. 95 96 ``memory-encryption=`` 97 Memory encryption object to use. The default is none. 98 99 ``hmat=on|off`` 100 Enables or disables ACPI Heterogeneous Memory Attribute Table 101 (HMAT) support. The default is off. 102 103 ``memory-backend='id'`` 104 An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options. 105 Allows to use a memory backend as main RAM. 106 107 For example: 108 :: 109 110 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on 111 -machine memory-backend=pc.ram 112 -m 512M 113 114 Migration compatibility note: 115 116 * as backend id one shall use value of 'default-ram-id', advertised by 117 machine type (available via ``query-machines`` QMP command), if migration 118 to/from old QEMU (<5.0) is expected. 119 * for machine types 4.0 and older, user shall 120 use ``x-use-canonical-path-for-ramblock-id=off`` backend option 121 if migration to/from old QEMU (<5.0) is expected. 122 123 For example: 124 :: 125 126 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off 127 -machine memory-backend=pc.ram 128 -m 512M 129 130 ``cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]`` 131 Define a CXL Fixed Memory Window (CFMW). 132 133 Described in the CXL 2.0 ECN: CEDT CFMWS & QTG _DSM. 134 135 They are regions of Host Physical Addresses (HPA) on a system which 136 may be interleaved across one or more CXL host bridges. The system 137 software will assign particular devices into these windows and 138 configure the downstream Host-managed Device Memory (HDM) decoders 139 in root ports, switch ports and devices appropriately to meet the 140 interleave requirements before enabling the memory devices. 141 142 ``targets.X=target`` provides the mapping to CXL host bridges 143 which may be identified by the id provied in the -device entry. 144 Multiple entries are needed to specify all the targets when 145 the fixed memory window represents interleaved memory. X is the 146 target index from 0. 147 148 ``size=size`` sets the size of the CFMW. This must be a multiple of 149 256MiB. The region will be aligned to 256MiB but the location is 150 platform and configuration dependent. 151 152 ``interleave-granularity=granularity`` sets the granularity of 153 interleave. Default 256KiB. Only 256KiB, 512KiB, 1024KiB, 2048KiB 154 4096KiB, 8192KiB and 16384KiB granularities supported. 155 156 Example: 157 158 :: 159 160 -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 161 162 ``zpcii-disable=on|off`` 163 Disables zPCI interpretation facilties on s390-ccw hosts. 164 This feature can be used to disable hardware virtual assists 165 related to zPCI devices. The default is off. 166ERST 167 168DEF("M", HAS_ARG, QEMU_OPTION_M, 169 " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n", 170 QEMU_ARCH_ALL) 171 172SRST 173``sgx-epc.0.memdev=@var{memid},sgx-epc.0.node=@var{numaid}`` 174 Define an SGX EPC section. 175ERST 176 177DEF("cpu", HAS_ARG, QEMU_OPTION_cpu, 178 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL) 179SRST 180``-cpu model`` 181 Select CPU model (``-cpu help`` for list and additional feature 182 selection) 183ERST 184 185DEF("accel", HAS_ARG, QEMU_OPTION_accel, 186 "-accel [accel=]accelerator[,prop[=value][,...]]\n" 187 " select accelerator (kvm, xen, hax, hvf, nvmm, whpx or tcg; use 'help' for a list)\n" 188 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n" 189 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n" 190 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n" 191 " split-wx=on|off (enable TCG split w^x mapping)\n" 192 " tb-size=n (TCG translation block cache size)\n" 193 " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n" 194 " notify-vmexit=run|internal-error|disable,notify-window=n (enable notify VM exit and set notify window, x86 only)\n" 195 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL) 196SRST 197``-accel name[,prop=value[,...]]`` 198 This is used to enable an accelerator. Depending on the target 199 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available. By 200 default, tcg is used. If there is more than one accelerator 201 specified, the next one is used if the previous one fails to 202 initialize. 203 204 ``igd-passthru=on|off`` 205 When Xen is in use, this option controls whether Intel 206 integrated graphics devices can be passed through to the guest 207 (default=off) 208 209 ``kernel-irqchip=on|off|split`` 210 Controls KVM in-kernel irqchip support. The default is full 211 acceleration of the interrupt controllers. On x86, split irqchip 212 reduces the kernel attack surface, at a performance cost for 213 non-MSI interrupts. Disabling the in-kernel irqchip completely 214 is not recommended except for debugging purposes. 215 216 ``kvm-shadow-mem=size`` 217 Defines the size of the KVM shadow MMU. 218 219 ``split-wx=on|off`` 220 Controls the use of split w^x mapping for the TCG code generation 221 buffer. Some operating systems require this to be enabled, and in 222 such a case this will default on. On other operating systems, this 223 will default off, but one may enable this for testing or debugging. 224 225 ``tb-size=n`` 226 Controls the size (in MiB) of the TCG translation block cache. 227 228 ``thread=single|multi`` 229 Controls number of TCG threads. When the TCG is multi-threaded 230 there will be one thread per vCPU therefore taking advantage of 231 additional host cores. The default is to enable multi-threading 232 where both the back-end and front-ends support it and no 233 incompatible TCG features have been enabled (e.g. 234 icount/replay). 235 236 ``dirty-ring-size=n`` 237 When the KVM accelerator is used, it controls the size of the per-vCPU 238 dirty page ring buffer (number of entries for each vCPU). It should 239 be a value that is power of two, and it should be 1024 or bigger (but 240 still less than the maximum value that the kernel supports). 4096 241 could be a good initial value if you have no idea which is the best. 242 Set this value to 0 to disable the feature. By default, this feature 243 is disabled (dirty-ring-size=0). When enabled, KVM will instead 244 record dirty pages in a bitmap. 245 246 ``notify-vmexit=run|internal-error|disable,notify-window=n`` 247 Enables or disables notify VM exit support on x86 host and specify 248 the corresponding notify window to trigger the VM exit if enabled. 249 ``run`` option enables the feature. It does nothing and continue 250 if the exit happens. ``internal-error`` option enables the feature. 251 It raises a internal error. ``disable`` option doesn't enable the feature. 252 This feature can mitigate the CPU stuck issue due to event windows don't 253 open up for a specified of time (i.e. notify-window). 254 Default: notify-vmexit=run,notify-window=0. 255 256ERST 257 258DEF("smp", HAS_ARG, QEMU_OPTION_smp, 259 "-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]\n" 260 " set the number of initial CPUs to 'n' [default=1]\n" 261 " maxcpus= maximum number of total CPUs, including\n" 262 " offline CPUs for hotplug, etc\n" 263 " sockets= number of sockets on the machine board\n" 264 " dies= number of dies in one socket\n" 265 " clusters= number of clusters in one die\n" 266 " cores= number of cores in one cluster\n" 267 " threads= number of threads in one core\n" 268 "Note: Different machines may have different subsets of the CPU topology\n" 269 " parameters supported, so the actual meaning of the supported parameters\n" 270 " will vary accordingly. For example, for a machine type that supports a\n" 271 " three-level CPU hierarchy of sockets/cores/threads, the parameters will\n" 272 " sequentially mean as below:\n" 273 " sockets means the number of sockets on the machine board\n" 274 " cores means the number of cores in one socket\n" 275 " threads means the number of threads in one core\n" 276 " For a particular machine type board, an expected CPU topology hierarchy\n" 277 " can be defined through the supported sub-option. Unsupported parameters\n" 278 " can also be provided in addition to the sub-option, but their values\n" 279 " must be set as 1 in the purpose of correct parsing.\n", 280 QEMU_ARCH_ALL) 281SRST 282``-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]`` 283 Simulate a SMP system with '\ ``n``\ ' CPUs initially present on 284 the machine type board. On boards supporting CPU hotplug, the optional 285 '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be 286 added at runtime. When both parameters are omitted, the maximum number 287 of CPUs will be calculated from the provided topology members and the 288 initial CPU count will match the maximum number. When only one of them 289 is given then the omitted one will be set to its counterpart's value. 290 Both parameters may be specified, but the maximum number of CPUs must 291 be equal to or greater than the initial CPU count. Product of the 292 CPU topology hierarchy must be equal to the maximum number of CPUs. 293 Both parameters are subject to an upper limit that is determined by 294 the specific machine type chosen. 295 296 To control reporting of CPU topology information, values of the topology 297 parameters can be specified. Machines may only support a subset of the 298 parameters and different machines may have different subsets supported 299 which vary depending on capacity of the corresponding CPU targets. So 300 for a particular machine type board, an expected topology hierarchy can 301 be defined through the supported sub-option. Unsupported parameters can 302 also be provided in addition to the sub-option, but their values must be 303 set as 1 in the purpose of correct parsing. 304 305 Either the initial CPU count, or at least one of the topology parameters 306 must be specified. The specified parameters must be greater than zero, 307 explicit configuration like "cpus=0" is not allowed. Values for any 308 omitted parameters will be computed from those which are given. 309 310 For example, the following sub-option defines a CPU topology hierarchy 311 (2 sockets totally on the machine, 2 cores per socket, 2 threads per 312 core) for a machine that only supports sockets/cores/threads. 313 Some members of the option can be omitted but their values will be 314 automatically computed: 315 316 :: 317 318 -smp 8,sockets=2,cores=2,threads=2,maxcpus=8 319 320 The following sub-option defines a CPU topology hierarchy (2 sockets 321 totally on the machine, 2 dies per socket, 2 cores per die, 2 threads 322 per core) for PC machines which support sockets/dies/cores/threads. 323 Some members of the option can be omitted but their values will be 324 automatically computed: 325 326 :: 327 328 -smp 16,sockets=2,dies=2,cores=2,threads=2,maxcpus=16 329 330 The following sub-option defines a CPU topology hierarchy (2 sockets 331 totally on the machine, 2 clusters per socket, 2 cores per cluster, 332 2 threads per core) for ARM virt machines which support sockets/clusters 333 /cores/threads. Some members of the option can be omitted but their values 334 will be automatically computed: 335 336 :: 337 338 -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16 339 340 Historically preference was given to the coarsest topology parameters 341 when computing missing values (ie sockets preferred over cores, which 342 were preferred over threads), however, this behaviour is considered 343 liable to change. Prior to 6.2 the preference was sockets over cores 344 over threads. Since 6.2 the preference is cores over sockets over threads. 345 346 For example, the following option defines a machine board with 2 sockets 347 of 1 core before 6.2 and 1 socket of 2 cores after 6.2: 348 349 :: 350 351 -smp 2 352ERST 353 354DEF("numa", HAS_ARG, QEMU_OPTION_numa, 355 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 356 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 357 "-numa dist,src=source,dst=destination,val=distance\n" 358 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n" 359 "-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" 360 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n", 361 QEMU_ARCH_ALL) 362SRST 363``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 364 \ 365``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 366 \ 367``-numa dist,src=source,dst=destination,val=distance`` 368 \ 369``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]`` 370 \ 371``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=tpye[,latency=lat][,bandwidth=bw]`` 372 \ 373``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]`` 374 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA 375 distance from a source node to a destination node. Set the ACPI 376 Heterogeneous Memory Attributes for the given nodes. 377 378 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and 379 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a 380 contiguous range of CPU indexes (or a single VCPU if lastcpu is 381 omitted). A non-contiguous set of VCPUs can be represented by 382 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is 383 omitted on all nodes, VCPUs are automatically split between them. 384 385 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a 386 NUMA node: 387 388 :: 389 390 -numa node,cpus=0-2,cpus=5 391 392 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option 393 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to 394 assign CPU objects to a node using topology layout properties of 395 CPU. The set of properties is machine specific, and depends on used 396 machine type/'\ ``smp``\ ' options. It could be queried with 397 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ ' 398 property specifies node to which CPU object will be assigned, it's 399 required for node to be declared with '\ ``node``\ ' option before 400 it's used with '\ ``cpu``\ ' option. 401 402 For example: 403 404 :: 405 406 -M pc \ 407 -smp 1,sockets=2,maxcpus=2 \ 408 -numa node,nodeid=0 -numa node,nodeid=1 \ 409 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 410 411 Legacy '\ ``mem``\ ' assigns a given RAM amount to a node (not supported 412 for 5.1 and newer machine types). '\ ``memdev``\ ' assigns RAM from 413 a given memory backend device to a node. If '\ ``mem``\ ' and 414 '\ ``memdev``\ ' are omitted in all nodes, RAM is split equally between them. 415 416 417 '\ ``mem``\ ' and '\ ``memdev``\ ' are mutually exclusive. 418 Furthermore, if one node uses '\ ``memdev``\ ', all of them have to 419 use it. 420 421 '\ ``initiator``\ ' is an additional option that points to an 422 initiator NUMA node that has best performance (the lowest latency or 423 largest bandwidth) to this NUMA node. Note that this option can be 424 set only when the machine property 'hmat' is set to 'on'. 425 426 Following example creates a machine with 2 NUMA nodes, node 0 has 427 CPU. node 1 has only memory, and its initiator is node 0. Note that 428 because node 0 has CPU, by default the initiator of node 0 is itself 429 and must be itself. 430 431 :: 432 433 -machine hmat=on \ 434 -m 2G,slots=2,maxmem=4G \ 435 -object memory-backend-ram,size=1G,id=m0 \ 436 -object memory-backend-ram,size=1G,id=m1 \ 437 -numa node,nodeid=0,memdev=m0 \ 438 -numa node,nodeid=1,memdev=m1,initiator=0 \ 439 -smp 2,sockets=2,maxcpus=2 \ 440 -numa cpu,node-id=0,socket-id=0 \ 441 -numa cpu,node-id=0,socket-id=1 442 443 source and destination are NUMA node IDs. distance is the NUMA 444 distance from source to destination. The distance from a node to 445 itself is always 10. If any pair of nodes is given a distance, then 446 all pairs must be given distances. Although, when distances are only 447 given in one direction for each pair of nodes, then the distances in 448 the opposite directions are assumed to be the same. If, however, an 449 asymmetrical pair of distances is given for even one node pair, then 450 all node pairs must be provided distance values for both directions, 451 even when they are symmetrical. When a node is unreachable from 452 another node, set the pair's distance to 255. 453 454 Note that the -``numa`` option doesn't allocate any of the specified 455 resources, it just assigns existing resources to NUMA nodes. This 456 means that one still has to use the ``-m``, ``-smp`` options to 457 allocate RAM and VCPUs respectively. 458 459 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth 460 Information between initiator and target NUMA nodes in ACPI 461 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can 462 create memory requests, usually it has one or more processors. 463 Target NUMA node contains addressable memory. 464 465 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is 466 the memory hierarchy of the target NUMA node: if hierarchy is 467 'memory', the structure represents the memory performance; if 468 hierarchy is 'first-level\|second-level\|third-level', this 469 structure represents aggregated performance of memory side caches 470 for each domain. type of 'data-type' is type of data represented by 471 this structure instance: if 'hierarchy' is 'memory', 'data-type' is 472 'access\|read\|write' latency or 'access\|read\|write' bandwidth of 473 the target memory; if 'hierarchy' is 474 'first-level\|second-level\|third-level', 'data-type' is 475 'access\|read\|write' hit latency or 'access\|read\|write' hit 476 bandwidth of the target memory side cache. 477 478 lat is latency value in nanoseconds. bw is bandwidth value, the 479 possible value and units are NUM[M\|G\|T], mean that the bandwidth 480 value are NUM byte per second (or MB/s, GB/s or TB/s depending on 481 used suffix). Note that if latency or bandwidth value is 0, means 482 the corresponding latency or bandwidth information is not provided. 483 484 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory 485 belongs. size is the size of memory side cache in bytes. level is 486 the cache level described in this structure, note that the cache 487 level 0 should not be used with '\ ``hmat-cache``\ ' option. 488 associativity is the cache associativity, the possible value is 489 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy 490 is the write policy. line is the cache Line size in bytes. 491 492 For example, the following options describe 2 NUMA nodes. Node 0 has 493 2 cpus and a ram, node 1 has only a ram. The processors in node 0 494 access memory in node 0 with access-latency 5 nanoseconds, 495 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access 496 memory in NUMA node 1 with access-latency 10 nanoseconds, 497 access-bandwidth is 100 MB/s. And for memory side cache information, 498 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB, 499 policy is write-back, the cache Line size is 8 bytes: 500 501 :: 502 503 -machine hmat=on \ 504 -m 2G \ 505 -object memory-backend-ram,size=1G,id=m0 \ 506 -object memory-backend-ram,size=1G,id=m1 \ 507 -smp 2,sockets=2,maxcpus=2 \ 508 -numa node,nodeid=0,memdev=m0 \ 509 -numa node,nodeid=1,memdev=m1,initiator=0 \ 510 -numa cpu,node-id=0,socket-id=0 \ 511 -numa cpu,node-id=0,socket-id=1 \ 512 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \ 513 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \ 514 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \ 515 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \ 516 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \ 517 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8 518ERST 519 520DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd, 521 "-add-fd fd=fd,set=set[,opaque=opaque]\n" 522 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL) 523SRST 524``-add-fd fd=fd,set=set[,opaque=opaque]`` 525 Add a file descriptor to an fd set. Valid options are: 526 527 ``fd=fd`` 528 This option defines the file descriptor of which a duplicate is 529 added to fd set. The file descriptor cannot be stdin, stdout, or 530 stderr. 531 532 ``set=set`` 533 This option defines the ID of the fd set to add the file 534 descriptor to. 535 536 ``opaque=opaque`` 537 This option defines a free-form string that can be used to 538 describe fd. 539 540 You can open an image using pre-opened file descriptors from an fd 541 set: 542 543 .. parsed-literal:: 544 545 |qemu_system| \\ 546 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 547 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 548 -drive file=/dev/fdset/2,index=0,media=disk 549ERST 550 551DEF("set", HAS_ARG, QEMU_OPTION_set, 552 "-set group.id.arg=value\n" 553 " set <arg> parameter for item <id> of type <group>\n" 554 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL) 555SRST 556``-set group.id.arg=value`` 557 Set parameter arg for item id of type group 558ERST 559 560DEF("global", HAS_ARG, QEMU_OPTION_global, 561 "-global driver.property=value\n" 562 "-global driver=driver,property=property,value=value\n" 563 " set a global default for a driver property\n", 564 QEMU_ARCH_ALL) 565SRST 566``-global driver.prop=value`` 567 \ 568``-global driver=driver,property=property,value=value`` 569 Set default value of driver's property prop to value, e.g.: 570 571 .. parsed-literal:: 572 573 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img 574 575 In particular, you can use this to set driver properties for devices 576 which are created automatically by the machine model. To create a 577 device which is not created automatically and set properties on it, 578 use -``device``. 579 580 -global driver.prop=value is shorthand for -global 581 driver=driver,property=prop,value=value. The longhand syntax works 582 even when driver contains a dot. 583ERST 584 585DEF("boot", HAS_ARG, QEMU_OPTION_boot, 586 "-boot [order=drives][,once=drives][,menu=on|off]\n" 587 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n" 588 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n" 589 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n" 590 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n" 591 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n", 592 QEMU_ARCH_ALL) 593SRST 594``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]`` 595 Specify boot order drives as a string of drive letters. Valid drive 596 letters depend on the target architecture. The x86 PC uses: a, b 597 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p 598 (Etherboot from network adapter 1-4), hard disk boot is the default. 599 To apply a particular boot order only on the first startup, specify 600 it via ``once``. Note that the ``order`` or ``once`` parameter 601 should not be used together with the ``bootindex`` property of 602 devices, since the firmware implementations normally do not support 603 both at the same time. 604 605 Interactive boot menus/prompts can be enabled via ``menu=on`` as far 606 as firmware/BIOS supports them. The default is non-interactive boot. 607 608 A splash picture could be passed to bios, enabling user to show it 609 as logo, when option splash=sp\_name is given and menu=on, If 610 firmware/BIOS supports them. Currently Seabios for X86 system 611 support it. limitation: The splash file could be a jpeg file or a 612 BMP file in 24 BPP format(true color). The resolution should be 613 supported by the SVGA mode, so the recommended is 320x240, 640x480, 614 800x640. 615 616 A timeout could be passed to bios, guest will pause for rb\_timeout 617 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will 618 not reboot, qemu passes '-1' to bios by default. Currently Seabios 619 for X86 system support it. 620 621 Do strict boot via ``strict=on`` as far as firmware/BIOS supports 622 it. This only effects when boot priority is changed by bootindex 623 options. The default is non-strict boot. 624 625 .. parsed-literal:: 626 627 # try to boot from network first, then from hard disk 628 |qemu_system_x86| -boot order=nc 629 # boot from CD-ROM first, switch back to default order after reboot 630 |qemu_system_x86| -boot once=d 631 # boot with a splash picture for 5 seconds. 632 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000 633 634 Note: The legacy format '-boot drives' is still supported but its 635 use is discouraged as it may be removed from future versions. 636ERST 637 638DEF("m", HAS_ARG, QEMU_OPTION_m, 639 "-m [size=]megs[,slots=n,maxmem=size]\n" 640 " configure guest RAM\n" 641 " size: initial amount of guest memory\n" 642 " slots: number of hotplug slots (default: none)\n" 643 " maxmem: maximum amount of guest memory (default: none)\n" 644 "NOTE: Some architectures might enforce a specific granularity\n", 645 QEMU_ARCH_ALL) 646SRST 647``-m [size=]megs[,slots=n,maxmem=size]`` 648 Sets guest startup RAM size to megs megabytes. Default is 128 MiB. 649 Optionally, a suffix of "M" or "G" can be used to signify a value in 650 megabytes or gigabytes respectively. Optional pair slots, maxmem 651 could be used to set amount of hotpluggable memory slots and maximum 652 amount of memory. Note that maxmem must be aligned to the page size. 653 654 For example, the following command-line sets the guest startup RAM 655 size to 1GB, creates 3 slots to hotplug additional memory and sets 656 the maximum memory the guest can reach to 4GB: 657 658 .. parsed-literal:: 659 660 |qemu_system| -m 1G,slots=3,maxmem=4G 661 662 If slots and maxmem are not specified, memory hotplug won't be 663 enabled and the guest startup RAM will never increase. 664ERST 665 666DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath, 667 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL) 668SRST 669``-mem-path path`` 670 Allocate guest RAM from a temporarily created file in path. 671ERST 672 673DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc, 674 "-mem-prealloc preallocate guest memory (use with -mem-path)\n", 675 QEMU_ARCH_ALL) 676SRST 677``-mem-prealloc`` 678 Preallocate memory when using -mem-path. 679ERST 680 681DEF("k", HAS_ARG, QEMU_OPTION_k, 682 "-k language use keyboard layout (for example 'fr' for French)\n", 683 QEMU_ARCH_ALL) 684SRST 685``-k language`` 686 Use keyboard layout language (for example ``fr`` for French). This 687 option is only needed where it is not easy to get raw PC keycodes 688 (e.g. on Macs, with some X11 servers or with a VNC or curses 689 display). You don't normally need to use it on PC/Linux or 690 PC/Windows hosts. 691 692 The available layouts are: 693 694 :: 695 696 ar de-ch es fo fr-ca hu ja mk no pt-br sv 697 da en-gb et fr fr-ch is lt nl pl ru th 698 de en-us fi fr-be hr it lv nl-be pt sl tr 699 700 The default is ``en-us``. 701ERST 702 703 704HXCOMM Deprecated by -audiodev 705DEF("audio-help", 0, QEMU_OPTION_audio_help, 706 "-audio-help show -audiodev equivalent of the currently specified audio settings\n", 707 QEMU_ARCH_ALL) 708SRST 709``-audio-help`` 710 Will show the -audiodev equivalent of the currently specified 711 (deprecated) environment variables. 712ERST 713 714DEF("audio", HAS_ARG, QEMU_OPTION_audio, 715 "-audio [driver=]driver,model=value[,prop[=value][,...]]\n" 716 " specifies the audio backend and device to use;\n" 717 " apart from 'model', options are the same as for -audiodev.\n" 718 " use '-audio model=help' to show possible devices.\n", 719 QEMU_ARCH_ALL) 720SRST 721``-audio [driver=]driver,model=value[,prop[=value][,...]]`` 722 This option is a shortcut for configuring both the guest audio 723 hardware and the host audio backend in one go. 724 The driver option is the same as with the corresponding ``-audiodev`` option below. 725 The guest hardware model can be set with ``model=modelname``. 726 727 Use ``driver=help`` to list the available drivers, 728 and ``model=help`` to list the available device types. 729 730 The following two example do exactly the same, to show how ``-audio`` 731 can be used to shorten the command line length: 732 733 .. parsed-literal:: 734 735 |qemu_system| -audiodev pa,id=pa -device sb16,audiodev=pa 736 |qemu_system| -audio pa,model=sb16 737ERST 738 739DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, 740 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n" 741 " specifies the audio backend to use\n" 742 " Use ``-audiodev help`` to list the available drivers\n" 743 " id= identifier of the backend\n" 744 " timer-period= timer period in microseconds\n" 745 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n" 746 " in|out.fixed-settings= use fixed settings for host audio\n" 747 " in|out.frequency= frequency to use with fixed settings\n" 748 " in|out.channels= number of channels to use with fixed settings\n" 749 " in|out.format= sample format to use with fixed settings\n" 750 " valid values: s8, s16, s32, u8, u16, u32, f32\n" 751 " in|out.voices= number of voices to use\n" 752 " in|out.buffer-length= length of buffer in microseconds\n" 753 "-audiodev none,id=id,[,prop[=value][,...]]\n" 754 " dummy driver that discards all output\n" 755#ifdef CONFIG_AUDIO_ALSA 756 "-audiodev alsa,id=id[,prop[=value][,...]]\n" 757 " in|out.dev= name of the audio device to use\n" 758 " in|out.period-length= length of period in microseconds\n" 759 " in|out.try-poll= attempt to use poll mode\n" 760 " threshold= threshold (in microseconds) when playback starts\n" 761#endif 762#ifdef CONFIG_AUDIO_COREAUDIO 763 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n" 764 " in|out.buffer-count= number of buffers\n" 765#endif 766#ifdef CONFIG_AUDIO_DSOUND 767 "-audiodev dsound,id=id[,prop[=value][,...]]\n" 768 " latency= add extra latency to playback in microseconds\n" 769#endif 770#ifdef CONFIG_AUDIO_OSS 771 "-audiodev oss,id=id[,prop[=value][,...]]\n" 772 " in|out.dev= path of the audio device to use\n" 773 " in|out.buffer-count= number of buffers\n" 774 " in|out.try-poll= attempt to use poll mode\n" 775 " try-mmap= try using memory mapped access\n" 776 " exclusive= open device in exclusive mode\n" 777 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n" 778#endif 779#ifdef CONFIG_AUDIO_PA 780 "-audiodev pa,id=id[,prop[=value][,...]]\n" 781 " server= PulseAudio server address\n" 782 " in|out.name= source/sink device name\n" 783 " in|out.latency= desired latency in microseconds\n" 784#endif 785#ifdef CONFIG_AUDIO_SDL 786 "-audiodev sdl,id=id[,prop[=value][,...]]\n" 787 " in|out.buffer-count= number of buffers\n" 788#endif 789#ifdef CONFIG_AUDIO_SNDIO 790 "-audiodev sndio,id=id[,prop[=value][,...]]\n" 791#endif 792#ifdef CONFIG_SPICE 793 "-audiodev spice,id=id[,prop[=value][,...]]\n" 794#endif 795#ifdef CONFIG_DBUS_DISPLAY 796 "-audiodev dbus,id=id[,prop[=value][,...]]\n" 797#endif 798 "-audiodev wav,id=id[,prop[=value][,...]]\n" 799 " path= path of wav file to record\n", 800 QEMU_ARCH_ALL) 801SRST 802``-audiodev [driver=]driver,id=id[,prop[=value][,...]]`` 803 Adds a new audio backend driver identified by id. There are global 804 and driver specific properties. Some values can be set differently 805 for input and output, they're marked with ``in|out.``. You can set 806 the input's property with ``in.prop`` and the output's property with 807 ``out.prop``. For example: 808 809 :: 810 811 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000 812 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified 813 814 NOTE: parameter validation is known to be incomplete, in many cases 815 specifying an invalid option causes QEMU to print an error message 816 and continue emulation without sound. 817 818 Valid global options are: 819 820 ``id=identifier`` 821 Identifies the audio backend. 822 823 ``timer-period=period`` 824 Sets the timer period used by the audio subsystem in 825 microseconds. Default is 10000 (10 ms). 826 827 ``in|out.mixing-engine=on|off`` 828 Use QEMU's mixing engine to mix all streams inside QEMU and 829 convert audio formats when not supported by the backend. When 830 off, fixed-settings must be off too. Note that disabling this 831 option means that the selected backend must support multiple 832 streams and the audio formats used by the virtual cards, 833 otherwise you'll get no sound. It's not recommended to disable 834 this option unless you want to use 5.1 or 7.1 audio, as mixing 835 engine only supports mono and stereo audio. Default is on. 836 837 ``in|out.fixed-settings=on|off`` 838 Use fixed settings for host audio. When off, it will change 839 based on how the guest opens the sound card. In this case you 840 must not specify frequency, channels or format. Default is on. 841 842 ``in|out.frequency=frequency`` 843 Specify the frequency to use when using fixed-settings. Default 844 is 44100Hz. 845 846 ``in|out.channels=channels`` 847 Specify the number of channels to use when using fixed-settings. 848 Default is 2 (stereo). 849 850 ``in|out.format=format`` 851 Specify the sample format to use when using fixed-settings. 852 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``, 853 ``u32``, ``f32``. Default is ``s16``. 854 855 ``in|out.voices=voices`` 856 Specify the number of voices to use. Default is 1. 857 858 ``in|out.buffer-length=usecs`` 859 Sets the size of the buffer in microseconds. 860 861``-audiodev none,id=id[,prop[=value][,...]]`` 862 Creates a dummy backend that discards all outputs. This backend has 863 no backend specific properties. 864 865``-audiodev alsa,id=id[,prop[=value][,...]]`` 866 Creates backend using the ALSA. This backend is only available on 867 Linux. 868 869 ALSA specific options are: 870 871 ``in|out.dev=device`` 872 Specify the ALSA device to use for input and/or output. Default 873 is ``default``. 874 875 ``in|out.period-length=usecs`` 876 Sets the period length in microseconds. 877 878 ``in|out.try-poll=on|off`` 879 Attempt to use poll mode with the device. Default is on. 880 881 ``threshold=threshold`` 882 Threshold (in microseconds) when playback starts. Default is 0. 883 884``-audiodev coreaudio,id=id[,prop[=value][,...]]`` 885 Creates a backend using Apple's Core Audio. This backend is only 886 available on Mac OS and only supports playback. 887 888 Core Audio specific options are: 889 890 ``in|out.buffer-count=count`` 891 Sets the count of the buffers. 892 893``-audiodev dsound,id=id[,prop[=value][,...]]`` 894 Creates a backend using Microsoft's DirectSound. This backend is 895 only available on Windows and only supports playback. 896 897 DirectSound specific options are: 898 899 ``latency=usecs`` 900 Add extra usecs microseconds latency to playback. Default is 901 10000 (10 ms). 902 903``-audiodev oss,id=id[,prop[=value][,...]]`` 904 Creates a backend using OSS. This backend is available on most 905 Unix-like systems. 906 907 OSS specific options are: 908 909 ``in|out.dev=device`` 910 Specify the file name of the OSS device to use. Default is 911 ``/dev/dsp``. 912 913 ``in|out.buffer-count=count`` 914 Sets the count of the buffers. 915 916 ``in|out.try-poll=on|of`` 917 Attempt to use poll mode with the device. Default is on. 918 919 ``try-mmap=on|off`` 920 Try using memory mapped device access. Default is off. 921 922 ``exclusive=on|off`` 923 Open the device in exclusive mode (vmix won't work in this 924 case). Default is off. 925 926 ``dsp-policy=policy`` 927 Sets the timing policy (between 0 and 10, where smaller number 928 means smaller latency but higher CPU usage). Use -1 to use 929 buffer sizes specified by ``buffer`` and ``buffer-count``. This 930 option is ignored if you do not have OSS 4. Default is 5. 931 932``-audiodev pa,id=id[,prop[=value][,...]]`` 933 Creates a backend using PulseAudio. This backend is available on 934 most systems. 935 936 PulseAudio specific options are: 937 938 ``server=server`` 939 Sets the PulseAudio server to connect to. 940 941 ``in|out.name=sink`` 942 Use the specified source/sink for recording/playback. 943 944 ``in|out.latency=usecs`` 945 Desired latency in microseconds. The PulseAudio server will try 946 to honor this value but actual latencies may be lower or higher. 947 948``-audiodev sdl,id=id[,prop[=value][,...]]`` 949 Creates a backend using SDL. This backend is available on most 950 systems, but you should use your platform's native backend if 951 possible. 952 953 SDL specific options are: 954 955 ``in|out.buffer-count=count`` 956 Sets the count of the buffers. 957 958``-audiodev sndio,id=id[,prop[=value][,...]]`` 959 Creates a backend using SNDIO. This backend is available on 960 OpenBSD and most other Unix-like systems. 961 962 Sndio specific options are: 963 964 ``in|out.dev=device`` 965 Specify the sndio device to use for input and/or output. Default 966 is ``default``. 967 968 ``in|out.latency=usecs`` 969 Sets the desired period length in microseconds. 970 971``-audiodev spice,id=id[,prop[=value][,...]]`` 972 Creates a backend that sends audio through SPICE. This backend 973 requires ``-spice`` and automatically selected in that case, so 974 usually you can ignore this option. This backend has no backend 975 specific properties. 976 977``-audiodev wav,id=id[,prop[=value][,...]]`` 978 Creates a backend that writes audio to a WAV file. 979 980 Backend specific options are: 981 982 ``path=path`` 983 Write recorded audio into the specified file. Default is 984 ``qemu.wav``. 985ERST 986 987DEF("device", HAS_ARG, QEMU_OPTION_device, 988 "-device driver[,prop[=value][,...]]\n" 989 " add device (based on driver)\n" 990 " prop=value,... sets driver properties\n" 991 " use '-device help' to print all possible drivers\n" 992 " use '-device driver,help' to print all possible properties\n", 993 QEMU_ARCH_ALL) 994SRST 995``-device driver[,prop[=value][,...]]`` 996 Add device driver. prop=value sets driver properties. Valid 997 properties depend on the driver. To get help on possible drivers and 998 properties, use ``-device help`` and ``-device driver,help``. 999 1000 Some drivers are: 1001 1002``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]`` 1003 Add an IPMI BMC. This is a simulation of a hardware management 1004 interface processor that normally sits on a system. It provides a 1005 watchdog and the ability to reset and power control the system. You 1006 need to connect this to an IPMI interface to make it useful 1007 1008 The IPMI slave address to use for the BMC. The default is 0x20. This 1009 address is the BMC's address on the I2C network of management 1010 controllers. If you don't know what this means, it is safe to ignore 1011 it. 1012 1013 ``id=id`` 1014 The BMC id for interfaces to use this device. 1015 1016 ``slave_addr=val`` 1017 Define slave address to use for the BMC. The default is 0x20. 1018 1019 ``sdrfile=file`` 1020 file containing raw Sensor Data Records (SDR) data. The default 1021 is none. 1022 1023 ``fruareasize=val`` 1024 size of a Field Replaceable Unit (FRU) area. The default is 1025 1024. 1026 1027 ``frudatafile=file`` 1028 file containing raw Field Replaceable Unit (FRU) inventory data. 1029 The default is none. 1030 1031 ``guid=uuid`` 1032 value for the GUID for the BMC, in standard UUID format. If this 1033 is set, get "Get GUID" command to the BMC will return it. 1034 Otherwise "Get GUID" will return an error. 1035 1036``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]`` 1037 Add a connection to an external IPMI BMC simulator. Instead of 1038 locally emulating the BMC like the above item, instead connect to an 1039 external entity that provides the IPMI services. 1040 1041 A connection is made to an external BMC simulator. If you do this, 1042 it is strongly recommended that you use the "reconnect=" chardev 1043 option to reconnect to the simulator if the connection is lost. Note 1044 that if this is not used carefully, it can be a security issue, as 1045 the interface has the ability to send resets, NMIs, and power off 1046 the VM. It's best if QEMU makes a connection to an external 1047 simulator running on a secure port on localhost, so neither the 1048 simulator nor QEMU is exposed to any outside network. 1049 1050 See the "lanserv/README.vm" file in the OpenIPMI library for more 1051 details on the external interface. 1052 1053``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]`` 1054 Add a KCS IPMI interafce on the ISA bus. This also adds a 1055 corresponding ACPI and SMBIOS entries, if appropriate. 1056 1057 ``bmc=id`` 1058 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern 1059 above. 1060 1061 ``ioport=val`` 1062 Define the I/O address of the interface. The default is 0xca0 1063 for KCS. 1064 1065 ``irq=val`` 1066 Define the interrupt to use. The default is 5. To disable 1067 interrupts, set this to 0. 1068 1069``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]`` 1070 Like the KCS interface, but defines a BT interface. The default port 1071 is 0xe4 and the default interrupt is 5. 1072 1073``-device pci-ipmi-kcs,bmc=id`` 1074 Add a KCS IPMI interafce on the PCI bus. 1075 1076 ``bmc=id`` 1077 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above. 1078 1079``-device pci-ipmi-bt,bmc=id`` 1080 Like the KCS interface, but defines a BT interface on the PCI bus. 1081 1082``-device intel-iommu[,option=...]`` 1083 This is only supported by ``-machine q35``, which will enable Intel VT-d 1084 emulation within the guest. It supports below options: 1085 1086 ``intremap=on|off`` (default: auto) 1087 This enables interrupt remapping feature. It's required to enable 1088 complete x2apic. Currently it only supports kvm kernel-irqchip modes 1089 ``off`` or ``split``, while full kernel-irqchip is not yet supported. 1090 The default value is "auto", which will be decided by the mode of 1091 kernel-irqchip. 1092 1093 ``caching-mode=on|off`` (default: off) 1094 This enables caching mode for the VT-d emulated device. When 1095 caching-mode is enabled, each guest DMA buffer mapping will generate an 1096 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in 1097 a synchronous way. It is required for ``-device vfio-pci`` to work 1098 with the VT-d device, because host assigned devices requires to setup 1099 the DMA mapping on the host before guest DMA starts. 1100 1101 ``device-iotlb=on|off`` (default: off) 1102 This enables device-iotlb capability for the emulated VT-d device. So 1103 far virtio/vhost should be the only real user for this parameter, 1104 paired with ats=on configured for the device. 1105 1106 ``aw-bits=39|48`` (default: 39) 1107 This decides the address width of IOVA address space. The address 1108 space has 39 bits width for 3-level IOMMU page tables, and 48 bits for 1109 4-level IOMMU page tables. 1110 1111 Please also refer to the wiki page for general scenarios of VT-d 1112 emulation in QEMU: https://wiki.qemu.org/Features/VT-d. 1113 1114ERST 1115 1116DEF("name", HAS_ARG, QEMU_OPTION_name, 1117 "-name string1[,process=string2][,debug-threads=on|off]\n" 1118 " set the name of the guest\n" 1119 " string1 sets the window title and string2 the process name\n" 1120 " When debug-threads is enabled, individual threads are given a separate name\n" 1121 " NOTE: The thread names are for debugging and not a stable API.\n", 1122 QEMU_ARCH_ALL) 1123SRST 1124``-name name`` 1125 Sets the name of the guest. This name will be displayed in the SDL 1126 window caption. The name will also be used for the VNC server. Also 1127 optionally set the top visible process name in Linux. Naming of 1128 individual threads can also be enabled on Linux to aid debugging. 1129ERST 1130 1131DEF("uuid", HAS_ARG, QEMU_OPTION_uuid, 1132 "-uuid %08x-%04x-%04x-%04x-%012x\n" 1133 " specify machine UUID\n", QEMU_ARCH_ALL) 1134SRST 1135``-uuid uuid`` 1136 Set system UUID. 1137ERST 1138 1139DEFHEADING() 1140 1141DEFHEADING(Block device options:) 1142 1143SRST 1144The QEMU block device handling options have a long history and 1145have gone through several iterations as the feature set and complexity 1146of the block layer have grown. Many online guides to QEMU often 1147reference older and deprecated options, which can lead to confusion. 1148 1149The recommended modern way to describe disks is to use a combination of 1150``-device`` to specify the hardware device and ``-blockdev`` to 1151describe the backend. The device defines what the guest sees and the 1152backend describes how QEMU handles the data. 1153 1154ERST 1155 1156DEF("fda", HAS_ARG, QEMU_OPTION_fda, 1157 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL) 1158DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL) 1159SRST 1160``-fda file`` 1161 \ 1162``-fdb file`` 1163 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in 1164 the System Emulation Users Guide). 1165ERST 1166 1167DEF("hda", HAS_ARG, QEMU_OPTION_hda, 1168 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL) 1169DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL) 1170DEF("hdc", HAS_ARG, QEMU_OPTION_hdc, 1171 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL) 1172DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL) 1173SRST 1174``-hda file`` 1175 \ 1176``-hdb file`` 1177 \ 1178``-hdc file`` 1179 \ 1180``-hdd file`` 1181 Use file as hard disk 0, 1, 2 or 3 image (see the :ref:`disk images` 1182 chapter in the System Emulation Users Guide). 1183ERST 1184 1185DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom, 1186 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n", 1187 QEMU_ARCH_ALL) 1188SRST 1189``-cdrom file`` 1190 Use file as CD-ROM image (you cannot use ``-hdc`` and ``-cdrom`` at 1191 the same time). You can use the host CD-ROM by using ``/dev/cdrom`` 1192 as filename. 1193ERST 1194 1195DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev, 1196 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n" 1197 " [,cache.direct=on|off][,cache.no-flush=on|off]\n" 1198 " [,read-only=on|off][,auto-read-only=on|off]\n" 1199 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n" 1200 " [,driver specific parameters...]\n" 1201 " configure a block backend\n", QEMU_ARCH_ALL) 1202SRST 1203``-blockdev option[,option[,option[,...]]]`` 1204 Define a new block driver node. Some of the options apply to all 1205 block drivers, other options are only accepted for a specific block 1206 driver. See below for a list of generic options and options for the 1207 most common block drivers. 1208 1209 Options that expect a reference to another node (e.g. ``file``) can 1210 be given in two ways. Either you specify the node name of an already 1211 existing node (file=node-name), or you define a new node inline, 1212 adding options for the referenced node after a dot 1213 (file.filename=path,file.aio=native). 1214 1215 A block driver node created with ``-blockdev`` can be used for a 1216 guest device by specifying its node name for the ``drive`` property 1217 in a ``-device`` argument that defines a block device. 1218 1219 ``Valid options for any block driver node:`` 1220 ``driver`` 1221 Specifies the block driver to use for the given node. 1222 1223 ``node-name`` 1224 This defines the name of the block driver node by which it 1225 will be referenced later. The name must be unique, i.e. it 1226 must not match the name of a different block driver node, or 1227 (if you use ``-drive`` as well) the ID of a drive. 1228 1229 If no node name is specified, it is automatically generated. 1230 The generated node name is not intended to be predictable 1231 and changes between QEMU invocations. For the top level, an 1232 explicit node name must be specified. 1233 1234 ``read-only`` 1235 Open the node read-only. Guest write attempts will fail. 1236 1237 Note that some block drivers support only read-only access, 1238 either generally or in certain configurations. In this case, 1239 the default value ``read-only=off`` does not work and the 1240 option must be specified explicitly. 1241 1242 ``auto-read-only`` 1243 If ``auto-read-only=on`` is set, QEMU may fall back to 1244 read-only usage even when ``read-only=off`` is requested, or 1245 even switch between modes as needed, e.g. depending on 1246 whether the image file is writable or whether a writing user 1247 is attached to the node. 1248 1249 ``force-share`` 1250 Override the image locking system of QEMU by forcing the 1251 node to utilize weaker shared access for permissions where 1252 it would normally request exclusive access. When there is 1253 the potential for multiple instances to have the same file 1254 open (whether this invocation of QEMU is the first or the 1255 second instance), both instances must permit shared access 1256 for the second instance to succeed at opening the file. 1257 1258 Enabling ``force-share=on`` requires ``read-only=on``. 1259 1260 ``cache.direct`` 1261 The host page cache can be avoided with ``cache.direct=on``. 1262 This will attempt to do disk IO directly to the guest's 1263 memory. QEMU may still perform an internal copy of the data. 1264 1265 ``cache.no-flush`` 1266 In case you don't care about data integrity over host 1267 failures, you can use ``cache.no-flush=on``. This option 1268 tells QEMU that it never needs to write any data to the disk 1269 but can instead keep things in cache. If anything goes 1270 wrong, like your host losing power, the disk storage getting 1271 disconnected accidentally, etc. your image will most 1272 probably be rendered unusable. 1273 1274 ``discard=discard`` 1275 discard is one of "ignore" (or "off") or "unmap" (or "on") 1276 and controls whether ``discard`` (also known as ``trim`` or 1277 ``unmap``) requests are ignored or passed to the filesystem. 1278 Some machine types may not support discard requests. 1279 1280 ``detect-zeroes=detect-zeroes`` 1281 detect-zeroes is "off", "on" or "unmap" and enables the 1282 automatic conversion of plain zero writes by the OS to 1283 driver specific optimized zero write commands. You may even 1284 choose "unmap" if discard is set to "unmap" to allow a zero 1285 write to be converted to an ``unmap`` operation. 1286 1287 ``Driver-specific options for file`` 1288 This is the protocol-level block driver for accessing regular 1289 files. 1290 1291 ``filename`` 1292 The path to the image file in the local filesystem 1293 1294 ``aio`` 1295 Specifies the AIO backend (threads/native/io_uring, 1296 default: threads) 1297 1298 ``locking`` 1299 Specifies whether the image file is protected with Linux OFD 1300 / POSIX locks. The default is to use the Linux Open File 1301 Descriptor API if available, otherwise no lock is applied. 1302 (auto/on/off, default: auto) 1303 1304 Example: 1305 1306 :: 1307 1308 -blockdev driver=file,node-name=disk,filename=disk.img 1309 1310 ``Driver-specific options for raw`` 1311 This is the image format block driver for raw images. It is 1312 usually stacked on top of a protocol level block driver such as 1313 ``file``. 1314 1315 ``file`` 1316 Reference to or definition of the data source block driver 1317 node (e.g. a ``file`` driver node) 1318 1319 Example 1: 1320 1321 :: 1322 1323 -blockdev driver=file,node-name=disk_file,filename=disk.img 1324 -blockdev driver=raw,node-name=disk,file=disk_file 1325 1326 Example 2: 1327 1328 :: 1329 1330 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img 1331 1332 ``Driver-specific options for qcow2`` 1333 This is the image format block driver for qcow2 images. It is 1334 usually stacked on top of a protocol level block driver such as 1335 ``file``. 1336 1337 ``file`` 1338 Reference to or definition of the data source block driver 1339 node (e.g. a ``file`` driver node) 1340 1341 ``backing`` 1342 Reference to or definition of the backing file block device 1343 (default is taken from the image file). It is allowed to 1344 pass ``null`` here in order to disable the default backing 1345 file. 1346 1347 ``lazy-refcounts`` 1348 Whether to enable the lazy refcounts feature (on/off; 1349 default is taken from the image file) 1350 1351 ``cache-size`` 1352 The maximum total size of the L2 table and refcount block 1353 caches in bytes (default: the sum of l2-cache-size and 1354 refcount-cache-size) 1355 1356 ``l2-cache-size`` 1357 The maximum size of the L2 table cache in bytes (default: if 1358 cache-size is not specified - 32M on Linux platforms, and 8M 1359 on non-Linux platforms; otherwise, as large as possible 1360 within the cache-size, while permitting the requested or the 1361 minimal refcount cache size) 1362 1363 ``refcount-cache-size`` 1364 The maximum size of the refcount block cache in bytes 1365 (default: 4 times the cluster size; or if cache-size is 1366 specified, the part of it which is not used for the L2 1367 cache) 1368 1369 ``cache-clean-interval`` 1370 Clean unused entries in the L2 and refcount caches. The 1371 interval is in seconds. The default value is 600 on 1372 supporting platforms, and 0 on other platforms. Setting it 1373 to 0 disables this feature. 1374 1375 ``pass-discard-request`` 1376 Whether discard requests to the qcow2 device should be 1377 forwarded to the data source (on/off; default: on if 1378 discard=unmap is specified, off otherwise) 1379 1380 ``pass-discard-snapshot`` 1381 Whether discard requests for the data source should be 1382 issued when a snapshot operation (e.g. deleting a snapshot) 1383 frees clusters in the qcow2 file (on/off; default: on) 1384 1385 ``pass-discard-other`` 1386 Whether discard requests for the data source should be 1387 issued on other occasions where a cluster gets freed 1388 (on/off; default: off) 1389 1390 ``overlap-check`` 1391 Which overlap checks to perform for writes to the image 1392 (none/constant/cached/all; default: cached). For details or 1393 finer granularity control refer to the QAPI documentation of 1394 ``blockdev-add``. 1395 1396 Example 1: 1397 1398 :: 1399 1400 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2 1401 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216 1402 1403 Example 2: 1404 1405 :: 1406 1407 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2 1408 1409 ``Driver-specific options for other drivers`` 1410 Please refer to the QAPI documentation of the ``blockdev-add`` 1411 QMP command. 1412ERST 1413 1414DEF("drive", HAS_ARG, QEMU_OPTION_drive, 1415 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n" 1416 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n" 1417 " [,snapshot=on|off][,rerror=ignore|stop|report]\n" 1418 " [,werror=ignore|stop|report|enospc][,id=name]\n" 1419 " [,aio=threads|native|io_uring]\n" 1420 " [,readonly=on|off][,copy-on-read=on|off]\n" 1421 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n" 1422 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n" 1423 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n" 1424 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n" 1425 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n" 1426 " [[,iops_size=is]]\n" 1427 " [[,group=g]]\n" 1428 " use 'file' as a drive image\n", QEMU_ARCH_ALL) 1429SRST 1430``-drive option[,option[,option[,...]]]`` 1431 Define a new drive. This includes creating a block driver node (the 1432 backend) as well as a guest device, and is mostly a shortcut for 1433 defining the corresponding ``-blockdev`` and ``-device`` options. 1434 1435 ``-drive`` accepts all options that are accepted by ``-blockdev``. 1436 In addition, it knows the following options: 1437 1438 ``file=file`` 1439 This option defines which disk image (see the :ref:`disk images` 1440 chapter in the System Emulation Users Guide) to use with this drive. 1441 If the filename contains comma, you must double it (for instance, 1442 "file=my,,file" to use file "my,file"). 1443 1444 Special files such as iSCSI devices can be specified using 1445 protocol specific URLs. See the section for "Device URL Syntax" 1446 for more information. 1447 1448 ``if=interface`` 1449 This option defines on which type on interface the drive is 1450 connected. Available types are: ide, scsi, sd, mtd, floppy, 1451 pflash, virtio, none. 1452 1453 ``bus=bus,unit=unit`` 1454 These options define where is connected the drive by defining 1455 the bus number and the unit id. 1456 1457 ``index=index`` 1458 This option defines where the drive is connected by using an 1459 index in the list of available connectors of a given interface 1460 type. 1461 1462 ``media=media`` 1463 This option defines the type of the media: disk or cdrom. 1464 1465 ``snapshot=snapshot`` 1466 snapshot is "on" or "off" and controls snapshot mode for the 1467 given drive (see ``-snapshot``). 1468 1469 ``cache=cache`` 1470 cache is "none", "writeback", "unsafe", "directsync" or 1471 "writethrough" and controls how the host cache is used to access 1472 block data. This is a shortcut that sets the ``cache.direct`` 1473 and ``cache.no-flush`` options (as in ``-blockdev``), and 1474 additionally ``cache.writeback``, which provides a default for 1475 the ``write-cache`` option of block guest devices (as in 1476 ``-device``). The modes correspond to the following settings: 1477 1478 ============= =============== ============ ============== 1479 \ cache.writeback cache.direct cache.no-flush 1480 ============= =============== ============ ============== 1481 writeback on off off 1482 none on on off 1483 writethrough off off off 1484 directsync off on off 1485 unsafe on off on 1486 ============= =============== ============ ============== 1487 1488 The default mode is ``cache=writeback``. 1489 1490 ``aio=aio`` 1491 aio is "threads", "native", or "io_uring" and selects between pthread 1492 based disk I/O, native Linux AIO, or Linux io_uring API. 1493 1494 ``format=format`` 1495 Specify which disk format will be used rather than detecting the 1496 format. Can be used to specify format=raw to avoid interpreting 1497 an untrusted format header. 1498 1499 ``werror=action,rerror=action`` 1500 Specify which action to take on write and read errors. Valid 1501 actions are: "ignore" (ignore the error and try to continue), 1502 "stop" (pause QEMU), "report" (report the error to the guest), 1503 "enospc" (pause QEMU only if the host disk is full; report the 1504 error to the guest otherwise). The default setting is 1505 ``werror=enospc`` and ``rerror=report``. 1506 1507 ``copy-on-read=copy-on-read`` 1508 copy-on-read is "on" or "off" and enables whether to copy read 1509 backing file sectors into the image file. 1510 1511 ``bps=b,bps_rd=r,bps_wr=w`` 1512 Specify bandwidth throttling limits in bytes per second, either 1513 for all request types or for reads or writes only. Small values 1514 can lead to timeouts or hangs inside the guest. A safe minimum 1515 for disks is 2 MB/s. 1516 1517 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm`` 1518 Specify bursts in bytes per second, either for all request types 1519 or for reads or writes only. Bursts allow the guest I/O to spike 1520 above the limit temporarily. 1521 1522 ``iops=i,iops_rd=r,iops_wr=w`` 1523 Specify request rate limits in requests per second, either for 1524 all request types or for reads or writes only. 1525 1526 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm`` 1527 Specify bursts in requests per second, either for all request 1528 types or for reads or writes only. Bursts allow the guest I/O to 1529 spike above the limit temporarily. 1530 1531 ``iops_size=is`` 1532 Let every is bytes of a request count as a new request for iops 1533 throttling purposes. Use this option to prevent guests from 1534 circumventing iops limits by sending fewer but larger requests. 1535 1536 ``group=g`` 1537 Join a throttling quota group with given name g. All drives that 1538 are members of the same group are accounted for together. Use 1539 this option to prevent guests from circumventing throttling 1540 limits by using many small disks instead of a single larger 1541 disk. 1542 1543 By default, the ``cache.writeback=on`` mode is used. It will report 1544 data writes as completed as soon as the data is present in the host 1545 page cache. This is safe as long as your guest OS makes sure to 1546 correctly flush disk caches where needed. If your guest OS does not 1547 handle volatile disk write caches correctly and your host crashes or 1548 loses power, then the guest may experience data corruption. 1549 1550 For such guests, you should consider using ``cache.writeback=off``. 1551 This means that the host page cache will be used to read and write 1552 data, but write notification will be sent to the guest only after 1553 QEMU has made sure to flush each write to the disk. Be aware that 1554 this has a major impact on performance. 1555 1556 When using the ``-snapshot`` option, unsafe caching is always used. 1557 1558 Copy-on-read avoids accessing the same backing file sectors 1559 repeatedly and is useful when the backing file is over a slow 1560 network. By default copy-on-read is off. 1561 1562 Instead of ``-cdrom`` you can use: 1563 1564 .. parsed-literal:: 1565 1566 |qemu_system| -drive file=file,index=2,media=cdrom 1567 1568 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use: 1569 1570 .. parsed-literal:: 1571 1572 |qemu_system| -drive file=file,index=0,media=disk 1573 |qemu_system| -drive file=file,index=1,media=disk 1574 |qemu_system| -drive file=file,index=2,media=disk 1575 |qemu_system| -drive file=file,index=3,media=disk 1576 1577 You can open an image using pre-opened file descriptors from an fd 1578 set: 1579 1580 .. parsed-literal:: 1581 1582 |qemu_system| \\ 1583 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 1584 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 1585 -drive file=/dev/fdset/2,index=0,media=disk 1586 1587 You can connect a CDROM to the slave of ide0: 1588 1589 .. parsed-literal:: 1590 1591 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom 1592 1593 If you don't specify the "file=" argument, you define an empty 1594 drive: 1595 1596 .. parsed-literal:: 1597 1598 |qemu_system_x86| -drive if=ide,index=1,media=cdrom 1599 1600 Instead of ``-fda``, ``-fdb``, you can use: 1601 1602 .. parsed-literal:: 1603 1604 |qemu_system_x86| -drive file=file,index=0,if=floppy 1605 |qemu_system_x86| -drive file=file,index=1,if=floppy 1606 1607 By default, interface is "ide" and index is automatically 1608 incremented: 1609 1610 .. parsed-literal:: 1611 1612 |qemu_system_x86| -drive file=a -drive file=b" 1613 1614 is interpreted like: 1615 1616 .. parsed-literal:: 1617 1618 |qemu_system_x86| -hda a -hdb b 1619ERST 1620 1621DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock, 1622 "-mtdblock file use 'file' as on-board Flash memory image\n", 1623 QEMU_ARCH_ALL) 1624SRST 1625``-mtdblock file`` 1626 Use file as on-board Flash memory image. 1627ERST 1628 1629DEF("sd", HAS_ARG, QEMU_OPTION_sd, 1630 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL) 1631SRST 1632``-sd file`` 1633 Use file as SecureDigital card image. 1634ERST 1635 1636DEF("snapshot", 0, QEMU_OPTION_snapshot, 1637 "-snapshot write to temporary files instead of disk image files\n", 1638 QEMU_ARCH_ALL) 1639SRST 1640``-snapshot`` 1641 Write to temporary files instead of disk image files. In this case, 1642 the raw disk image you use is not written back. You can however 1643 force the write back by pressing C-a s (see the :ref:`disk images` 1644 chapter in the System Emulation Users Guide). 1645ERST 1646 1647DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev, 1648 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1649 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n" 1650 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n" 1651 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n" 1652 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n" 1653 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n" 1654 " [[,throttling.iops-size=is]]\n" 1655 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n" 1656 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n" 1657 "-fsdev synth,id=id\n", 1658 QEMU_ARCH_ALL) 1659 1660SRST 1661``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]`` 1662 \ 1663``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]`` 1664 \ 1665``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]`` 1666 \ 1667``-fsdev synth,id=id[,readonly=on]`` 1668 Define a new file system device. Valid options are: 1669 1670 ``local`` 1671 Accesses to the filesystem are done by QEMU. 1672 1673 ``proxy`` 1674 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1675 1676 ``synth`` 1677 Synthetic filesystem, only used by QTests. 1678 1679 ``id=id`` 1680 Specifies identifier for this device. 1681 1682 ``path=path`` 1683 Specifies the export path for the file system device. Files 1684 under this path will be available to the 9p client on the guest. 1685 1686 ``security_model=security_model`` 1687 Specifies the security model to be used for this export path. 1688 Supported security models are "passthrough", "mapped-xattr", 1689 "mapped-file" and "none". In "passthrough" security model, files 1690 are stored using the same credentials as they are created on the 1691 guest. This requires QEMU to run as root. In "mapped-xattr" 1692 security model, some of the file attributes like uid, gid, mode 1693 bits and link target are stored as file attributes. For 1694 "mapped-file" these attributes are stored in the hidden 1695 .virtfs\_metadata directory. Directories exported by this 1696 security model cannot interact with other unix tools. "none" 1697 security model is same as passthrough except the sever won't 1698 report failures if it fails to set file attributes like 1699 ownership. Security model is mandatory only for local fsdriver. 1700 Other fsdrivers (like proxy) don't take security model as a 1701 parameter. 1702 1703 ``writeout=writeout`` 1704 This is an optional argument. The only supported value is 1705 "immediate". This means that host page cache will be used to 1706 read and write data but write notification will be sent to the 1707 guest only when the data has been reported as written by the 1708 storage subsystem. 1709 1710 ``readonly=on`` 1711 Enables exporting 9p share as a readonly mount for guests. By 1712 default read-write access is given. 1713 1714 ``socket=socket`` 1715 Enables proxy filesystem driver to use passed socket file for 1716 communicating with virtfs-proxy-helper(1). 1717 1718 ``sock_fd=sock_fd`` 1719 Enables proxy filesystem driver to use passed socket descriptor 1720 for communicating with virtfs-proxy-helper(1). Usually a helper 1721 like libvirt will create socketpair and pass one of the fds as 1722 sock\_fd. 1723 1724 ``fmode=fmode`` 1725 Specifies the default mode for newly created files on the host. 1726 Works only with security models "mapped-xattr" and 1727 "mapped-file". 1728 1729 ``dmode=dmode`` 1730 Specifies the default mode for newly created directories on the 1731 host. Works only with security models "mapped-xattr" and 1732 "mapped-file". 1733 1734 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w`` 1735 Specify bandwidth throttling limits in bytes per second, either 1736 for all request types or for reads or writes only. 1737 1738 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm`` 1739 Specify bursts in bytes per second, either for all request types 1740 or for reads or writes only. Bursts allow the guest I/O to spike 1741 above the limit temporarily. 1742 1743 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w`` 1744 Specify request rate limits in requests per second, either for 1745 all request types or for reads or writes only. 1746 1747 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm`` 1748 Specify bursts in requests per second, either for all request 1749 types or for reads or writes only. Bursts allow the guest I/O to 1750 spike above the limit temporarily. 1751 1752 ``throttling.iops-size=is`` 1753 Let every is bytes of a request count as a new request for iops 1754 throttling purposes. 1755 1756 -fsdev option is used along with -device driver "virtio-9p-...". 1757 1758``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag`` 1759 Options for virtio-9p-... driver are: 1760 1761 ``type`` 1762 Specifies the variant to be used. Supported values are "pci", 1763 "ccw" or "device", depending on the machine type. 1764 1765 ``fsdev=id`` 1766 Specifies the id value specified along with -fsdev option. 1767 1768 ``mount_tag=mount_tag`` 1769 Specifies the tag name to be used by the guest to mount this 1770 export point. 1771ERST 1772 1773DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs, 1774 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1775 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n" 1776 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n" 1777 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n" 1778 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n", 1779 QEMU_ARCH_ALL) 1780 1781SRST 1782``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]`` 1783 \ 1784``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1785 \ 1786``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1787 \ 1788``-virtfs synth,mount_tag=mount_tag`` 1789 Define a new virtual filesystem device and expose it to the guest using 1790 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain 1791 directory on host is made directly accessible by guest as a pass-through 1792 file system by using the 9P network protocol for communication between 1793 host and guests, if desired even accessible, shared by several guests 1794 simultaniously. 1795 1796 Note that ``-virtfs`` is actually just a convenience shortcut for its 1797 generalized form ``-fsdev -device virtio-9p-pci``. 1798 1799 The general form of pass-through file system options are: 1800 1801 ``local`` 1802 Accesses to the filesystem are done by QEMU. 1803 1804 ``proxy`` 1805 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1806 1807 ``synth`` 1808 Synthetic filesystem, only used by QTests. 1809 1810 ``id=id`` 1811 Specifies identifier for the filesystem device 1812 1813 ``path=path`` 1814 Specifies the export path for the file system device. Files 1815 under this path will be available to the 9p client on the guest. 1816 1817 ``security_model=security_model`` 1818 Specifies the security model to be used for this export path. 1819 Supported security models are "passthrough", "mapped-xattr", 1820 "mapped-file" and "none". In "passthrough" security model, files 1821 are stored using the same credentials as they are created on the 1822 guest. This requires QEMU to run as root. In "mapped-xattr" 1823 security model, some of the file attributes like uid, gid, mode 1824 bits and link target are stored as file attributes. For 1825 "mapped-file" these attributes are stored in the hidden 1826 .virtfs\_metadata directory. Directories exported by this 1827 security model cannot interact with other unix tools. "none" 1828 security model is same as passthrough except the sever won't 1829 report failures if it fails to set file attributes like 1830 ownership. Security model is mandatory only for local fsdriver. 1831 Other fsdrivers (like proxy) don't take security model as a 1832 parameter. 1833 1834 ``writeout=writeout`` 1835 This is an optional argument. The only supported value is 1836 "immediate". This means that host page cache will be used to 1837 read and write data but write notification will be sent to the 1838 guest only when the data has been reported as written by the 1839 storage subsystem. 1840 1841 ``readonly=on`` 1842 Enables exporting 9p share as a readonly mount for guests. By 1843 default read-write access is given. 1844 1845 ``socket=socket`` 1846 Enables proxy filesystem driver to use passed socket file for 1847 communicating with virtfs-proxy-helper(1). Usually a helper like 1848 libvirt will create socketpair and pass one of the fds as 1849 sock\_fd. 1850 1851 ``sock_fd`` 1852 Enables proxy filesystem driver to use passed 'sock\_fd' as the 1853 socket descriptor for interfacing with virtfs-proxy-helper(1). 1854 1855 ``fmode=fmode`` 1856 Specifies the default mode for newly created files on the host. 1857 Works only with security models "mapped-xattr" and 1858 "mapped-file". 1859 1860 ``dmode=dmode`` 1861 Specifies the default mode for newly created directories on the 1862 host. Works only with security models "mapped-xattr" and 1863 "mapped-file". 1864 1865 ``mount_tag=mount_tag`` 1866 Specifies the tag name to be used by the guest to mount this 1867 export point. 1868 1869 ``multidevs=multidevs`` 1870 Specifies how to deal with multiple devices being shared with a 1871 9p export. Supported behaviours are either "remap", "forbid" or 1872 "warn". The latter is the default behaviour on which virtfs 9p 1873 expects only one device to be shared with the same export, and 1874 if more than one device is shared and accessed via the same 9p 1875 export then only a warning message is logged (once) by qemu on 1876 host side. In order to avoid file ID collisions on guest you 1877 should either create a separate virtfs export for each device to 1878 be shared with guests (recommended way) or you might use "remap" 1879 instead which allows you to share multiple devices with only one 1880 export instead, which is achieved by remapping the original 1881 inode numbers from host to guest in a way that would prevent 1882 such collisions. Remapping inodes in such use cases is required 1883 because the original device IDs from host are never passed and 1884 exposed on guest. Instead all files of an export shared with 1885 virtfs always share the same device id on guest. So two files 1886 with identical inode numbers but from actually different devices 1887 on host would otherwise cause a file ID collision and hence 1888 potential misbehaviours on guest. "forbid" on the other hand 1889 assumes like "warn" that only one device is shared by the same 1890 export, however it will not only log a warning message but also 1891 deny access to additional devices on guest. Note though that 1892 "forbid" does currently not block all possible file access 1893 operations (e.g. readdir() would still return entries from other 1894 devices). 1895ERST 1896 1897DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi, 1898 "-iscsi [user=user][,password=password]\n" 1899 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE\n" 1900 " [,initiator-name=initiator-iqn][,id=target-iqn]\n" 1901 " [,timeout=timeout]\n" 1902 " iSCSI session parameters\n", QEMU_ARCH_ALL) 1903 1904SRST 1905``-iscsi`` 1906 Configure iSCSI session parameters. 1907ERST 1908 1909DEFHEADING() 1910 1911DEFHEADING(USB convenience options:) 1912 1913DEF("usb", 0, QEMU_OPTION_usb, 1914 "-usb enable on-board USB host controller (if not enabled by default)\n", 1915 QEMU_ARCH_ALL) 1916SRST 1917``-usb`` 1918 Enable USB emulation on machine types with an on-board USB host 1919 controller (if not enabled by default). Note that on-board USB host 1920 controllers may not support USB 3.0. In this case 1921 ``-device qemu-xhci`` can be used instead on machines with PCI. 1922ERST 1923 1924DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice, 1925 "-usbdevice name add the host or guest USB device 'name'\n", 1926 QEMU_ARCH_ALL) 1927SRST 1928``-usbdevice devname`` 1929 Add the USB device devname, and enable an on-board USB controller 1930 if possible and necessary (just like it can be done via 1931 ``-machine usb=on``). Note that this option is mainly intended for 1932 the user's convenience only. More fine-grained control can be 1933 achieved by selecting a USB host controller (if necessary) and the 1934 desired USB device via the ``-device`` option instead. For example, 1935 instead of using ``-usbdevice mouse`` it is possible to use 1936 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse 1937 to a USB 3.0 controller instead (at least on machines that support 1938 PCI and do not have an USB controller enabled by default yet). 1939 For more details, see the chapter about 1940 :ref:`Connecting USB devices` in the System Emulation Users Guide. 1941 Possible devices for devname are: 1942 1943 ``braille`` 1944 Braille device. This will use BrlAPI to display the braille 1945 output on a real or fake device (i.e. it also creates a 1946 corresponding ``braille`` chardev automatically beside the 1947 ``usb-braille`` USB device). 1948 1949 ``keyboard`` 1950 Standard USB keyboard. Will override the PS/2 keyboard (if present). 1951 1952 ``mouse`` 1953 Virtual Mouse. This will override the PS/2 mouse emulation when 1954 activated. 1955 1956 ``tablet`` 1957 Pointer device that uses absolute coordinates (like a 1958 touchscreen). This means QEMU is able to report the mouse 1959 position without having to grab the mouse. Also overrides the 1960 PS/2 mouse emulation when activated. 1961 1962 ``wacom-tablet`` 1963 Wacom PenPartner USB tablet. 1964 1965 1966ERST 1967 1968DEFHEADING() 1969 1970DEFHEADING(Display options:) 1971 1972DEF("display", HAS_ARG, QEMU_OPTION_display, 1973#if defined(CONFIG_SPICE) 1974 "-display spice-app[,gl=on|off]\n" 1975#endif 1976#if defined(CONFIG_SDL) 1977 "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n" 1978 " [,window-close=on|off]\n" 1979#endif 1980#if defined(CONFIG_GTK) 1981 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n" 1982 " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n" 1983 " [,show-menubar=on|off]\n" 1984#endif 1985#if defined(CONFIG_VNC) 1986 "-display vnc=<display>[,<optargs>]\n" 1987#endif 1988#if defined(CONFIG_CURSES) 1989 "-display curses[,charset=<encoding>]\n" 1990#endif 1991#if defined(CONFIG_COCOA) 1992 "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n" 1993#endif 1994#if defined(CONFIG_OPENGL) 1995 "-display egl-headless[,rendernode=<file>]\n" 1996#endif 1997#if defined(CONFIG_DBUS_DISPLAY) 1998 "-display dbus[,addr=<dbusaddr>]\n" 1999 " [,gl=on|core|es|off][,rendernode=<file>]\n" 2000#endif 2001#if defined(CONFIG_COCOA) 2002 "-display cocoa[,show-cursor=on|off][,left-command-key=on|off]\n" 2003#endif 2004 "-display none\n" 2005 " select display backend type\n" 2006 " The default display is equivalent to\n " 2007#if defined(CONFIG_GTK) 2008 "\"-display gtk\"\n" 2009#elif defined(CONFIG_SDL) 2010 "\"-display sdl\"\n" 2011#elif defined(CONFIG_COCOA) 2012 "\"-display cocoa\"\n" 2013#elif defined(CONFIG_VNC) 2014 "\"-vnc localhost:0,to=99,id=default\"\n" 2015#else 2016 "\"-display none\"\n" 2017#endif 2018 , QEMU_ARCH_ALL) 2019SRST 2020``-display type`` 2021 Select type of display to use. Use ``-display help`` to list the available 2022 display types. Valid values for type are 2023 2024 ``spice-app[,gl=on|off]`` 2025 Start QEMU as a Spice server and launch the default Spice client 2026 application. The Spice server will redirect the serial consoles 2027 and QEMU monitors. (Since 4.0) 2028 2029 ``dbus`` 2030 Export the display over D-Bus interfaces. (Since 7.0) 2031 2032 The connection is registered with the "org.qemu" name (and queued when 2033 already owned). 2034 2035 ``addr=<dbusaddr>`` : D-Bus bus address to connect to. 2036 2037 ``p2p=yes|no`` : Use peer-to-peer connection, accepted via QMP ``add_client``. 2038 2039 ``gl=on|off|core|es`` : Use OpenGL for rendering (the D-Bus interface 2040 will share framebuffers with DMABUF file descriptors). 2041 2042 ``sdl`` 2043 Display video output via SDL (usually in a separate graphics 2044 window; see the SDL documentation for other possibilities). 2045 Valid parameters are: 2046 2047 ``grab-mod=<mods>`` : Used to select the modifier keys for toggling 2048 the mouse grabbing in conjunction with the "g" key. ``<mods>`` can be 2049 either ``lshift-lctrl-lalt`` or ``rctrl``. 2050 2051 ``gl=on|off|core|es`` : Use OpenGL for displaying 2052 2053 ``show-cursor=on|off`` : Force showing the mouse cursor 2054 2055 ``window-close=on|off`` : Allow to quit qemu with window close button 2056 2057 ``gtk`` 2058 Display video output in a GTK window. This interface provides 2059 drop-down menus and other UI elements to configure and control 2060 the VM during runtime. Valid parameters are: 2061 2062 ``full-screen=on|off`` : Start in fullscreen mode 2063 2064 ``gl=on|off`` : Use OpenGL for displaying 2065 2066 ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover 2067 2068 ``show-tabs=on|off`` : Display the tab bar for switching between the 2069 various graphical interfaces (e.g. VGA and 2070 virtual console character devices) by default. 2071 2072 ``show-cursor=on|off`` : Force showing the mouse cursor 2073 2074 ``window-close=on|off`` : Allow to quit qemu with window close button 2075 2076 ``show-menubar=on|off`` : Display the main window menubar, defaults to "on" 2077 2078 ``curses[,charset=<encoding>]`` 2079 Display video output via curses. For graphics device models 2080 which support a text mode, QEMU can display this output using a 2081 curses/ncurses interface. Nothing is displayed when the graphics 2082 device is in graphical mode or if the graphics device does not 2083 support a text mode. Generally only the VGA device models 2084 support text mode. The font charset used by the guest can be 2085 specified with the ``charset`` option, for example 2086 ``charset=CP850`` for IBM CP850 encoding. The default is 2087 ``CP437``. 2088 2089 ``cocoa`` 2090 Display video output in a Cocoa window. Mac only. This interface 2091 provides drop-down menus and other UI elements to configure and 2092 control the VM during runtime. Valid parameters are: 2093 2094 ``show-cursor=on|off`` : Force showing the mouse cursor 2095 2096 ``left-command-key=on|off`` : Disable forwarding left command key to host 2097 2098 ``egl-headless[,rendernode=<file>]`` 2099 Offload all OpenGL operations to a local DRI device. For any 2100 graphical display, this display needs to be paired with either 2101 VNC or SPICE displays. 2102 2103 ``vnc=<display>`` 2104 Start a VNC server on display <display> 2105 2106 ``none`` 2107 Do not display video output. The guest will still see an 2108 emulated graphics card, but its output will not be displayed to 2109 the QEMU user. This option differs from the -nographic option in 2110 that it only affects what is done with video output; -nographic 2111 also changes the destination of the serial and parallel port 2112 data. 2113ERST 2114 2115DEF("nographic", 0, QEMU_OPTION_nographic, 2116 "-nographic disable graphical output and redirect serial I/Os to console\n", 2117 QEMU_ARCH_ALL) 2118SRST 2119``-nographic`` 2120 Normally, if QEMU is compiled with graphical window support, it 2121 displays output such as guest graphics, guest console, and the QEMU 2122 monitor in a window. With this option, you can totally disable 2123 graphical output so that QEMU is a simple command line application. 2124 The emulated serial port is redirected on the console and muxed with 2125 the monitor (unless redirected elsewhere explicitly). Therefore, you 2126 can still use QEMU to debug a Linux kernel with a serial console. 2127 Use C-a h for help on switching between the console and monitor. 2128ERST 2129 2130#ifdef CONFIG_SPICE 2131DEF("spice", HAS_ARG, QEMU_OPTION_spice, 2132 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n" 2133 " [,x509-key-file=<file>][,x509-key-password=<file>]\n" 2134 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n" 2135 " [,x509-dh-key-file=<file>][,addr=addr]\n" 2136 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n" 2137 " [,tls-ciphers=<list>]\n" 2138 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n" 2139 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n" 2140 " [,sasl=on|off][,disable-ticketing=on|off]\n" 2141 " [,password=<string>][,password-secret=<secret-id>]\n" 2142 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n" 2143 " [,jpeg-wan-compression=[auto|never|always]]\n" 2144 " [,zlib-glz-wan-compression=[auto|never|always]]\n" 2145 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n" 2146 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n" 2147 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n" 2148 " [,gl=[on|off]][,rendernode=<file>]\n" 2149 " enable spice\n" 2150 " at least one of {port, tls-port} is mandatory\n", 2151 QEMU_ARCH_ALL) 2152#endif 2153SRST 2154``-spice option[,option[,...]]`` 2155 Enable the spice remote desktop protocol. Valid options are 2156 2157 ``port=<nr>`` 2158 Set the TCP port spice is listening on for plaintext channels. 2159 2160 ``addr=<addr>`` 2161 Set the IP address spice is listening on. Default is any 2162 address. 2163 2164 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off`` 2165 Force using the specified IP version. 2166 2167 ``password=<string>`` 2168 Set the password you need to authenticate. 2169 2170 This option is deprecated and insecure because it leaves the 2171 password visible in the process listing. Use ``password-secret`` 2172 instead. 2173 2174 ``password-secret=<secret-id>`` 2175 Set the ID of the ``secret`` object containing the password 2176 you need to authenticate. 2177 2178 ``sasl=on|off`` 2179 Require that the client use SASL to authenticate with the spice. 2180 The exact choice of authentication method used is controlled 2181 from the system / user's SASL configuration file for the 'qemu' 2182 service. This is typically found in /etc/sasl2/qemu.conf. If 2183 running QEMU as an unprivileged user, an environment variable 2184 SASL\_CONF\_PATH can be used to make it search alternate 2185 locations for the service config. While some SASL auth methods 2186 can also provide data encryption (eg GSSAPI), it is recommended 2187 that SASL always be combined with the 'tls' and 'x509' settings 2188 to enable use of SSL and server certificates. This ensures a 2189 data encryption preventing compromise of authentication 2190 credentials. 2191 2192 ``disable-ticketing=on|off`` 2193 Allow client connects without authentication. 2194 2195 ``disable-copy-paste=on|off`` 2196 Disable copy paste between the client and the guest. 2197 2198 ``disable-agent-file-xfer=on|off`` 2199 Disable spice-vdagent based file-xfer between the client and the 2200 guest. 2201 2202 ``tls-port=<nr>`` 2203 Set the TCP port spice is listening on for encrypted channels. 2204 2205 ``x509-dir=<dir>`` 2206 Set the x509 file directory. Expects same filenames as -vnc 2207 $display,x509=$dir 2208 2209 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>`` 2210 The x509 file names can also be configured individually. 2211 2212 ``tls-ciphers=<list>`` 2213 Specify which ciphers to use. 2214 2215 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]`` 2216 Force specific channel to be used with or without TLS 2217 encryption. The options can be specified multiple times to 2218 configure multiple channels. The special name "default" can be 2219 used to set the default mode. For channels which are not 2220 explicitly forced into one mode the spice client is allowed to 2221 pick tls/plaintext as he pleases. 2222 2223 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]`` 2224 Configure image compression (lossless). Default is auto\_glz. 2225 2226 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]`` 2227 Configure wan image compression (lossy for slow links). Default 2228 is auto. 2229 2230 ``streaming-video=[off|all|filter]`` 2231 Configure video stream detection. Default is off. 2232 2233 ``agent-mouse=[on|off]`` 2234 Enable/disable passing mouse events via vdagent. Default is on. 2235 2236 ``playback-compression=[on|off]`` 2237 Enable/disable audio stream compression (using celt 0.5.1). 2238 Default is on. 2239 2240 ``seamless-migration=[on|off]`` 2241 Enable/disable spice seamless migration. Default is off. 2242 2243 ``gl=[on|off]`` 2244 Enable/disable OpenGL context. Default is off. 2245 2246 ``rendernode=<file>`` 2247 DRM render node for OpenGL rendering. If not specified, it will 2248 pick the first available. (Since 2.9) 2249ERST 2250 2251DEF("portrait", 0, QEMU_OPTION_portrait, 2252 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n", 2253 QEMU_ARCH_ALL) 2254SRST 2255``-portrait`` 2256 Rotate graphical output 90 deg left (only PXA LCD). 2257ERST 2258 2259DEF("rotate", HAS_ARG, QEMU_OPTION_rotate, 2260 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n", 2261 QEMU_ARCH_ALL) 2262SRST 2263``-rotate deg`` 2264 Rotate graphical output some deg left (only PXA LCD). 2265ERST 2266 2267DEF("vga", HAS_ARG, QEMU_OPTION_vga, 2268 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n" 2269 " select video card type\n", QEMU_ARCH_ALL) 2270SRST 2271``-vga type`` 2272 Select type of VGA card to emulate. Valid values for type are 2273 2274 ``cirrus`` 2275 Cirrus Logic GD5446 Video card. All Windows versions starting 2276 from Windows 95 should recognize and use this graphic card. For 2277 optimal performances, use 16 bit color depth in the guest and 2278 the host OS. (This card was the default before QEMU 2.2) 2279 2280 ``std`` 2281 Standard VGA card with Bochs VBE extensions. If your guest OS 2282 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if 2283 you want to use high resolution modes (>= 1280x1024x16) then you 2284 should use this option. (This card is the default since QEMU 2285 2.2) 2286 2287 ``vmware`` 2288 VMWare SVGA-II compatible adapter. Use it if you have 2289 sufficiently recent XFree86/XOrg server or Windows guest with a 2290 driver for this card. 2291 2292 ``qxl`` 2293 QXL paravirtual graphic card. It is VGA compatible (including 2294 VESA 2.0 VBE support). Works best with qxl guest drivers 2295 installed though. Recommended choice when using the spice 2296 protocol. 2297 2298 ``tcx`` 2299 (sun4m only) Sun TCX framebuffer. This is the default 2300 framebuffer for sun4m machines and offers both 8-bit and 24-bit 2301 colour depths at a fixed resolution of 1024x768. 2302 2303 ``cg3`` 2304 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit 2305 framebuffer for sun4m machines available in both 1024x768 2306 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people 2307 wishing to run older Solaris versions. 2308 2309 ``virtio`` 2310 Virtio VGA card. 2311 2312 ``none`` 2313 Disable VGA card. 2314ERST 2315 2316DEF("full-screen", 0, QEMU_OPTION_full_screen, 2317 "-full-screen start in full screen\n", QEMU_ARCH_ALL) 2318SRST 2319``-full-screen`` 2320 Start in full screen. 2321ERST 2322 2323DEF("g", HAS_ARG, QEMU_OPTION_g , 2324 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n", 2325 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K) 2326SRST 2327``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]`` 2328 Set the initial graphical resolution and depth (PPC, SPARC only). 2329 2330 For PPC the default is 800x600x32. 2331 2332 For SPARC with the TCX graphics device, the default is 1024x768x8 2333 with the option of 1024x768x24. For cgthree, the default is 2334 1024x768x8 with the option of 1152x900x8 for people who wish to use 2335 OBP. 2336ERST 2337 2338DEF("vnc", HAS_ARG, QEMU_OPTION_vnc , 2339 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL) 2340SRST 2341``-vnc display[,option[,option[,...]]]`` 2342 Normally, if QEMU is compiled with graphical window support, it 2343 displays output such as guest graphics, guest console, and the QEMU 2344 monitor in a window. With this option, you can have QEMU listen on 2345 VNC display display and redirect the VGA display over the VNC 2346 session. It is very useful to enable the usb tablet device when 2347 using this option (option ``-device usb-tablet``). When using the 2348 VNC display, you must use the ``-k`` parameter to set the keyboard 2349 layout if you are not using en-us. Valid syntax for the display is 2350 2351 ``to=L`` 2352 With this option, QEMU will try next available VNC displays, 2353 until the number L, if the origianlly defined "-vnc display" is 2354 not available, e.g. port 5900+display is already used by another 2355 application. By default, to=0. 2356 2357 ``host:d`` 2358 TCP connections will only be allowed from host on display d. By 2359 convention the TCP port is 5900+d. Optionally, host can be 2360 omitted in which case the server will accept connections from 2361 any host. 2362 2363 ``unix:path`` 2364 Connections will be allowed over UNIX domain sockets where path 2365 is the location of a unix socket to listen for connections on. 2366 2367 ``none`` 2368 VNC is initialized but not started. The monitor ``change`` 2369 command can be used to later start the VNC server. 2370 2371 Following the display value there may be one or more option flags 2372 separated by commas. Valid options are 2373 2374 ``reverse=on|off`` 2375 Connect to a listening VNC client via a "reverse" connection. 2376 The client is specified by the display. For reverse network 2377 connections (host:d,``reverse``), the d argument is a TCP port 2378 number, not a display number. 2379 2380 ``websocket=on|off`` 2381 Opens an additional TCP listening port dedicated to VNC 2382 Websocket connections. If a bare websocket option is given, the 2383 Websocket port is 5700+display. An alternative port can be 2384 specified with the syntax ``websocket``\ =port. 2385 2386 If host is specified connections will only be allowed from this 2387 host. It is possible to control the websocket listen address 2388 independently, using the syntax ``websocket``\ =host:port. 2389 2390 If no TLS credentials are provided, the websocket connection 2391 runs in unencrypted mode. If TLS credentials are provided, the 2392 websocket connection requires encrypted client connections. 2393 2394 ``password=on|off`` 2395 Require that password based authentication is used for client 2396 connections. 2397 2398 The password must be set separately using the ``set_password`` 2399 command in the :ref:`QEMU monitor`. The 2400 syntax to change your password is: 2401 ``set_password <protocol> <password>`` where <protocol> could be 2402 either "vnc" or "spice". 2403 2404 If you would like to change <protocol> password expiration, you 2405 should use ``expire_password <protocol> <expiration-time>`` 2406 where expiration time could be one of the following options: 2407 now, never, +seconds or UNIX time of expiration, e.g. +60 to 2408 make password expire in 60 seconds, or 1335196800 to make 2409 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for 2410 this date and time). 2411 2412 You can also use keywords "now" or "never" for the expiration 2413 time to allow <protocol> password to expire immediately or never 2414 expire. 2415 2416 ``password-secret=<secret-id>`` 2417 Require that password based authentication is used for client 2418 connections, using the password provided by the ``secret`` 2419 object identified by ``secret-id``. 2420 2421 ``tls-creds=ID`` 2422 Provides the ID of a set of TLS credentials to use to secure the 2423 VNC server. They will apply to both the normal VNC server socket 2424 and the websocket socket (if enabled). Setting TLS credentials 2425 will cause the VNC server socket to enable the VeNCrypt auth 2426 mechanism. The credentials should have been previously created 2427 using the ``-object tls-creds`` argument. 2428 2429 ``tls-authz=ID`` 2430 Provides the ID of the QAuthZ authorization object against which 2431 the client's x509 distinguished name will validated. This object 2432 is only resolved at time of use, so can be deleted and recreated 2433 on the fly while the VNC server is active. If missing, it will 2434 default to denying access. 2435 2436 ``sasl=on|off`` 2437 Require that the client use SASL to authenticate with the VNC 2438 server. The exact choice of authentication method used is 2439 controlled from the system / user's SASL configuration file for 2440 the 'qemu' service. This is typically found in 2441 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user, 2442 an environment variable SASL\_CONF\_PATH can be used to make it 2443 search alternate locations for the service config. While some 2444 SASL auth methods can also provide data encryption (eg GSSAPI), 2445 it is recommended that SASL always be combined with the 'tls' 2446 and 'x509' settings to enable use of SSL and server 2447 certificates. This ensures a data encryption preventing 2448 compromise of authentication credentials. See the 2449 :ref:`VNC security` section in the System Emulation Users Guide 2450 for details on using SASL authentication. 2451 2452 ``sasl-authz=ID`` 2453 Provides the ID of the QAuthZ authorization object against which 2454 the client's SASL username will validated. This object is only 2455 resolved at time of use, so can be deleted and recreated on the 2456 fly while the VNC server is active. If missing, it will default 2457 to denying access. 2458 2459 ``acl=on|off`` 2460 Legacy method for enabling authorization of clients against the 2461 x509 distinguished name and SASL username. It results in the 2462 creation of two ``authz-list`` objects with IDs of 2463 ``vnc.username`` and ``vnc.x509dname``. The rules for these 2464 objects must be configured with the HMP ACL commands. 2465 2466 This option is deprecated and should no longer be used. The new 2467 ``sasl-authz`` and ``tls-authz`` options are a replacement. 2468 2469 ``lossy=on|off`` 2470 Enable lossy compression methods (gradient, JPEG, ...). If this 2471 option is set, VNC client may receive lossy framebuffer updates 2472 depending on its encoding settings. Enabling this option can 2473 save a lot of bandwidth at the expense of quality. 2474 2475 ``non-adaptive=on|off`` 2476 Disable adaptive encodings. Adaptive encodings are enabled by 2477 default. An adaptive encoding will try to detect frequently 2478 updated screen regions, and send updates in these regions using 2479 a lossy encoding (like JPEG). This can be really helpful to save 2480 bandwidth when playing videos. Disabling adaptive encodings 2481 restores the original static behavior of encodings like Tight. 2482 2483 ``share=[allow-exclusive|force-shared|ignore]`` 2484 Set display sharing policy. 'allow-exclusive' allows clients to 2485 ask for exclusive access. As suggested by the rfb spec this is 2486 implemented by dropping other connections. Connecting multiple 2487 clients in parallel requires all clients asking for a shared 2488 session (vncviewer: -shared switch). This is the default. 2489 'force-shared' disables exclusive client access. Useful for 2490 shared desktop sessions, where you don't want someone forgetting 2491 specify -shared disconnect everybody else. 'ignore' completely 2492 ignores the shared flag and allows everybody connect 2493 unconditionally. Doesn't conform to the rfb spec but is 2494 traditional QEMU behavior. 2495 2496 ``key-delay-ms`` 2497 Set keyboard delay, for key down and key up events, in 2498 milliseconds. Default is 10. Keyboards are low-bandwidth 2499 devices, so this slowdown can help the device and guest to keep 2500 up and not lose events in case events are arriving in bulk. 2501 Possible causes for the latter are flaky network connections, or 2502 scripts for automated testing. 2503 2504 ``audiodev=audiodev`` 2505 Use the specified audiodev when the VNC client requests audio 2506 transmission. When not using an -audiodev argument, this option 2507 must be omitted, otherwise is must be present and specify a 2508 valid audiodev. 2509 2510 ``power-control=on|off`` 2511 Permit the remote client to issue shutdown, reboot or reset power 2512 control requests. 2513ERST 2514 2515ARCHHEADING(, QEMU_ARCH_I386) 2516 2517ARCHHEADING(i386 target only:, QEMU_ARCH_I386) 2518 2519DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack, 2520 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n", 2521 QEMU_ARCH_I386) 2522SRST 2523``-win2k-hack`` 2524 Use it when installing Windows 2000 to avoid a disk full bug. After 2525 Windows 2000 is installed, you no longer need this option (this 2526 option slows down the IDE transfers). 2527ERST 2528 2529DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk, 2530 "-no-fd-bootchk disable boot signature checking for floppy disks\n", 2531 QEMU_ARCH_I386) 2532SRST 2533``-no-fd-bootchk`` 2534 Disable boot signature checking for floppy disks in BIOS. May be 2535 needed to boot from old floppy disks. 2536ERST 2537 2538DEF("no-acpi", 0, QEMU_OPTION_no_acpi, 2539 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2540SRST 2541``-no-acpi`` 2542 Disable ACPI (Advanced Configuration and Power Interface) support. 2543 Use it if your guest OS complains about ACPI problems (PC target 2544 machine only). 2545ERST 2546 2547DEF("no-hpet", 0, QEMU_OPTION_no_hpet, 2548 "-no-hpet disable HPET\n", QEMU_ARCH_I386) 2549SRST 2550``-no-hpet`` 2551 Disable HPET support. 2552ERST 2553 2554DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable, 2555 "-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" 2556 " ACPI table description\n", QEMU_ARCH_I386) 2557SRST 2558``-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]...]`` 2559 Add ACPI table with specified header fields and context from 2560 specified files. For file=, take whole ACPI table from the specified 2561 files, including all ACPI headers (possible overridden by other 2562 options). For data=, only data portion of the table is used, all 2563 header information is specified in the command line. If a SLIC table 2564 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id 2565 fields will override the same in the RSDT and the FADT (a.k.a. 2566 FACP), in order to ensure the field matches required by the 2567 Microsoft SLIC spec and the ACPI spec. 2568ERST 2569 2570DEF("smbios", HAS_ARG, QEMU_OPTION_smbios, 2571 "-smbios file=binary\n" 2572 " load SMBIOS entry from binary file\n" 2573 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n" 2574 " [,uefi=on|off]\n" 2575 " specify SMBIOS type 0 fields\n" 2576 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2577 " [,uuid=uuid][,sku=str][,family=str]\n" 2578 " specify SMBIOS type 1 fields\n" 2579 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2580 " [,asset=str][,location=str]\n" 2581 " specify SMBIOS type 2 fields\n" 2582 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n" 2583 " [,sku=str]\n" 2584 " specify SMBIOS type 3 fields\n" 2585 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n" 2586 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n" 2587 " [,processor-id=%d]\n" 2588 " specify SMBIOS type 4 fields\n" 2589 "-smbios type=8[,external_reference=str][,internal_reference=str][,connector_type=%d][,port_type=%d]\n" 2590 " specify SMBIOS type 8 fields\n" 2591 "-smbios type=11[,value=str][,path=filename]\n" 2592 " specify SMBIOS type 11 fields\n" 2593 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n" 2594 " [,asset=str][,part=str][,speed=%d]\n" 2595 " specify SMBIOS type 17 fields\n" 2596 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n" 2597 " specify SMBIOS type 41 fields\n", 2598 QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2599SRST 2600``-smbios file=binary`` 2601 Load SMBIOS entry from binary file. 2602 2603``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]`` 2604 Specify SMBIOS type 0 fields 2605 2606``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]`` 2607 Specify SMBIOS type 1 fields 2608 2609``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]`` 2610 Specify SMBIOS type 2 fields 2611 2612``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]`` 2613 Specify SMBIOS type 3 fields 2614 2615``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-id=%d]`` 2616 Specify SMBIOS type 4 fields 2617 2618``-smbios type=11[,value=str][,path=filename]`` 2619 Specify SMBIOS type 11 fields 2620 2621 This argument can be repeated multiple times, and values are added in the order they are parsed. 2622 Applications intending to use OEM strings data are encouraged to use their application name as 2623 a prefix for the value string. This facilitates passing information for multiple applications 2624 concurrently. 2625 2626 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax 2627 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes. 2628 2629 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to 2630 the SMBIOS table in the order in which they appear. 2631 2632 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535 2633 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the 2634 guest. Instead it should be used as a indicator to inform the guest where to locate the real 2635 data set, for example, by specifying the serial ID of a block device. 2636 2637 An example passing three strings is 2638 2639 .. parsed-literal:: 2640 2641 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\ 2642 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\ 2643 path=/some/file/with/oemstringsdata.txt 2644 2645 In the guest OS this is visible with the ``dmidecode`` command 2646 2647 .. parsed-literal:: 2648 2649 $ dmidecode -t 11 2650 Handle 0x0E00, DMI type 11, 5 bytes 2651 OEM Strings 2652 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/ 2653 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os 2654 String 3: myapp:some extra data 2655 2656 2657``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]`` 2658 Specify SMBIOS type 17 fields 2659 2660``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]`` 2661 Specify SMBIOS type 41 fields 2662 2663 This argument can be repeated multiple times. Its main use is to allow network interfaces be created 2664 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface 2665 position on the PCI bus. 2666 2667 Here is an example of use: 2668 2669 .. parsed-literal:: 2670 2671 -netdev user,id=internet \\ 2672 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\ 2673 -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev 2674 2675 In the guest OS, the device should then appear as ``eno1``: 2676 2677 ..parsed-literal:: 2678 2679 $ ip -brief l 2680 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP> 2681 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP> 2682 2683 Currently, the PCI device has to be attached to the root bus. 2684 2685ERST 2686 2687DEFHEADING() 2688 2689DEFHEADING(Network options:) 2690 2691DEF("netdev", HAS_ARG, QEMU_OPTION_netdev, 2692#ifdef CONFIG_SLIRP 2693 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n" 2694 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n" 2695 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n" 2696 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n" 2697 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]" 2698#ifndef _WIN32 2699 "[,smb=dir[,smbserver=addr]]\n" 2700#endif 2701 " configure a user mode network backend with ID 'str',\n" 2702 " its DHCP server and optional services\n" 2703#endif 2704#ifdef _WIN32 2705 "-netdev tap,id=str,ifname=name\n" 2706 " configure a host TAP network backend with ID 'str'\n" 2707#else 2708 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n" 2709 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n" 2710 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n" 2711 " [,poll-us=n]\n" 2712 " configure a host TAP network backend with ID 'str'\n" 2713 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2714 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n" 2715 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n" 2716 " to deconfigure it\n" 2717 " use '[down]script=no' to disable script execution\n" 2718 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n" 2719 " configure it\n" 2720 " use 'fd=h' to connect to an already opened TAP interface\n" 2721 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n" 2722 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n" 2723 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n" 2724 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n" 2725 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n" 2726 " use vhost=on to enable experimental in kernel accelerator\n" 2727 " (only has effect for virtio guests which use MSIX)\n" 2728 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n" 2729 " use 'vhostfd=h' to connect to an already opened vhost net device\n" 2730 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n" 2731 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n" 2732 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n" 2733 " spent on busy polling for vhost net\n" 2734 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n" 2735 " configure a host TAP network backend with ID 'str' that is\n" 2736 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2737 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n" 2738#endif 2739#ifdef __linux__ 2740 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n" 2741 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n" 2742 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n" 2743 " [,rxcookie=rxcookie][,offset=offset]\n" 2744 " configure a network backend with ID 'str' connected to\n" 2745 " an Ethernet over L2TPv3 pseudowire.\n" 2746 " Linux kernel 3.3+ as well as most routers can talk\n" 2747 " L2TPv3. This transport allows connecting a VM to a VM,\n" 2748 " VM to a router and even VM to Host. It is a nearly-universal\n" 2749 " standard (RFC3931). Note - this implementation uses static\n" 2750 " pre-configured tunnels (same as the Linux kernel).\n" 2751 " use 'src=' to specify source address\n" 2752 " use 'dst=' to specify destination address\n" 2753 " use 'udp=on' to specify udp encapsulation\n" 2754 " use 'srcport=' to specify source udp port\n" 2755 " use 'dstport=' to specify destination udp port\n" 2756 " use 'ipv6=on' to force v6\n" 2757 " L2TPv3 uses cookies to prevent misconfiguration as\n" 2758 " well as a weak security measure\n" 2759 " use 'rxcookie=0x012345678' to specify a rxcookie\n" 2760 " use 'txcookie=0x012345678' to specify a txcookie\n" 2761 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n" 2762 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n" 2763 " use 'pincounter=on' to work around broken counter handling in peer\n" 2764 " use 'offset=X' to add an extra offset between header and data\n" 2765#endif 2766 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n" 2767 " configure a network backend to connect to another network\n" 2768 " using a socket connection\n" 2769 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n" 2770 " configure a network backend to connect to a multicast maddr and port\n" 2771 " use 'localaddr=addr' to specify the host address to send packets from\n" 2772 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n" 2773 " configure a network backend to connect to another network\n" 2774 " using an UDP tunnel\n" 2775 "-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]\n" 2776 "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off]\n" 2777 "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor\n" 2778 " configure a network backend to connect to another network\n" 2779 " using a socket connection in stream mode.\n" 2780 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n" 2781 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]\n" 2782 " configure a network backend to connect to a multicast maddr and port\n" 2783 " use ``local.host=addr`` to specify the host address to send packets from\n" 2784 "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n" 2785 "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n" 2786 "-netdev dgram,id=str,local.type=fd,local.str=file-descriptor\n" 2787 " configure a network backend to connect to another network\n" 2788 " using an UDP tunnel\n" 2789#ifdef CONFIG_VDE 2790 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n" 2791 " configure a network backend to connect to port 'n' of a vde switch\n" 2792 " running on host and listening for incoming connections on 'socketpath'.\n" 2793 " Use group 'groupname' and mode 'octalmode' to change default\n" 2794 " ownership and permissions for communication port.\n" 2795#endif 2796#ifdef CONFIG_NETMAP 2797 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n" 2798 " attach to the existing netmap-enabled network interface 'name', or to a\n" 2799 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n" 2800 " netmap device, defaults to '/dev/netmap')\n" 2801#endif 2802#ifdef CONFIG_POSIX 2803 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n" 2804 " configure a vhost-user network, backed by a chardev 'dev'\n" 2805#endif 2806#ifdef __linux__ 2807 "-netdev vhost-vdpa,id=str[,vhostdev=/path/to/dev][,vhostfd=h]\n" 2808 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n" 2809 " use 'vhostdev=/path/to/dev' to open a vhost vdpa device\n" 2810 " use 'vhostfd=h' to connect to an already opened vhost vdpa device\n" 2811#endif 2812#ifdef CONFIG_VMNET 2813 "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n" 2814 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" 2815 " configure a vmnet network backend in host mode with ID 'str',\n" 2816 " isolate this interface from others with 'isolated',\n" 2817 " configure the address range and choose a subnet mask,\n" 2818 " specify network UUID 'uuid' to disable DHCP and interact with\n" 2819 " vmnet-host interfaces within this isolated network\n" 2820 "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n" 2821 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" 2822 " configure a vmnet network backend in shared mode with ID 'str',\n" 2823 " configure the address range and choose a subnet mask,\n" 2824 " set IPv6 ULA prefix (of length 64) to use for internal network,\n" 2825 " isolate this interface from others with 'isolated'\n" 2826 "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n" 2827 " configure a vmnet network backend in bridged mode with ID 'str',\n" 2828 " use 'ifname=name' to select a physical network interface to be bridged,\n" 2829 " isolate this interface from others with 'isolated'\n" 2830#endif 2831 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n" 2832 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL) 2833DEF("nic", HAS_ARG, QEMU_OPTION_nic, 2834 "-nic [tap|bridge|" 2835#ifdef CONFIG_SLIRP 2836 "user|" 2837#endif 2838#ifdef __linux__ 2839 "l2tpv3|" 2840#endif 2841#ifdef CONFIG_VDE 2842 "vde|" 2843#endif 2844#ifdef CONFIG_NETMAP 2845 "netmap|" 2846#endif 2847#ifdef CONFIG_POSIX 2848 "vhost-user|" 2849#endif 2850#ifdef CONFIG_VMNET 2851 "vmnet-host|vmnet-shared|vmnet-bridged|" 2852#endif 2853 "socket][,option][,...][mac=macaddr]\n" 2854 " initialize an on-board / default host NIC (using MAC address\n" 2855 " macaddr) and connect it to the given host network backend\n" 2856 "-nic none use it alone to have zero network devices (the default is to\n" 2857 " provided a 'user' network connection)\n", 2858 QEMU_ARCH_ALL) 2859DEF("net", HAS_ARG, QEMU_OPTION_net, 2860 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n" 2861 " configure or create an on-board (or machine default) NIC and\n" 2862 " connect it to hub 0 (please use -nic unless you need a hub)\n" 2863 "-net [" 2864#ifdef CONFIG_SLIRP 2865 "user|" 2866#endif 2867 "tap|" 2868 "bridge|" 2869#ifdef CONFIG_VDE 2870 "vde|" 2871#endif 2872#ifdef CONFIG_NETMAP 2873 "netmap|" 2874#endif 2875#ifdef CONFIG_VMNET 2876 "vmnet-host|vmnet-shared|vmnet-bridged|" 2877#endif 2878 "socket][,option][,option][,...]\n" 2879 " old way to initialize a host network interface\n" 2880 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL) 2881SRST 2882``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]`` 2883 This option is a shortcut for configuring both the on-board 2884 (default) guest NIC hardware and the host network backend in one go. 2885 The host backend options are the same as with the corresponding 2886 ``-netdev`` options below. The guest NIC model can be set with 2887 ``model=modelname``. Use ``model=help`` to list the available device 2888 types. The hardware MAC address can be set with ``mac=macaddr``. 2889 2890 The following two example do exactly the same, to show how ``-nic`` 2891 can be used to shorten the command line length: 2892 2893 .. parsed-literal:: 2894 2895 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32 2896 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32 2897 2898``-nic none`` 2899 Indicate that no network devices should be configured. It is used to 2900 override the default configuration (default NIC with "user" host 2901 network backend) which is activated if no other networking options 2902 are provided. 2903 2904``-netdev user,id=id[,option][,option][,...]`` 2905 Configure user mode host network backend which requires no 2906 administrator privilege to run. Valid options are: 2907 2908 ``id=id`` 2909 Assign symbolic name for use in monitor commands. 2910 2911 ``ipv4=on|off and ipv6=on|off`` 2912 Specify that either IPv4 or IPv6 must be enabled. If neither is 2913 specified both protocols are enabled. 2914 2915 ``net=addr[/mask]`` 2916 Set IP network address the guest will see. Optionally specify 2917 the netmask, either in the form a.b.c.d or as number of valid 2918 top-most bits. Default is 10.0.2.0/24. 2919 2920 ``host=addr`` 2921 Specify the guest-visible address of the host. Default is the 2922 2nd IP in the guest network, i.e. x.x.x.2. 2923 2924 ``ipv6-net=addr[/int]`` 2925 Set IPv6 network address the guest will see (default is 2926 fec0::/64). The network prefix is given in the usual hexadecimal 2927 IPv6 address notation. The prefix size is optional, and is given 2928 as the number of valid top-most bits (default is 64). 2929 2930 ``ipv6-host=addr`` 2931 Specify the guest-visible IPv6 address of the host. Default is 2932 the 2nd IPv6 in the guest network, i.e. xxxx::2. 2933 2934 ``restrict=on|off`` 2935 If this option is enabled, the guest will be isolated, i.e. it 2936 will not be able to contact the host and no guest IP packets 2937 will be routed over the host to the outside. This option does 2938 not affect any explicitly set forwarding rules. 2939 2940 ``hostname=name`` 2941 Specifies the client hostname reported by the built-in DHCP 2942 server. 2943 2944 ``dhcpstart=addr`` 2945 Specify the first of the 16 IPs the built-in DHCP server can 2946 assign. Default is the 15th to 31st IP in the guest network, 2947 i.e. x.x.x.15 to x.x.x.31. 2948 2949 ``dns=addr`` 2950 Specify the guest-visible address of the virtual nameserver. The 2951 address must be different from the host address. Default is the 2952 3rd IP in the guest network, i.e. x.x.x.3. 2953 2954 ``ipv6-dns=addr`` 2955 Specify the guest-visible address of the IPv6 virtual 2956 nameserver. The address must be different from the host address. 2957 Default is the 3rd IP in the guest network, i.e. xxxx::3. 2958 2959 ``dnssearch=domain`` 2960 Provides an entry for the domain-search list sent by the 2961 built-in DHCP server. More than one domain suffix can be 2962 transmitted by specifying this option multiple times. If 2963 supported, this will cause the guest to automatically try to 2964 append the given domain suffix(es) in case a domain name can not 2965 be resolved. 2966 2967 Example: 2968 2969 .. parsed-literal:: 2970 2971 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org 2972 2973 ``domainname=domain`` 2974 Specifies the client domain name reported by the built-in DHCP 2975 server. 2976 2977 ``tftp=dir`` 2978 When using the user mode network stack, activate a built-in TFTP 2979 server. The files in dir will be exposed as the root of a TFTP 2980 server. The TFTP client on the guest must be configured in 2981 binary mode (use the command ``bin`` of the Unix TFTP client). 2982 2983 ``tftp-server-name=name`` 2984 In BOOTP reply, broadcast name as the "TFTP server name" 2985 (RFC2132 option 66). This can be used to advise the guest to 2986 load boot files or configurations from a different server than 2987 the host address. 2988 2989 ``bootfile=file`` 2990 When using the user mode network stack, broadcast file as the 2991 BOOTP filename. In conjunction with ``tftp``, this can be used 2992 to network boot a guest from a local directory. 2993 2994 Example (using pxelinux): 2995 2996 .. parsed-literal:: 2997 2998 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\ 2999 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0 3000 3001 ``smb=dir[,smbserver=addr]`` 3002 When using the user mode network stack, activate a built-in SMB 3003 server so that Windows OSes can access to the host files in 3004 ``dir`` transparently. The IP address of the SMB server can be 3005 set to addr. By default the 4th IP in the guest network is used, 3006 i.e. x.x.x.4. 3007 3008 In the guest Windows OS, the line: 3009 3010 :: 3011 3012 10.0.2.4 smbserver 3013 3014 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows 3015 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows 3016 NT/2000). 3017 3018 Then ``dir`` can be accessed in ``\\smbserver\qemu``. 3019 3020 Note that a SAMBA server must be installed on the host OS. 3021 3022 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport`` 3023 Redirect incoming TCP or UDP connections to the host port 3024 hostport to the guest IP address guestaddr on guest port 3025 guestport. If guestaddr is not specified, its value is x.x.x.15 3026 (default first address given by the built-in DHCP server). By 3027 specifying hostaddr, the rule can be bound to a specific host 3028 interface. If no connection type is set, TCP is used. This 3029 option can be given multiple times. 3030 3031 For example, to redirect host X11 connection from screen 1 to 3032 guest screen 0, use the following: 3033 3034 .. parsed-literal:: 3035 3036 # on the host 3037 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000 3038 # this host xterm should open in the guest X11 server 3039 xterm -display :1 3040 3041 To redirect telnet connections from host port 5555 to telnet 3042 port on the guest, use the following: 3043 3044 .. parsed-literal:: 3045 3046 # on the host 3047 |qemu_system| -nic user,hostfwd=tcp::5555-:23 3048 telnet localhost 5555 3049 3050 Then when you use on the host ``telnet localhost 5555``, you 3051 connect to the guest telnet server. 3052 3053 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command`` 3054 Forward guest TCP connections to the IP address server on port 3055 port to the character device dev or to a program executed by 3056 cmd:command which gets spawned for each connection. This option 3057 can be given multiple times. 3058 3059 You can either use a chardev directly and have that one used 3060 throughout QEMU's lifetime, like in the following example: 3061 3062 .. parsed-literal:: 3063 3064 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever 3065 # the guest accesses it 3066 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321 3067 3068 Or you can execute a command on every TCP connection established 3069 by the guest, so that QEMU behaves similar to an inetd process 3070 for that virtual server: 3071 3072 .. parsed-literal:: 3073 3074 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234 3075 # and connect the TCP stream to its stdin/stdout 3076 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321' 3077 3078``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]`` 3079 Configure a host TAP network backend with ID id. 3080 3081 Use the network script file to configure it and the network script 3082 dfile to deconfigure it. If name is not provided, the OS 3083 automatically provides one. The default network configure script is 3084 ``/etc/qemu-ifup`` and the default network deconfigure script is 3085 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to 3086 disable script execution. 3087 3088 If running QEMU as an unprivileged user, use the network helper 3089 to configure the TAP interface and attach it to the bridge. 3090 The default network helper executable is 3091 ``/path/to/qemu-bridge-helper`` and the default bridge device is 3092 ``br0``. 3093 3094 ``fd``\ =h can be used to specify the handle of an already opened 3095 host TAP interface. 3096 3097 Examples: 3098 3099 .. parsed-literal:: 3100 3101 #launch a QEMU instance with the default network script 3102 |qemu_system| linux.img -nic tap 3103 3104 .. parsed-literal:: 3105 3106 #launch a QEMU instance with two NICs, each one connected 3107 #to a TAP device 3108 |qemu_system| linux.img \\ 3109 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\ 3110 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1 3111 3112 .. parsed-literal:: 3113 3114 #launch a QEMU instance with the default network helper to 3115 #connect a TAP device to bridge br0 3116 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3117 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper" 3118 3119``-netdev bridge,id=id[,br=bridge][,helper=helper]`` 3120 Connect a host TAP network interface to a host bridge device. 3121 3122 Use the network helper helper to configure the TAP interface and 3123 attach it to the bridge. The default network helper executable is 3124 ``/path/to/qemu-bridge-helper`` and the default bridge device is 3125 ``br0``. 3126 3127 Examples: 3128 3129 .. parsed-literal:: 3130 3131 #launch a QEMU instance with the default network helper to 3132 #connect a TAP device to bridge br0 3133 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1 3134 3135 .. parsed-literal:: 3136 3137 #launch a QEMU instance with the default network helper to 3138 #connect a TAP device to bridge qemubr0 3139 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1 3140 3141``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]`` 3142 This host network backend can be used to connect the guest's network 3143 to another QEMU virtual machine using a TCP socket connection. If 3144 ``listen`` is specified, QEMU waits for incoming connections on port 3145 (host is optional). ``connect`` is used to connect to another QEMU 3146 instance using the ``listen`` option. ``fd``\ =h specifies an 3147 already opened TCP socket. 3148 3149 Example: 3150 3151 .. parsed-literal:: 3152 3153 # launch a first QEMU instance 3154 |qemu_system| linux.img \\ 3155 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3156 -netdev socket,id=n1,listen=:1234 3157 # connect the network of this instance to the network of the first instance 3158 |qemu_system| linux.img \\ 3159 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3160 -netdev socket,id=n2,connect=127.0.0.1:1234 3161 3162``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]`` 3163 Configure a socket host network backend to share the guest's network 3164 traffic with another QEMU virtual machines using a UDP multicast 3165 socket, effectively making a bus for every QEMU with same multicast 3166 address maddr and port. NOTES: 3167 3168 1. Several QEMU can be running on different hosts and share same bus 3169 (assuming correct multicast setup for these hosts). 3170 3171 2. mcast support is compatible with User Mode Linux (argument 3172 ``ethN=mcast``), see http://user-mode-linux.sf.net. 3173 3174 3. Use ``fd=h`` to specify an already opened UDP multicast socket. 3175 3176 Example: 3177 3178 .. parsed-literal:: 3179 3180 # launch one QEMU instance 3181 |qemu_system| linux.img \\ 3182 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3183 -netdev socket,id=n1,mcast=230.0.0.1:1234 3184 # launch another QEMU instance on same "bus" 3185 |qemu_system| linux.img \\ 3186 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3187 -netdev socket,id=n2,mcast=230.0.0.1:1234 3188 # launch yet another QEMU instance on same "bus" 3189 |qemu_system| linux.img \\ 3190 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\ 3191 -netdev socket,id=n3,mcast=230.0.0.1:1234 3192 3193 Example (User Mode Linux compat.): 3194 3195 .. parsed-literal:: 3196 3197 # launch QEMU instance (note mcast address selected is UML's default) 3198 |qemu_system| linux.img \\ 3199 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3200 -netdev socket,id=n1,mcast=239.192.168.1:1102 3201 # launch UML 3202 /path/to/linux ubd0=/path/to/root_fs eth0=mcast 3203 3204 Example (send packets from host's 1.2.3.4): 3205 3206 .. parsed-literal:: 3207 3208 |qemu_system| linux.img \\ 3209 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3210 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4 3211 3212``-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]`` 3213 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931) 3214 is a popular protocol to transport Ethernet (and other Layer 2) data 3215 frames between two systems. It is present in routers, firewalls and 3216 the Linux kernel (from version 3.3 onwards). 3217 3218 This transport allows a VM to communicate to another VM, router or 3219 firewall directly. 3220 3221 ``src=srcaddr`` 3222 source address (mandatory) 3223 3224 ``dst=dstaddr`` 3225 destination address (mandatory) 3226 3227 ``udp`` 3228 select udp encapsulation (default is ip). 3229 3230 ``srcport=srcport`` 3231 source udp port. 3232 3233 ``dstport=dstport`` 3234 destination udp port. 3235 3236 ``ipv6`` 3237 force v6, otherwise defaults to v4. 3238 3239 ``rxcookie=rxcookie``; \ ``txcookie=txcookie`` 3240 Cookies are a weak form of security in the l2tpv3 specification. 3241 Their function is mostly to prevent misconfiguration. By default 3242 they are 32 bit. 3243 3244 ``cookie64`` 3245 Set cookie size to 64 bit instead of the default 32 3246 3247 ``counter=off`` 3248 Force a 'cut-down' L2TPv3 with no counter as in 3249 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00 3250 3251 ``pincounter=on`` 3252 Work around broken counter handling in peer. This may also help 3253 on networks which have packet reorder. 3254 3255 ``offset=offset`` 3256 Add an extra offset between header and data 3257 3258 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to 3259 the bridge br-lan on the remote Linux host 1.2.3.4: 3260 3261 .. parsed-literal:: 3262 3263 # Setup tunnel on linux host using raw ip as encapsulation 3264 # on 1.2.3.4 3265 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\ 3266 encap udp udp_sport 16384 udp_dport 16384 3267 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\ 3268 0xFFFFFFFF peer_session_id 0xFFFFFFFF 3269 ifconfig vmtunnel0 mtu 1500 3270 ifconfig vmtunnel0 up 3271 brctl addif br-lan vmtunnel0 3272 3273 3274 # on 4.3.2.1 3275 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter 3276 3277 |qemu_system| linux.img -device e1000,netdev=n1 \\ 3278 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter 3279 3280``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]`` 3281 Configure VDE backend to connect to PORT n of a vde switch running 3282 on host and listening for incoming connections on socketpath. Use 3283 GROUP groupname and MODE octalmode to change default ownership and 3284 permissions for communication port. This option is only available if 3285 QEMU has been compiled with vde support enabled. 3286 3287 Example: 3288 3289 .. parsed-literal:: 3290 3291 # launch vde switch 3292 vde_switch -F -sock /tmp/myswitch 3293 # launch QEMU instance 3294 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch 3295 3296``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]`` 3297 Establish a vhost-user netdev, backed by a chardev id. The chardev 3298 should be a unix domain socket backed one. The vhost-user uses a 3299 specifically defined protocol to pass vhost ioctl replacement 3300 messages to an application on the other end of the socket. On 3301 non-MSIX guests, the feature can be forced with vhostforce. Use 3302 'queues=n' to specify the number of queues to be created for 3303 multiqueue vhost-user. 3304 3305 Example: 3306 3307 :: 3308 3309 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \ 3310 -numa node,memdev=mem \ 3311 -chardev socket,id=chr0,path=/path/to/socket \ 3312 -netdev type=vhost-user,id=net0,chardev=chr0 \ 3313 -device virtio-net-pci,netdev=net0 3314 3315``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]`` 3316 Establish a vhost-vdpa netdev. 3317 3318 vDPA device is a device that uses a datapath which complies with 3319 the virtio specifications with a vendor specific control path. 3320 vDPA devices can be both physically located on the hardware or 3321 emulated by software. 3322 3323``-netdev hubport,id=id,hubid=hubid[,netdev=nd]`` 3324 Create a hub port on the emulated hub with ID hubid. 3325 3326 The hubport netdev lets you connect a NIC to a QEMU emulated hub 3327 instead of a single netdev. Alternatively, you can also connect the 3328 hubport to another netdev with ID nd by using the ``netdev=nd`` 3329 option. 3330 3331``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]`` 3332 Legacy option to configure or create an on-board (or machine 3333 default) Network Interface Card(NIC) and connect it either to the 3334 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd. 3335 If model is omitted, then the default NIC model associated with the 3336 machine type is used. Note that the default NIC model may change in 3337 future QEMU releases, so it is highly recommended to always specify 3338 a model. Optionally, the MAC address can be changed to mac, the 3339 device address set to addr (PCI cards only), and a name can be 3340 assigned for use in monitor commands. Optionally, for PCI cards, you 3341 can specify the number v of MSI-X vectors that the card should have; 3342 this option currently only affects virtio cards; set v = 0 to 3343 disable MSI-X. If no ``-net`` option is specified, a single NIC is 3344 created. QEMU can emulate several different models of network card. 3345 Use ``-net nic,model=help`` for a list of available devices for your 3346 target. 3347 3348``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]`` 3349 Configure a host network backend (with the options corresponding to 3350 the same ``-netdev`` option) and connect it to the emulated hub 0 3351 (the default hub). Use name to specify the name of the hub port. 3352ERST 3353 3354DEFHEADING() 3355 3356DEFHEADING(Character device options:) 3357 3358DEF("chardev", HAS_ARG, QEMU_OPTION_chardev, 3359 "-chardev help\n" 3360 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3361 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n" 3362 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n" 3363 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n" 3364 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n" 3365 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n" 3366 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n" 3367 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n" 3368 " [,logfile=PATH][,logappend=on|off]\n" 3369 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3370 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n" 3371 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3372 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n" 3373 "-chardev file,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3374 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3375#ifdef _WIN32 3376 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3377 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3378#else 3379 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3380 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n" 3381#endif 3382#ifdef CONFIG_BRLAPI 3383 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3384#endif 3385#if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \ 3386 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) 3387 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3388 "-chardev tty,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3389#endif 3390#if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__) 3391 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3392 "-chardev parport,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3393#endif 3394#if defined(CONFIG_SPICE) 3395 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3396 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3397#endif 3398 , QEMU_ARCH_ALL 3399) 3400 3401SRST 3402The general form of a character device option is: 3403 3404``-chardev backend,id=id[,mux=on|off][,options]`` 3405 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``, 3406 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``, 3407 ``pty``, ``stdio``, ``braille``, ``tty``, ``parallel``, ``parport``, 3408 ``spicevmc``, ``spiceport``. The specific backend will determine the 3409 applicable options. 3410 3411 Use ``-chardev help`` to print all available chardev backend types. 3412 3413 All devices must have an id, which can be any string up to 127 3414 characters long. It is used to uniquely identify this device in 3415 other command line directives. 3416 3417 A character device may be used in multiplexing mode by multiple 3418 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is 3419 a "1:N" device, and here the "1" end is your specified chardev 3420 backend, and the "N" end is the various parts of QEMU that can talk 3421 to a chardev. If you create a chardev with ``id=myid`` and 3422 ``mux=on``, QEMU will create a multiplexer with your specified ID, 3423 and you can then configure multiple front ends to use that chardev 3424 ID for their input/output. Up to four different front ends can be 3425 connected to a single multiplexed chardev. (Without multiplexing 3426 enabled, a chardev can only be used by a single front end.) For 3427 instance you could use this to allow a single stdio chardev to be 3428 used by two serial ports and the QEMU monitor: 3429 3430 :: 3431 3432 -chardev stdio,mux=on,id=char0 \ 3433 -mon chardev=char0,mode=readline \ 3434 -serial chardev:char0 \ 3435 -serial chardev:char0 3436 3437 You can have more than one multiplexer in a system configuration; 3438 for instance you could have a TCP port multiplexed between UART 0 3439 and UART 1, and stdio multiplexed between the QEMU monitor and a 3440 parallel port: 3441 3442 :: 3443 3444 -chardev stdio,mux=on,id=char0 \ 3445 -mon chardev=char0,mode=readline \ 3446 -parallel chardev:char0 \ 3447 -chardev tcp,...,mux=on,id=char1 \ 3448 -serial chardev:char1 \ 3449 -serial chardev:char1 3450 3451 When you're using a multiplexed character device, some escape 3452 sequences are interpreted in the input. See the chapter about 3453 :ref:`keys in the character backend multiplexer` in the 3454 System Emulation Users Guide for more details. 3455 3456 Note that some other command line options may implicitly create 3457 multiplexed character backends; for instance ``-serial mon:stdio`` 3458 creates a multiplexed stdio backend connected to the serial port and 3459 the QEMU monitor, and ``-nographic`` also multiplexes the console 3460 and the monitor to stdio. 3461 3462 There is currently no support for multiplexing in the other 3463 direction (where a single QEMU front end takes input and output from 3464 multiple chardevs). 3465 3466 Every backend supports the ``logfile`` option, which supplies the 3467 path to a file to record all data transmitted via the backend. The 3468 ``logappend`` option controls whether the log file will be truncated 3469 or appended to when opened. 3470 3471The available backends are: 3472 3473``-chardev null,id=id`` 3474 A void device. This device will not emit any data, and will drop any 3475 data it receives. The null backend does not take any options. 3476 3477``-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]`` 3478 Create a two-way stream socket, which can be either a TCP or a unix 3479 socket. A unix socket will be created if ``path`` is specified. 3480 Behaviour is undefined if TCP options are specified for a unix 3481 socket. 3482 3483 ``server=on|off`` specifies that the socket shall be a listening socket. 3484 3485 ``wait=on|off`` specifies that QEMU should not block waiting for a client 3486 to connect to a listening socket. 3487 3488 ``telnet=on|off`` specifies that traffic on the socket should interpret 3489 telnet escape sequences. 3490 3491 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for 3492 communication. 3493 3494 ``reconnect`` sets the timeout for reconnecting on non-server 3495 sockets when the remote end goes away. qemu will delay this many 3496 seconds and then attempt to reconnect. Zero disables reconnecting, 3497 and is the default. 3498 3499 ``tls-creds`` requests enablement of the TLS protocol for 3500 encryption, and specifies the id of the TLS credentials to use for 3501 the handshake. The credentials must be previously created with the 3502 ``-object tls-creds`` argument. 3503 3504 ``tls-auth`` provides the ID of the QAuthZ authorization object 3505 against which the client's x509 distinguished name will be 3506 validated. This object is only resolved at time of use, so can be 3507 deleted and recreated on the fly while the chardev server is active. 3508 If missing, it will default to denying access. 3509 3510 TCP and unix socket options are given below: 3511 3512 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]`` 3513 ``host`` for a listening socket specifies the local address to 3514 be bound. For a connecting socket species the remote host to 3515 connect to. ``host`` is optional for listening sockets. If not 3516 specified it defaults to ``0.0.0.0``. 3517 3518 ``port`` for a listening socket specifies the local port to be 3519 bound. For a connecting socket specifies the port on the remote 3520 host to connect to. ``port`` can be given as either a port 3521 number or a service name. ``port`` is required. 3522 3523 ``to`` is only relevant to listening sockets. If it is 3524 specified, and ``port`` cannot be bound, QEMU will attempt to 3525 bind to subsequent ports up to and including ``to`` until it 3526 succeeds. ``to`` must be specified as a port number. 3527 3528 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 3529 or IPv6 must be used. If neither is specified the socket may 3530 use either protocol. 3531 3532 ``nodelay=on|off`` disables the Nagle algorithm. 3533 3534 ``unix options: path=path[,abstract=on|off][,tight=on|off]`` 3535 ``path`` specifies the local path of the unix socket. ``path`` 3536 is required. 3537 ``abstract=on|off`` specifies the use of the abstract socket namespace, 3538 rather than the filesystem. Optional, defaults to false. 3539 ``tight=on|off`` sets the socket length of abstract sockets to their minimum, 3540 rather than the full sun_path length. Optional, defaults to true. 3541 3542``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]`` 3543 Sends all traffic from the guest to a remote host over UDP. 3544 3545 ``host`` specifies the remote host to connect to. If not specified 3546 it defaults to ``localhost``. 3547 3548 ``port`` specifies the port on the remote host to connect to. 3549 ``port`` is required. 3550 3551 ``localaddr`` specifies the local address to bind to. If not 3552 specified it defaults to ``0.0.0.0``. 3553 3554 ``localport`` specifies the local port to bind to. If not specified 3555 any available local port will be used. 3556 3557 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used. 3558 If neither is specified the device may use either protocol. 3559 3560``-chardev msmouse,id=id`` 3561 Forward QEMU's emulated msmouse events to the guest. ``msmouse`` 3562 does not take any options. 3563 3564``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]`` 3565 Connect to a QEMU text console. ``vc`` may optionally be given a 3566 specific size. 3567 3568 ``width`` and ``height`` specify the width and height respectively 3569 of the console, in pixels. 3570 3571 ``cols`` and ``rows`` specify that the console be sized to fit a 3572 text console with the given dimensions. 3573 3574``-chardev ringbuf,id=id[,size=size]`` 3575 Create a ring buffer with fixed size ``size``. size must be a power 3576 of two and defaults to ``64K``. 3577 3578``-chardev file,id=id,path=path`` 3579 Log all traffic received from the guest to a file. 3580 3581 ``path`` specifies the path of the file to be opened. This file will 3582 be created if it does not already exist, and overwritten if it does. 3583 ``path`` is required. 3584 3585``-chardev pipe,id=id,path=path`` 3586 Create a two-way connection to the guest. The behaviour differs 3587 slightly between Windows hosts and other hosts: 3588 3589 On Windows, a single duplex pipe will be created at 3590 ``\\.pipe\path``. 3591 3592 On other hosts, 2 pipes will be created called ``path.in`` and 3593 ``path.out``. Data written to ``path.in`` will be received by the 3594 guest. Data written by the guest can be read from ``path.out``. QEMU 3595 will not create these fifos, and requires them to be present. 3596 3597 ``path`` forms part of the pipe path as described above. ``path`` is 3598 required. 3599 3600``-chardev console,id=id`` 3601 Send traffic from the guest to QEMU's standard output. ``console`` 3602 does not take any options. 3603 3604 ``console`` is only available on Windows hosts. 3605 3606``-chardev serial,id=id,path=path`` 3607 Send traffic from the guest to a serial device on the host. 3608 3609 On Unix hosts serial will actually accept any tty device, not only 3610 serial lines. 3611 3612 ``path`` specifies the name of the serial device to open. 3613 3614``-chardev pty,id=id`` 3615 Create a new pseudo-terminal on the host and connect to it. ``pty`` 3616 does not take any options. 3617 3618 ``pty`` is not available on Windows hosts. 3619 3620``-chardev stdio,id=id[,signal=on|off]`` 3621 Connect to standard input and standard output of the QEMU process. 3622 3623 ``signal`` controls if signals are enabled on the terminal, that 3624 includes exiting QEMU with the key sequence Control-c. This option 3625 is enabled by default, use ``signal=off`` to disable it. 3626 3627``-chardev braille,id=id`` 3628 Connect to a local BrlAPI server. ``braille`` does not take any 3629 options. 3630 3631``-chardev tty,id=id,path=path`` 3632 ``tty`` is only available on Linux, Sun, FreeBSD, NetBSD, OpenBSD 3633 and DragonFlyBSD hosts. It is an alias for ``serial``. 3634 3635 ``path`` specifies the path to the tty. ``path`` is required. 3636 3637``-chardev parallel,id=id,path=path`` 3638 \ 3639``-chardev parport,id=id,path=path`` 3640 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD 3641 hosts. 3642 3643 Connect to a local parallel port. 3644 3645 ``path`` specifies the path to the parallel port device. ``path`` is 3646 required. 3647 3648``-chardev spicevmc,id=id,debug=debug,name=name`` 3649 ``spicevmc`` is only available when spice support is built in. 3650 3651 ``debug`` debug level for spicevmc 3652 3653 ``name`` name of spice channel to connect to 3654 3655 Connect to a spice virtual machine channel, such as vdiport. 3656 3657``-chardev spiceport,id=id,debug=debug,name=name`` 3658 ``spiceport`` is only available when spice support is built in. 3659 3660 ``debug`` debug level for spicevmc 3661 3662 ``name`` name of spice port to connect to 3663 3664 Connect to a spice port, allowing a Spice client to handle the 3665 traffic identified by a name (preferably a fqdn). 3666ERST 3667 3668DEFHEADING() 3669 3670#ifdef CONFIG_TPM 3671DEFHEADING(TPM device options:) 3672 3673DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \ 3674 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n" 3675 " use path to provide path to a character device; default is /dev/tpm0\n" 3676 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n" 3677 " not provided it will be searched for in /sys/class/misc/tpm?/device\n" 3678 "-tpmdev emulator,id=id,chardev=dev\n" 3679 " configure the TPM device using chardev backend\n", 3680 QEMU_ARCH_ALL) 3681SRST 3682The general form of a TPM device option is: 3683 3684``-tpmdev backend,id=id[,options]`` 3685 The specific backend type will determine the applicable options. The 3686 ``-tpmdev`` option creates the TPM backend and requires a 3687 ``-device`` option that specifies the TPM frontend interface model. 3688 3689 Use ``-tpmdev help`` to print all available TPM backend types. 3690 3691The available backends are: 3692 3693``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path`` 3694 (Linux-host only) Enable access to the host's TPM using the 3695 passthrough driver. 3696 3697 ``path`` specifies the path to the host's TPM device, i.e., on a 3698 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by 3699 default ``/dev/tpm0`` is used. 3700 3701 ``cancel-path`` specifies the path to the host TPM device's sysfs 3702 entry allowing for cancellation of an ongoing TPM command. 3703 ``cancel-path`` is optional and by default QEMU will search for the 3704 sysfs entry to use. 3705 3706 Some notes about using the host's TPM with the passthrough driver: 3707 3708 The TPM device accessed by the passthrough driver must not be used 3709 by any other application on the host. 3710 3711 Since the host's firmware (BIOS/UEFI) has already initialized the 3712 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize 3713 the TPM again and may therefore not show a TPM-specific menu that 3714 would otherwise allow the user to configure the TPM, e.g., allow the 3715 user to enable/disable or activate/deactivate the TPM. Further, if 3716 TPM ownership is released from within a VM then the host's TPM will 3717 get disabled and deactivated. To enable and activate the TPM again 3718 afterwards, the host has to be rebooted and the user is required to 3719 enter the firmware's menu to enable and activate the TPM. If the TPM 3720 is left disabled and/or deactivated most TPM commands will fail. 3721 3722 To create a passthrough TPM use the following two options: 3723 3724 :: 3725 3726 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0 3727 3728 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by 3729 ``tpmdev=tpm0`` in the device option. 3730 3731``-tpmdev emulator,id=id,chardev=dev`` 3732 (Linux-host only) Enable access to a TPM emulator using Unix domain 3733 socket based chardev backend. 3734 3735 ``chardev`` specifies the unique ID of a character device backend 3736 that provides connection to the software TPM server. 3737 3738 To create a TPM emulator backend device with chardev socket backend: 3739 3740 :: 3741 3742 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0 3743ERST 3744 3745DEFHEADING() 3746 3747#endif 3748 3749DEFHEADING(Boot Image or Kernel specific:) 3750SRST 3751There are broadly 4 ways you can boot a system with QEMU. 3752 3753 - specify a firmware and let it control finding a kernel 3754 - specify a firmware and pass a hint to the kernel to boot 3755 - direct kernel image boot 3756 - manually load files into the guest's address space 3757 3758The third method is useful for quickly testing kernels but as there is 3759no firmware to pass configuration information to the kernel the 3760hardware must either be probeable, the kernel built for the exact 3761configuration or passed some configuration data (e.g. a DTB blob) 3762which tells the kernel what drivers it needs. This exact details are 3763often hardware specific. 3764 3765The final method is the most generic way of loading images into the 3766guest address space and used mostly for ``bare metal`` type 3767development where the reset vectors of the processor are taken into 3768account. 3769 3770ERST 3771 3772SRST 3773 3774For x86 machines and some other architectures ``-bios`` will generally 3775do the right thing with whatever it is given. For other machines the 3776more strict ``-pflash`` option needs an image that is sized for the 3777flash device for the given machine type. 3778 3779Please see the :ref:`system-targets-ref` section of the manual for 3780more detailed documentation. 3781 3782ERST 3783 3784DEF("bios", HAS_ARG, QEMU_OPTION_bios, \ 3785 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL) 3786SRST 3787``-bios file`` 3788 Set the filename for the BIOS. 3789ERST 3790 3791DEF("pflash", HAS_ARG, QEMU_OPTION_pflash, 3792 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL) 3793SRST 3794``-pflash file`` 3795 Use file as a parallel flash image. 3796ERST 3797 3798SRST 3799 3800The kernel options were designed to work with Linux kernels although 3801other things (like hypervisors) can be packaged up as a kernel 3802executable image. The exact format of a executable image is usually 3803architecture specific. 3804 3805The way in which the kernel is started (what address it is loaded at, 3806what if any information is passed to it via CPU registers, the state 3807of the hardware when it is started, and so on) is also architecture 3808specific. Typically it follows the specification laid down by the 3809Linux kernel for how kernels for that architecture must be started. 3810 3811ERST 3812 3813DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \ 3814 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL) 3815SRST 3816``-kernel bzImage`` 3817 Use bzImage as kernel image. The kernel can be either a Linux kernel 3818 or in multiboot format. 3819ERST 3820 3821DEF("append", HAS_ARG, QEMU_OPTION_append, \ 3822 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL) 3823SRST 3824``-append cmdline`` 3825 Use cmdline as kernel command line 3826ERST 3827 3828DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \ 3829 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL) 3830SRST 3831``-initrd file`` 3832 Use file as initial ram disk. 3833 3834``-initrd "file1 arg=foo,file2"`` 3835 This syntax is only available with multiboot. 3836 3837 Use file1 and file2 as modules and pass arg=foo as parameter to the 3838 first module. 3839ERST 3840 3841DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \ 3842 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL) 3843SRST 3844``-dtb file`` 3845 Use file as a device tree binary (dtb) image and pass it to the 3846 kernel on boot. 3847ERST 3848 3849SRST 3850 3851Finally you can also manually load images directly into the address 3852space of the guest. This is most useful for developers who already 3853know the layout of their guest and take care to ensure something sane 3854will happen when the reset vector executes. 3855 3856The generic loader can be invoked by using the loader device: 3857 3858``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]`` 3859 3860there is also the guest loader which operates in a similar way but 3861tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where 3862the guest image is: 3863 3864``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]`` 3865 3866ERST 3867 3868DEFHEADING() 3869 3870DEFHEADING(Debug/Expert options:) 3871 3872DEF("compat", HAS_ARG, QEMU_OPTION_compat, 3873 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n" 3874 " Policy for handling deprecated management interfaces\n" 3875 "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n" 3876 " Policy for handling unstable management interfaces\n", 3877 QEMU_ARCH_ALL) 3878SRST 3879``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]`` 3880 Set policy for handling deprecated management interfaces (experimental): 3881 3882 ``deprecated-input=accept`` (default) 3883 Accept deprecated commands and arguments 3884 ``deprecated-input=reject`` 3885 Reject deprecated commands and arguments 3886 ``deprecated-input=crash`` 3887 Crash on deprecated commands and arguments 3888 ``deprecated-output=accept`` (default) 3889 Emit deprecated command results and events 3890 ``deprecated-output=hide`` 3891 Suppress deprecated command results and events 3892 3893 Limitation: covers only syntactic aspects of QMP. 3894 3895``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]`` 3896 Set policy for handling unstable management interfaces (experimental): 3897 3898 ``unstable-input=accept`` (default) 3899 Accept unstable commands and arguments 3900 ``unstable-input=reject`` 3901 Reject unstable commands and arguments 3902 ``unstable-input=crash`` 3903 Crash on unstable commands and arguments 3904 ``unstable-output=accept`` (default) 3905 Emit unstable command results and events 3906 ``unstable-output=hide`` 3907 Suppress unstable command results and events 3908 3909 Limitation: covers only syntactic aspects of QMP. 3910ERST 3911 3912DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg, 3913 "-fw_cfg [name=]<name>,file=<file>\n" 3914 " add named fw_cfg entry with contents from file\n" 3915 "-fw_cfg [name=]<name>,string=<str>\n" 3916 " add named fw_cfg entry with contents from string\n", 3917 QEMU_ARCH_ALL) 3918SRST 3919``-fw_cfg [name=]name,file=file`` 3920 Add named fw\_cfg entry with contents from file file. 3921 3922``-fw_cfg [name=]name,string=str`` 3923 Add named fw\_cfg entry with contents from string str. 3924 3925 The terminating NUL character of the contents of str will not be 3926 included as part of the fw\_cfg item data. To insert contents with 3927 embedded NUL characters, you have to use the file parameter. 3928 3929 The fw\_cfg entries are passed by QEMU through to the guest. 3930 3931 Example: 3932 3933 :: 3934 3935 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin 3936 3937 creates an fw\_cfg entry named opt/com.mycompany/blob with contents 3938 from ./my\_blob.bin. 3939ERST 3940 3941DEF("serial", HAS_ARG, QEMU_OPTION_serial, \ 3942 "-serial dev redirect the serial port to char device 'dev'\n", 3943 QEMU_ARCH_ALL) 3944SRST 3945``-serial dev`` 3946 Redirect the virtual serial port to host character device dev. The 3947 default device is ``vc`` in graphical mode and ``stdio`` in non 3948 graphical mode. 3949 3950 This option can be used several times to simulate up to 4 serial 3951 ports. 3952 3953 Use ``-serial none`` to disable all serial ports. 3954 3955 Available character devices are: 3956 3957 ``vc[:WxH]`` 3958 Virtual console. Optionally, a width and height can be given in 3959 pixel with 3960 3961 :: 3962 3963 vc:800x600 3964 3965 It is also possible to specify width or height in characters: 3966 3967 :: 3968 3969 vc:80Cx24C 3970 3971 ``pty`` 3972 [Linux only] Pseudo TTY (a new PTY is automatically allocated) 3973 3974 ``none`` 3975 No device is allocated. 3976 3977 ``null`` 3978 void device 3979 3980 ``chardev:id`` 3981 Use a named character device defined with the ``-chardev`` 3982 option. 3983 3984 ``/dev/XXX`` 3985 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial 3986 port parameters are set according to the emulated ones. 3987 3988 ``/dev/parportN`` 3989 [Linux only, parallel port only] Use host parallel port N. 3990 Currently SPP and EPP parallel port features can be used. 3991 3992 ``file:filename`` 3993 Write output to filename. No character can be read. 3994 3995 ``stdio`` 3996 [Unix only] standard input/output 3997 3998 ``pipe:filename`` 3999 name pipe filename 4000 4001 ``COMn`` 4002 [Windows only] Use host serial port n 4003 4004 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]`` 4005 This implements UDP Net Console. When remote\_host or src\_ip 4006 are not specified they default to ``0.0.0.0``. When not using a 4007 specified src\_port a random port is automatically chosen. 4008 4009 If you just want a simple readonly console you can use 4010 ``netcat`` or ``nc``, by starting QEMU with: 4011 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time 4012 QEMU writes something to that port it will appear in the 4013 netconsole session. 4014 4015 If you plan to send characters back via netconsole or you want 4016 to stop and start QEMU a lot of times, you should have QEMU use 4017 the same source port each time by using something like ``-serial 4018 udp::4555@:4556`` to QEMU. Another approach is to use a patched 4019 version of netcat which can listen to a TCP port and send and 4020 receive characters via udp. If you have a patched version of 4021 netcat which activates telnet remote echo and single char 4022 transfer, then you can use the following options to set up a 4023 netcat redirector to allow telnet on port 5555 to access the 4024 QEMU port. 4025 4026 ``QEMU Options:`` 4027 -serial udp::4555@:4556 4028 4029 ``netcat options:`` 4030 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T 4031 4032 ``telnet options:`` 4033 localhost 5555 4034 4035 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]`` 4036 The TCP Net Console has two modes of operation. It can send the 4037 serial I/O to a location or wait for a connection from a 4038 location. By default the TCP Net Console is sent to host at the 4039 port. If you use the ``server=on`` option QEMU will wait for a client 4040 socket application to connect to the port before continuing, 4041 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off`` 4042 option disables the Nagle buffering algorithm. The ``reconnect=on`` 4043 option only applies if ``server=no`` is set, if the connection goes 4044 down it will attempt to reconnect at the given interval. If host 4045 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a 4046 time is accepted. You can use ``telnet=on`` to connect to the 4047 corresponding character device. 4048 4049 ``Example to send tcp console to 192.168.0.2 port 4444`` 4050 -serial tcp:192.168.0.2:4444 4051 4052 ``Example to listen and wait on port 4444 for connection`` 4053 -serial tcp::4444,server=on 4054 4055 ``Example to not wait and listen on ip 192.168.0.100 port 4444`` 4056 -serial tcp:192.168.0.100:4444,server=on,wait=off 4057 4058 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]`` 4059 The telnet protocol is used instead of raw tcp sockets. The 4060 options work the same as if you had specified ``-serial tcp``. 4061 The difference is that the port acts like a telnet server or 4062 client using telnet option negotiation. This will also allow you 4063 to send the MAGIC\_SYSRQ sequence if you use a telnet that 4064 supports sending the break sequence. Typically in unix telnet 4065 you do it with Control-] and then type "send break" followed by 4066 pressing the enter key. 4067 4068 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]`` 4069 The WebSocket protocol is used instead of raw tcp socket. The 4070 port acts as a WebSocket server. Client mode is not supported. 4071 4072 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]`` 4073 A unix domain socket is used instead of a tcp socket. The option 4074 works the same as if you had specified ``-serial tcp`` except 4075 the unix domain socket path is used for connections. 4076 4077 ``mon:dev_string`` 4078 This is a special option to allow the monitor to be multiplexed 4079 onto another serial port. The monitor is accessed with key 4080 sequence of Control-a and then pressing c. dev\_string should be 4081 any one of the serial devices specified above. An example to 4082 multiplex the monitor onto a telnet server listening on port 4083 4444 would be: 4084 4085 ``-serial mon:telnet::4444,server=on,wait=off`` 4086 4087 When the monitor is multiplexed to stdio in this way, Ctrl+C 4088 will not terminate QEMU any more but will be passed to the guest 4089 instead. 4090 4091 ``braille`` 4092 Braille device. This will use BrlAPI to display the braille 4093 output on a real or fake device. 4094 4095 ``msmouse`` 4096 Three button serial mouse. Configure the guest to use Microsoft 4097 protocol. 4098ERST 4099 4100DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \ 4101 "-parallel dev redirect the parallel port to char device 'dev'\n", 4102 QEMU_ARCH_ALL) 4103SRST 4104``-parallel dev`` 4105 Redirect the virtual parallel port to host device dev (same devices 4106 as the serial port). On Linux hosts, ``/dev/parportN`` can be used 4107 to use hardware devices connected on the corresponding host parallel 4108 port. 4109 4110 This option can be used several times to simulate up to 3 parallel 4111 ports. 4112 4113 Use ``-parallel none`` to disable all parallel ports. 4114ERST 4115 4116DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \ 4117 "-monitor dev redirect the monitor to char device 'dev'\n", 4118 QEMU_ARCH_ALL) 4119SRST 4120``-monitor dev`` 4121 Redirect the monitor to host device dev (same devices as the serial 4122 port). The default device is ``vc`` in graphical mode and ``stdio`` 4123 in non graphical mode. Use ``-monitor none`` to disable the default 4124 monitor. 4125ERST 4126DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \ 4127 "-qmp dev like -monitor but opens in 'control' mode\n", 4128 QEMU_ARCH_ALL) 4129SRST 4130``-qmp dev`` 4131 Like -monitor but opens in 'control' mode. 4132ERST 4133DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \ 4134 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n", 4135 QEMU_ARCH_ALL) 4136SRST 4137``-qmp-pretty dev`` 4138 Like -qmp but uses pretty JSON formatting. 4139ERST 4140 4141DEF("mon", HAS_ARG, QEMU_OPTION_mon, \ 4142 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL) 4143SRST 4144``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]`` 4145 Setup monitor on chardev name. ``mode=control`` configures 4146 a QMP monitor (a JSON RPC-style protocol) and it is not the 4147 same as HMP, the human monitor that has a "(qemu)" prompt. 4148 ``pretty`` is only valid when ``mode=control``, 4149 turning on JSON pretty printing to ease 4150 human reading and debugging. 4151ERST 4152 4153DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \ 4154 "-debugcon dev redirect the debug console to char device 'dev'\n", 4155 QEMU_ARCH_ALL) 4156SRST 4157``-debugcon dev`` 4158 Redirect the debug console to host device dev (same devices as the 4159 serial port). The debug console is an I/O port which is typically 4160 port 0xe9; writing to that I/O port sends output to this device. The 4161 default device is ``vc`` in graphical mode and ``stdio`` in non 4162 graphical mode. 4163ERST 4164 4165DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \ 4166 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL) 4167SRST 4168``-pidfile file`` 4169 Store the QEMU process PID in file. It is useful if you launch QEMU 4170 from a script. 4171ERST 4172 4173DEF("singlestep", 0, QEMU_OPTION_singlestep, \ 4174 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL) 4175SRST 4176``-singlestep`` 4177 Run the emulation in single step mode. 4178ERST 4179 4180DEF("preconfig", 0, QEMU_OPTION_preconfig, \ 4181 "--preconfig pause QEMU before machine is initialized (experimental)\n", 4182 QEMU_ARCH_ALL) 4183SRST 4184``--preconfig`` 4185 Pause QEMU for interactive configuration before the machine is 4186 created, which allows querying and configuring properties that will 4187 affect machine initialization. Use QMP command 'x-exit-preconfig' to 4188 exit the preconfig state and move to the next state (i.e. run guest 4189 if -S isn't used or pause the second time if -S is used). This 4190 option is experimental. 4191ERST 4192 4193DEF("S", 0, QEMU_OPTION_S, \ 4194 "-S freeze CPU at startup (use 'c' to start execution)\n", 4195 QEMU_ARCH_ALL) 4196SRST 4197``-S`` 4198 Do not start CPU at startup (you must type 'c' in the monitor). 4199ERST 4200 4201DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit, 4202 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n" 4203 " run qemu with overcommit hints\n" 4204 " mem-lock=on|off controls memory lock support (default: off)\n" 4205 " cpu-pm=on|off controls cpu power management (default: off)\n", 4206 QEMU_ARCH_ALL) 4207SRST 4208``-overcommit mem-lock=on|off`` 4209 \ 4210``-overcommit cpu-pm=on|off`` 4211 Run qemu with hints about host resource overcommit. The default is 4212 to assume that host overcommits all resources. 4213 4214 Locking qemu and guest memory can be enabled via ``mem-lock=on`` 4215 (disabled by default). This works when host memory is not 4216 overcommitted and reduces the worst-case latency for guest. 4217 4218 Guest ability to manage power state of host cpus (increasing latency 4219 for other processes on the same host cpu, but decreasing latency for 4220 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This 4221 works best when host CPU is not overcommitted. When used, host 4222 estimates of CPU cycle and power utilization will be incorrect, not 4223 taking into account guest idle time. 4224ERST 4225 4226DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \ 4227 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n" 4228 " the guest without waiting for gdb to connect; use -S too\n" 4229 " if you want it to not start execution.)\n", 4230 QEMU_ARCH_ALL) 4231SRST 4232``-gdb dev`` 4233 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter 4234 in the System Emulation Users Guide). Note that this option does not pause QEMU 4235 execution -- if you want QEMU to not start the guest until you 4236 connect with gdb and issue a ``continue`` command, you will need to 4237 also pass the ``-S`` option to QEMU. 4238 4239 The most usual configuration is to listen on a local TCP socket:: 4240 4241 -gdb tcp::3117 4242 4243 but you can specify other backends; UDP, pseudo TTY, or even stdio 4244 are all reasonable use cases. For example, a stdio connection 4245 allows you to start QEMU from within gdb and establish the 4246 connection via a pipe: 4247 4248 .. parsed-literal:: 4249 4250 (gdb) target remote | exec |qemu_system| -gdb stdio ... 4251ERST 4252 4253DEF("s", 0, QEMU_OPTION_s, \ 4254 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n", 4255 QEMU_ARCH_ALL) 4256SRST 4257``-s`` 4258 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234 4259 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide). 4260ERST 4261 4262DEF("d", HAS_ARG, QEMU_OPTION_d, \ 4263 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n", 4264 QEMU_ARCH_ALL) 4265SRST 4266``-d item1[,...]`` 4267 Enable logging of specified items. Use '-d help' for a list of log 4268 items. 4269ERST 4270 4271DEF("D", HAS_ARG, QEMU_OPTION_D, \ 4272 "-D logfile output log to logfile (default stderr)\n", 4273 QEMU_ARCH_ALL) 4274SRST 4275``-D logfile`` 4276 Output log in logfile instead of to stderr 4277ERST 4278 4279DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \ 4280 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n", 4281 QEMU_ARCH_ALL) 4282SRST 4283``-dfilter range1[,...]`` 4284 Filter debug output to that relevant to a range of target addresses. 4285 The filter spec can be either start+size, start-size or start..end 4286 where start end and size are the addresses and sizes required. For 4287 example: 4288 4289 :: 4290 4291 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000 4292 4293 Will dump output for any code in the 0x1000 sized block starting at 4294 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and 4295 another 0x1000 sized block starting at 0xffffffc00005f000. 4296ERST 4297 4298DEF("seed", HAS_ARG, QEMU_OPTION_seed, \ 4299 "-seed number seed the pseudo-random number generator\n", 4300 QEMU_ARCH_ALL) 4301SRST 4302``-seed number`` 4303 Force the guest to use a deterministic pseudo-random number 4304 generator, seeded with number. This does not affect crypto routines 4305 within the host. 4306ERST 4307 4308DEF("L", HAS_ARG, QEMU_OPTION_L, \ 4309 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n", 4310 QEMU_ARCH_ALL) 4311SRST 4312``-L path`` 4313 Set the directory for the BIOS, VGA BIOS and keymaps. 4314 4315 To list all the data directories, use ``-L help``. 4316ERST 4317 4318DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \ 4319 "-enable-kvm enable KVM full virtualization support\n", 4320 QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC | 4321 QEMU_ARCH_RISCV | QEMU_ARCH_S390X) 4322SRST 4323``-enable-kvm`` 4324 Enable KVM full virtualization support. This option is only 4325 available if KVM support is enabled when compiling. 4326ERST 4327 4328DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, 4329 "-xen-domid id specify xen guest domain id\n", 4330 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4331DEF("xen-attach", 0, QEMU_OPTION_xen_attach, 4332 "-xen-attach attach to existing xen domain\n" 4333 " libxl will use this when starting QEMU\n", 4334 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4335DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, 4336 "-xen-domid-restrict restrict set of available xen operations\n" 4337 " to specified domain id. (Does not affect\n" 4338 " xenpv machine type).\n", 4339 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4340SRST 4341``-xen-domid id`` 4342 Specify xen guest domain id (XEN only). 4343 4344``-xen-attach`` 4345 Attach to existing xen domain. libxl will use this when starting 4346 QEMU (XEN only). Restrict set of available xen operations to 4347 specified domain id (XEN only). 4348ERST 4349 4350DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \ 4351 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL) 4352SRST 4353``-no-reboot`` 4354 Exit instead of rebooting. 4355ERST 4356 4357DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \ 4358 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL) 4359SRST 4360``-no-shutdown`` 4361 Don't exit QEMU on guest shutdown, but instead only stop the 4362 emulation. This allows for instance switching to monitor to commit 4363 changes to the disk image. 4364ERST 4365 4366DEF("action", HAS_ARG, QEMU_OPTION_action, 4367 "-action reboot=reset|shutdown\n" 4368 " action when guest reboots [default=reset]\n" 4369 "-action shutdown=poweroff|pause\n" 4370 " action when guest shuts down [default=poweroff]\n" 4371 "-action panic=pause|shutdown|exit-failure|none\n" 4372 " action when guest panics [default=shutdown]\n" 4373 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" 4374 " action when watchdog fires [default=reset]\n", 4375 QEMU_ARCH_ALL) 4376SRST 4377``-action event=action`` 4378 The action parameter serves to modify QEMU's default behavior when 4379 certain guest events occur. It provides a generic method for specifying the 4380 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown`` 4381 parameters. 4382 4383 Examples: 4384 4385 ``-action panic=none`` 4386 ``-action reboot=shutdown,shutdown=pause`` 4387 ``-device i6300esb -action watchdog=pause`` 4388 4389ERST 4390 4391DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \ 4392 "-loadvm [tag|id]\n" \ 4393 " start right away with a saved state (loadvm in monitor)\n", 4394 QEMU_ARCH_ALL) 4395SRST 4396``-loadvm file`` 4397 Start right away with a saved state (``loadvm`` in monitor) 4398ERST 4399 4400#ifndef _WIN32 4401DEF("daemonize", 0, QEMU_OPTION_daemonize, \ 4402 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL) 4403#endif 4404SRST 4405``-daemonize`` 4406 Daemonize the QEMU process after initialization. QEMU will not 4407 detach from standard IO until it is ready to receive connections on 4408 any of its devices. This option is a useful way for external 4409 programs to launch QEMU without having to cope with initialization 4410 race conditions. 4411ERST 4412 4413DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \ 4414 "-option-rom rom load a file, rom, into the option ROM space\n", 4415 QEMU_ARCH_ALL) 4416SRST 4417``-option-rom file`` 4418 Load the contents of file as an option ROM. This option is useful to 4419 load things like EtherBoot. 4420ERST 4421 4422DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \ 4423 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \ 4424 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n", 4425 QEMU_ARCH_ALL) 4426 4427SRST 4428``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]`` 4429 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at 4430 the current UTC or local time, respectively. ``localtime`` is 4431 required for correct date in MS-DOS or Windows. To start at a 4432 specific point in time, provide datetime in the format 4433 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC. 4434 4435 By default the RTC is driven by the host system time. This allows 4436 using of the RTC as accurate reference clock inside the guest, 4437 specifically if the host time is smoothly following an accurate 4438 external reference clock, e.g. via NTP. If you want to isolate the 4439 guest time from the host, you can set ``clock`` to ``rt`` instead, 4440 which provides a host monotonic clock if host support it. To even 4441 prevent the RTC from progressing during suspension, you can set 4442 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is 4443 recommended especially in icount mode in order to preserve 4444 determinism; however, note that in icount mode the speed of the 4445 virtual clock is variable and can in general differ from the host 4446 clock. 4447 4448 Enable ``driftfix`` (i386 targets only) if you experience time drift 4449 problems, specifically with Windows' ACPI HAL. This option will try 4450 to figure out how many timer interrupts were not processed by the 4451 Windows guest and will re-inject them. 4452ERST 4453 4454DEF("icount", HAS_ARG, QEMU_OPTION_icount, \ 4455 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \ 4456 " enable virtual instruction counter with 2^N clock ticks per\n" \ 4457 " instruction, enable aligning the host and virtual clocks\n" \ 4458 " or disable real time cpu sleeping, and optionally enable\n" \ 4459 " record-and-replay mode\n", QEMU_ARCH_ALL) 4460SRST 4461``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]`` 4462 Enable virtual instruction counter. The virtual cpu will execute one 4463 instruction every 2^N ns of virtual time. If ``auto`` is specified 4464 then the virtual cpu speed will be automatically adjusted to keep 4465 virtual time within a few seconds of real time. 4466 4467 Note that while this option can give deterministic behavior, it does 4468 not provide cycle accurate emulation. Modern CPUs contain 4469 superscalar out of order cores with complex cache hierarchies. The 4470 number of instructions executed often has little or no correlation 4471 with actual performance. 4472 4473 When the virtual cpu is sleeping, the virtual time will advance at 4474 default speed unless ``sleep=on`` is specified. With 4475 ``sleep=on``, the virtual time will jump to the next timer 4476 deadline instantly whenever the virtual cpu goes to sleep mode and 4477 will not advance if no timer is enabled. This behavior gives 4478 deterministic execution times from the guest point of view. 4479 The default if icount is enabled is ``sleep=off``. 4480 ``sleep=on`` cannot be used together with either ``shift=auto`` 4481 or ``align=on``. 4482 4483 ``align=on`` will activate the delay algorithm which will try to 4484 synchronise the host clock and the virtual clock. The goal is to 4485 have a guest running at the real frequency imposed by the shift 4486 option. Whenever the guest clock is behind the host clock and if 4487 ``align=on`` is specified then we print a message to the user to 4488 inform about the delay. Currently this option does not work when 4489 ``shift`` is ``auto``. Note: The sync algorithm will work for those 4490 shift values for which the guest clock runs ahead of the host clock. 4491 Typically this happens when the shift value is high (how high 4492 depends on the host machine). The default if icount is enabled 4493 is ``align=off``. 4494 4495 When the ``rr`` option is specified deterministic record/replay is 4496 enabled. The ``rrfile=`` option must also be provided to 4497 specify the path to the replay log. In record mode data is written 4498 to this file, and in replay mode it is read back. 4499 If the ``rrsnapshot`` option is given then it specifies a VM snapshot 4500 name. In record mode, a new VM snapshot with the given name is created 4501 at the start of execution recording. In replay mode this option 4502 specifies the snapshot name used to load the initial VM state. 4503ERST 4504 4505DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \ 4506 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \ 4507 " action when watchdog fires [default=reset]\n", 4508 QEMU_ARCH_ALL) 4509SRST 4510``-watchdog-action action`` 4511 The action controls what QEMU will do when the watchdog timer 4512 expires. The default is ``reset`` (forcefully reset the guest). 4513 Other possible actions are: ``shutdown`` (attempt to gracefully 4514 shutdown the guest), ``poweroff`` (forcefully poweroff the guest), 4515 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the 4516 guest), ``debug`` (print a debug message and continue), or ``none`` 4517 (do nothing). 4518 4519 Note that the ``shutdown`` action requires that the guest responds 4520 to ACPI signals, which it may not be able to do in the sort of 4521 situations where the watchdog would have expired, and thus 4522 ``-watchdog-action shutdown`` is not recommended for production use. 4523 4524 Examples: 4525 4526 ``-device i6300esb -watchdog-action pause`` 4527 4528ERST 4529 4530DEF("echr", HAS_ARG, QEMU_OPTION_echr, \ 4531 "-echr chr set terminal escape character instead of ctrl-a\n", 4532 QEMU_ARCH_ALL) 4533SRST 4534``-echr numeric_ascii_value`` 4535 Change the escape character used for switching to the monitor when 4536 using monitor and serial sharing. The default is ``0x01`` when using 4537 the ``-nographic`` option. ``0x01`` is equal to pressing 4538 ``Control-a``. You can select a different character from the ascii 4539 control keys where 1 through 26 map to Control-a through Control-z. 4540 For instance you could use the either of the following to change the 4541 escape character to Control-t. 4542 4543 ``-echr 0x14``; \ ``-echr 20`` 4544 4545ERST 4546 4547DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \ 4548 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \ 4549 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \ 4550 "-incoming unix:socketpath\n" \ 4551 " prepare for incoming migration, listen on\n" \ 4552 " specified protocol and socket address\n" \ 4553 "-incoming fd:fd\n" \ 4554 "-incoming exec:cmdline\n" \ 4555 " accept incoming migration on given file descriptor\n" \ 4556 " or from given external command\n" \ 4557 "-incoming defer\n" \ 4558 " wait for the URI to be specified via migrate_incoming\n", 4559 QEMU_ARCH_ALL) 4560SRST 4561``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]`` 4562 \ 4563``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]`` 4564 Prepare for incoming migration, listen on a given tcp port. 4565 4566``-incoming unix:socketpath`` 4567 Prepare for incoming migration, listen on a given unix socket. 4568 4569``-incoming fd:fd`` 4570 Accept incoming migration from a given filedescriptor. 4571 4572``-incoming exec:cmdline`` 4573 Accept incoming migration as an output from specified external 4574 command. 4575 4576``-incoming defer`` 4577 Wait for the URI to be specified via migrate\_incoming. The monitor 4578 can be used to change settings (such as migration parameters) prior 4579 to issuing the migrate\_incoming to allow the migration to begin. 4580ERST 4581 4582DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \ 4583 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL) 4584SRST 4585``-only-migratable`` 4586 Only allow migratable devices. Devices will not be allowed to enter 4587 an unmigratable state. 4588ERST 4589 4590DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \ 4591 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL) 4592SRST 4593``-nodefaults`` 4594 Don't create default devices. Normally, QEMU sets the default 4595 devices like serial port, parallel port, virtual console, monitor 4596 device, VGA adapter, floppy and CD-ROM drive and others. The 4597 ``-nodefaults`` option will disable all those default devices. 4598ERST 4599 4600#ifndef _WIN32 4601DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \ 4602 "-chroot dir chroot to dir just before starting the VM\n", 4603 QEMU_ARCH_ALL) 4604#endif 4605SRST 4606``-chroot dir`` 4607 Immediately before starting guest execution, chroot to the specified 4608 directory. Especially useful in combination with -runas. 4609ERST 4610 4611#ifndef _WIN32 4612DEF("runas", HAS_ARG, QEMU_OPTION_runas, \ 4613 "-runas user change to user id user just before starting the VM\n" \ 4614 " user can be numeric uid:gid instead\n", 4615 QEMU_ARCH_ALL) 4616#endif 4617SRST 4618``-runas user`` 4619 Immediately before starting guest execution, drop root privileges, 4620 switching to the specified user. 4621ERST 4622 4623DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env, 4624 "-prom-env variable=value\n" 4625 " set OpenBIOS nvram variables\n", 4626 QEMU_ARCH_PPC | QEMU_ARCH_SPARC) 4627SRST 4628``-prom-env variable=value`` 4629 Set OpenBIOS nvram variable to given value (PPC, SPARC only). 4630 4631 :: 4632 4633 qemu-system-sparc -prom-env 'auto-boot?=false' \ 4634 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single' 4635 4636 :: 4637 4638 qemu-system-ppc -prom-env 'auto-boot?=false' \ 4639 -prom-env 'boot-device=hd:2,\yaboot' \ 4640 -prom-env 'boot-args=conf=hd:2,\yaboot.conf' 4641ERST 4642DEF("semihosting", 0, QEMU_OPTION_semihosting, 4643 "-semihosting semihosting mode\n", 4644 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 4645 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) 4646SRST 4647``-semihosting`` 4648 Enable semihosting mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only). 4649 4650 Note that this allows guest direct access to the host filesystem, so 4651 should only be used with a trusted guest OS. 4652 4653 See the -semihosting-config option documentation for further 4654 information about the facilities this enables. 4655ERST 4656DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config, 4657 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]\n" \ 4658 " semihosting configuration\n", 4659QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 4660QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) 4661SRST 4662``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]`` 4663 Enable and configure semihosting (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V 4664 only). 4665 4666 Note that this allows guest direct access to the host filesystem, so 4667 should only be used with a trusted guest OS. 4668 4669 On Arm this implements the standard semihosting API, version 2.0. 4670 4671 On M68K this implements the "ColdFire GDB" interface used by 4672 libgloss. 4673 4674 Xtensa semihosting provides basic file IO calls, such as 4675 open/read/write/seek/select. Tensilica baremetal libc for ISS and 4676 linux platform "sim" use this interface. 4677 4678 On RISC-V this implements the standard semihosting API, version 0.2. 4679 4680 ``target=native|gdb|auto`` 4681 Defines where the semihosting calls will be addressed, to QEMU 4682 (``native``) or to GDB (``gdb``). The default is ``auto``, which 4683 means ``gdb`` during debug sessions and ``native`` otherwise. 4684 4685 ``chardev=str1`` 4686 Send the output to a chardev backend output for native or auto 4687 output when not in gdb 4688 4689 ``userspace=on|off`` 4690 Allows code running in guest userspace to access the semihosting 4691 interface. The default is that only privileged guest code can 4692 make semihosting calls. Note that setting ``userspace=on`` should 4693 only be used if all guest code is trusted (for example, in 4694 bare-metal test case code). 4695 4696 ``arg=str1,arg=str2,...`` 4697 Allows the user to pass input arguments, and can be used 4698 multiple times to build up a list. The old-style 4699 ``-kernel``/``-append`` method of passing a command line is 4700 still supported for backward compatibility. If both the 4701 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are 4702 specified, the former is passed to semihosting as it always 4703 takes precedence. 4704ERST 4705DEF("old-param", 0, QEMU_OPTION_old_param, 4706 "-old-param old param mode\n", QEMU_ARCH_ARM) 4707SRST 4708``-old-param`` 4709 Old param mode (ARM only). 4710ERST 4711 4712DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \ 4713 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \ 4714 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \ 4715 " Enable seccomp mode 2 system call filter (default 'off').\n" \ 4716 " use 'obsolete' to allow obsolete system calls that are provided\n" \ 4717 " by the kernel, but typically no longer used by modern\n" \ 4718 " C library implementations.\n" \ 4719 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \ 4720 " to elevate privileges using set*uid|gid system calls.\n" \ 4721 " The value 'children' will deny set*uid|gid system calls for\n" \ 4722 " main QEMU process but will allow forks and execves to run unprivileged\n" \ 4723 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \ 4724 " blocking *fork and execve\n" \ 4725 " use 'resourcecontrol' to disable process affinity and schedular priority\n", 4726 QEMU_ARCH_ALL) 4727SRST 4728``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]`` 4729 Enable Seccomp mode 2 system call filter. 'on' will enable syscall 4730 filtering and 'off' will disable it. The default is 'off'. 4731 4732 ``obsolete=string`` 4733 Enable Obsolete system calls 4734 4735 ``elevateprivileges=string`` 4736 Disable set\*uid\|gid system calls 4737 4738 ``spawn=string`` 4739 Disable \*fork and execve 4740 4741 ``resourcecontrol=string`` 4742 Disable process affinity and schedular priority 4743ERST 4744 4745DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig, 4746 "-readconfig <file>\n" 4747 " read config file\n", QEMU_ARCH_ALL) 4748SRST 4749``-readconfig file`` 4750 Read device configuration from file. This approach is useful when 4751 you want to spawn QEMU process with many command line options but 4752 you don't want to exceed the command line character limit. 4753ERST 4754 4755DEF("no-user-config", 0, QEMU_OPTION_nouserconfig, 4756 "-no-user-config\n" 4757 " do not load default user-provided config files at startup\n", 4758 QEMU_ARCH_ALL) 4759SRST 4760``-no-user-config`` 4761 The ``-no-user-config`` option makes QEMU not load any of the 4762 user-provided config files on sysconfdir. 4763ERST 4764 4765DEF("trace", HAS_ARG, QEMU_OPTION_trace, 4766 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n" 4767 " specify tracing options\n", 4768 QEMU_ARCH_ALL) 4769SRST 4770``-trace [[enable=]pattern][,events=file][,file=file]`` 4771 .. include:: ../qemu-option-trace.rst.inc 4772 4773ERST 4774DEF("plugin", HAS_ARG, QEMU_OPTION_plugin, 4775 "-plugin [file=]<file>[,<argname>=<argvalue>]\n" 4776 " load a plugin\n", 4777 QEMU_ARCH_ALL) 4778SRST 4779``-plugin file=file[,argname=argvalue]`` 4780 Load a plugin. 4781 4782 ``file=file`` 4783 Load the given plugin from a shared library file. 4784 4785 ``argname=argvalue`` 4786 Argument passed to the plugin. (Can be given multiple times.) 4787ERST 4788 4789HXCOMM Internal use 4790DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL) 4791DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL) 4792 4793DEF("msg", HAS_ARG, QEMU_OPTION_msg, 4794 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n" 4795 " control error message format\n" 4796 " timestamp=on enables timestamps (default: off)\n" 4797 " guest-name=on enables guest name prefix but only if\n" 4798 " -name guest option is set (default: off)\n", 4799 QEMU_ARCH_ALL) 4800SRST 4801``-msg [timestamp[=on|off]][,guest-name[=on|off]]`` 4802 Control error message format. 4803 4804 ``timestamp=on|off`` 4805 Prefix messages with a timestamp. Default is off. 4806 4807 ``guest-name=on|off`` 4808 Prefix messages with guest name but only if -name guest option is set 4809 otherwise the option is ignored. Default is off. 4810ERST 4811 4812DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate, 4813 "-dump-vmstate <file>\n" 4814 " Output vmstate information in JSON format to file.\n" 4815 " Use the scripts/vmstate-static-checker.py file to\n" 4816 " check for possible regressions in migration code\n" 4817 " by comparing two such vmstate dumps.\n", 4818 QEMU_ARCH_ALL) 4819SRST 4820``-dump-vmstate file`` 4821 Dump json-encoded vmstate information for current machine type to 4822 file in file 4823ERST 4824 4825DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile, 4826 "-enable-sync-profile\n" 4827 " enable synchronization profiling\n", 4828 QEMU_ARCH_ALL) 4829SRST 4830``-enable-sync-profile`` 4831 Enable synchronization profiling. 4832ERST 4833 4834DEFHEADING() 4835 4836DEFHEADING(Generic object creation:) 4837 4838DEF("object", HAS_ARG, QEMU_OPTION_object, 4839 "-object TYPENAME[,PROP1=VALUE1,...]\n" 4840 " create a new object of type TYPENAME setting properties\n" 4841 " in the order they are specified. Note that the 'id'\n" 4842 " property must be set. These objects are placed in the\n" 4843 " '/objects' path.\n", 4844 QEMU_ARCH_ALL) 4845SRST 4846``-object typename[,prop1=value1,...]`` 4847 Create a new object of type typename setting properties in the order 4848 they are specified. Note that the 'id' property must be set. These 4849 objects are placed in the '/objects' path. 4850 4851 ``-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,readonly=on|off`` 4852 Creates a memory file backend object, which can be used to back 4853 the guest RAM with huge pages. 4854 4855 The ``id`` parameter is a unique ID that will be used to 4856 reference this memory region in other parameters, e.g. ``-numa``, 4857 ``-device nvdimm``, etc. 4858 4859 The ``size`` option provides the size of the memory region, and 4860 accepts common suffixes, e.g. ``500M``. 4861 4862 The ``mem-path`` provides the path to either a shared memory or 4863 huge page filesystem mount. 4864 4865 The ``share`` boolean option determines whether the memory 4866 region is marked as private to QEMU, or shared. The latter 4867 allows a co-operating external process to access the QEMU memory 4868 region. 4869 4870 The ``share`` is also required for pvrdma devices due to 4871 limitations in the RDMA API provided by Linux. 4872 4873 Setting share=on might affect the ability to configure NUMA 4874 bindings for the memory backend under some circumstances, see 4875 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel 4876 source tree for additional details. 4877 4878 Setting the ``discard-data`` boolean option to on indicates that 4879 file contents can be destroyed when QEMU exits, to avoid 4880 unnecessarily flushing data to the backing file. Note that 4881 ``discard-data`` is only an optimization, and QEMU might not 4882 discard file contents if it aborts unexpectedly or is terminated 4883 using SIGKILL. 4884 4885 The ``merge`` boolean option enables memory merge, also known as 4886 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider 4887 the pages for memory deduplication. 4888 4889 Setting the ``dump`` boolean option to off excludes the memory 4890 from core dumps. This feature is also known as MADV\_DONTDUMP. 4891 4892 The ``prealloc`` boolean option enables memory preallocation. 4893 4894 The ``host-nodes`` option binds the memory range to a list of 4895 NUMA host nodes. 4896 4897 The ``policy`` option sets the NUMA policy to one of the 4898 following values: 4899 4900 ``default`` 4901 default host policy 4902 4903 ``preferred`` 4904 prefer the given host node list for allocation 4905 4906 ``bind`` 4907 restrict memory allocation to the given host node list 4908 4909 ``interleave`` 4910 interleave memory allocations across the given host node 4911 list 4912 4913 The ``align`` option specifies the base address alignment when 4914 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg 4915 ``2M``. Some backend store specified by ``mem-path`` requires an 4916 alignment different than the default one used by QEMU, eg the 4917 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In 4918 such cases, users can specify the required alignment via this 4919 option. 4920 4921 The ``pmem`` option specifies whether the backing file specified 4922 by ``mem-path`` is in host persistent memory that can be 4923 accessed using the SNIA NVM programming model (e.g. Intel 4924 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary 4925 operations to guarantee the persistence of its own writes to 4926 ``mem-path`` (e.g. in vNVDIMM label emulation and live 4927 migration). Also, we will map the backend-file with MAP\_SYNC 4928 flag, which ensures the file metadata is in sync for 4929 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC 4930 requires support from both the host kernel (since Linux kernel 4931 4.15) and the filesystem of ``mem-path`` mounted with DAX 4932 option. 4933 4934 The ``readonly`` option specifies whether the backing file is opened 4935 read-only or read-write (default). 4936 4937 ``-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`` 4938 Creates a memory backend object, which can be used to back the 4939 guest RAM. Memory backend objects offer more control than the 4940 ``-m`` option that is traditionally used to define guest RAM. 4941 Please refer to ``memory-backend-file`` for a description of the 4942 options. 4943 4944 ``-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`` 4945 Creates an anonymous memory file backend object, which allows 4946 QEMU to share the memory with an external process (e.g. when 4947 using vhost-user). The memory is allocated with memfd and 4948 optional sealing. (Linux only) 4949 4950 The ``seal`` option creates a sealed-file, that will block 4951 further resizing the memory ('on' by default). 4952 4953 The ``hugetlb`` option specify the file to be created resides in 4954 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction 4955 with the ``hugetlb`` option, the ``hugetlbsize`` option specify 4956 the hugetlb page size on systems that support multiple hugetlb 4957 page sizes (it must be a power of 2 value supported by the 4958 system). 4959 4960 In some versions of Linux, the ``hugetlb`` option is 4961 incompatible with the ``seal`` option (requires at least Linux 4962 4.16). 4963 4964 Please refer to ``memory-backend-file`` for a description of the 4965 other options. 4966 4967 The ``share`` boolean option is on by default with memfd. 4968 4969 ``-object rng-builtin,id=id`` 4970 Creates a random number generator backend which obtains entropy 4971 from QEMU builtin functions. The ``id`` parameter is a unique ID 4972 that will be used to reference this entropy backend from the 4973 ``virtio-rng`` device. By default, the ``virtio-rng`` device 4974 uses this RNG backend. 4975 4976 ``-object rng-random,id=id,filename=/dev/random`` 4977 Creates a random number generator backend which obtains entropy 4978 from a device on the host. The ``id`` parameter is a unique ID 4979 that will be used to reference this entropy backend from the 4980 ``virtio-rng`` device. The ``filename`` parameter specifies 4981 which file to obtain entropy from and if omitted defaults to 4982 ``/dev/urandom``. 4983 4984 ``-object rng-egd,id=id,chardev=chardevid`` 4985 Creates a random number generator backend which obtains entropy 4986 from an external daemon running on the host. The ``id`` 4987 parameter is a unique ID that will be used to reference this 4988 entropy backend from the ``virtio-rng`` device. The ``chardev`` 4989 parameter is the unique ID of a character device backend that 4990 provides the connection to the RNG daemon. 4991 4992 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off`` 4993 Creates a TLS anonymous credentials object, which can be used to 4994 provide TLS support on network backends. The ``id`` parameter is 4995 a unique ID which network backends will use to access the 4996 credentials. The ``endpoint`` is either ``server`` or ``client`` 4997 depending on whether the QEMU network backend that uses the 4998 credentials will be acting as a client or as a server. If 4999 ``verify-peer`` is enabled (the default) then once the handshake 5000 is completed, the peer credentials will be verified, though this 5001 is a no-op for anonymous credentials. 5002 5003 The dir parameter tells QEMU where to find the credential files. 5004 For server endpoints, this directory may contain a file 5005 dh-params.pem providing diffie-hellman parameters to use for the 5006 TLS server. If the file is missing, QEMU will generate a set of 5007 DH parameters at startup. This is a computationally expensive 5008 operation that consumes random pool entropy, so it is 5009 recommended that a persistent set of parameters be generated 5010 upfront and saved. 5011 5012 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]`` 5013 Creates a TLS Pre-Shared Keys (PSK) credentials object, which 5014 can be used to provide TLS support on network backends. The 5015 ``id`` parameter is a unique ID which network backends will use 5016 to access the credentials. The ``endpoint`` is either ``server`` 5017 or ``client`` depending on whether the QEMU network backend that 5018 uses the credentials will be acting as a client or as a server. 5019 For clients only, ``username`` is the username which will be 5020 sent to the server. If omitted it defaults to "qemu". 5021 5022 The dir parameter tells QEMU where to find the keys file. It is 5023 called "dir/keys.psk" and contains "username:key" pairs. This 5024 file can most easily be created using the GnuTLS ``psktool`` 5025 program. 5026 5027 For server endpoints, dir may also contain a file dh-params.pem 5028 providing diffie-hellman parameters to use for the TLS server. 5029 If the file is missing, QEMU will generate a set of DH 5030 parameters at startup. This is a computationally expensive 5031 operation that consumes random pool entropy, so it is 5032 recommended that a persistent set of parameters be generated up 5033 front and saved. 5034 5035 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id`` 5036 Creates a TLS anonymous credentials object, which can be used to 5037 provide TLS support on network backends. The ``id`` parameter is 5038 a unique ID which network backends will use to access the 5039 credentials. The ``endpoint`` is either ``server`` or ``client`` 5040 depending on whether the QEMU network backend that uses the 5041 credentials will be acting as a client or as a server. If 5042 ``verify-peer`` is enabled (the default) then once the handshake 5043 is completed, the peer credentials will be verified. With x509 5044 certificates, this implies that the clients must be provided 5045 with valid client certificates too. 5046 5047 The dir parameter tells QEMU where to find the credential files. 5048 For server endpoints, this directory may contain a file 5049 dh-params.pem providing diffie-hellman parameters to use for the 5050 TLS server. If the file is missing, QEMU will generate a set of 5051 DH parameters at startup. This is a computationally expensive 5052 operation that consumes random pool entropy, so it is 5053 recommended that a persistent set of parameters be generated 5054 upfront and saved. 5055 5056 For x509 certificate credentials the directory will contain 5057 further files providing the x509 certificates. The certificates 5058 must be stored in PEM format, in filenames ca-cert.pem, 5059 ca-crl.pem (optional), server-cert.pem (only servers), 5060 server-key.pem (only servers), client-cert.pem (only clients), 5061 and client-key.pem (only clients). 5062 5063 For the server-key.pem and client-key.pem files which contain 5064 sensitive private keys, it is possible to use an encrypted 5065 version by providing the passwordid parameter. This provides the 5066 ID of a previously created ``secret`` object containing the 5067 password for decryption. 5068 5069 The priority parameter allows to override the global default 5070 priority used by gnutls. This can be useful if the system 5071 administrator needs to use a weaker set of crypto priorities for 5072 QEMU without potentially forcing the weakness onto all 5073 applications. Or conversely if one wants wants a stronger 5074 default for QEMU than for all other applications, they can do 5075 this through this parameter. Its format is a gnutls priority 5076 string as described at 5077 https://gnutls.org/manual/html_node/Priority-Strings.html. 5078 5079 ``-object tls-cipher-suites,id=id,priority=priority`` 5080 Creates a TLS cipher suites object, which can be used to control 5081 the TLS cipher/protocol algorithms that applications are permitted 5082 to use. 5083 5084 The ``id`` parameter is a unique ID which frontends will use to 5085 access the ordered list of permitted TLS cipher suites from the 5086 host. 5087 5088 The ``priority`` parameter allows to override the global default 5089 priority used by gnutls. This can be useful if the system 5090 administrator needs to use a weaker set of crypto priorities for 5091 QEMU without potentially forcing the weakness onto all 5092 applications. Or conversely if one wants wants a stronger 5093 default for QEMU than for all other applications, they can do 5094 this through this parameter. Its format is a gnutls priority 5095 string as described at 5096 https://gnutls.org/manual/html_node/Priority-Strings.html. 5097 5098 An example of use of this object is to control UEFI HTTPS Boot. 5099 The tls-cipher-suites object exposes the ordered list of permitted 5100 TLS cipher suites from the host side to the guest firmware, via 5101 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER 5102 objects. The firmware uses the IANA_TLS_CIPHER array for configuring 5103 guest-side TLS. 5104 5105 In the following example, the priority at which the host-side policy 5106 is retrieved is given by the ``priority`` property. 5107 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to 5108 refer to /etc/crypto-policies/back-ends/gnutls.config. 5109 5110 .. parsed-literal:: 5111 5112 # |qemu_system| \\ 5113 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\ 5114 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0 5115 5116 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]`` 5117 Interval t can't be 0, this filter batches the packet delivery: 5118 all packets arriving in a given interval on netdev netdevid are 5119 delayed until the end of the interval. Interval is in 5120 microseconds. ``status`` is optional that indicate whether the 5121 netfilter is on (enabled) or off (disabled), the default status 5122 for netfilter will be 'on'. 5123 5124 queue all\|rx\|tx is an option that can be applied to any 5125 netfilter. 5126 5127 ``all``: the filter is attached both to the receive and the 5128 transmit queue of the netdev (default). 5129 5130 ``rx``: the filter is attached to the receive queue of the 5131 netdev, where it will receive packets sent to the netdev. 5132 5133 ``tx``: the filter is attached to the transmit queue of the 5134 netdev, where it will receive packets sent by the netdev. 5135 5136 position head\|tail\|id=<id> is an option to specify where the 5137 filter should be inserted in the filter list. It can be applied 5138 to any netfilter. 5139 5140 ``head``: the filter is inserted at the head of the filter list, 5141 before any existing filters. 5142 5143 ``tail``: the filter is inserted at the tail of the filter list, 5144 behind any existing filters (default). 5145 5146 ``id=<id>``: the filter is inserted before or behind the filter 5147 specified by <id>, see the insert option below. 5148 5149 insert behind\|before is an option to specify where to insert 5150 the new filter relative to the one specified with 5151 position=id=<id>. It can be applied to any netfilter. 5152 5153 ``before``: insert before the specified filter. 5154 5155 ``behind``: insert behind the specified filter (default). 5156 5157 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5158 filter-mirror on netdev netdevid,mirror net packet to 5159 chardevchardevid, if it has the vnet\_hdr\_support flag, 5160 filter-mirror will mirror packet with vnet\_hdr\_len. 5161 5162 ``-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]`` 5163 filter-redirector on netdev netdevid,redirect filter's net 5164 packet to chardev chardevid,and redirect indev's packet to 5165 filter.if it has the vnet\_hdr\_support flag, filter-redirector 5166 will redirect packet with vnet\_hdr\_len. Create a 5167 filter-redirector we need to differ outdev id from indev id, id 5168 can not be the same. we can just use indev or outdev, but at 5169 least one of indev or outdev need to be specified. 5170 5171 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5172 Filter-rewriter is a part of COLO project.It will rewrite tcp 5173 packet to secondary from primary to keep secondary tcp 5174 connection,and rewrite tcp packet to primary from secondary make 5175 tcp packet can be handled by client.if it has the 5176 vnet\_hdr\_support flag, we can parse packet with vnet header. 5177 5178 usage: colo secondary: -object 5179 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object 5180 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object 5181 filter-rewriter,id=rew0,netdev=hn0,queue=all 5182 5183 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]`` 5184 Dump the network traffic on netdev dev to the file specified by 5185 filename. At most len bytes (64k by default) per packet are 5186 stored. The file format is libpcap, so it can be analyzed with 5187 tools such as tcpdump or Wireshark. 5188 5189 ``-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}]`` 5190 Colo-compare gets packet from primary\_in chardevid and 5191 secondary\_in, then compare whether the payload of primary packet 5192 and secondary packet are the same. If same, it will output 5193 primary packet to out\_dev, else it will notify COLO-framework to do 5194 checkpoint and send primary packet to out\_dev. In order to 5195 improve efficiency, we need to put the task of comparison in 5196 another iothread. If it has the vnet\_hdr\_support flag, 5197 colo compare will send/recv packet with vnet\_hdr\_len. 5198 The compare\_timeout=@var{ms} determines the maximum time of the 5199 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms} 5200 is to set the period of scanning expired primary node network packets. 5201 The max\_queue\_size=@var{size} is to set the max compare queue 5202 size depend on user environment. 5203 If user want to use Xen COLO, need to add the notify\_dev to 5204 notify Xen colo-frame to do checkpoint. 5205 5206 COLO-compare must be used with the help of filter-mirror, 5207 filter-redirector and filter-rewriter. 5208 5209 :: 5210 5211 KVM COLO 5212 5213 primary: 5214 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 5215 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5216 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5217 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5218 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5219 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5220 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5221 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5222 -object iothread,id=iothread1 5223 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5224 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5225 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5226 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1 5227 5228 secondary: 5229 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 5230 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5231 -chardev socket,id=red0,host=3.3.3.3,port=9003 5232 -chardev socket,id=red1,host=3.3.3.3,port=9004 5233 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5234 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5235 5236 5237 Xen COLO 5238 5239 primary: 5240 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 5241 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5242 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5243 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5244 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5245 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5246 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5247 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5248 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off 5249 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5250 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5251 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5252 -object iothread,id=iothread1 5253 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1 5254 5255 secondary: 5256 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 5257 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5258 -chardev socket,id=red0,host=3.3.3.3,port=9003 5259 -chardev socket,id=red1,host=3.3.3.3,port=9004 5260 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5261 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5262 5263 If you want to know the detail of above command line, you can 5264 read the colo-compare git log. 5265 5266 ``-object cryptodev-backend-builtin,id=id[,queues=queues]`` 5267 Creates a cryptodev backend which executes crypto opreation from 5268 the QEMU cipher APIS. The id parameter is a unique ID that will 5269 be used to reference this cryptodev backend from the 5270 ``virtio-crypto`` device. The queues parameter is optional, 5271 which specify the queue number of cryptodev backend, the default 5272 of queues is 1. 5273 5274 .. parsed-literal:: 5275 5276 # |qemu_system| \\ 5277 [...] \\ 5278 -object cryptodev-backend-builtin,id=cryptodev0 \\ 5279 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5280 [...] 5281 5282 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]`` 5283 Creates a vhost-user cryptodev backend, backed by a chardev 5284 chardevid. The id parameter is a unique ID that will be used to 5285 reference this cryptodev backend from the ``virtio-crypto`` 5286 device. The chardev should be a unix domain socket backed one. 5287 The vhost-user uses a specifically defined protocol to pass 5288 vhost ioctl replacement messages to an application on the other 5289 end of the socket. The queues parameter is optional, which 5290 specify the queue number of cryptodev backend for multiqueue 5291 vhost-user, the default of queues is 1. 5292 5293 .. parsed-literal:: 5294 5295 # |qemu_system| \\ 5296 [...] \\ 5297 -chardev socket,id=chardev0,path=/path/to/socket \\ 5298 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\ 5299 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5300 [...] 5301 5302 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]`` 5303 \ 5304 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]`` 5305 Defines a secret to store a password, encryption key, or some 5306 other sensitive data. The sensitive data can either be passed 5307 directly via the data parameter, or indirectly via the file 5308 parameter. Using the data parameter is insecure unless the 5309 sensitive data is encrypted. 5310 5311 The sensitive data can be provided in raw format (the default), 5312 or base64. When encoded as JSON, the raw format only supports 5313 valid UTF-8 characters, so base64 is recommended for sending 5314 binary data. QEMU will convert from which ever format is 5315 provided to the format it needs internally. eg, an RBD password 5316 can be provided in raw format, even though it will be base64 5317 encoded when passed onto the RBD sever. 5318 5319 For added protection, it is possible to encrypt the data 5320 associated with a secret using the AES-256-CBC cipher. Use of 5321 encryption is indicated by providing the keyid and iv 5322 parameters. The keyid parameter provides the ID of a previously 5323 defined secret that contains the AES-256 decryption key. This 5324 key should be 32-bytes long and be base64 encoded. The iv 5325 parameter provides the random initialization vector used for 5326 encryption of this particular secret and should be a base64 5327 encrypted string of the 16-byte IV. 5328 5329 The simplest (insecure) usage is to provide the secret inline 5330 5331 .. parsed-literal:: 5332 5333 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw 5334 5335 The simplest secure usage is to provide the secret via a file 5336 5337 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object 5338 secret,id=sec0,file=mypasswd.txt,format=raw 5339 5340 For greater security, AES-256-CBC should be used. To illustrate 5341 usage, consider the openssl command line tool which can encrypt 5342 the data. Note that when encrypting, the plaintext must be 5343 padded to the cipher block size (32 bytes) using the standard 5344 PKCS#5/6 compatible padding algorithm. 5345 5346 First a master key needs to be created in base64 encoding: 5347 5348 :: 5349 5350 # openssl rand -base64 32 > key.b64 5351 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"') 5352 5353 Each secret to be encrypted needs to have a random 5354 initialization vector generated. These do not need to be kept 5355 secret 5356 5357 :: 5358 5359 # openssl rand -base64 16 > iv.b64 5360 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"') 5361 5362 The secret to be defined can now be encrypted, in this case 5363 we're telling openssl to base64 encode the result, but it could 5364 be left as raw bytes if desired. 5365 5366 :: 5367 5368 # SECRET=$(printf "letmein" | 5369 openssl enc -aes-256-cbc -a -K $KEY -iv $IV) 5370 5371 When launching QEMU, create a master secret pointing to 5372 ``key.b64`` and specify that to be used to decrypt the user 5373 password. Pass the contents of ``iv.b64`` to the second secret 5374 5375 .. parsed-literal:: 5376 5377 # |qemu_system| \\ 5378 -object secret,id=secmaster0,format=base64,file=key.b64 \\ 5379 -object secret,id=sec0,keyid=secmaster0,format=base64,\\ 5380 data=$SECRET,iv=$(<iv.b64) 5381 5382 ``-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]`` 5383 Create a Secure Encrypted Virtualization (SEV) guest object, 5384 which can be used to provide the guest memory encryption support 5385 on AMD processors. 5386 5387 When memory encryption is enabled, one of the physical address 5388 bit (aka the C-bit) is utilized to mark if a memory page is 5389 protected. The ``cbitpos`` is used to provide the C-bit 5390 position. The C-bit position is Host family dependent hence user 5391 must provide this value. On EPYC, the value should be 47. 5392 5393 When memory encryption is enabled, we loose certain bits in 5394 physical address space. The ``reduced-phys-bits`` is used to 5395 provide the number of bits we loose in physical address space. 5396 Similar to C-bit, the value is Host family dependent. On EPYC, 5397 the value should be 5. 5398 5399 The ``sev-device`` provides the device file to use for 5400 communicating with the SEV firmware running inside AMD Secure 5401 Processor. The default device is '/dev/sev'. If hardware 5402 supports memory encryption then /dev/sev devices are created by 5403 CCP driver. 5404 5405 The ``policy`` provides the guest policy to be enforced by the 5406 SEV firmware and restrict what configuration and operational 5407 commands can be performed on this guest by the hypervisor. The 5408 policy should be provided by the guest owner and is bound to the 5409 guest and cannot be changed throughout the lifetime of the 5410 guest. The default is 0. 5411 5412 If guest ``policy`` allows sharing the key with another SEV 5413 guest then ``handle`` can be use to provide handle of the guest 5414 from which to share the key. 5415 5416 The ``dh-cert-file`` and ``session-file`` provides the guest 5417 owner's Public Diffie-Hillman key defined in SEV spec. The PDH 5418 and session parameters are used for establishing a cryptographic 5419 session with the guest owner to negotiate keys used for 5420 attestation. The file must be encoded in base64. 5421 5422 The ``kernel-hashes`` adds the hashes of given kernel/initrd/ 5423 cmdline to a designated guest firmware page for measured Linux 5424 boot with -kernel. The default is off. (Since 6.2) 5425 5426 e.g to launch a SEV guest 5427 5428 .. parsed-literal:: 5429 5430 # |qemu_system_x86| \\ 5431 ...... \\ 5432 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \\ 5433 -machine ...,memory-encryption=sev0 \\ 5434 ..... 5435 5436 ``-object authz-simple,id=id,identity=string`` 5437 Create an authorization object that will control access to 5438 network services. 5439 5440 The ``identity`` parameter is identifies the user and its format 5441 depends on the network service that authorization object is 5442 associated with. For authorizing based on TLS x509 certificates, 5443 the identity must be the x509 distinguished name. Note that care 5444 must be taken to escape any commas in the distinguished name. 5445 5446 An example authorization object to validate a x509 distinguished 5447 name would look like: 5448 5449 .. parsed-literal:: 5450 5451 # |qemu_system| \\ 5452 ... \\ 5453 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\ 5454 ... 5455 5456 Note the use of quotes due to the x509 distinguished name 5457 containing whitespace, and escaping of ','. 5458 5459 ``-object authz-listfile,id=id,filename=path,refresh=on|off`` 5460 Create an authorization object that will control access to 5461 network services. 5462 5463 The ``filename`` parameter is the fully qualified path to a file 5464 containing the access control list rules in JSON format. 5465 5466 An example set of rules that match against SASL usernames might 5467 look like: 5468 5469 :: 5470 5471 { 5472 "rules": [ 5473 { "match": "fred", "policy": "allow", "format": "exact" }, 5474 { "match": "bob", "policy": "allow", "format": "exact" }, 5475 { "match": "danb", "policy": "deny", "format": "glob" }, 5476 { "match": "dan*", "policy": "allow", "format": "exact" }, 5477 ], 5478 "policy": "deny" 5479 } 5480 5481 When checking access the object will iterate over all the rules 5482 and the first rule to match will have its ``policy`` value 5483 returned as the result. If no rules match, then the default 5484 ``policy`` value is returned. 5485 5486 The rules can either be an exact string match, or they can use 5487 the simple UNIX glob pattern matching to allow wildcards to be 5488 used. 5489 5490 If ``refresh`` is set to true the file will be monitored and 5491 automatically reloaded whenever its content changes. 5492 5493 As with the ``authz-simple`` object, the format of the identity 5494 strings being matched depends on the network service, but is 5495 usually a TLS x509 distinguished name, or a SASL username. 5496 5497 An example authorization object to validate a SASL username 5498 would look like: 5499 5500 .. parsed-literal:: 5501 5502 # |qemu_system| \\ 5503 ... \\ 5504 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\ 5505 ... 5506 5507 ``-object authz-pam,id=id,service=string`` 5508 Create an authorization object that will control access to 5509 network services. 5510 5511 The ``service`` parameter provides the name of a PAM service to 5512 use for authorization. It requires that a file 5513 ``/etc/pam.d/service`` exist to provide the configuration for 5514 the ``account`` subsystem. 5515 5516 An example authorization object to validate a TLS x509 5517 distinguished name would look like: 5518 5519 .. parsed-literal:: 5520 5521 # |qemu_system| \\ 5522 ... \\ 5523 -object authz-pam,id=auth0,service=qemu-vnc \\ 5524 ... 5525 5526 There would then be a corresponding config file for PAM at 5527 ``/etc/pam.d/qemu-vnc`` that contains: 5528 5529 :: 5530 5531 account requisite pam_listfile.so item=user sense=allow \ 5532 file=/etc/qemu/vnc.allow 5533 5534 Finally the ``/etc/qemu/vnc.allow`` file would contain the list 5535 of x509 distingished names that are permitted access 5536 5537 :: 5538 5539 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB 5540 5541 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch`` 5542 Creates a dedicated event loop thread that devices can be 5543 assigned to. This is known as an IOThread. By default device 5544 emulation happens in vCPU threads or the main event loop thread. 5545 This can become a scalability bottleneck. IOThreads allow device 5546 emulation and I/O to run on other host CPUs. 5547 5548 The ``id`` parameter is a unique ID that will be used to 5549 reference this IOThread from ``-device ...,iothread=id``. 5550 Multiple devices can be assigned to an IOThread. Note that not 5551 all devices support an ``iothread`` parameter. 5552 5553 The ``query-iothreads`` QMP command lists IOThreads and reports 5554 their thread IDs so that the user can configure host CPU 5555 pinning/affinity. 5556 5557 IOThreads use an adaptive polling algorithm to reduce event loop 5558 latency. Instead of entering a blocking system call to monitor 5559 file descriptors and then pay the cost of being woken up when an 5560 event occurs, the polling algorithm spins waiting for events for 5561 a short time. The algorithm's default parameters are suitable 5562 for many cases but can be adjusted based on knowledge of the 5563 workload and/or host device latency. 5564 5565 The ``poll-max-ns`` parameter is the maximum number of 5566 nanoseconds to busy wait for events. Polling can be disabled by 5567 setting this value to 0. 5568 5569 The ``poll-grow`` parameter is the multiplier used to increase 5570 the polling time when the algorithm detects it is missing events 5571 due to not polling long enough. 5572 5573 The ``poll-shrink`` parameter is the divisor used to decrease 5574 the polling time when the algorithm detects it is spending too 5575 long polling without encountering events. 5576 5577 The ``aio-max-batch`` parameter is the maximum number of requests 5578 in a batch for the AIO engine, 0 means that the engine will use 5579 its default. 5580 5581 The IOThread parameters can be modified at run-time using the 5582 ``qom-set`` command (where ``iothread1`` is the IOThread's 5583 ``id``): 5584 5585 :: 5586 5587 (qemu) qom-set /objects/iothread1 poll-max-ns 100000 5588ERST 5589 5590 5591HXCOMM This is the last statement. Insert new options before this line! 5592 5593#undef DEF 5594#undef DEFHEADING 5595#undef ARCHHEADING 5596
