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 QEMU_ARCH_ALL) 42SRST 43``-machine [type=]name[,prop=value[,...]]`` 44 Select the emulated machine by name. Use ``-machine help`` to list 45 available machines. 46 47 For architectures which aim to support live migration compatibility 48 across releases, each release will introduce a new versioned machine 49 type. For example, the 2.8.0 release introduced machine types 50 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures. 51 52 To allow live migration of guests from QEMU version 2.8.0, to QEMU 53 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8" 54 and "pc-q35-2.8" machines too. To allow users live migrating VMs to 55 skip multiple intermediate releases when upgrading, new releases of 56 QEMU will support machine types from many previous versions. 57 58 Supported machine properties are: 59 60 ``accel=accels1[:accels2[:...]]`` 61 This is used to enable an accelerator. Depending on the target 62 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available. 63 By default, tcg is used. If there is more than one accelerator 64 specified, the next one is used if the previous one fails to 65 initialize. 66 67 ``vmport=on|off|auto`` 68 Enables emulation of VMWare IO port, for vmmouse etc. auto says 69 to select the value based on accel. For accel=xen the default is 70 off otherwise the default is on. 71 72 ``dump-guest-core=on|off`` 73 Include guest memory in a core dump. The default is on. 74 75 ``mem-merge=on|off`` 76 Enables or disables memory merge support. This feature, when 77 supported by the host, de-duplicates identical memory pages 78 among VMs instances (enabled by default). 79 80 ``aes-key-wrap=on|off`` 81 Enables or disables AES key wrapping support on s390-ccw hosts. 82 This feature controls whether AES wrapping keys will be created 83 to allow execution of AES cryptographic functions. The default 84 is on. 85 86 ``dea-key-wrap=on|off`` 87 Enables or disables DEA key wrapping support on s390-ccw hosts. 88 This feature controls whether DEA wrapping keys will be created 89 to allow execution of DEA cryptographic functions. The default 90 is on. 91 92 ``nvdimm=on|off`` 93 Enables or disables NVDIMM support. The default is off. 94 95 ``memory-encryption=`` 96 Memory encryption object to use. The default is none. 97 98 ``hmat=on|off`` 99 Enables or disables ACPI Heterogeneous Memory Attribute Table 100 (HMAT) support. The default is off. 101 102 ``memory-backend='id'`` 103 An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options. 104 Allows to use a memory backend as main RAM. 105 106 For example: 107 :: 108 109 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on 110 -machine memory-backend=pc.ram 111 -m 512M 112 113 Migration compatibility note: 114 115 * as backend id one shall use value of 'default-ram-id', advertised by 116 machine type (available via ``query-machines`` QMP command), if migration 117 to/from old QEMU (<5.0) is expected. 118 * for machine types 4.0 and older, user shall 119 use ``x-use-canonical-path-for-ramblock-id=off`` backend option 120 if migration to/from old QEMU (<5.0) is expected. 121 122 For example: 123 :: 124 125 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off 126 -machine memory-backend=pc.ram 127 -m 512M 128 129 ``cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]`` 130 Define a CXL Fixed Memory Window (CFMW). 131 132 Described in the CXL 2.0 ECN: CEDT CFMWS & QTG _DSM. 133 134 They are regions of Host Physical Addresses (HPA) on a system which 135 may be interleaved across one or more CXL host bridges. The system 136 software will assign particular devices into these windows and 137 configure the downstream Host-managed Device Memory (HDM) decoders 138 in root ports, switch ports and devices appropriately to meet the 139 interleave requirements before enabling the memory devices. 140 141 ``targets.X=target`` provides the mapping to CXL host bridges 142 which may be identified by the id provided in the -device entry. 143 Multiple entries are needed to specify all the targets when 144 the fixed memory window represents interleaved memory. X is the 145 target index from 0. 146 147 ``size=size`` sets the size of the CFMW. This must be a multiple of 148 256MiB. The region will be aligned to 256MiB but the location is 149 platform and configuration dependent. 150 151 ``interleave-granularity=granularity`` sets the granularity of 152 interleave. Default 256KiB. Only 256KiB, 512KiB, 1024KiB, 2048KiB 153 4096KiB, 8192KiB and 16384KiB granularities supported. 154 155 Example: 156 157 :: 158 159 -machine cxl-fmw.0.targets.0=cxl.0,cxl-fmw.0.targets.1=cxl.1,cxl-fmw.0.size=128G,cxl-fmw.0.interleave-granularity=512k 160ERST 161 162DEF("M", HAS_ARG, QEMU_OPTION_M, 163 " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n", 164 QEMU_ARCH_ALL) 165 166SRST 167``sgx-epc.0.memdev=@var{memid},sgx-epc.0.node=@var{numaid}`` 168 Define an SGX EPC section. 169ERST 170 171DEF("cpu", HAS_ARG, QEMU_OPTION_cpu, 172 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL) 173SRST 174``-cpu model`` 175 Select CPU model (``-cpu help`` for list and additional feature 176 selection) 177ERST 178 179DEF("accel", HAS_ARG, QEMU_OPTION_accel, 180 "-accel [accel=]accelerator[,prop[=value][,...]]\n" 181 " select accelerator (kvm, xen, hax, hvf, nvmm, whpx or tcg; use 'help' for a list)\n" 182 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n" 183 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n" 184 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n" 185 " one-insn-per-tb=on|off (one guest instruction per TCG translation block)\n" 186 " split-wx=on|off (enable TCG split w^x mapping)\n" 187 " tb-size=n (TCG translation block cache size)\n" 188 " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n" 189 " notify-vmexit=run|internal-error|disable,notify-window=n (enable notify VM exit and set notify window, x86 only)\n" 190 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL) 191SRST 192``-accel name[,prop=value[,...]]`` 193 This is used to enable an accelerator. Depending on the target 194 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available. By 195 default, tcg is used. If there is more than one accelerator 196 specified, the next one is used if the previous one fails to 197 initialize. 198 199 ``igd-passthru=on|off`` 200 When Xen is in use, this option controls whether Intel 201 integrated graphics devices can be passed through to the guest 202 (default=off) 203 204 ``kernel-irqchip=on|off|split`` 205 Controls KVM in-kernel irqchip support. The default is full 206 acceleration of the interrupt controllers. On x86, split irqchip 207 reduces the kernel attack surface, at a performance cost for 208 non-MSI interrupts. Disabling the in-kernel irqchip completely 209 is not recommended except for debugging purposes. 210 211 ``kvm-shadow-mem=size`` 212 Defines the size of the KVM shadow MMU. 213 214 ``one-insn-per-tb=on|off`` 215 Makes the TCG accelerator put only one guest instruction into 216 each translation block. This slows down emulation a lot, but 217 can be useful in some situations, such as when trying to analyse 218 the logs produced by the ``-d`` option. 219 220 ``split-wx=on|off`` 221 Controls the use of split w^x mapping for the TCG code generation 222 buffer. Some operating systems require this to be enabled, and in 223 such a case this will default on. On other operating systems, this 224 will default off, but one may enable this for testing or debugging. 225 226 ``tb-size=n`` 227 Controls the size (in MiB) of the TCG translation block cache. 228 229 ``thread=single|multi`` 230 Controls number of TCG threads. When the TCG is multi-threaded 231 there will be one thread per vCPU therefore taking advantage of 232 additional host cores. The default is to enable multi-threading 233 where both the back-end and front-ends support it and no 234 incompatible TCG features have been enabled (e.g. 235 icount/replay). 236 237 ``dirty-ring-size=n`` 238 When the KVM accelerator is used, it controls the size of the per-vCPU 239 dirty page ring buffer (number of entries for each vCPU). It should 240 be a value that is power of two, and it should be 1024 or bigger (but 241 still less than the maximum value that the kernel supports). 4096 242 could be a good initial value if you have no idea which is the best. 243 Set this value to 0 to disable the feature. By default, this feature 244 is disabled (dirty-ring-size=0). When enabled, KVM will instead 245 record dirty pages in a bitmap. 246 247 ``notify-vmexit=run|internal-error|disable,notify-window=n`` 248 Enables or disables notify VM exit support on x86 host and specify 249 the corresponding notify window to trigger the VM exit if enabled. 250 ``run`` option enables the feature. It does nothing and continue 251 if the exit happens. ``internal-error`` option enables the feature. 252 It raises a internal error. ``disable`` option doesn't enable the feature. 253 This feature can mitigate the CPU stuck issue due to event windows don't 254 open up for a specified of time (i.e. notify-window). 255 Default: notify-vmexit=run,notify-window=0. 256 257ERST 258 259DEF("smp", HAS_ARG, QEMU_OPTION_smp, 260 "-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]\n" 261 " set the number of initial CPUs to 'n' [default=1]\n" 262 " maxcpus= maximum number of total CPUs, including\n" 263 " offline CPUs for hotplug, etc\n" 264 " sockets= number of sockets on the machine board\n" 265 " dies= number of dies in one socket\n" 266 " clusters= number of clusters in one die\n" 267 " cores= number of cores in one cluster\n" 268 " threads= number of threads in one core\n" 269 "Note: Different machines may have different subsets of the CPU topology\n" 270 " parameters supported, so the actual meaning of the supported parameters\n" 271 " will vary accordingly. For example, for a machine type that supports a\n" 272 " three-level CPU hierarchy of sockets/cores/threads, the parameters will\n" 273 " sequentially mean as below:\n" 274 " sockets means the number of sockets on the machine board\n" 275 " cores means the number of cores in one socket\n" 276 " threads means the number of threads in one core\n" 277 " For a particular machine type board, an expected CPU topology hierarchy\n" 278 " can be defined through the supported sub-option. Unsupported parameters\n" 279 " can also be provided in addition to the sub-option, but their values\n" 280 " must be set as 1 in the purpose of correct parsing.\n", 281 QEMU_ARCH_ALL) 282SRST 283``-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]`` 284 Simulate a SMP system with '\ ``n``\ ' CPUs initially present on 285 the machine type board. On boards supporting CPU hotplug, the optional 286 '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be 287 added at runtime. When both parameters are omitted, the maximum number 288 of CPUs will be calculated from the provided topology members and the 289 initial CPU count will match the maximum number. When only one of them 290 is given then the omitted one will be set to its counterpart's value. 291 Both parameters may be specified, but the maximum number of CPUs must 292 be equal to or greater than the initial CPU count. Product of the 293 CPU topology hierarchy must be equal to the maximum number of CPUs. 294 Both parameters are subject to an upper limit that is determined by 295 the specific machine type chosen. 296 297 To control reporting of CPU topology information, values of the topology 298 parameters can be specified. Machines may only support a subset of the 299 parameters and different machines may have different subsets supported 300 which vary depending on capacity of the corresponding CPU targets. So 301 for a particular machine type board, an expected topology hierarchy can 302 be defined through the supported sub-option. Unsupported parameters can 303 also be provided in addition to the sub-option, but their values must be 304 set as 1 in the purpose of correct parsing. 305 306 Either the initial CPU count, or at least one of the topology parameters 307 must be specified. The specified parameters must be greater than zero, 308 explicit configuration like "cpus=0" is not allowed. Values for any 309 omitted parameters will be computed from those which are given. 310 311 For example, the following sub-option defines a CPU topology hierarchy 312 (2 sockets totally on the machine, 2 cores per socket, 2 threads per 313 core) for a machine that only supports sockets/cores/threads. 314 Some members of the option can be omitted but their values will be 315 automatically computed: 316 317 :: 318 319 -smp 8,sockets=2,cores=2,threads=2,maxcpus=8 320 321 The following sub-option defines a CPU topology hierarchy (2 sockets 322 totally on the machine, 2 dies per socket, 2 cores per die, 2 threads 323 per core) for PC machines which support sockets/dies/cores/threads. 324 Some members of the option can be omitted but their values will be 325 automatically computed: 326 327 :: 328 329 -smp 16,sockets=2,dies=2,cores=2,threads=2,maxcpus=16 330 331 The following sub-option defines a CPU topology hierarchy (2 sockets 332 totally on the machine, 2 clusters per socket, 2 cores per cluster, 333 2 threads per core) for ARM virt machines which support sockets/clusters 334 /cores/threads. Some members of the option can be omitted but their values 335 will be automatically computed: 336 337 :: 338 339 -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16 340 341 Historically preference was given to the coarsest topology parameters 342 when computing missing values (ie sockets preferred over cores, which 343 were preferred over threads), however, this behaviour is considered 344 liable to change. Prior to 6.2 the preference was sockets over cores 345 over threads. Since 6.2 the preference is cores over sockets over threads. 346 347 For example, the following option defines a machine board with 2 sockets 348 of 1 core before 6.2 and 1 socket of 2 cores after 6.2: 349 350 :: 351 352 -smp 2 353 354 Note: The cluster topology will only be generated in ACPI and exposed 355 to guest if it's explicitly specified in -smp. 356ERST 357 358DEF("numa", HAS_ARG, QEMU_OPTION_numa, 359 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 360 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 361 "-numa dist,src=source,dst=destination,val=distance\n" 362 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n" 363 "-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" 364 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n", 365 QEMU_ARCH_ALL) 366SRST 367``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 368 \ 369``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 370 \ 371``-numa dist,src=source,dst=destination,val=distance`` 372 \ 373``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]`` 374 \ 375``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=type[,latency=lat][,bandwidth=bw]`` 376 \ 377``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]`` 378 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA 379 distance from a source node to a destination node. Set the ACPI 380 Heterogeneous Memory Attributes for the given nodes. 381 382 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and 383 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a 384 contiguous range of CPU indexes (or a single VCPU if lastcpu is 385 omitted). A non-contiguous set of VCPUs can be represented by 386 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is 387 omitted on all nodes, VCPUs are automatically split between them. 388 389 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a 390 NUMA node: 391 392 :: 393 394 -numa node,cpus=0-2,cpus=5 395 396 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option 397 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to 398 assign CPU objects to a node using topology layout properties of 399 CPU. The set of properties is machine specific, and depends on used 400 machine type/'\ ``smp``\ ' options. It could be queried with 401 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ ' 402 property specifies node to which CPU object will be assigned, it's 403 required for node to be declared with '\ ``node``\ ' option before 404 it's used with '\ ``cpu``\ ' option. 405 406 For example: 407 408 :: 409 410 -M pc \ 411 -smp 1,sockets=2,maxcpus=2 \ 412 -numa node,nodeid=0 -numa node,nodeid=1 \ 413 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 414 415 '\ ``memdev``\ ' option assigns RAM from a given memory backend 416 device to a node. It is recommended to use '\ ``memdev``\ ' option 417 over legacy '\ ``mem``\ ' option. This is because '\ ``memdev``\ ' 418 option provides better performance and more control over the 419 backend's RAM (e.g. '\ ``prealloc``\ ' parameter of 420 '\ ``-memory-backend-ram``\ ' allows memory preallocation). 421 422 For compatibility reasons, legacy '\ ``mem``\ ' option is 423 supported in 5.0 and older machine types. Note that '\ ``mem``\ ' 424 and '\ ``memdev``\ ' are mutually exclusive. If one node uses 425 '\ ``memdev``\ ', the rest nodes have to use '\ ``memdev``\ ' 426 option, and vice versa. 427 428 Users must specify memory for all NUMA nodes by '\ ``memdev``\ ' 429 (or legacy '\ ``mem``\ ' if available). In QEMU 5.2, the support 430 for '\ ``-numa node``\ ' without memory specified was removed. 431 432 '\ ``initiator``\ ' is an additional option that points to an 433 initiator NUMA node that has best performance (the lowest latency or 434 largest bandwidth) to this NUMA node. Note that this option can be 435 set only when the machine property 'hmat' is set to 'on'. 436 437 Following example creates a machine with 2 NUMA nodes, node 0 has 438 CPU. node 1 has only memory, and its initiator is node 0. Note that 439 because node 0 has CPU, by default the initiator of node 0 is itself 440 and must be itself. 441 442 :: 443 444 -machine hmat=on \ 445 -m 2G,slots=2,maxmem=4G \ 446 -object memory-backend-ram,size=1G,id=m0 \ 447 -object memory-backend-ram,size=1G,id=m1 \ 448 -numa node,nodeid=0,memdev=m0 \ 449 -numa node,nodeid=1,memdev=m1,initiator=0 \ 450 -smp 2,sockets=2,maxcpus=2 \ 451 -numa cpu,node-id=0,socket-id=0 \ 452 -numa cpu,node-id=0,socket-id=1 453 454 source and destination are NUMA node IDs. distance is the NUMA 455 distance from source to destination. The distance from a node to 456 itself is always 10. If any pair of nodes is given a distance, then 457 all pairs must be given distances. Although, when distances are only 458 given in one direction for each pair of nodes, then the distances in 459 the opposite directions are assumed to be the same. If, however, an 460 asymmetrical pair of distances is given for even one node pair, then 461 all node pairs must be provided distance values for both directions, 462 even when they are symmetrical. When a node is unreachable from 463 another node, set the pair's distance to 255. 464 465 Note that the -``numa`` option doesn't allocate any of the specified 466 resources, it just assigns existing resources to NUMA nodes. This 467 means that one still has to use the ``-m``, ``-smp`` options to 468 allocate RAM and VCPUs respectively. 469 470 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth 471 Information between initiator and target NUMA nodes in ACPI 472 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can 473 create memory requests, usually it has one or more processors. 474 Target NUMA node contains addressable memory. 475 476 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is 477 the memory hierarchy of the target NUMA node: if hierarchy is 478 'memory', the structure represents the memory performance; if 479 hierarchy is 'first-level\|second-level\|third-level', this 480 structure represents aggregated performance of memory side caches 481 for each domain. type of 'data-type' is type of data represented by 482 this structure instance: if 'hierarchy' is 'memory', 'data-type' is 483 'access\|read\|write' latency or 'access\|read\|write' bandwidth of 484 the target memory; if 'hierarchy' is 485 'first-level\|second-level\|third-level', 'data-type' is 486 'access\|read\|write' hit latency or 'access\|read\|write' hit 487 bandwidth of the target memory side cache. 488 489 lat is latency value in nanoseconds. bw is bandwidth value, the 490 possible value and units are NUM[M\|G\|T], mean that the bandwidth 491 value are NUM byte per second (or MB/s, GB/s or TB/s depending on 492 used suffix). Note that if latency or bandwidth value is 0, means 493 the corresponding latency or bandwidth information is not provided. 494 495 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory 496 belongs. size is the size of memory side cache in bytes. level is 497 the cache level described in this structure, note that the cache 498 level 0 should not be used with '\ ``hmat-cache``\ ' option. 499 associativity is the cache associativity, the possible value is 500 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy 501 is the write policy. line is the cache Line size in bytes. 502 503 For example, the following options describe 2 NUMA nodes. Node 0 has 504 2 cpus and a ram, node 1 has only a ram. The processors in node 0 505 access memory in node 0 with access-latency 5 nanoseconds, 506 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access 507 memory in NUMA node 1 with access-latency 10 nanoseconds, 508 access-bandwidth is 100 MB/s. And for memory side cache information, 509 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB, 510 policy is write-back, the cache Line size is 8 bytes: 511 512 :: 513 514 -machine hmat=on \ 515 -m 2G \ 516 -object memory-backend-ram,size=1G,id=m0 \ 517 -object memory-backend-ram,size=1G,id=m1 \ 518 -smp 2,sockets=2,maxcpus=2 \ 519 -numa node,nodeid=0,memdev=m0 \ 520 -numa node,nodeid=1,memdev=m1,initiator=0 \ 521 -numa cpu,node-id=0,socket-id=0 \ 522 -numa cpu,node-id=0,socket-id=1 \ 523 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \ 524 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \ 525 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \ 526 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \ 527 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \ 528 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8 529ERST 530 531DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd, 532 "-add-fd fd=fd,set=set[,opaque=opaque]\n" 533 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL) 534SRST 535``-add-fd fd=fd,set=set[,opaque=opaque]`` 536 Add a file descriptor to an fd set. Valid options are: 537 538 ``fd=fd`` 539 This option defines the file descriptor of which a duplicate is 540 added to fd set. The file descriptor cannot be stdin, stdout, or 541 stderr. 542 543 ``set=set`` 544 This option defines the ID of the fd set to add the file 545 descriptor to. 546 547 ``opaque=opaque`` 548 This option defines a free-form string that can be used to 549 describe fd. 550 551 You can open an image using pre-opened file descriptors from an fd 552 set: 553 554 .. parsed-literal:: 555 556 |qemu_system| \\ 557 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 558 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 559 -drive file=/dev/fdset/2,index=0,media=disk 560ERST 561 562DEF("set", HAS_ARG, QEMU_OPTION_set, 563 "-set group.id.arg=value\n" 564 " set <arg> parameter for item <id> of type <group>\n" 565 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL) 566SRST 567``-set group.id.arg=value`` 568 Set parameter arg for item id of type group 569ERST 570 571DEF("global", HAS_ARG, QEMU_OPTION_global, 572 "-global driver.property=value\n" 573 "-global driver=driver,property=property,value=value\n" 574 " set a global default for a driver property\n", 575 QEMU_ARCH_ALL) 576SRST 577``-global driver.prop=value`` 578 \ 579``-global driver=driver,property=property,value=value`` 580 Set default value of driver's property prop to value, e.g.: 581 582 .. parsed-literal:: 583 584 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img 585 586 In particular, you can use this to set driver properties for devices 587 which are created automatically by the machine model. To create a 588 device which is not created automatically and set properties on it, 589 use -``device``. 590 591 -global driver.prop=value is shorthand for -global 592 driver=driver,property=prop,value=value. The longhand syntax works 593 even when driver contains a dot. 594ERST 595 596DEF("boot", HAS_ARG, QEMU_OPTION_boot, 597 "-boot [order=drives][,once=drives][,menu=on|off]\n" 598 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n" 599 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n" 600 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n" 601 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n" 602 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n", 603 QEMU_ARCH_ALL) 604SRST 605``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]`` 606 Specify boot order drives as a string of drive letters. Valid drive 607 letters depend on the target architecture. The x86 PC uses: a, b 608 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p 609 (Etherboot from network adapter 1-4), hard disk boot is the default. 610 To apply a particular boot order only on the first startup, specify 611 it via ``once``. Note that the ``order`` or ``once`` parameter 612 should not be used together with the ``bootindex`` property of 613 devices, since the firmware implementations normally do not support 614 both at the same time. 615 616 Interactive boot menus/prompts can be enabled via ``menu=on`` as far 617 as firmware/BIOS supports them. The default is non-interactive boot. 618 619 A splash picture could be passed to bios, enabling user to show it 620 as logo, when option splash=sp\_name is given and menu=on, If 621 firmware/BIOS supports them. Currently Seabios for X86 system 622 support it. limitation: The splash file could be a jpeg file or a 623 BMP file in 24 BPP format(true color). The resolution should be 624 supported by the SVGA mode, so the recommended is 320x240, 640x480, 625 800x640. 626 627 A timeout could be passed to bios, guest will pause for rb\_timeout 628 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will 629 not reboot, qemu passes '-1' to bios by default. Currently Seabios 630 for X86 system support it. 631 632 Do strict boot via ``strict=on`` as far as firmware/BIOS supports 633 it. This only effects when boot priority is changed by bootindex 634 options. The default is non-strict boot. 635 636 .. parsed-literal:: 637 638 # try to boot from network first, then from hard disk 639 |qemu_system_x86| -boot order=nc 640 # boot from CD-ROM first, switch back to default order after reboot 641 |qemu_system_x86| -boot once=d 642 # boot with a splash picture for 5 seconds. 643 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000 644 645 Note: The legacy format '-boot drives' is still supported but its 646 use is discouraged as it may be removed from future versions. 647ERST 648 649DEF("m", HAS_ARG, QEMU_OPTION_m, 650 "-m [size=]megs[,slots=n,maxmem=size]\n" 651 " configure guest RAM\n" 652 " size: initial amount of guest memory\n" 653 " slots: number of hotplug slots (default: none)\n" 654 " maxmem: maximum amount of guest memory (default: none)\n" 655 "NOTE: Some architectures might enforce a specific granularity\n", 656 QEMU_ARCH_ALL) 657SRST 658``-m [size=]megs[,slots=n,maxmem=size]`` 659 Sets guest startup RAM size to megs megabytes. Default is 128 MiB. 660 Optionally, a suffix of "M" or "G" can be used to signify a value in 661 megabytes or gigabytes respectively. Optional pair slots, maxmem 662 could be used to set amount of hotpluggable memory slots and maximum 663 amount of memory. Note that maxmem must be aligned to the page size. 664 665 For example, the following command-line sets the guest startup RAM 666 size to 1GB, creates 3 slots to hotplug additional memory and sets 667 the maximum memory the guest can reach to 4GB: 668 669 .. parsed-literal:: 670 671 |qemu_system| -m 1G,slots=3,maxmem=4G 672 673 If slots and maxmem are not specified, memory hotplug won't be 674 enabled and the guest startup RAM will never increase. 675ERST 676 677DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath, 678 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL) 679SRST 680``-mem-path path`` 681 Allocate guest RAM from a temporarily created file in path. 682ERST 683 684DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc, 685 "-mem-prealloc preallocate guest memory (use with -mem-path)\n", 686 QEMU_ARCH_ALL) 687SRST 688``-mem-prealloc`` 689 Preallocate memory when using -mem-path. 690ERST 691 692DEF("k", HAS_ARG, QEMU_OPTION_k, 693 "-k language use keyboard layout (for example 'fr' for French)\n", 694 QEMU_ARCH_ALL) 695SRST 696``-k language`` 697 Use keyboard layout language (for example ``fr`` for French). This 698 option is only needed where it is not easy to get raw PC keycodes 699 (e.g. on Macs, with some X11 servers or with a VNC or curses 700 display). You don't normally need to use it on PC/Linux or 701 PC/Windows hosts. 702 703 The available layouts are: 704 705 :: 706 707 ar de-ch es fo fr-ca hu ja mk no pt-br sv 708 da en-gb et fr fr-ch is lt nl pl ru th 709 de en-us fi fr-be hr it lv nl-be pt sl tr 710 711 The default is ``en-us``. 712ERST 713 714 715HXCOMM Deprecated by -audiodev 716DEF("audio-help", 0, QEMU_OPTION_audio_help, 717 "-audio-help show -audiodev equivalent of the currently specified audio settings\n", 718 QEMU_ARCH_ALL) 719SRST 720``-audio-help`` 721 Will show the -audiodev equivalent of the currently specified 722 (deprecated) environment variables. 723ERST 724 725DEF("audio", HAS_ARG, QEMU_OPTION_audio, 726 "-audio [driver=]driver,model=value[,prop[=value][,...]]\n" 727 " specifies the audio backend and device to use;\n" 728 " apart from 'model', options are the same as for -audiodev.\n" 729 " use '-audio model=help' to show possible devices.\n", 730 QEMU_ARCH_ALL) 731SRST 732``-audio [driver=]driver,model=value[,prop[=value][,...]]`` 733 This option is a shortcut for configuring both the guest audio 734 hardware and the host audio backend in one go. 735 The driver option is the same as with the corresponding ``-audiodev`` option below. 736 The guest hardware model can be set with ``model=modelname``. 737 738 Use ``driver=help`` to list the available drivers, 739 and ``model=help`` to list the available device types. 740 741 The following two example do exactly the same, to show how ``-audio`` 742 can be used to shorten the command line length: 743 744 .. parsed-literal:: 745 746 |qemu_system| -audiodev pa,id=pa -device sb16,audiodev=pa 747 |qemu_system| -audio pa,model=sb16 748ERST 749 750DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, 751 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n" 752 " specifies the audio backend to use\n" 753 " Use ``-audiodev help`` to list the available drivers\n" 754 " id= identifier of the backend\n" 755 " timer-period= timer period in microseconds\n" 756 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n" 757 " in|out.fixed-settings= use fixed settings for host audio\n" 758 " in|out.frequency= frequency to use with fixed settings\n" 759 " in|out.channels= number of channels to use with fixed settings\n" 760 " in|out.format= sample format to use with fixed settings\n" 761 " valid values: s8, s16, s32, u8, u16, u32, f32\n" 762 " in|out.voices= number of voices to use\n" 763 " in|out.buffer-length= length of buffer in microseconds\n" 764 "-audiodev none,id=id,[,prop[=value][,...]]\n" 765 " dummy driver that discards all output\n" 766#ifdef CONFIG_AUDIO_ALSA 767 "-audiodev alsa,id=id[,prop[=value][,...]]\n" 768 " in|out.dev= name of the audio device to use\n" 769 " in|out.period-length= length of period in microseconds\n" 770 " in|out.try-poll= attempt to use poll mode\n" 771 " threshold= threshold (in microseconds) when playback starts\n" 772#endif 773#ifdef CONFIG_AUDIO_COREAUDIO 774 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n" 775 " in|out.buffer-count= number of buffers\n" 776#endif 777#ifdef CONFIG_AUDIO_DSOUND 778 "-audiodev dsound,id=id[,prop[=value][,...]]\n" 779 " latency= add extra latency to playback in microseconds\n" 780#endif 781#ifdef CONFIG_AUDIO_OSS 782 "-audiodev oss,id=id[,prop[=value][,...]]\n" 783 " in|out.dev= path of the audio device to use\n" 784 " in|out.buffer-count= number of buffers\n" 785 " in|out.try-poll= attempt to use poll mode\n" 786 " try-mmap= try using memory mapped access\n" 787 " exclusive= open device in exclusive mode\n" 788 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n" 789#endif 790#ifdef CONFIG_AUDIO_PA 791 "-audiodev pa,id=id[,prop[=value][,...]]\n" 792 " server= PulseAudio server address\n" 793 " in|out.name= source/sink device name\n" 794 " in|out.latency= desired latency in microseconds\n" 795#endif 796#ifdef CONFIG_AUDIO_SDL 797 "-audiodev sdl,id=id[,prop[=value][,...]]\n" 798 " in|out.buffer-count= number of buffers\n" 799#endif 800#ifdef CONFIG_AUDIO_SNDIO 801 "-audiodev sndio,id=id[,prop[=value][,...]]\n" 802#endif 803#ifdef CONFIG_SPICE 804 "-audiodev spice,id=id[,prop[=value][,...]]\n" 805#endif 806#ifdef CONFIG_DBUS_DISPLAY 807 "-audiodev dbus,id=id[,prop[=value][,...]]\n" 808#endif 809 "-audiodev wav,id=id[,prop[=value][,...]]\n" 810 " path= path of wav file to record\n", 811 QEMU_ARCH_ALL) 812SRST 813``-audiodev [driver=]driver,id=id[,prop[=value][,...]]`` 814 Adds a new audio backend driver identified by id. There are global 815 and driver specific properties. Some values can be set differently 816 for input and output, they're marked with ``in|out.``. You can set 817 the input's property with ``in.prop`` and the output's property with 818 ``out.prop``. For example: 819 820 :: 821 822 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000 823 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified 824 825 NOTE: parameter validation is known to be incomplete, in many cases 826 specifying an invalid option causes QEMU to print an error message 827 and continue emulation without sound. 828 829 Valid global options are: 830 831 ``id=identifier`` 832 Identifies the audio backend. 833 834 ``timer-period=period`` 835 Sets the timer period used by the audio subsystem in 836 microseconds. Default is 10000 (10 ms). 837 838 ``in|out.mixing-engine=on|off`` 839 Use QEMU's mixing engine to mix all streams inside QEMU and 840 convert audio formats when not supported by the backend. When 841 off, fixed-settings must be off too. Note that disabling this 842 option means that the selected backend must support multiple 843 streams and the audio formats used by the virtual cards, 844 otherwise you'll get no sound. It's not recommended to disable 845 this option unless you want to use 5.1 or 7.1 audio, as mixing 846 engine only supports mono and stereo audio. Default is on. 847 848 ``in|out.fixed-settings=on|off`` 849 Use fixed settings for host audio. When off, it will change 850 based on how the guest opens the sound card. In this case you 851 must not specify frequency, channels or format. Default is on. 852 853 ``in|out.frequency=frequency`` 854 Specify the frequency to use when using fixed-settings. Default 855 is 44100Hz. 856 857 ``in|out.channels=channels`` 858 Specify the number of channels to use when using fixed-settings. 859 Default is 2 (stereo). 860 861 ``in|out.format=format`` 862 Specify the sample format to use when using fixed-settings. 863 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``, 864 ``u32``, ``f32``. Default is ``s16``. 865 866 ``in|out.voices=voices`` 867 Specify the number of voices to use. Default is 1. 868 869 ``in|out.buffer-length=usecs`` 870 Sets the size of the buffer in microseconds. 871 872``-audiodev none,id=id[,prop[=value][,...]]`` 873 Creates a dummy backend that discards all outputs. This backend has 874 no backend specific properties. 875 876``-audiodev alsa,id=id[,prop[=value][,...]]`` 877 Creates backend using the ALSA. This backend is only available on 878 Linux. 879 880 ALSA specific options are: 881 882 ``in|out.dev=device`` 883 Specify the ALSA device to use for input and/or output. Default 884 is ``default``. 885 886 ``in|out.period-length=usecs`` 887 Sets the period length in microseconds. 888 889 ``in|out.try-poll=on|off`` 890 Attempt to use poll mode with the device. Default is on. 891 892 ``threshold=threshold`` 893 Threshold (in microseconds) when playback starts. Default is 0. 894 895``-audiodev coreaudio,id=id[,prop[=value][,...]]`` 896 Creates a backend using Apple's Core Audio. This backend is only 897 available on Mac OS and only supports playback. 898 899 Core Audio specific options are: 900 901 ``in|out.buffer-count=count`` 902 Sets the count of the buffers. 903 904``-audiodev dsound,id=id[,prop[=value][,...]]`` 905 Creates a backend using Microsoft's DirectSound. This backend is 906 only available on Windows and only supports playback. 907 908 DirectSound specific options are: 909 910 ``latency=usecs`` 911 Add extra usecs microseconds latency to playback. Default is 912 10000 (10 ms). 913 914``-audiodev oss,id=id[,prop[=value][,...]]`` 915 Creates a backend using OSS. This backend is available on most 916 Unix-like systems. 917 918 OSS specific options are: 919 920 ``in|out.dev=device`` 921 Specify the file name of the OSS device to use. Default is 922 ``/dev/dsp``. 923 924 ``in|out.buffer-count=count`` 925 Sets the count of the buffers. 926 927 ``in|out.try-poll=on|of`` 928 Attempt to use poll mode with the device. Default is on. 929 930 ``try-mmap=on|off`` 931 Try using memory mapped device access. Default is off. 932 933 ``exclusive=on|off`` 934 Open the device in exclusive mode (vmix won't work in this 935 case). Default is off. 936 937 ``dsp-policy=policy`` 938 Sets the timing policy (between 0 and 10, where smaller number 939 means smaller latency but higher CPU usage). Use -1 to use 940 buffer sizes specified by ``buffer`` and ``buffer-count``. This 941 option is ignored if you do not have OSS 4. Default is 5. 942 943``-audiodev pa,id=id[,prop[=value][,...]]`` 944 Creates a backend using PulseAudio. This backend is available on 945 most systems. 946 947 PulseAudio specific options are: 948 949 ``server=server`` 950 Sets the PulseAudio server to connect to. 951 952 ``in|out.name=sink`` 953 Use the specified source/sink for recording/playback. 954 955 ``in|out.latency=usecs`` 956 Desired latency in microseconds. The PulseAudio server will try 957 to honor this value but actual latencies may be lower or higher. 958 959``-audiodev sdl,id=id[,prop[=value][,...]]`` 960 Creates a backend using SDL. This backend is available on most 961 systems, but you should use your platform's native backend if 962 possible. 963 964 SDL specific options are: 965 966 ``in|out.buffer-count=count`` 967 Sets the count of the buffers. 968 969``-audiodev sndio,id=id[,prop[=value][,...]]`` 970 Creates a backend using SNDIO. This backend is available on 971 OpenBSD and most other Unix-like systems. 972 973 Sndio specific options are: 974 975 ``in|out.dev=device`` 976 Specify the sndio device to use for input and/or output. Default 977 is ``default``. 978 979 ``in|out.latency=usecs`` 980 Sets the desired period length in microseconds. 981 982``-audiodev spice,id=id[,prop[=value][,...]]`` 983 Creates a backend that sends audio through SPICE. This backend 984 requires ``-spice`` and automatically selected in that case, so 985 usually you can ignore this option. This backend has no backend 986 specific properties. 987 988``-audiodev wav,id=id[,prop[=value][,...]]`` 989 Creates a backend that writes audio to a WAV file. 990 991 Backend specific options are: 992 993 ``path=path`` 994 Write recorded audio into the specified file. Default is 995 ``qemu.wav``. 996ERST 997 998DEF("device", HAS_ARG, QEMU_OPTION_device, 999 "-device driver[,prop[=value][,...]]\n" 1000 " add device (based on driver)\n" 1001 " prop=value,... sets driver properties\n" 1002 " use '-device help' to print all possible drivers\n" 1003 " use '-device driver,help' to print all possible properties\n", 1004 QEMU_ARCH_ALL) 1005SRST 1006``-device driver[,prop[=value][,...]]`` 1007 Add device driver. prop=value sets driver properties. Valid 1008 properties depend on the driver. To get help on possible drivers and 1009 properties, use ``-device help`` and ``-device driver,help``. 1010 1011 Some drivers are: 1012 1013``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]`` 1014 Add an IPMI BMC. This is a simulation of a hardware management 1015 interface processor that normally sits on a system. It provides a 1016 watchdog and the ability to reset and power control the system. You 1017 need to connect this to an IPMI interface to make it useful 1018 1019 The IPMI slave address to use for the BMC. The default is 0x20. This 1020 address is the BMC's address on the I2C network of management 1021 controllers. If you don't know what this means, it is safe to ignore 1022 it. 1023 1024 ``id=id`` 1025 The BMC id for interfaces to use this device. 1026 1027 ``slave_addr=val`` 1028 Define slave address to use for the BMC. The default is 0x20. 1029 1030 ``sdrfile=file`` 1031 file containing raw Sensor Data Records (SDR) data. The default 1032 is none. 1033 1034 ``fruareasize=val`` 1035 size of a Field Replaceable Unit (FRU) area. The default is 1036 1024. 1037 1038 ``frudatafile=file`` 1039 file containing raw Field Replaceable Unit (FRU) inventory data. 1040 The default is none. 1041 1042 ``guid=uuid`` 1043 value for the GUID for the BMC, in standard UUID format. If this 1044 is set, get "Get GUID" command to the BMC will return it. 1045 Otherwise "Get GUID" will return an error. 1046 1047``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]`` 1048 Add a connection to an external IPMI BMC simulator. Instead of 1049 locally emulating the BMC like the above item, instead connect to an 1050 external entity that provides the IPMI services. 1051 1052 A connection is made to an external BMC simulator. If you do this, 1053 it is strongly recommended that you use the "reconnect=" chardev 1054 option to reconnect to the simulator if the connection is lost. Note 1055 that if this is not used carefully, it can be a security issue, as 1056 the interface has the ability to send resets, NMIs, and power off 1057 the VM. It's best if QEMU makes a connection to an external 1058 simulator running on a secure port on localhost, so neither the 1059 simulator nor QEMU is exposed to any outside network. 1060 1061 See the "lanserv/README.vm" file in the OpenIPMI library for more 1062 details on the external interface. 1063 1064``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]`` 1065 Add a KCS IPMI interface on the ISA bus. This also adds a 1066 corresponding ACPI and SMBIOS entries, if appropriate. 1067 1068 ``bmc=id`` 1069 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern 1070 above. 1071 1072 ``ioport=val`` 1073 Define the I/O address of the interface. The default is 0xca0 1074 for KCS. 1075 1076 ``irq=val`` 1077 Define the interrupt to use. The default is 5. To disable 1078 interrupts, set this to 0. 1079 1080``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]`` 1081 Like the KCS interface, but defines a BT interface. The default port 1082 is 0xe4 and the default interrupt is 5. 1083 1084``-device pci-ipmi-kcs,bmc=id`` 1085 Add a KCS IPMI interface on the PCI bus. 1086 1087 ``bmc=id`` 1088 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above. 1089 1090``-device pci-ipmi-bt,bmc=id`` 1091 Like the KCS interface, but defines a BT interface on the PCI bus. 1092 1093``-device intel-iommu[,option=...]`` 1094 This is only supported by ``-machine q35``, which will enable Intel VT-d 1095 emulation within the guest. It supports below options: 1096 1097 ``intremap=on|off`` (default: auto) 1098 This enables interrupt remapping feature. It's required to enable 1099 complete x2apic. Currently it only supports kvm kernel-irqchip modes 1100 ``off`` or ``split``, while full kernel-irqchip is not yet supported. 1101 The default value is "auto", which will be decided by the mode of 1102 kernel-irqchip. 1103 1104 ``caching-mode=on|off`` (default: off) 1105 This enables caching mode for the VT-d emulated device. When 1106 caching-mode is enabled, each guest DMA buffer mapping will generate an 1107 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in 1108 a synchronous way. It is required for ``-device vfio-pci`` to work 1109 with the VT-d device, because host assigned devices requires to setup 1110 the DMA mapping on the host before guest DMA starts. 1111 1112 ``device-iotlb=on|off`` (default: off) 1113 This enables device-iotlb capability for the emulated VT-d device. So 1114 far virtio/vhost should be the only real user for this parameter, 1115 paired with ats=on configured for the device. 1116 1117 ``aw-bits=39|48`` (default: 39) 1118 This decides the address width of IOVA address space. The address 1119 space has 39 bits width for 3-level IOMMU page tables, and 48 bits for 1120 4-level IOMMU page tables. 1121 1122 Please also refer to the wiki page for general scenarios of VT-d 1123 emulation in QEMU: https://wiki.qemu.org/Features/VT-d. 1124 1125ERST 1126 1127DEF("name", HAS_ARG, QEMU_OPTION_name, 1128 "-name string1[,process=string2][,debug-threads=on|off]\n" 1129 " set the name of the guest\n" 1130 " string1 sets the window title and string2 the process name\n" 1131 " When debug-threads is enabled, individual threads are given a separate name\n" 1132 " NOTE: The thread names are for debugging and not a stable API.\n", 1133 QEMU_ARCH_ALL) 1134SRST 1135``-name name`` 1136 Sets the name of the guest. This name will be displayed in the SDL 1137 window caption. The name will also be used for the VNC server. Also 1138 optionally set the top visible process name in Linux. Naming of 1139 individual threads can also be enabled on Linux to aid debugging. 1140ERST 1141 1142DEF("uuid", HAS_ARG, QEMU_OPTION_uuid, 1143 "-uuid %08x-%04x-%04x-%04x-%012x\n" 1144 " specify machine UUID\n", QEMU_ARCH_ALL) 1145SRST 1146``-uuid uuid`` 1147 Set system UUID. 1148ERST 1149 1150DEFHEADING() 1151 1152DEFHEADING(Block device options:) 1153 1154SRST 1155The QEMU block device handling options have a long history and 1156have gone through several iterations as the feature set and complexity 1157of the block layer have grown. Many online guides to QEMU often 1158reference older and deprecated options, which can lead to confusion. 1159 1160The most explicit way to describe disks is to use a combination of 1161``-device`` to specify the hardware device and ``-blockdev`` to 1162describe the backend. The device defines what the guest sees and the 1163backend describes how QEMU handles the data. It is the only guaranteed 1164stable interface for describing block devices and as such is 1165recommended for management tools and scripting. 1166 1167The ``-drive`` option combines the device and backend into a single 1168command line option which is a more human friendly. There is however no 1169interface stability guarantee although some older board models still 1170need updating to work with the modern blockdev forms. 1171 1172Older options like ``-hda`` are essentially macros which expand into 1173``-drive`` options for various drive interfaces. The original forms 1174bake in a lot of assumptions from the days when QEMU was emulating a 1175legacy PC, they are not recommended for modern configurations. 1176 1177ERST 1178 1179DEF("fda", HAS_ARG, QEMU_OPTION_fda, 1180 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL) 1181DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL) 1182SRST 1183``-fda file`` 1184 \ 1185``-fdb file`` 1186 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in 1187 the System Emulation Users Guide). 1188ERST 1189 1190DEF("hda", HAS_ARG, QEMU_OPTION_hda, 1191 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL) 1192DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL) 1193DEF("hdc", HAS_ARG, QEMU_OPTION_hdc, 1194 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL) 1195DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL) 1196SRST 1197``-hda file`` 1198 \ 1199``-hdb file`` 1200 \ 1201``-hdc file`` 1202 \ 1203``-hdd file`` 1204 Use file as hard disk 0, 1, 2 or 3 image (see the :ref:`disk images` 1205 chapter in the System Emulation Users Guide). 1206ERST 1207 1208DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom, 1209 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n", 1210 QEMU_ARCH_ALL) 1211SRST 1212``-cdrom file`` 1213 Use file as CD-ROM image (you cannot use ``-hdc`` and ``-cdrom`` at 1214 the same time). You can use the host CD-ROM by using ``/dev/cdrom`` 1215 as filename. 1216ERST 1217 1218DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev, 1219 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n" 1220 " [,cache.direct=on|off][,cache.no-flush=on|off]\n" 1221 " [,read-only=on|off][,auto-read-only=on|off]\n" 1222 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n" 1223 " [,driver specific parameters...]\n" 1224 " configure a block backend\n", QEMU_ARCH_ALL) 1225SRST 1226``-blockdev option[,option[,option[,...]]]`` 1227 Define a new block driver node. Some of the options apply to all 1228 block drivers, other options are only accepted for a specific block 1229 driver. See below for a list of generic options and options for the 1230 most common block drivers. 1231 1232 Options that expect a reference to another node (e.g. ``file``) can 1233 be given in two ways. Either you specify the node name of an already 1234 existing node (file=node-name), or you define a new node inline, 1235 adding options for the referenced node after a dot 1236 (file.filename=path,file.aio=native). 1237 1238 A block driver node created with ``-blockdev`` can be used for a 1239 guest device by specifying its node name for the ``drive`` property 1240 in a ``-device`` argument that defines a block device. 1241 1242 ``Valid options for any block driver node:`` 1243 ``driver`` 1244 Specifies the block driver to use for the given node. 1245 1246 ``node-name`` 1247 This defines the name of the block driver node by which it 1248 will be referenced later. The name must be unique, i.e. it 1249 must not match the name of a different block driver node, or 1250 (if you use ``-drive`` as well) the ID of a drive. 1251 1252 If no node name is specified, it is automatically generated. 1253 The generated node name is not intended to be predictable 1254 and changes between QEMU invocations. For the top level, an 1255 explicit node name must be specified. 1256 1257 ``read-only`` 1258 Open the node read-only. Guest write attempts will fail. 1259 1260 Note that some block drivers support only read-only access, 1261 either generally or in certain configurations. In this case, 1262 the default value ``read-only=off`` does not work and the 1263 option must be specified explicitly. 1264 1265 ``auto-read-only`` 1266 If ``auto-read-only=on`` is set, QEMU may fall back to 1267 read-only usage even when ``read-only=off`` is requested, or 1268 even switch between modes as needed, e.g. depending on 1269 whether the image file is writable or whether a writing user 1270 is attached to the node. 1271 1272 ``force-share`` 1273 Override the image locking system of QEMU by forcing the 1274 node to utilize weaker shared access for permissions where 1275 it would normally request exclusive access. When there is 1276 the potential for multiple instances to have the same file 1277 open (whether this invocation of QEMU is the first or the 1278 second instance), both instances must permit shared access 1279 for the second instance to succeed at opening the file. 1280 1281 Enabling ``force-share=on`` requires ``read-only=on``. 1282 1283 ``cache.direct`` 1284 The host page cache can be avoided with ``cache.direct=on``. 1285 This will attempt to do disk IO directly to the guest's 1286 memory. QEMU may still perform an internal copy of the data. 1287 1288 ``cache.no-flush`` 1289 In case you don't care about data integrity over host 1290 failures, you can use ``cache.no-flush=on``. This option 1291 tells QEMU that it never needs to write any data to the disk 1292 but can instead keep things in cache. If anything goes 1293 wrong, like your host losing power, the disk storage getting 1294 disconnected accidentally, etc. your image will most 1295 probably be rendered unusable. 1296 1297 ``discard=discard`` 1298 discard is one of "ignore" (or "off") or "unmap" (or "on") 1299 and controls whether ``discard`` (also known as ``trim`` or 1300 ``unmap``) requests are ignored or passed to the filesystem. 1301 Some machine types may not support discard requests. 1302 1303 ``detect-zeroes=detect-zeroes`` 1304 detect-zeroes is "off", "on" or "unmap" and enables the 1305 automatic conversion of plain zero writes by the OS to 1306 driver specific optimized zero write commands. You may even 1307 choose "unmap" if discard is set to "unmap" to allow a zero 1308 write to be converted to an ``unmap`` operation. 1309 1310 ``Driver-specific options for file`` 1311 This is the protocol-level block driver for accessing regular 1312 files. 1313 1314 ``filename`` 1315 The path to the image file in the local filesystem 1316 1317 ``aio`` 1318 Specifies the AIO backend (threads/native/io_uring, 1319 default: threads) 1320 1321 ``locking`` 1322 Specifies whether the image file is protected with Linux OFD 1323 / POSIX locks. The default is to use the Linux Open File 1324 Descriptor API if available, otherwise no lock is applied. 1325 (auto/on/off, default: auto) 1326 1327 Example: 1328 1329 :: 1330 1331 -blockdev driver=file,node-name=disk,filename=disk.img 1332 1333 ``Driver-specific options for raw`` 1334 This is the image format block driver for raw images. It is 1335 usually stacked on top of a protocol level block driver such as 1336 ``file``. 1337 1338 ``file`` 1339 Reference to or definition of the data source block driver 1340 node (e.g. a ``file`` driver node) 1341 1342 Example 1: 1343 1344 :: 1345 1346 -blockdev driver=file,node-name=disk_file,filename=disk.img 1347 -blockdev driver=raw,node-name=disk,file=disk_file 1348 1349 Example 2: 1350 1351 :: 1352 1353 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img 1354 1355 ``Driver-specific options for qcow2`` 1356 This is the image format block driver for qcow2 images. It is 1357 usually stacked on top of a protocol level block driver such as 1358 ``file``. 1359 1360 ``file`` 1361 Reference to or definition of the data source block driver 1362 node (e.g. a ``file`` driver node) 1363 1364 ``backing`` 1365 Reference to or definition of the backing file block device 1366 (default is taken from the image file). It is allowed to 1367 pass ``null`` here in order to disable the default backing 1368 file. 1369 1370 ``lazy-refcounts`` 1371 Whether to enable the lazy refcounts feature (on/off; 1372 default is taken from the image file) 1373 1374 ``cache-size`` 1375 The maximum total size of the L2 table and refcount block 1376 caches in bytes (default: the sum of l2-cache-size and 1377 refcount-cache-size) 1378 1379 ``l2-cache-size`` 1380 The maximum size of the L2 table cache in bytes (default: if 1381 cache-size is not specified - 32M on Linux platforms, and 8M 1382 on non-Linux platforms; otherwise, as large as possible 1383 within the cache-size, while permitting the requested or the 1384 minimal refcount cache size) 1385 1386 ``refcount-cache-size`` 1387 The maximum size of the refcount block cache in bytes 1388 (default: 4 times the cluster size; or if cache-size is 1389 specified, the part of it which is not used for the L2 1390 cache) 1391 1392 ``cache-clean-interval`` 1393 Clean unused entries in the L2 and refcount caches. The 1394 interval is in seconds. The default value is 600 on 1395 supporting platforms, and 0 on other platforms. Setting it 1396 to 0 disables this feature. 1397 1398 ``pass-discard-request`` 1399 Whether discard requests to the qcow2 device should be 1400 forwarded to the data source (on/off; default: on if 1401 discard=unmap is specified, off otherwise) 1402 1403 ``pass-discard-snapshot`` 1404 Whether discard requests for the data source should be 1405 issued when a snapshot operation (e.g. deleting a snapshot) 1406 frees clusters in the qcow2 file (on/off; default: on) 1407 1408 ``pass-discard-other`` 1409 Whether discard requests for the data source should be 1410 issued on other occasions where a cluster gets freed 1411 (on/off; default: off) 1412 1413 ``overlap-check`` 1414 Which overlap checks to perform for writes to the image 1415 (none/constant/cached/all; default: cached). For details or 1416 finer granularity control refer to the QAPI documentation of 1417 ``blockdev-add``. 1418 1419 Example 1: 1420 1421 :: 1422 1423 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2 1424 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216 1425 1426 Example 2: 1427 1428 :: 1429 1430 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2 1431 1432 ``Driver-specific options for other drivers`` 1433 Please refer to the QAPI documentation of the ``blockdev-add`` 1434 QMP command. 1435ERST 1436 1437DEF("drive", HAS_ARG, QEMU_OPTION_drive, 1438 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n" 1439 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n" 1440 " [,snapshot=on|off][,rerror=ignore|stop|report]\n" 1441 " [,werror=ignore|stop|report|enospc][,id=name]\n" 1442 " [,aio=threads|native|io_uring]\n" 1443 " [,readonly=on|off][,copy-on-read=on|off]\n" 1444 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n" 1445 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n" 1446 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n" 1447 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n" 1448 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n" 1449 " [[,iops_size=is]]\n" 1450 " [[,group=g]]\n" 1451 " use 'file' as a drive image\n", QEMU_ARCH_ALL) 1452SRST 1453``-drive option[,option[,option[,...]]]`` 1454 Define a new drive. This includes creating a block driver node (the 1455 backend) as well as a guest device, and is mostly a shortcut for 1456 defining the corresponding ``-blockdev`` and ``-device`` options. 1457 1458 ``-drive`` accepts all options that are accepted by ``-blockdev``. 1459 In addition, it knows the following options: 1460 1461 ``file=file`` 1462 This option defines which disk image (see the :ref:`disk images` 1463 chapter in the System Emulation Users Guide) to use with this drive. 1464 If the filename contains comma, you must double it (for instance, 1465 "file=my,,file" to use file "my,file"). 1466 1467 Special files such as iSCSI devices can be specified using 1468 protocol specific URLs. See the section for "Device URL Syntax" 1469 for more information. 1470 1471 ``if=interface`` 1472 This option defines on which type on interface the drive is 1473 connected. Available types are: ide, scsi, sd, mtd, floppy, 1474 pflash, virtio, none. 1475 1476 ``bus=bus,unit=unit`` 1477 These options define where is connected the drive by defining 1478 the bus number and the unit id. 1479 1480 ``index=index`` 1481 This option defines where the drive is connected by using an 1482 index in the list of available connectors of a given interface 1483 type. 1484 1485 ``media=media`` 1486 This option defines the type of the media: disk or cdrom. 1487 1488 ``snapshot=snapshot`` 1489 snapshot is "on" or "off" and controls snapshot mode for the 1490 given drive (see ``-snapshot``). 1491 1492 ``cache=cache`` 1493 cache is "none", "writeback", "unsafe", "directsync" or 1494 "writethrough" and controls how the host cache is used to access 1495 block data. This is a shortcut that sets the ``cache.direct`` 1496 and ``cache.no-flush`` options (as in ``-blockdev``), and 1497 additionally ``cache.writeback``, which provides a default for 1498 the ``write-cache`` option of block guest devices (as in 1499 ``-device``). The modes correspond to the following settings: 1500 1501 ============= =============== ============ ============== 1502 \ cache.writeback cache.direct cache.no-flush 1503 ============= =============== ============ ============== 1504 writeback on off off 1505 none on on off 1506 writethrough off off off 1507 directsync off on off 1508 unsafe on off on 1509 ============= =============== ============ ============== 1510 1511 The default mode is ``cache=writeback``. 1512 1513 ``aio=aio`` 1514 aio is "threads", "native", or "io_uring" and selects between pthread 1515 based disk I/O, native Linux AIO, or Linux io_uring API. 1516 1517 ``format=format`` 1518 Specify which disk format will be used rather than detecting the 1519 format. Can be used to specify format=raw to avoid interpreting 1520 an untrusted format header. 1521 1522 ``werror=action,rerror=action`` 1523 Specify which action to take on write and read errors. Valid 1524 actions are: "ignore" (ignore the error and try to continue), 1525 "stop" (pause QEMU), "report" (report the error to the guest), 1526 "enospc" (pause QEMU only if the host disk is full; report the 1527 error to the guest otherwise). The default setting is 1528 ``werror=enospc`` and ``rerror=report``. 1529 1530 ``copy-on-read=copy-on-read`` 1531 copy-on-read is "on" or "off" and enables whether to copy read 1532 backing file sectors into the image file. 1533 1534 ``bps=b,bps_rd=r,bps_wr=w`` 1535 Specify bandwidth throttling limits in bytes per second, either 1536 for all request types or for reads or writes only. Small values 1537 can lead to timeouts or hangs inside the guest. A safe minimum 1538 for disks is 2 MB/s. 1539 1540 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm`` 1541 Specify bursts in bytes per second, either for all request types 1542 or for reads or writes only. Bursts allow the guest I/O to spike 1543 above the limit temporarily. 1544 1545 ``iops=i,iops_rd=r,iops_wr=w`` 1546 Specify request rate limits in requests per second, either for 1547 all request types or for reads or writes only. 1548 1549 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm`` 1550 Specify bursts in requests per second, either for all request 1551 types or for reads or writes only. Bursts allow the guest I/O to 1552 spike above the limit temporarily. 1553 1554 ``iops_size=is`` 1555 Let every is bytes of a request count as a new request for iops 1556 throttling purposes. Use this option to prevent guests from 1557 circumventing iops limits by sending fewer but larger requests. 1558 1559 ``group=g`` 1560 Join a throttling quota group with given name g. All drives that 1561 are members of the same group are accounted for together. Use 1562 this option to prevent guests from circumventing throttling 1563 limits by using many small disks instead of a single larger 1564 disk. 1565 1566 By default, the ``cache.writeback=on`` mode is used. It will report 1567 data writes as completed as soon as the data is present in the host 1568 page cache. This is safe as long as your guest OS makes sure to 1569 correctly flush disk caches where needed. If your guest OS does not 1570 handle volatile disk write caches correctly and your host crashes or 1571 loses power, then the guest may experience data corruption. 1572 1573 For such guests, you should consider using ``cache.writeback=off``. 1574 This means that the host page cache will be used to read and write 1575 data, but write notification will be sent to the guest only after 1576 QEMU has made sure to flush each write to the disk. Be aware that 1577 this has a major impact on performance. 1578 1579 When using the ``-snapshot`` option, unsafe caching is always used. 1580 1581 Copy-on-read avoids accessing the same backing file sectors 1582 repeatedly and is useful when the backing file is over a slow 1583 network. By default copy-on-read is off. 1584 1585 Instead of ``-cdrom`` you can use: 1586 1587 .. parsed-literal:: 1588 1589 |qemu_system| -drive file=file,index=2,media=cdrom 1590 1591 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use: 1592 1593 .. parsed-literal:: 1594 1595 |qemu_system| -drive file=file,index=0,media=disk 1596 |qemu_system| -drive file=file,index=1,media=disk 1597 |qemu_system| -drive file=file,index=2,media=disk 1598 |qemu_system| -drive file=file,index=3,media=disk 1599 1600 You can open an image using pre-opened file descriptors from an fd 1601 set: 1602 1603 .. parsed-literal:: 1604 1605 |qemu_system| \\ 1606 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 1607 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 1608 -drive file=/dev/fdset/2,index=0,media=disk 1609 1610 You can connect a CDROM to the slave of ide0: 1611 1612 .. parsed-literal:: 1613 1614 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom 1615 1616 If you don't specify the "file=" argument, you define an empty 1617 drive: 1618 1619 .. parsed-literal:: 1620 1621 |qemu_system_x86| -drive if=ide,index=1,media=cdrom 1622 1623 Instead of ``-fda``, ``-fdb``, you can use: 1624 1625 .. parsed-literal:: 1626 1627 |qemu_system_x86| -drive file=file,index=0,if=floppy 1628 |qemu_system_x86| -drive file=file,index=1,if=floppy 1629 1630 By default, interface is "ide" and index is automatically 1631 incremented: 1632 1633 .. parsed-literal:: 1634 1635 |qemu_system_x86| -drive file=a -drive file=b 1636 1637 is interpreted like: 1638 1639 .. parsed-literal:: 1640 1641 |qemu_system_x86| -hda a -hdb b 1642ERST 1643 1644DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock, 1645 "-mtdblock file use 'file' as on-board Flash memory image\n", 1646 QEMU_ARCH_ALL) 1647SRST 1648``-mtdblock file`` 1649 Use file as on-board Flash memory image. 1650ERST 1651 1652DEF("sd", HAS_ARG, QEMU_OPTION_sd, 1653 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL) 1654SRST 1655``-sd file`` 1656 Use file as SecureDigital card image. 1657ERST 1658 1659DEF("snapshot", 0, QEMU_OPTION_snapshot, 1660 "-snapshot write to temporary files instead of disk image files\n", 1661 QEMU_ARCH_ALL) 1662SRST 1663``-snapshot`` 1664 Write to temporary files instead of disk image files. In this case, 1665 the raw disk image you use is not written back. You can however 1666 force the write back by pressing C-a s (see the :ref:`disk images` 1667 chapter in the System Emulation Users Guide). 1668 1669 .. warning:: 1670 snapshot is incompatible with ``-blockdev`` (instead use qemu-img 1671 to manually create snapshot images to attach to your blockdev). 1672 If you have mixed ``-blockdev`` and ``-drive`` declarations you 1673 can use the 'snapshot' property on your drive declarations 1674 instead of this global option. 1675 1676ERST 1677 1678DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev, 1679 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1680 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n" 1681 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n" 1682 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n" 1683 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n" 1684 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n" 1685 " [[,throttling.iops-size=is]]\n" 1686 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n" 1687 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n" 1688 "-fsdev synth,id=id\n", 1689 QEMU_ARCH_ALL) 1690 1691SRST 1692``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]`` 1693 \ 1694``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]`` 1695 \ 1696``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]`` 1697 \ 1698``-fsdev synth,id=id[,readonly=on]`` 1699 Define a new file system device. Valid options are: 1700 1701 ``local`` 1702 Accesses to the filesystem are done by QEMU. 1703 1704 ``proxy`` 1705 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1706 1707 ``synth`` 1708 Synthetic filesystem, only used by QTests. 1709 1710 ``id=id`` 1711 Specifies identifier for this device. 1712 1713 ``path=path`` 1714 Specifies the export path for the file system device. Files 1715 under this path will be available to the 9p client on the guest. 1716 1717 ``security_model=security_model`` 1718 Specifies the security model to be used for this export path. 1719 Supported security models are "passthrough", "mapped-xattr", 1720 "mapped-file" and "none". In "passthrough" security model, files 1721 are stored using the same credentials as they are created on the 1722 guest. This requires QEMU to run as root. In "mapped-xattr" 1723 security model, some of the file attributes like uid, gid, mode 1724 bits and link target are stored as file attributes. For 1725 "mapped-file" these attributes are stored in the hidden 1726 .virtfs\_metadata directory. Directories exported by this 1727 security model cannot interact with other unix tools. "none" 1728 security model is same as passthrough except the sever won't 1729 report failures if it fails to set file attributes like 1730 ownership. Security model is mandatory only for local fsdriver. 1731 Other fsdrivers (like proxy) don't take security model as a 1732 parameter. 1733 1734 ``writeout=writeout`` 1735 This is an optional argument. The only supported value is 1736 "immediate". This means that host page cache will be used to 1737 read and write data but write notification will be sent to the 1738 guest only when the data has been reported as written by the 1739 storage subsystem. 1740 1741 ``readonly=on`` 1742 Enables exporting 9p share as a readonly mount for guests. By 1743 default read-write access is given. 1744 1745 ``socket=socket`` 1746 Enables proxy filesystem driver to use passed socket file for 1747 communicating with virtfs-proxy-helper(1). 1748 1749 ``sock_fd=sock_fd`` 1750 Enables proxy filesystem driver to use passed socket descriptor 1751 for communicating with virtfs-proxy-helper(1). Usually a helper 1752 like libvirt will create socketpair and pass one of the fds as 1753 sock\_fd. 1754 1755 ``fmode=fmode`` 1756 Specifies the default mode for newly created files on the host. 1757 Works only with security models "mapped-xattr" and 1758 "mapped-file". 1759 1760 ``dmode=dmode`` 1761 Specifies the default mode for newly created directories on the 1762 host. Works only with security models "mapped-xattr" and 1763 "mapped-file". 1764 1765 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w`` 1766 Specify bandwidth throttling limits in bytes per second, either 1767 for all request types or for reads or writes only. 1768 1769 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm`` 1770 Specify bursts in bytes per second, either for all request types 1771 or for reads or writes only. Bursts allow the guest I/O to spike 1772 above the limit temporarily. 1773 1774 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w`` 1775 Specify request rate limits in requests per second, either for 1776 all request types or for reads or writes only. 1777 1778 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm`` 1779 Specify bursts in requests per second, either for all request 1780 types or for reads or writes only. Bursts allow the guest I/O to 1781 spike above the limit temporarily. 1782 1783 ``throttling.iops-size=is`` 1784 Let every is bytes of a request count as a new request for iops 1785 throttling purposes. 1786 1787 -fsdev option is used along with -device driver "virtio-9p-...". 1788 1789``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag`` 1790 Options for virtio-9p-... driver are: 1791 1792 ``type`` 1793 Specifies the variant to be used. Supported values are "pci", 1794 "ccw" or "device", depending on the machine type. 1795 1796 ``fsdev=id`` 1797 Specifies the id value specified along with -fsdev option. 1798 1799 ``mount_tag=mount_tag`` 1800 Specifies the tag name to be used by the guest to mount this 1801 export point. 1802ERST 1803 1804DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs, 1805 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1806 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n" 1807 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n" 1808 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n" 1809 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n", 1810 QEMU_ARCH_ALL) 1811 1812SRST 1813``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]`` 1814 \ 1815``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1816 \ 1817``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1818 \ 1819``-virtfs synth,mount_tag=mount_tag`` 1820 Define a new virtual filesystem device and expose it to the guest using 1821 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain 1822 directory on host is made directly accessible by guest as a pass-through 1823 file system by using the 9P network protocol for communication between 1824 host and guests, if desired even accessible, shared by several guests 1825 simultaneously. 1826 1827 Note that ``-virtfs`` is actually just a convenience shortcut for its 1828 generalized form ``-fsdev -device virtio-9p-pci``. 1829 1830 The general form of pass-through file system options are: 1831 1832 ``local`` 1833 Accesses to the filesystem are done by QEMU. 1834 1835 ``proxy`` 1836 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1837 1838 ``synth`` 1839 Synthetic filesystem, only used by QTests. 1840 1841 ``id=id`` 1842 Specifies identifier for the filesystem device 1843 1844 ``path=path`` 1845 Specifies the export path for the file system device. Files 1846 under this path will be available to the 9p client on the guest. 1847 1848 ``security_model=security_model`` 1849 Specifies the security model to be used for this export path. 1850 Supported security models are "passthrough", "mapped-xattr", 1851 "mapped-file" and "none". In "passthrough" security model, files 1852 are stored using the same credentials as they are created on the 1853 guest. This requires QEMU to run as root. In "mapped-xattr" 1854 security model, some of the file attributes like uid, gid, mode 1855 bits and link target are stored as file attributes. For 1856 "mapped-file" these attributes are stored in the hidden 1857 .virtfs\_metadata directory. Directories exported by this 1858 security model cannot interact with other unix tools. "none" 1859 security model is same as passthrough except the sever won't 1860 report failures if it fails to set file attributes like 1861 ownership. Security model is mandatory only for local fsdriver. 1862 Other fsdrivers (like proxy) don't take security model as a 1863 parameter. 1864 1865 ``writeout=writeout`` 1866 This is an optional argument. The only supported value is 1867 "immediate". This means that host page cache will be used to 1868 read and write data but write notification will be sent to the 1869 guest only when the data has been reported as written by the 1870 storage subsystem. 1871 1872 ``readonly=on`` 1873 Enables exporting 9p share as a readonly mount for guests. By 1874 default read-write access is given. 1875 1876 ``socket=socket`` 1877 Enables proxy filesystem driver to use passed socket file for 1878 communicating with virtfs-proxy-helper(1). Usually a helper like 1879 libvirt will create socketpair and pass one of the fds as 1880 sock\_fd. 1881 1882 ``sock_fd`` 1883 Enables proxy filesystem driver to use passed 'sock\_fd' as the 1884 socket descriptor for interfacing with virtfs-proxy-helper(1). 1885 1886 ``fmode=fmode`` 1887 Specifies the default mode for newly created files on the host. 1888 Works only with security models "mapped-xattr" and 1889 "mapped-file". 1890 1891 ``dmode=dmode`` 1892 Specifies the default mode for newly created directories on the 1893 host. Works only with security models "mapped-xattr" and 1894 "mapped-file". 1895 1896 ``mount_tag=mount_tag`` 1897 Specifies the tag name to be used by the guest to mount this 1898 export point. 1899 1900 ``multidevs=multidevs`` 1901 Specifies how to deal with multiple devices being shared with a 1902 9p export. Supported behaviours are either "remap", "forbid" or 1903 "warn". The latter is the default behaviour on which virtfs 9p 1904 expects only one device to be shared with the same export, and 1905 if more than one device is shared and accessed via the same 9p 1906 export then only a warning message is logged (once) by qemu on 1907 host side. In order to avoid file ID collisions on guest you 1908 should either create a separate virtfs export for each device to 1909 be shared with guests (recommended way) or you might use "remap" 1910 instead which allows you to share multiple devices with only one 1911 export instead, which is achieved by remapping the original 1912 inode numbers from host to guest in a way that would prevent 1913 such collisions. Remapping inodes in such use cases is required 1914 because the original device IDs from host are never passed and 1915 exposed on guest. Instead all files of an export shared with 1916 virtfs always share the same device id on guest. So two files 1917 with identical inode numbers but from actually different devices 1918 on host would otherwise cause a file ID collision and hence 1919 potential misbehaviours on guest. "forbid" on the other hand 1920 assumes like "warn" that only one device is shared by the same 1921 export, however it will not only log a warning message but also 1922 deny access to additional devices on guest. Note though that 1923 "forbid" does currently not block all possible file access 1924 operations (e.g. readdir() would still return entries from other 1925 devices). 1926ERST 1927 1928DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi, 1929 "-iscsi [user=user][,password=password][,password-secret=secret-id]\n" 1930 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE]\n" 1931 " [,initiator-name=initiator-iqn][,id=target-iqn]\n" 1932 " [,timeout=timeout]\n" 1933 " iSCSI session parameters\n", QEMU_ARCH_ALL) 1934 1935SRST 1936``-iscsi`` 1937 Configure iSCSI session parameters. 1938ERST 1939 1940DEFHEADING() 1941 1942DEFHEADING(USB convenience options:) 1943 1944DEF("usb", 0, QEMU_OPTION_usb, 1945 "-usb enable on-board USB host controller (if not enabled by default)\n", 1946 QEMU_ARCH_ALL) 1947SRST 1948``-usb`` 1949 Enable USB emulation on machine types with an on-board USB host 1950 controller (if not enabled by default). Note that on-board USB host 1951 controllers may not support USB 3.0. In this case 1952 ``-device qemu-xhci`` can be used instead on machines with PCI. 1953ERST 1954 1955DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice, 1956 "-usbdevice name add the host or guest USB device 'name'\n", 1957 QEMU_ARCH_ALL) 1958SRST 1959``-usbdevice devname`` 1960 Add the USB device devname, and enable an on-board USB controller 1961 if possible and necessary (just like it can be done via 1962 ``-machine usb=on``). Note that this option is mainly intended for 1963 the user's convenience only. More fine-grained control can be 1964 achieved by selecting a USB host controller (if necessary) and the 1965 desired USB device via the ``-device`` option instead. For example, 1966 instead of using ``-usbdevice mouse`` it is possible to use 1967 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse 1968 to a USB 3.0 controller instead (at least on machines that support 1969 PCI and do not have an USB controller enabled by default yet). 1970 For more details, see the chapter about 1971 :ref:`Connecting USB devices` in the System Emulation Users Guide. 1972 Possible devices for devname are: 1973 1974 ``braille`` 1975 Braille device. This will use BrlAPI to display the braille 1976 output on a real or fake device (i.e. it also creates a 1977 corresponding ``braille`` chardev automatically beside the 1978 ``usb-braille`` USB device). 1979 1980 ``keyboard`` 1981 Standard USB keyboard. Will override the PS/2 keyboard (if present). 1982 1983 ``mouse`` 1984 Virtual Mouse. This will override the PS/2 mouse emulation when 1985 activated. 1986 1987 ``tablet`` 1988 Pointer device that uses absolute coordinates (like a 1989 touchscreen). This means QEMU is able to report the mouse 1990 position without having to grab the mouse. Also overrides the 1991 PS/2 mouse emulation when activated. 1992 1993 ``wacom-tablet`` 1994 Wacom PenPartner USB tablet. 1995 1996 1997ERST 1998 1999DEFHEADING() 2000 2001DEFHEADING(Display options:) 2002 2003DEF("display", HAS_ARG, QEMU_OPTION_display, 2004#if defined(CONFIG_SPICE) 2005 "-display spice-app[,gl=on|off]\n" 2006#endif 2007#if defined(CONFIG_SDL) 2008 "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n" 2009 " [,window-close=on|off]\n" 2010#endif 2011#if defined(CONFIG_GTK) 2012 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n" 2013 " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n" 2014 " [,show-menubar=on|off]\n" 2015#endif 2016#if defined(CONFIG_VNC) 2017 "-display vnc=<display>[,<optargs>]\n" 2018#endif 2019#if defined(CONFIG_CURSES) 2020 "-display curses[,charset=<encoding>]\n" 2021#endif 2022#if defined(CONFIG_COCOA) 2023 "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n" 2024#endif 2025#if defined(CONFIG_OPENGL) 2026 "-display egl-headless[,rendernode=<file>]\n" 2027#endif 2028#if defined(CONFIG_DBUS_DISPLAY) 2029 "-display dbus[,addr=<dbusaddr>]\n" 2030 " [,gl=on|core|es|off][,rendernode=<file>]\n" 2031#endif 2032#if defined(CONFIG_COCOA) 2033 "-display cocoa[,show-cursor=on|off][,left-command-key=on|off]\n" 2034#endif 2035 "-display none\n" 2036 " select display backend type\n" 2037 " The default display is equivalent to\n " 2038#if defined(CONFIG_GTK) 2039 "\"-display gtk\"\n" 2040#elif defined(CONFIG_SDL) 2041 "\"-display sdl\"\n" 2042#elif defined(CONFIG_COCOA) 2043 "\"-display cocoa\"\n" 2044#elif defined(CONFIG_VNC) 2045 "\"-vnc localhost:0,to=99,id=default\"\n" 2046#else 2047 "\"-display none\"\n" 2048#endif 2049 , QEMU_ARCH_ALL) 2050SRST 2051``-display type`` 2052 Select type of display to use. Use ``-display help`` to list the available 2053 display types. Valid values for type are 2054 2055 ``spice-app[,gl=on|off]`` 2056 Start QEMU as a Spice server and launch the default Spice client 2057 application. The Spice server will redirect the serial consoles 2058 and QEMU monitors. (Since 4.0) 2059 2060 ``dbus`` 2061 Export the display over D-Bus interfaces. (Since 7.0) 2062 2063 The connection is registered with the "org.qemu" name (and queued when 2064 already owned). 2065 2066 ``addr=<dbusaddr>`` : D-Bus bus address to connect to. 2067 2068 ``p2p=yes|no`` : Use peer-to-peer connection, accepted via QMP ``add_client``. 2069 2070 ``gl=on|off|core|es`` : Use OpenGL for rendering (the D-Bus interface 2071 will share framebuffers with DMABUF file descriptors). 2072 2073 ``sdl`` 2074 Display video output via SDL (usually in a separate graphics 2075 window; see the SDL documentation for other possibilities). 2076 Valid parameters are: 2077 2078 ``grab-mod=<mods>`` : Used to select the modifier keys for toggling 2079 the mouse grabbing in conjunction with the "g" key. ``<mods>`` can be 2080 either ``lshift-lctrl-lalt`` or ``rctrl``. 2081 2082 ``gl=on|off|core|es`` : Use OpenGL for displaying 2083 2084 ``show-cursor=on|off`` : Force showing the mouse cursor 2085 2086 ``window-close=on|off`` : Allow to quit qemu with window close button 2087 2088 ``gtk`` 2089 Display video output in a GTK window. This interface provides 2090 drop-down menus and other UI elements to configure and control 2091 the VM during runtime. Valid parameters are: 2092 2093 ``full-screen=on|off`` : Start in fullscreen mode 2094 2095 ``gl=on|off`` : Use OpenGL for displaying 2096 2097 ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover 2098 2099 ``show-tabs=on|off`` : Display the tab bar for switching between the 2100 various graphical interfaces (e.g. VGA and 2101 virtual console character devices) by default. 2102 2103 ``show-cursor=on|off`` : Force showing the mouse cursor 2104 2105 ``window-close=on|off`` : Allow to quit qemu with window close button 2106 2107 ``show-menubar=on|off`` : Display the main window menubar, defaults to "on" 2108 2109 ``curses[,charset=<encoding>]`` 2110 Display video output via curses. For graphics device models 2111 which support a text mode, QEMU can display this output using a 2112 curses/ncurses interface. Nothing is displayed when the graphics 2113 device is in graphical mode or if the graphics device does not 2114 support a text mode. Generally only the VGA device models 2115 support text mode. The font charset used by the guest can be 2116 specified with the ``charset`` option, for example 2117 ``charset=CP850`` for IBM CP850 encoding. The default is 2118 ``CP437``. 2119 2120 ``cocoa`` 2121 Display video output in a Cocoa window. Mac only. This interface 2122 provides drop-down menus and other UI elements to configure and 2123 control the VM during runtime. Valid parameters are: 2124 2125 ``show-cursor=on|off`` : Force showing the mouse cursor 2126 2127 ``left-command-key=on|off`` : Disable forwarding left command key to host 2128 2129 ``egl-headless[,rendernode=<file>]`` 2130 Offload all OpenGL operations to a local DRI device. For any 2131 graphical display, this display needs to be paired with either 2132 VNC or SPICE displays. 2133 2134 ``vnc=<display>`` 2135 Start a VNC server on display <display> 2136 2137 ``none`` 2138 Do not display video output. The guest will still see an 2139 emulated graphics card, but its output will not be displayed to 2140 the QEMU user. This option differs from the -nographic option in 2141 that it only affects what is done with video output; -nographic 2142 also changes the destination of the serial and parallel port 2143 data. 2144ERST 2145 2146DEF("nographic", 0, QEMU_OPTION_nographic, 2147 "-nographic disable graphical output and redirect serial I/Os to console\n", 2148 QEMU_ARCH_ALL) 2149SRST 2150``-nographic`` 2151 Normally, if QEMU is compiled with graphical window support, it 2152 displays output such as guest graphics, guest console, and the QEMU 2153 monitor in a window. With this option, you can totally disable 2154 graphical output so that QEMU is a simple command line application. 2155 The emulated serial port is redirected on the console and muxed with 2156 the monitor (unless redirected elsewhere explicitly). Therefore, you 2157 can still use QEMU to debug a Linux kernel with a serial console. 2158 Use C-a h for help on switching between the console and monitor. 2159ERST 2160 2161#ifdef CONFIG_SPICE 2162DEF("spice", HAS_ARG, QEMU_OPTION_spice, 2163 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n" 2164 " [,x509-key-file=<file>][,x509-key-password=<file>]\n" 2165 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n" 2166 " [,x509-dh-key-file=<file>][,addr=addr]\n" 2167 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n" 2168 " [,tls-ciphers=<list>]\n" 2169 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n" 2170 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n" 2171 " [,sasl=on|off][,disable-ticketing=on|off]\n" 2172 " [,password-secret=<secret-id>]\n" 2173 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n" 2174 " [,jpeg-wan-compression=[auto|never|always]]\n" 2175 " [,zlib-glz-wan-compression=[auto|never|always]]\n" 2176 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n" 2177 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n" 2178 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n" 2179 " [,gl=[on|off]][,rendernode=<file>]\n" 2180 " enable spice\n" 2181 " at least one of {port, tls-port} is mandatory\n", 2182 QEMU_ARCH_ALL) 2183#endif 2184SRST 2185``-spice option[,option[,...]]`` 2186 Enable the spice remote desktop protocol. Valid options are 2187 2188 ``port=<nr>`` 2189 Set the TCP port spice is listening on for plaintext channels. 2190 2191 ``addr=<addr>`` 2192 Set the IP address spice is listening on. Default is any 2193 address. 2194 2195 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off`` 2196 Force using the specified IP version. 2197 2198 ``password-secret=<secret-id>`` 2199 Set the ID of the ``secret`` object containing the password 2200 you need to authenticate. 2201 2202 ``sasl=on|off`` 2203 Require that the client use SASL to authenticate with the spice. 2204 The exact choice of authentication method used is controlled 2205 from the system / user's SASL configuration file for the 'qemu' 2206 service. This is typically found in /etc/sasl2/qemu.conf. If 2207 running QEMU as an unprivileged user, an environment variable 2208 SASL\_CONF\_PATH can be used to make it search alternate 2209 locations for the service config. While some SASL auth methods 2210 can also provide data encryption (eg GSSAPI), it is recommended 2211 that SASL always be combined with the 'tls' and 'x509' settings 2212 to enable use of SSL and server certificates. This ensures a 2213 data encryption preventing compromise of authentication 2214 credentials. 2215 2216 ``disable-ticketing=on|off`` 2217 Allow client connects without authentication. 2218 2219 ``disable-copy-paste=on|off`` 2220 Disable copy paste between the client and the guest. 2221 2222 ``disable-agent-file-xfer=on|off`` 2223 Disable spice-vdagent based file-xfer between the client and the 2224 guest. 2225 2226 ``tls-port=<nr>`` 2227 Set the TCP port spice is listening on for encrypted channels. 2228 2229 ``x509-dir=<dir>`` 2230 Set the x509 file directory. Expects same filenames as -vnc 2231 $display,x509=$dir 2232 2233 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>`` 2234 The x509 file names can also be configured individually. 2235 2236 ``tls-ciphers=<list>`` 2237 Specify which ciphers to use. 2238 2239 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]`` 2240 Force specific channel to be used with or without TLS 2241 encryption. The options can be specified multiple times to 2242 configure multiple channels. The special name "default" can be 2243 used to set the default mode. For channels which are not 2244 explicitly forced into one mode the spice client is allowed to 2245 pick tls/plaintext as he pleases. 2246 2247 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]`` 2248 Configure image compression (lossless). Default is auto\_glz. 2249 2250 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]`` 2251 Configure wan image compression (lossy for slow links). Default 2252 is auto. 2253 2254 ``streaming-video=[off|all|filter]`` 2255 Configure video stream detection. Default is off. 2256 2257 ``agent-mouse=[on|off]`` 2258 Enable/disable passing mouse events via vdagent. Default is on. 2259 2260 ``playback-compression=[on|off]`` 2261 Enable/disable audio stream compression (using celt 0.5.1). 2262 Default is on. 2263 2264 ``seamless-migration=[on|off]`` 2265 Enable/disable spice seamless migration. Default is off. 2266 2267 ``gl=[on|off]`` 2268 Enable/disable OpenGL context. Default is off. 2269 2270 ``rendernode=<file>`` 2271 DRM render node for OpenGL rendering. If not specified, it will 2272 pick the first available. (Since 2.9) 2273ERST 2274 2275DEF("portrait", 0, QEMU_OPTION_portrait, 2276 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n", 2277 QEMU_ARCH_ALL) 2278SRST 2279``-portrait`` 2280 Rotate graphical output 90 deg left (only PXA LCD). 2281ERST 2282 2283DEF("rotate", HAS_ARG, QEMU_OPTION_rotate, 2284 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n", 2285 QEMU_ARCH_ALL) 2286SRST 2287``-rotate deg`` 2288 Rotate graphical output some deg left (only PXA LCD). 2289ERST 2290 2291DEF("vga", HAS_ARG, QEMU_OPTION_vga, 2292 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n" 2293 " select video card type\n", QEMU_ARCH_ALL) 2294SRST 2295``-vga type`` 2296 Select type of VGA card to emulate. Valid values for type are 2297 2298 ``cirrus`` 2299 Cirrus Logic GD5446 Video card. All Windows versions starting 2300 from Windows 95 should recognize and use this graphic card. For 2301 optimal performances, use 16 bit color depth in the guest and 2302 the host OS. (This card was the default before QEMU 2.2) 2303 2304 ``std`` 2305 Standard VGA card with Bochs VBE extensions. If your guest OS 2306 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if 2307 you want to use high resolution modes (>= 1280x1024x16) then you 2308 should use this option. (This card is the default since QEMU 2309 2.2) 2310 2311 ``vmware`` 2312 VMWare SVGA-II compatible adapter. Use it if you have 2313 sufficiently recent XFree86/XOrg server or Windows guest with a 2314 driver for this card. 2315 2316 ``qxl`` 2317 QXL paravirtual graphic card. It is VGA compatible (including 2318 VESA 2.0 VBE support). Works best with qxl guest drivers 2319 installed though. Recommended choice when using the spice 2320 protocol. 2321 2322 ``tcx`` 2323 (sun4m only) Sun TCX framebuffer. This is the default 2324 framebuffer for sun4m machines and offers both 8-bit and 24-bit 2325 colour depths at a fixed resolution of 1024x768. 2326 2327 ``cg3`` 2328 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit 2329 framebuffer for sun4m machines available in both 1024x768 2330 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people 2331 wishing to run older Solaris versions. 2332 2333 ``virtio`` 2334 Virtio VGA card. 2335 2336 ``none`` 2337 Disable VGA card. 2338ERST 2339 2340DEF("full-screen", 0, QEMU_OPTION_full_screen, 2341 "-full-screen start in full screen\n", QEMU_ARCH_ALL) 2342SRST 2343``-full-screen`` 2344 Start in full screen. 2345ERST 2346 2347DEF("g", HAS_ARG, QEMU_OPTION_g , 2348 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n", 2349 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K) 2350SRST 2351``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]`` 2352 Set the initial graphical resolution and depth (PPC, SPARC only). 2353 2354 For PPC the default is 800x600x32. 2355 2356 For SPARC with the TCX graphics device, the default is 1024x768x8 2357 with the option of 1024x768x24. For cgthree, the default is 2358 1024x768x8 with the option of 1152x900x8 for people who wish to use 2359 OBP. 2360ERST 2361 2362DEF("vnc", HAS_ARG, QEMU_OPTION_vnc , 2363 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL) 2364SRST 2365``-vnc display[,option[,option[,...]]]`` 2366 Normally, if QEMU is compiled with graphical window support, it 2367 displays output such as guest graphics, guest console, and the QEMU 2368 monitor in a window. With this option, you can have QEMU listen on 2369 VNC display display and redirect the VGA display over the VNC 2370 session. It is very useful to enable the usb tablet device when 2371 using this option (option ``-device usb-tablet``). When using the 2372 VNC display, you must use the ``-k`` parameter to set the keyboard 2373 layout if you are not using en-us. Valid syntax for the display is 2374 2375 ``to=L`` 2376 With this option, QEMU will try next available VNC displays, 2377 until the number L, if the origianlly defined "-vnc display" is 2378 not available, e.g. port 5900+display is already used by another 2379 application. By default, to=0. 2380 2381 ``host:d`` 2382 TCP connections will only be allowed from host on display d. By 2383 convention the TCP port is 5900+d. Optionally, host can be 2384 omitted in which case the server will accept connections from 2385 any host. 2386 2387 ``unix:path`` 2388 Connections will be allowed over UNIX domain sockets where path 2389 is the location of a unix socket to listen for connections on. 2390 2391 ``none`` 2392 VNC is initialized but not started. The monitor ``change`` 2393 command can be used to later start the VNC server. 2394 2395 Following the display value there may be one or more option flags 2396 separated by commas. Valid options are 2397 2398 ``reverse=on|off`` 2399 Connect to a listening VNC client via a "reverse" connection. 2400 The client is specified by the display. For reverse network 2401 connections (host:d,``reverse``), the d argument is a TCP port 2402 number, not a display number. 2403 2404 ``websocket=on|off`` 2405 Opens an additional TCP listening port dedicated to VNC 2406 Websocket connections. If a bare websocket option is given, the 2407 Websocket port is 5700+display. An alternative port can be 2408 specified with the syntax ``websocket``\ =port. 2409 2410 If host is specified connections will only be allowed from this 2411 host. It is possible to control the websocket listen address 2412 independently, using the syntax ``websocket``\ =host:port. 2413 2414 If no TLS credentials are provided, the websocket connection 2415 runs in unencrypted mode. If TLS credentials are provided, the 2416 websocket connection requires encrypted client connections. 2417 2418 ``password=on|off`` 2419 Require that password based authentication is used for client 2420 connections. 2421 2422 The password must be set separately using the ``set_password`` 2423 command in the :ref:`QEMU monitor`. The 2424 syntax to change your password is: 2425 ``set_password <protocol> <password>`` where <protocol> could be 2426 either "vnc" or "spice". 2427 2428 If you would like to change <protocol> password expiration, you 2429 should use ``expire_password <protocol> <expiration-time>`` 2430 where expiration time could be one of the following options: 2431 now, never, +seconds or UNIX time of expiration, e.g. +60 to 2432 make password expire in 60 seconds, or 1335196800 to make 2433 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for 2434 this date and time). 2435 2436 You can also use keywords "now" or "never" for the expiration 2437 time to allow <protocol> password to expire immediately or never 2438 expire. 2439 2440 ``password-secret=<secret-id>`` 2441 Require that password based authentication is used for client 2442 connections, using the password provided by the ``secret`` 2443 object identified by ``secret-id``. 2444 2445 ``tls-creds=ID`` 2446 Provides the ID of a set of TLS credentials to use to secure the 2447 VNC server. They will apply to both the normal VNC server socket 2448 and the websocket socket (if enabled). Setting TLS credentials 2449 will cause the VNC server socket to enable the VeNCrypt auth 2450 mechanism. The credentials should have been previously created 2451 using the ``-object tls-creds`` argument. 2452 2453 ``tls-authz=ID`` 2454 Provides the ID of the QAuthZ authorization object against which 2455 the client's x509 distinguished name will validated. This object 2456 is only resolved at time of use, so can be deleted and recreated 2457 on the fly while the VNC server is active. If missing, it will 2458 default to denying access. 2459 2460 ``sasl=on|off`` 2461 Require that the client use SASL to authenticate with the VNC 2462 server. The exact choice of authentication method used is 2463 controlled from the system / user's SASL configuration file for 2464 the 'qemu' service. This is typically found in 2465 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user, 2466 an environment variable SASL\_CONF\_PATH can be used to make it 2467 search alternate locations for the service config. While some 2468 SASL auth methods can also provide data encryption (eg GSSAPI), 2469 it is recommended that SASL always be combined with the 'tls' 2470 and 'x509' settings to enable use of SSL and server 2471 certificates. This ensures a data encryption preventing 2472 compromise of authentication credentials. See the 2473 :ref:`VNC security` section in the System Emulation Users Guide 2474 for details on using SASL authentication. 2475 2476 ``sasl-authz=ID`` 2477 Provides the ID of the QAuthZ authorization object against which 2478 the client's SASL username will validated. This object is only 2479 resolved at time of use, so can be deleted and recreated on the 2480 fly while the VNC server is active. If missing, it will default 2481 to denying access. 2482 2483 ``acl=on|off`` 2484 Legacy method for enabling authorization of clients against the 2485 x509 distinguished name and SASL username. It results in the 2486 creation of two ``authz-list`` objects with IDs of 2487 ``vnc.username`` and ``vnc.x509dname``. The rules for these 2488 objects must be configured with the HMP ACL commands. 2489 2490 This option is deprecated and should no longer be used. The new 2491 ``sasl-authz`` and ``tls-authz`` options are a replacement. 2492 2493 ``lossy=on|off`` 2494 Enable lossy compression methods (gradient, JPEG, ...). If this 2495 option is set, VNC client may receive lossy framebuffer updates 2496 depending on its encoding settings. Enabling this option can 2497 save a lot of bandwidth at the expense of quality. 2498 2499 ``non-adaptive=on|off`` 2500 Disable adaptive encodings. Adaptive encodings are enabled by 2501 default. An adaptive encoding will try to detect frequently 2502 updated screen regions, and send updates in these regions using 2503 a lossy encoding (like JPEG). This can be really helpful to save 2504 bandwidth when playing videos. Disabling adaptive encodings 2505 restores the original static behavior of encodings like Tight. 2506 2507 ``share=[allow-exclusive|force-shared|ignore]`` 2508 Set display sharing policy. 'allow-exclusive' allows clients to 2509 ask for exclusive access. As suggested by the rfb spec this is 2510 implemented by dropping other connections. Connecting multiple 2511 clients in parallel requires all clients asking for a shared 2512 session (vncviewer: -shared switch). This is the default. 2513 'force-shared' disables exclusive client access. Useful for 2514 shared desktop sessions, where you don't want someone forgetting 2515 specify -shared disconnect everybody else. 'ignore' completely 2516 ignores the shared flag and allows everybody connect 2517 unconditionally. Doesn't conform to the rfb spec but is 2518 traditional QEMU behavior. 2519 2520 ``key-delay-ms`` 2521 Set keyboard delay, for key down and key up events, in 2522 milliseconds. Default is 10. Keyboards are low-bandwidth 2523 devices, so this slowdown can help the device and guest to keep 2524 up and not lose events in case events are arriving in bulk. 2525 Possible causes for the latter are flaky network connections, or 2526 scripts for automated testing. 2527 2528 ``audiodev=audiodev`` 2529 Use the specified audiodev when the VNC client requests audio 2530 transmission. When not using an -audiodev argument, this option 2531 must be omitted, otherwise is must be present and specify a 2532 valid audiodev. 2533 2534 ``power-control=on|off`` 2535 Permit the remote client to issue shutdown, reboot or reset power 2536 control requests. 2537ERST 2538 2539ARCHHEADING(, QEMU_ARCH_I386) 2540 2541ARCHHEADING(i386 target only:, QEMU_ARCH_I386) 2542 2543DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack, 2544 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n", 2545 QEMU_ARCH_I386) 2546SRST 2547``-win2k-hack`` 2548 Use it when installing Windows 2000 to avoid a disk full bug. After 2549 Windows 2000 is installed, you no longer need this option (this 2550 option slows down the IDE transfers). 2551ERST 2552 2553DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk, 2554 "-no-fd-bootchk disable boot signature checking for floppy disks\n", 2555 QEMU_ARCH_I386) 2556SRST 2557``-no-fd-bootchk`` 2558 Disable boot signature checking for floppy disks in BIOS. May be 2559 needed to boot from old floppy disks. 2560ERST 2561 2562DEF("no-acpi", 0, QEMU_OPTION_no_acpi, 2563 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2564SRST 2565``-no-acpi`` 2566 Disable ACPI (Advanced Configuration and Power Interface) support. 2567 Use it if your guest OS complains about ACPI problems (PC target 2568 machine only). 2569ERST 2570 2571DEF("no-hpet", 0, QEMU_OPTION_no_hpet, 2572 "-no-hpet disable HPET\n", QEMU_ARCH_I386) 2573SRST 2574``-no-hpet`` 2575 Disable HPET support. Deprecated, use '-machine hpet=off' instead. 2576ERST 2577 2578DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable, 2579 "-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" 2580 " ACPI table description\n", QEMU_ARCH_I386) 2581SRST 2582``-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]...]`` 2583 Add ACPI table with specified header fields and context from 2584 specified files. For file=, take whole ACPI table from the specified 2585 files, including all ACPI headers (possible overridden by other 2586 options). For data=, only data portion of the table is used, all 2587 header information is specified in the command line. If a SLIC table 2588 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id 2589 fields will override the same in the RSDT and the FADT (a.k.a. 2590 FACP), in order to ensure the field matches required by the 2591 Microsoft SLIC spec and the ACPI spec. 2592ERST 2593 2594DEF("smbios", HAS_ARG, QEMU_OPTION_smbios, 2595 "-smbios file=binary\n" 2596 " load SMBIOS entry from binary file\n" 2597 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n" 2598 " [,uefi=on|off]\n" 2599 " specify SMBIOS type 0 fields\n" 2600 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2601 " [,uuid=uuid][,sku=str][,family=str]\n" 2602 " specify SMBIOS type 1 fields\n" 2603 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2604 " [,asset=str][,location=str]\n" 2605 " specify SMBIOS type 2 fields\n" 2606 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n" 2607 " [,sku=str]\n" 2608 " specify SMBIOS type 3 fields\n" 2609 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n" 2610 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n" 2611 " [,processor-id=%d]\n" 2612 " specify SMBIOS type 4 fields\n" 2613 "-smbios type=8[,external_reference=str][,internal_reference=str][,connector_type=%d][,port_type=%d]\n" 2614 " specify SMBIOS type 8 fields\n" 2615 "-smbios type=11[,value=str][,path=filename]\n" 2616 " specify SMBIOS type 11 fields\n" 2617 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n" 2618 " [,asset=str][,part=str][,speed=%d]\n" 2619 " specify SMBIOS type 17 fields\n" 2620 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n" 2621 " specify SMBIOS type 41 fields\n", 2622 QEMU_ARCH_I386 | QEMU_ARCH_ARM | QEMU_ARCH_LOONGARCH) 2623SRST 2624``-smbios file=binary`` 2625 Load SMBIOS entry from binary file. 2626 2627``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]`` 2628 Specify SMBIOS type 0 fields 2629 2630``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]`` 2631 Specify SMBIOS type 1 fields 2632 2633``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]`` 2634 Specify SMBIOS type 2 fields 2635 2636``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]`` 2637 Specify SMBIOS type 3 fields 2638 2639``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-id=%d]`` 2640 Specify SMBIOS type 4 fields 2641 2642``-smbios type=11[,value=str][,path=filename]`` 2643 Specify SMBIOS type 11 fields 2644 2645 This argument can be repeated multiple times, and values are added in the order they are parsed. 2646 Applications intending to use OEM strings data are encouraged to use their application name as 2647 a prefix for the value string. This facilitates passing information for multiple applications 2648 concurrently. 2649 2650 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax 2651 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes. 2652 2653 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to 2654 the SMBIOS table in the order in which they appear. 2655 2656 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535 2657 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the 2658 guest. Instead it should be used as a indicator to inform the guest where to locate the real 2659 data set, for example, by specifying the serial ID of a block device. 2660 2661 An example passing three strings is 2662 2663 .. parsed-literal:: 2664 2665 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\ 2666 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\ 2667 path=/some/file/with/oemstringsdata.txt 2668 2669 In the guest OS this is visible with the ``dmidecode`` command 2670 2671 .. parsed-literal:: 2672 2673 $ dmidecode -t 11 2674 Handle 0x0E00, DMI type 11, 5 bytes 2675 OEM Strings 2676 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/ 2677 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os 2678 String 3: myapp:some extra data 2679 2680 2681``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]`` 2682 Specify SMBIOS type 17 fields 2683 2684``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]`` 2685 Specify SMBIOS type 41 fields 2686 2687 This argument can be repeated multiple times. Its main use is to allow network interfaces be created 2688 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface 2689 position on the PCI bus. 2690 2691 Here is an example of use: 2692 2693 .. parsed-literal:: 2694 2695 -netdev user,id=internet \\ 2696 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\ 2697 -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev 2698 2699 In the guest OS, the device should then appear as ``eno1``: 2700 2701 ..parsed-literal:: 2702 2703 $ ip -brief l 2704 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP> 2705 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP> 2706 2707 Currently, the PCI device has to be attached to the root bus. 2708 2709ERST 2710 2711DEFHEADING() 2712 2713DEFHEADING(Network options:) 2714 2715DEF("netdev", HAS_ARG, QEMU_OPTION_netdev, 2716#ifdef CONFIG_SLIRP 2717 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n" 2718 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n" 2719 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n" 2720 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n" 2721 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]" 2722#ifndef _WIN32 2723 "[,smb=dir[,smbserver=addr]]\n" 2724#endif 2725 " configure a user mode network backend with ID 'str',\n" 2726 " its DHCP server and optional services\n" 2727#endif 2728#ifdef _WIN32 2729 "-netdev tap,id=str,ifname=name\n" 2730 " configure a host TAP network backend with ID 'str'\n" 2731#else 2732 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n" 2733 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n" 2734 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n" 2735 " [,poll-us=n]\n" 2736 " configure a host TAP network backend with ID 'str'\n" 2737 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2738 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n" 2739 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n" 2740 " to deconfigure it\n" 2741 " use '[down]script=no' to disable script execution\n" 2742 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n" 2743 " configure it\n" 2744 " use 'fd=h' to connect to an already opened TAP interface\n" 2745 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n" 2746 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n" 2747 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n" 2748 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n" 2749 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n" 2750 " use vhost=on to enable experimental in kernel accelerator\n" 2751 " (only has effect for virtio guests which use MSIX)\n" 2752 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n" 2753 " use 'vhostfd=h' to connect to an already opened vhost net device\n" 2754 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n" 2755 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n" 2756 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n" 2757 " spent on busy polling for vhost net\n" 2758 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n" 2759 " configure a host TAP network backend with ID 'str' that is\n" 2760 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2761 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n" 2762#endif 2763#ifdef __linux__ 2764 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n" 2765 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n" 2766 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n" 2767 " [,rxcookie=rxcookie][,offset=offset]\n" 2768 " configure a network backend with ID 'str' connected to\n" 2769 " an Ethernet over L2TPv3 pseudowire.\n" 2770 " Linux kernel 3.3+ as well as most routers can talk\n" 2771 " L2TPv3. This transport allows connecting a VM to a VM,\n" 2772 " VM to a router and even VM to Host. It is a nearly-universal\n" 2773 " standard (RFC3931). Note - this implementation uses static\n" 2774 " pre-configured tunnels (same as the Linux kernel).\n" 2775 " use 'src=' to specify source address\n" 2776 " use 'dst=' to specify destination address\n" 2777 " use 'udp=on' to specify udp encapsulation\n" 2778 " use 'srcport=' to specify source udp port\n" 2779 " use 'dstport=' to specify destination udp port\n" 2780 " use 'ipv6=on' to force v6\n" 2781 " L2TPv3 uses cookies to prevent misconfiguration as\n" 2782 " well as a weak security measure\n" 2783 " use 'rxcookie=0x012345678' to specify a rxcookie\n" 2784 " use 'txcookie=0x012345678' to specify a txcookie\n" 2785 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n" 2786 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n" 2787 " use 'pincounter=on' to work around broken counter handling in peer\n" 2788 " use 'offset=X' to add an extra offset between header and data\n" 2789#endif 2790 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n" 2791 " configure a network backend to connect to another network\n" 2792 " using a socket connection\n" 2793 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n" 2794 " configure a network backend to connect to a multicast maddr and port\n" 2795 " use 'localaddr=addr' to specify the host address to send packets from\n" 2796 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n" 2797 " configure a network backend to connect to another network\n" 2798 " using an UDP tunnel\n" 2799 "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port[,to=maxport][,numeric=on|off][,keep-alive=on|off][,mptcp=on|off][,addr.ipv4=on|off][,addr.ipv6=on|off][,reconnect=seconds]\n" 2800 "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off][,reconnect=seconds]\n" 2801 "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor[,reconnect=seconds]\n" 2802 " configure a network backend to connect to another network\n" 2803 " using a socket connection in stream mode.\n" 2804 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n" 2805 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]\n" 2806 " configure a network backend to connect to a multicast maddr and port\n" 2807 " use ``local.host=addr`` to specify the host address to send packets from\n" 2808 "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n" 2809 "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n" 2810 "-netdev dgram,id=str,local.type=fd,local.str=file-descriptor\n" 2811 " configure a network backend to connect to another network\n" 2812 " using an UDP tunnel\n" 2813#ifdef CONFIG_VDE 2814 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n" 2815 " configure a network backend to connect to port 'n' of a vde switch\n" 2816 " running on host and listening for incoming connections on 'socketpath'.\n" 2817 " Use group 'groupname' and mode 'octalmode' to change default\n" 2818 " ownership and permissions for communication port.\n" 2819#endif 2820#ifdef CONFIG_NETMAP 2821 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n" 2822 " attach to the existing netmap-enabled network interface 'name', or to a\n" 2823 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n" 2824 " netmap device, defaults to '/dev/netmap')\n" 2825#endif 2826#ifdef CONFIG_POSIX 2827 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n" 2828 " configure a vhost-user network, backed by a chardev 'dev'\n" 2829#endif 2830#ifdef __linux__ 2831 "-netdev vhost-vdpa,id=str[,vhostdev=/path/to/dev][,vhostfd=h]\n" 2832 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n" 2833 " use 'vhostdev=/path/to/dev' to open a vhost vdpa device\n" 2834 " use 'vhostfd=h' to connect to an already opened vhost vdpa device\n" 2835#endif 2836#ifdef CONFIG_VMNET 2837 "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n" 2838 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" 2839 " configure a vmnet network backend in host mode with ID 'str',\n" 2840 " isolate this interface from others with 'isolated',\n" 2841 " configure the address range and choose a subnet mask,\n" 2842 " specify network UUID 'uuid' to disable DHCP and interact with\n" 2843 " vmnet-host interfaces within this isolated network\n" 2844 "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n" 2845 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" 2846 " configure a vmnet network backend in shared mode with ID 'str',\n" 2847 " configure the address range and choose a subnet mask,\n" 2848 " set IPv6 ULA prefix (of length 64) to use for internal network,\n" 2849 " isolate this interface from others with 'isolated'\n" 2850 "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n" 2851 " configure a vmnet network backend in bridged mode with ID 'str',\n" 2852 " use 'ifname=name' to select a physical network interface to be bridged,\n" 2853 " isolate this interface from others with 'isolated'\n" 2854#endif 2855 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n" 2856 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL) 2857DEF("nic", HAS_ARG, QEMU_OPTION_nic, 2858 "-nic [tap|bridge|" 2859#ifdef CONFIG_SLIRP 2860 "user|" 2861#endif 2862#ifdef __linux__ 2863 "l2tpv3|" 2864#endif 2865#ifdef CONFIG_VDE 2866 "vde|" 2867#endif 2868#ifdef CONFIG_NETMAP 2869 "netmap|" 2870#endif 2871#ifdef CONFIG_POSIX 2872 "vhost-user|" 2873#endif 2874#ifdef CONFIG_VMNET 2875 "vmnet-host|vmnet-shared|vmnet-bridged|" 2876#endif 2877 "socket][,option][,...][mac=macaddr]\n" 2878 " initialize an on-board / default host NIC (using MAC address\n" 2879 " macaddr) and connect it to the given host network backend\n" 2880 "-nic none use it alone to have zero network devices (the default is to\n" 2881 " provided a 'user' network connection)\n", 2882 QEMU_ARCH_ALL) 2883DEF("net", HAS_ARG, QEMU_OPTION_net, 2884 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n" 2885 " configure or create an on-board (or machine default) NIC and\n" 2886 " connect it to hub 0 (please use -nic unless you need a hub)\n" 2887 "-net [" 2888#ifdef CONFIG_SLIRP 2889 "user|" 2890#endif 2891 "tap|" 2892 "bridge|" 2893#ifdef CONFIG_VDE 2894 "vde|" 2895#endif 2896#ifdef CONFIG_NETMAP 2897 "netmap|" 2898#endif 2899#ifdef CONFIG_VMNET 2900 "vmnet-host|vmnet-shared|vmnet-bridged|" 2901#endif 2902 "socket][,option][,option][,...]\n" 2903 " old way to initialize a host network interface\n" 2904 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL) 2905SRST 2906``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]`` 2907 This option is a shortcut for configuring both the on-board 2908 (default) guest NIC hardware and the host network backend in one go. 2909 The host backend options are the same as with the corresponding 2910 ``-netdev`` options below. The guest NIC model can be set with 2911 ``model=modelname``. Use ``model=help`` to list the available device 2912 types. The hardware MAC address can be set with ``mac=macaddr``. 2913 2914 The following two example do exactly the same, to show how ``-nic`` 2915 can be used to shorten the command line length: 2916 2917 .. parsed-literal:: 2918 2919 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32 2920 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32 2921 2922``-nic none`` 2923 Indicate that no network devices should be configured. It is used to 2924 override the default configuration (default NIC with "user" host 2925 network backend) which is activated if no other networking options 2926 are provided. 2927 2928``-netdev user,id=id[,option][,option][,...]`` 2929 Configure user mode host network backend which requires no 2930 administrator privilege to run. Valid options are: 2931 2932 ``id=id`` 2933 Assign symbolic name for use in monitor commands. 2934 2935 ``ipv4=on|off and ipv6=on|off`` 2936 Specify that either IPv4 or IPv6 must be enabled. If neither is 2937 specified both protocols are enabled. 2938 2939 ``net=addr[/mask]`` 2940 Set IP network address the guest will see. Optionally specify 2941 the netmask, either in the form a.b.c.d or as number of valid 2942 top-most bits. Default is 10.0.2.0/24. 2943 2944 ``host=addr`` 2945 Specify the guest-visible address of the host. Default is the 2946 2nd IP in the guest network, i.e. x.x.x.2. 2947 2948 ``ipv6-net=addr[/int]`` 2949 Set IPv6 network address the guest will see (default is 2950 fec0::/64). The network prefix is given in the usual hexadecimal 2951 IPv6 address notation. The prefix size is optional, and is given 2952 as the number of valid top-most bits (default is 64). 2953 2954 ``ipv6-host=addr`` 2955 Specify the guest-visible IPv6 address of the host. Default is 2956 the 2nd IPv6 in the guest network, i.e. xxxx::2. 2957 2958 ``restrict=on|off`` 2959 If this option is enabled, the guest will be isolated, i.e. it 2960 will not be able to contact the host and no guest IP packets 2961 will be routed over the host to the outside. This option does 2962 not affect any explicitly set forwarding rules. 2963 2964 ``hostname=name`` 2965 Specifies the client hostname reported by the built-in DHCP 2966 server. 2967 2968 ``dhcpstart=addr`` 2969 Specify the first of the 16 IPs the built-in DHCP server can 2970 assign. Default is the 15th to 31st IP in the guest network, 2971 i.e. x.x.x.15 to x.x.x.31. 2972 2973 ``dns=addr`` 2974 Specify the guest-visible address of the virtual nameserver. The 2975 address must be different from the host address. Default is the 2976 3rd IP in the guest network, i.e. x.x.x.3. 2977 2978 ``ipv6-dns=addr`` 2979 Specify the guest-visible address of the IPv6 virtual 2980 nameserver. The address must be different from the host address. 2981 Default is the 3rd IP in the guest network, i.e. xxxx::3. 2982 2983 ``dnssearch=domain`` 2984 Provides an entry for the domain-search list sent by the 2985 built-in DHCP server. More than one domain suffix can be 2986 transmitted by specifying this option multiple times. If 2987 supported, this will cause the guest to automatically try to 2988 append the given domain suffix(es) in case a domain name can not 2989 be resolved. 2990 2991 Example: 2992 2993 .. parsed-literal:: 2994 2995 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org 2996 2997 ``domainname=domain`` 2998 Specifies the client domain name reported by the built-in DHCP 2999 server. 3000 3001 ``tftp=dir`` 3002 When using the user mode network stack, activate a built-in TFTP 3003 server. The files in dir will be exposed as the root of a TFTP 3004 server. The TFTP client on the guest must be configured in 3005 binary mode (use the command ``bin`` of the Unix TFTP client). 3006 3007 ``tftp-server-name=name`` 3008 In BOOTP reply, broadcast name as the "TFTP server name" 3009 (RFC2132 option 66). This can be used to advise the guest to 3010 load boot files or configurations from a different server than 3011 the host address. 3012 3013 ``bootfile=file`` 3014 When using the user mode network stack, broadcast file as the 3015 BOOTP filename. In conjunction with ``tftp``, this can be used 3016 to network boot a guest from a local directory. 3017 3018 Example (using pxelinux): 3019 3020 .. parsed-literal:: 3021 3022 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\ 3023 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0 3024 3025 ``smb=dir[,smbserver=addr]`` 3026 When using the user mode network stack, activate a built-in SMB 3027 server so that Windows OSes can access to the host files in 3028 ``dir`` transparently. The IP address of the SMB server can be 3029 set to addr. By default the 4th IP in the guest network is used, 3030 i.e. x.x.x.4. 3031 3032 In the guest Windows OS, the line: 3033 3034 :: 3035 3036 10.0.2.4 smbserver 3037 3038 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows 3039 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows 3040 NT/2000). 3041 3042 Then ``dir`` can be accessed in ``\\smbserver\qemu``. 3043 3044 Note that a SAMBA server must be installed on the host OS. 3045 3046 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport`` 3047 Redirect incoming TCP or UDP connections to the host port 3048 hostport to the guest IP address guestaddr on guest port 3049 guestport. If guestaddr is not specified, its value is x.x.x.15 3050 (default first address given by the built-in DHCP server). By 3051 specifying hostaddr, the rule can be bound to a specific host 3052 interface. If no connection type is set, TCP is used. This 3053 option can be given multiple times. 3054 3055 For example, to redirect host X11 connection from screen 1 to 3056 guest screen 0, use the following: 3057 3058 .. parsed-literal:: 3059 3060 # on the host 3061 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000 3062 # this host xterm should open in the guest X11 server 3063 xterm -display :1 3064 3065 To redirect telnet connections from host port 5555 to telnet 3066 port on the guest, use the following: 3067 3068 .. parsed-literal:: 3069 3070 # on the host 3071 |qemu_system| -nic user,hostfwd=tcp::5555-:23 3072 telnet localhost 5555 3073 3074 Then when you use on the host ``telnet localhost 5555``, you 3075 connect to the guest telnet server. 3076 3077 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command`` 3078 Forward guest TCP connections to the IP address server on port 3079 port to the character device dev or to a program executed by 3080 cmd:command which gets spawned for each connection. This option 3081 can be given multiple times. 3082 3083 You can either use a chardev directly and have that one used 3084 throughout QEMU's lifetime, like in the following example: 3085 3086 .. parsed-literal:: 3087 3088 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever 3089 # the guest accesses it 3090 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321 3091 3092 Or you can execute a command on every TCP connection established 3093 by the guest, so that QEMU behaves similar to an inetd process 3094 for that virtual server: 3095 3096 .. parsed-literal:: 3097 3098 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234 3099 # and connect the TCP stream to its stdin/stdout 3100 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321' 3101 3102``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]`` 3103 Configure a host TAP network backend with ID id. 3104 3105 Use the network script file to configure it and the network script 3106 dfile to deconfigure it. If name is not provided, the OS 3107 automatically provides one. The default network configure script is 3108 ``/etc/qemu-ifup`` and the default network deconfigure script is 3109 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to 3110 disable script execution. 3111 3112 If running QEMU as an unprivileged user, use the network helper 3113 to configure the TAP interface and attach it to the bridge. 3114 The default network helper executable is 3115 ``/path/to/qemu-bridge-helper`` and the default bridge device is 3116 ``br0``. 3117 3118 ``fd``\ =h can be used to specify the handle of an already opened 3119 host TAP interface. 3120 3121 Examples: 3122 3123 .. parsed-literal:: 3124 3125 #launch a QEMU instance with the default network script 3126 |qemu_system| linux.img -nic tap 3127 3128 .. parsed-literal:: 3129 3130 #launch a QEMU instance with two NICs, each one connected 3131 #to a TAP device 3132 |qemu_system| linux.img \\ 3133 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\ 3134 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1 3135 3136 .. parsed-literal:: 3137 3138 #launch a QEMU instance with the default network helper to 3139 #connect a TAP device to bridge br0 3140 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 3141 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper" 3142 3143``-netdev bridge,id=id[,br=bridge][,helper=helper]`` 3144 Connect a host TAP network interface to a host bridge device. 3145 3146 Use the network helper helper to configure the TAP interface and 3147 attach it to the bridge. The default network helper executable is 3148 ``/path/to/qemu-bridge-helper`` and the default bridge device is 3149 ``br0``. 3150 3151 Examples: 3152 3153 .. parsed-literal:: 3154 3155 #launch a QEMU instance with the default network helper to 3156 #connect a TAP device to bridge br0 3157 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1 3158 3159 .. parsed-literal:: 3160 3161 #launch a QEMU instance with the default network helper to 3162 #connect a TAP device to bridge qemubr0 3163 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1 3164 3165``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]`` 3166 This host network backend can be used to connect the guest's network 3167 to another QEMU virtual machine using a TCP socket connection. If 3168 ``listen`` is specified, QEMU waits for incoming connections on port 3169 (host is optional). ``connect`` is used to connect to another QEMU 3170 instance using the ``listen`` option. ``fd``\ =h specifies an 3171 already opened TCP socket. 3172 3173 Example: 3174 3175 .. parsed-literal:: 3176 3177 # launch a first QEMU instance 3178 |qemu_system| linux.img \\ 3179 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3180 -netdev socket,id=n1,listen=:1234 3181 # connect the network of this instance to the network of the first instance 3182 |qemu_system| linux.img \\ 3183 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3184 -netdev socket,id=n2,connect=127.0.0.1:1234 3185 3186``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]`` 3187 Configure a socket host network backend to share the guest's network 3188 traffic with another QEMU virtual machines using a UDP multicast 3189 socket, effectively making a bus for every QEMU with same multicast 3190 address maddr and port. NOTES: 3191 3192 1. Several QEMU can be running on different hosts and share same bus 3193 (assuming correct multicast setup for these hosts). 3194 3195 2. mcast support is compatible with User Mode Linux (argument 3196 ``ethN=mcast``), see http://user-mode-linux.sf.net. 3197 3198 3. Use ``fd=h`` to specify an already opened UDP multicast socket. 3199 3200 Example: 3201 3202 .. parsed-literal:: 3203 3204 # launch one QEMU instance 3205 |qemu_system| linux.img \\ 3206 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3207 -netdev socket,id=n1,mcast=230.0.0.1:1234 3208 # launch another QEMU instance on same "bus" 3209 |qemu_system| linux.img \\ 3210 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3211 -netdev socket,id=n2,mcast=230.0.0.1:1234 3212 # launch yet another QEMU instance on same "bus" 3213 |qemu_system| linux.img \\ 3214 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\ 3215 -netdev socket,id=n3,mcast=230.0.0.1:1234 3216 3217 Example (User Mode Linux compat.): 3218 3219 .. parsed-literal:: 3220 3221 # launch QEMU instance (note mcast address selected is UML's default) 3222 |qemu_system| linux.img \\ 3223 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3224 -netdev socket,id=n1,mcast=239.192.168.1:1102 3225 # launch UML 3226 /path/to/linux ubd0=/path/to/root_fs eth0=mcast 3227 3228 Example (send packets from host's 1.2.3.4): 3229 3230 .. parsed-literal:: 3231 3232 |qemu_system| linux.img \\ 3233 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3234 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4 3235 3236``-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]`` 3237 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931) 3238 is a popular protocol to transport Ethernet (and other Layer 2) data 3239 frames between two systems. It is present in routers, firewalls and 3240 the Linux kernel (from version 3.3 onwards). 3241 3242 This transport allows a VM to communicate to another VM, router or 3243 firewall directly. 3244 3245 ``src=srcaddr`` 3246 source address (mandatory) 3247 3248 ``dst=dstaddr`` 3249 destination address (mandatory) 3250 3251 ``udp`` 3252 select udp encapsulation (default is ip). 3253 3254 ``srcport=srcport`` 3255 source udp port. 3256 3257 ``dstport=dstport`` 3258 destination udp port. 3259 3260 ``ipv6`` 3261 force v6, otherwise defaults to v4. 3262 3263 ``rxcookie=rxcookie``; \ ``txcookie=txcookie`` 3264 Cookies are a weak form of security in the l2tpv3 specification. 3265 Their function is mostly to prevent misconfiguration. By default 3266 they are 32 bit. 3267 3268 ``cookie64`` 3269 Set cookie size to 64 bit instead of the default 32 3270 3271 ``counter=off`` 3272 Force a 'cut-down' L2TPv3 with no counter as in 3273 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00 3274 3275 ``pincounter=on`` 3276 Work around broken counter handling in peer. This may also help 3277 on networks which have packet reorder. 3278 3279 ``offset=offset`` 3280 Add an extra offset between header and data 3281 3282 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to 3283 the bridge br-lan on the remote Linux host 1.2.3.4: 3284 3285 .. parsed-literal:: 3286 3287 # Setup tunnel on linux host using raw ip as encapsulation 3288 # on 1.2.3.4 3289 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\ 3290 encap udp udp_sport 16384 udp_dport 16384 3291 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\ 3292 0xFFFFFFFF peer_session_id 0xFFFFFFFF 3293 ifconfig vmtunnel0 mtu 1500 3294 ifconfig vmtunnel0 up 3295 brctl addif br-lan vmtunnel0 3296 3297 3298 # on 4.3.2.1 3299 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter 3300 3301 |qemu_system| linux.img -device e1000,netdev=n1 \\ 3302 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter 3303 3304``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]`` 3305 Configure VDE backend to connect to PORT n of a vde switch running 3306 on host and listening for incoming connections on socketpath. Use 3307 GROUP groupname and MODE octalmode to change default ownership and 3308 permissions for communication port. This option is only available if 3309 QEMU has been compiled with vde support enabled. 3310 3311 Example: 3312 3313 .. parsed-literal:: 3314 3315 # launch vde switch 3316 vde_switch -F -sock /tmp/myswitch 3317 # launch QEMU instance 3318 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch 3319 3320``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]`` 3321 Establish a vhost-user netdev, backed by a chardev id. The chardev 3322 should be a unix domain socket backed one. The vhost-user uses a 3323 specifically defined protocol to pass vhost ioctl replacement 3324 messages to an application on the other end of the socket. On 3325 non-MSIX guests, the feature can be forced with vhostforce. Use 3326 'queues=n' to specify the number of queues to be created for 3327 multiqueue vhost-user. 3328 3329 Example: 3330 3331 :: 3332 3333 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \ 3334 -numa node,memdev=mem \ 3335 -chardev socket,id=chr0,path=/path/to/socket \ 3336 -netdev type=vhost-user,id=net0,chardev=chr0 \ 3337 -device virtio-net-pci,netdev=net0 3338 3339``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]`` 3340 Establish a vhost-vdpa netdev. 3341 3342 vDPA device is a device that uses a datapath which complies with 3343 the virtio specifications with a vendor specific control path. 3344 vDPA devices can be both physically located on the hardware or 3345 emulated by software. 3346 3347``-netdev hubport,id=id,hubid=hubid[,netdev=nd]`` 3348 Create a hub port on the emulated hub with ID hubid. 3349 3350 The hubport netdev lets you connect a NIC to a QEMU emulated hub 3351 instead of a single netdev. Alternatively, you can also connect the 3352 hubport to another netdev with ID nd by using the ``netdev=nd`` 3353 option. 3354 3355``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]`` 3356 Legacy option to configure or create an on-board (or machine 3357 default) Network Interface Card(NIC) and connect it either to the 3358 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd. 3359 If model is omitted, then the default NIC model associated with the 3360 machine type is used. Note that the default NIC model may change in 3361 future QEMU releases, so it is highly recommended to always specify 3362 a model. Optionally, the MAC address can be changed to mac, the 3363 device address set to addr (PCI cards only), and a name can be 3364 assigned for use in monitor commands. Optionally, for PCI cards, you 3365 can specify the number v of MSI-X vectors that the card should have; 3366 this option currently only affects virtio cards; set v = 0 to 3367 disable MSI-X. If no ``-net`` option is specified, a single NIC is 3368 created. QEMU can emulate several different models of network card. 3369 Use ``-net nic,model=help`` for a list of available devices for your 3370 target. 3371 3372``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]`` 3373 Configure a host network backend (with the options corresponding to 3374 the same ``-netdev`` option) and connect it to the emulated hub 0 3375 (the default hub). Use name to specify the name of the hub port. 3376ERST 3377 3378DEFHEADING() 3379 3380DEFHEADING(Character device options:) 3381 3382DEF("chardev", HAS_ARG, QEMU_OPTION_chardev, 3383 "-chardev help\n" 3384 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3385 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n" 3386 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n" 3387 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n" 3388 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n" 3389 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n" 3390 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n" 3391 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n" 3392 " [,logfile=PATH][,logappend=on|off]\n" 3393 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3394 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n" 3395 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3396 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n" 3397 "-chardev file,id=id,path=path[,input-path=input-file][,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3398 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3399#ifdef _WIN32 3400 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3401 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3402#else 3403 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3404 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n" 3405#endif 3406#ifdef CONFIG_BRLAPI 3407 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3408#endif 3409#if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \ 3410 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) 3411 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3412#endif 3413#if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__) 3414 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3415#endif 3416#if defined(CONFIG_SPICE) 3417 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3418 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3419#endif 3420 , QEMU_ARCH_ALL 3421) 3422 3423SRST 3424The general form of a character device option is: 3425 3426``-chardev backend,id=id[,mux=on|off][,options]`` 3427 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``, 3428 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``, 3429 ``pty``, ``stdio``, ``braille``, ``parallel``, 3430 ``spicevmc``, ``spiceport``. The specific backend will determine the 3431 applicable options. 3432 3433 Use ``-chardev help`` to print all available chardev backend types. 3434 3435 All devices must have an id, which can be any string up to 127 3436 characters long. It is used to uniquely identify this device in 3437 other command line directives. 3438 3439 A character device may be used in multiplexing mode by multiple 3440 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is 3441 a "1:N" device, and here the "1" end is your specified chardev 3442 backend, and the "N" end is the various parts of QEMU that can talk 3443 to a chardev. If you create a chardev with ``id=myid`` and 3444 ``mux=on``, QEMU will create a multiplexer with your specified ID, 3445 and you can then configure multiple front ends to use that chardev 3446 ID for their input/output. Up to four different front ends can be 3447 connected to a single multiplexed chardev. (Without multiplexing 3448 enabled, a chardev can only be used by a single front end.) For 3449 instance you could use this to allow a single stdio chardev to be 3450 used by two serial ports and the QEMU monitor: 3451 3452 :: 3453 3454 -chardev stdio,mux=on,id=char0 \ 3455 -mon chardev=char0,mode=readline \ 3456 -serial chardev:char0 \ 3457 -serial chardev:char0 3458 3459 You can have more than one multiplexer in a system configuration; 3460 for instance you could have a TCP port multiplexed between UART 0 3461 and UART 1, and stdio multiplexed between the QEMU monitor and a 3462 parallel port: 3463 3464 :: 3465 3466 -chardev stdio,mux=on,id=char0 \ 3467 -mon chardev=char0,mode=readline \ 3468 -parallel chardev:char0 \ 3469 -chardev tcp,...,mux=on,id=char1 \ 3470 -serial chardev:char1 \ 3471 -serial chardev:char1 3472 3473 When you're using a multiplexed character device, some escape 3474 sequences are interpreted in the input. See the chapter about 3475 :ref:`keys in the character backend multiplexer` in the 3476 System Emulation Users Guide for more details. 3477 3478 Note that some other command line options may implicitly create 3479 multiplexed character backends; for instance ``-serial mon:stdio`` 3480 creates a multiplexed stdio backend connected to the serial port and 3481 the QEMU monitor, and ``-nographic`` also multiplexes the console 3482 and the monitor to stdio. 3483 3484 There is currently no support for multiplexing in the other 3485 direction (where a single QEMU front end takes input and output from 3486 multiple chardevs). 3487 3488 Every backend supports the ``logfile`` option, which supplies the 3489 path to a file to record all data transmitted via the backend. The 3490 ``logappend`` option controls whether the log file will be truncated 3491 or appended to when opened. 3492 3493The available backends are: 3494 3495``-chardev null,id=id`` 3496 A void device. This device will not emit any data, and will drop any 3497 data it receives. The null backend does not take any options. 3498 3499``-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]`` 3500 Create a two-way stream socket, which can be either a TCP or a unix 3501 socket. A unix socket will be created if ``path`` is specified. 3502 Behaviour is undefined if TCP options are specified for a unix 3503 socket. 3504 3505 ``server=on|off`` specifies that the socket shall be a listening socket. 3506 3507 ``wait=on|off`` specifies that QEMU should not block waiting for a client 3508 to connect to a listening socket. 3509 3510 ``telnet=on|off`` specifies that traffic on the socket should interpret 3511 telnet escape sequences. 3512 3513 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for 3514 communication. 3515 3516 ``reconnect`` sets the timeout for reconnecting on non-server 3517 sockets when the remote end goes away. qemu will delay this many 3518 seconds and then attempt to reconnect. Zero disables reconnecting, 3519 and is the default. 3520 3521 ``tls-creds`` requests enablement of the TLS protocol for 3522 encryption, and specifies the id of the TLS credentials to use for 3523 the handshake. The credentials must be previously created with the 3524 ``-object tls-creds`` argument. 3525 3526 ``tls-auth`` provides the ID of the QAuthZ authorization object 3527 against which the client's x509 distinguished name will be 3528 validated. This object is only resolved at time of use, so can be 3529 deleted and recreated on the fly while the chardev server is active. 3530 If missing, it will default to denying access. 3531 3532 TCP and unix socket options are given below: 3533 3534 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]`` 3535 ``host`` for a listening socket specifies the local address to 3536 be bound. For a connecting socket species the remote host to 3537 connect to. ``host`` is optional for listening sockets. If not 3538 specified it defaults to ``0.0.0.0``. 3539 3540 ``port`` for a listening socket specifies the local port to be 3541 bound. For a connecting socket specifies the port on the remote 3542 host to connect to. ``port`` can be given as either a port 3543 number or a service name. ``port`` is required. 3544 3545 ``to`` is only relevant to listening sockets. If it is 3546 specified, and ``port`` cannot be bound, QEMU will attempt to 3547 bind to subsequent ports up to and including ``to`` until it 3548 succeeds. ``to`` must be specified as a port number. 3549 3550 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 3551 or IPv6 must be used. If neither is specified the socket may 3552 use either protocol. 3553 3554 ``nodelay=on|off`` disables the Nagle algorithm. 3555 3556 ``unix options: path=path[,abstract=on|off][,tight=on|off]`` 3557 ``path`` specifies the local path of the unix socket. ``path`` 3558 is required. 3559 ``abstract=on|off`` specifies the use of the abstract socket namespace, 3560 rather than the filesystem. Optional, defaults to false. 3561 ``tight=on|off`` sets the socket length of abstract sockets to their minimum, 3562 rather than the full sun_path length. Optional, defaults to true. 3563 3564``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]`` 3565 Sends all traffic from the guest to a remote host over UDP. 3566 3567 ``host`` specifies the remote host to connect to. If not specified 3568 it defaults to ``localhost``. 3569 3570 ``port`` specifies the port on the remote host to connect to. 3571 ``port`` is required. 3572 3573 ``localaddr`` specifies the local address to bind to. If not 3574 specified it defaults to ``0.0.0.0``. 3575 3576 ``localport`` specifies the local port to bind to. If not specified 3577 any available local port will be used. 3578 3579 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used. 3580 If neither is specified the device may use either protocol. 3581 3582``-chardev msmouse,id=id`` 3583 Forward QEMU's emulated msmouse events to the guest. ``msmouse`` 3584 does not take any options. 3585 3586``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]`` 3587 Connect to a QEMU text console. ``vc`` may optionally be given a 3588 specific size. 3589 3590 ``width`` and ``height`` specify the width and height respectively 3591 of the console, in pixels. 3592 3593 ``cols`` and ``rows`` specify that the console be sized to fit a 3594 text console with the given dimensions. 3595 3596``-chardev ringbuf,id=id[,size=size]`` 3597 Create a ring buffer with fixed size ``size``. size must be a power 3598 of two and defaults to ``64K``. 3599 3600``-chardev file,id=id,path=path[,input-path=input-path]`` 3601 Log all traffic received from the guest to a file. 3602 3603 ``path`` specifies the path of the file to be opened. This file will 3604 be created if it does not already exist, and overwritten if it does. 3605 ``path`` is required. 3606 3607 If ``input-path`` is specified, this is the path of a second file 3608 which will be used for input. If ``input-path`` is not specified, 3609 no input will be available from the chardev. 3610 3611 Note that ``input-path`` is not supported on Windows hosts. 3612 3613``-chardev pipe,id=id,path=path`` 3614 Create a two-way connection to the guest. The behaviour differs 3615 slightly between Windows hosts and other hosts: 3616 3617 On Windows, a single duplex pipe will be created at 3618 ``\\.pipe\path``. 3619 3620 On other hosts, 2 pipes will be created called ``path.in`` and 3621 ``path.out``. Data written to ``path.in`` will be received by the 3622 guest. Data written by the guest can be read from ``path.out``. QEMU 3623 will not create these fifos, and requires them to be present. 3624 3625 ``path`` forms part of the pipe path as described above. ``path`` is 3626 required. 3627 3628``-chardev console,id=id`` 3629 Send traffic from the guest to QEMU's standard output. ``console`` 3630 does not take any options. 3631 3632 ``console`` is only available on Windows hosts. 3633 3634``-chardev serial,id=id,path=path`` 3635 Send traffic from the guest to a serial device on the host. 3636 3637 On Unix hosts serial will actually accept any tty device, not only 3638 serial lines. 3639 3640 ``path`` specifies the name of the serial device to open. 3641 3642``-chardev pty,id=id`` 3643 Create a new pseudo-terminal on the host and connect to it. ``pty`` 3644 does not take any options. 3645 3646 ``pty`` is not available on Windows hosts. 3647 3648``-chardev stdio,id=id[,signal=on|off]`` 3649 Connect to standard input and standard output of the QEMU process. 3650 3651 ``signal`` controls if signals are enabled on the terminal, that 3652 includes exiting QEMU with the key sequence Control-c. This option 3653 is enabled by default, use ``signal=off`` to disable it. 3654 3655``-chardev braille,id=id`` 3656 Connect to a local BrlAPI server. ``braille`` does not take any 3657 options. 3658 3659``-chardev parallel,id=id,path=path`` 3660 \ 3661 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD 3662 hosts. 3663 3664 Connect to a local parallel port. 3665 3666 ``path`` specifies the path to the parallel port device. ``path`` is 3667 required. 3668 3669``-chardev spicevmc,id=id,debug=debug,name=name`` 3670 ``spicevmc`` is only available when spice support is built in. 3671 3672 ``debug`` debug level for spicevmc 3673 3674 ``name`` name of spice channel to connect to 3675 3676 Connect to a spice virtual machine channel, such as vdiport. 3677 3678``-chardev spiceport,id=id,debug=debug,name=name`` 3679 ``spiceport`` is only available when spice support is built in. 3680 3681 ``debug`` debug level for spicevmc 3682 3683 ``name`` name of spice port to connect to 3684 3685 Connect to a spice port, allowing a Spice client to handle the 3686 traffic identified by a name (preferably a fqdn). 3687ERST 3688 3689DEFHEADING() 3690 3691#ifdef CONFIG_TPM 3692DEFHEADING(TPM device options:) 3693 3694DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \ 3695 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n" 3696 " use path to provide path to a character device; default is /dev/tpm0\n" 3697 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n" 3698 " not provided it will be searched for in /sys/class/misc/tpm?/device\n" 3699 "-tpmdev emulator,id=id,chardev=dev\n" 3700 " configure the TPM device using chardev backend\n", 3701 QEMU_ARCH_ALL) 3702SRST 3703The general form of a TPM device option is: 3704 3705``-tpmdev backend,id=id[,options]`` 3706 The specific backend type will determine the applicable options. The 3707 ``-tpmdev`` option creates the TPM backend and requires a 3708 ``-device`` option that specifies the TPM frontend interface model. 3709 3710 Use ``-tpmdev help`` to print all available TPM backend types. 3711 3712The available backends are: 3713 3714``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path`` 3715 (Linux-host only) Enable access to the host's TPM using the 3716 passthrough driver. 3717 3718 ``path`` specifies the path to the host's TPM device, i.e., on a 3719 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by 3720 default ``/dev/tpm0`` is used. 3721 3722 ``cancel-path`` specifies the path to the host TPM device's sysfs 3723 entry allowing for cancellation of an ongoing TPM command. 3724 ``cancel-path`` is optional and by default QEMU will search for the 3725 sysfs entry to use. 3726 3727 Some notes about using the host's TPM with the passthrough driver: 3728 3729 The TPM device accessed by the passthrough driver must not be used 3730 by any other application on the host. 3731 3732 Since the host's firmware (BIOS/UEFI) has already initialized the 3733 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize 3734 the TPM again and may therefore not show a TPM-specific menu that 3735 would otherwise allow the user to configure the TPM, e.g., allow the 3736 user to enable/disable or activate/deactivate the TPM. Further, if 3737 TPM ownership is released from within a VM then the host's TPM will 3738 get disabled and deactivated. To enable and activate the TPM again 3739 afterwards, the host has to be rebooted and the user is required to 3740 enter the firmware's menu to enable and activate the TPM. If the TPM 3741 is left disabled and/or deactivated most TPM commands will fail. 3742 3743 To create a passthrough TPM use the following two options: 3744 3745 :: 3746 3747 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0 3748 3749 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by 3750 ``tpmdev=tpm0`` in the device option. 3751 3752``-tpmdev emulator,id=id,chardev=dev`` 3753 (Linux-host only) Enable access to a TPM emulator using Unix domain 3754 socket based chardev backend. 3755 3756 ``chardev`` specifies the unique ID of a character device backend 3757 that provides connection to the software TPM server. 3758 3759 To create a TPM emulator backend device with chardev socket backend: 3760 3761 :: 3762 3763 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0 3764ERST 3765 3766DEFHEADING() 3767 3768#endif 3769 3770DEFHEADING(Boot Image or Kernel specific:) 3771SRST 3772There are broadly 4 ways you can boot a system with QEMU. 3773 3774 - specify a firmware and let it control finding a kernel 3775 - specify a firmware and pass a hint to the kernel to boot 3776 - direct kernel image boot 3777 - manually load files into the guest's address space 3778 3779The third method is useful for quickly testing kernels but as there is 3780no firmware to pass configuration information to the kernel the 3781hardware must either be probeable, the kernel built for the exact 3782configuration or passed some configuration data (e.g. a DTB blob) 3783which tells the kernel what drivers it needs. This exact details are 3784often hardware specific. 3785 3786The final method is the most generic way of loading images into the 3787guest address space and used mostly for ``bare metal`` type 3788development where the reset vectors of the processor are taken into 3789account. 3790 3791ERST 3792 3793SRST 3794 3795For x86 machines and some other architectures ``-bios`` will generally 3796do the right thing with whatever it is given. For other machines the 3797more strict ``-pflash`` option needs an image that is sized for the 3798flash device for the given machine type. 3799 3800Please see the :ref:`system-targets-ref` section of the manual for 3801more detailed documentation. 3802 3803ERST 3804 3805DEF("bios", HAS_ARG, QEMU_OPTION_bios, \ 3806 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL) 3807SRST 3808``-bios file`` 3809 Set the filename for the BIOS. 3810ERST 3811 3812DEF("pflash", HAS_ARG, QEMU_OPTION_pflash, 3813 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL) 3814SRST 3815``-pflash file`` 3816 Use file as a parallel flash image. 3817ERST 3818 3819SRST 3820 3821The kernel options were designed to work with Linux kernels although 3822other things (like hypervisors) can be packaged up as a kernel 3823executable image. The exact format of a executable image is usually 3824architecture specific. 3825 3826The way in which the kernel is started (what address it is loaded at, 3827what if any information is passed to it via CPU registers, the state 3828of the hardware when it is started, and so on) is also architecture 3829specific. Typically it follows the specification laid down by the 3830Linux kernel for how kernels for that architecture must be started. 3831 3832ERST 3833 3834DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \ 3835 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL) 3836SRST 3837``-kernel bzImage`` 3838 Use bzImage as kernel image. The kernel can be either a Linux kernel 3839 or in multiboot format. 3840ERST 3841 3842DEF("append", HAS_ARG, QEMU_OPTION_append, \ 3843 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL) 3844SRST 3845``-append cmdline`` 3846 Use cmdline as kernel command line 3847ERST 3848 3849DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \ 3850 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL) 3851SRST 3852``-initrd file`` 3853 Use file as initial ram disk. 3854 3855``-initrd "file1 arg=foo,file2"`` 3856 This syntax is only available with multiboot. 3857 3858 Use file1 and file2 as modules and pass arg=foo as parameter to the 3859 first module. 3860ERST 3861 3862DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \ 3863 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL) 3864SRST 3865``-dtb file`` 3866 Use file as a device tree binary (dtb) image and pass it to the 3867 kernel on boot. 3868ERST 3869 3870SRST 3871 3872Finally you can also manually load images directly into the address 3873space of the guest. This is most useful for developers who already 3874know the layout of their guest and take care to ensure something sane 3875will happen when the reset vector executes. 3876 3877The generic loader can be invoked by using the loader device: 3878 3879``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]`` 3880 3881there is also the guest loader which operates in a similar way but 3882tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where 3883the guest image is: 3884 3885``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]`` 3886 3887ERST 3888 3889DEFHEADING() 3890 3891DEFHEADING(Debug/Expert options:) 3892 3893DEF("compat", HAS_ARG, QEMU_OPTION_compat, 3894 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n" 3895 " Policy for handling deprecated management interfaces\n" 3896 "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n" 3897 " Policy for handling unstable management interfaces\n", 3898 QEMU_ARCH_ALL) 3899SRST 3900``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]`` 3901 Set policy for handling deprecated management interfaces (experimental): 3902 3903 ``deprecated-input=accept`` (default) 3904 Accept deprecated commands and arguments 3905 ``deprecated-input=reject`` 3906 Reject deprecated commands and arguments 3907 ``deprecated-input=crash`` 3908 Crash on deprecated commands and arguments 3909 ``deprecated-output=accept`` (default) 3910 Emit deprecated command results and events 3911 ``deprecated-output=hide`` 3912 Suppress deprecated command results and events 3913 3914 Limitation: covers only syntactic aspects of QMP. 3915 3916``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]`` 3917 Set policy for handling unstable management interfaces (experimental): 3918 3919 ``unstable-input=accept`` (default) 3920 Accept unstable commands and arguments 3921 ``unstable-input=reject`` 3922 Reject unstable commands and arguments 3923 ``unstable-input=crash`` 3924 Crash on unstable commands and arguments 3925 ``unstable-output=accept`` (default) 3926 Emit unstable command results and events 3927 ``unstable-output=hide`` 3928 Suppress unstable command results and events 3929 3930 Limitation: covers only syntactic aspects of QMP. 3931ERST 3932 3933DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg, 3934 "-fw_cfg [name=]<name>,file=<file>\n" 3935 " add named fw_cfg entry with contents from file\n" 3936 "-fw_cfg [name=]<name>,string=<str>\n" 3937 " add named fw_cfg entry with contents from string\n", 3938 QEMU_ARCH_ALL) 3939SRST 3940``-fw_cfg [name=]name,file=file`` 3941 Add named fw\_cfg entry with contents from file file. 3942 3943``-fw_cfg [name=]name,string=str`` 3944 Add named fw\_cfg entry with contents from string str. 3945 3946 The terminating NUL character of the contents of str will not be 3947 included as part of the fw\_cfg item data. To insert contents with 3948 embedded NUL characters, you have to use the file parameter. 3949 3950 The fw\_cfg entries are passed by QEMU through to the guest. 3951 3952 Example: 3953 3954 :: 3955 3956 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin 3957 3958 creates an fw\_cfg entry named opt/com.mycompany/blob with contents 3959 from ./my\_blob.bin. 3960ERST 3961 3962DEF("serial", HAS_ARG, QEMU_OPTION_serial, \ 3963 "-serial dev redirect the serial port to char device 'dev'\n", 3964 QEMU_ARCH_ALL) 3965SRST 3966``-serial dev`` 3967 Redirect the virtual serial port to host character device dev. The 3968 default device is ``vc`` in graphical mode and ``stdio`` in non 3969 graphical mode. 3970 3971 This option can be used several times to simulate up to 4 serial 3972 ports. 3973 3974 Use ``-serial none`` to disable all serial ports. 3975 3976 Available character devices are: 3977 3978 ``vc[:WxH]`` 3979 Virtual console. Optionally, a width and height can be given in 3980 pixel with 3981 3982 :: 3983 3984 vc:800x600 3985 3986 It is also possible to specify width or height in characters: 3987 3988 :: 3989 3990 vc:80Cx24C 3991 3992 ``pty`` 3993 [Linux only] Pseudo TTY (a new PTY is automatically allocated) 3994 3995 ``none`` 3996 No device is allocated. 3997 3998 ``null`` 3999 void device 4000 4001 ``chardev:id`` 4002 Use a named character device defined with the ``-chardev`` 4003 option. 4004 4005 ``/dev/XXX`` 4006 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial 4007 port parameters are set according to the emulated ones. 4008 4009 ``/dev/parportN`` 4010 [Linux only, parallel port only] Use host parallel port N. 4011 Currently SPP and EPP parallel port features can be used. 4012 4013 ``file:filename`` 4014 Write output to filename. No character can be read. 4015 4016 ``stdio`` 4017 [Unix only] standard input/output 4018 4019 ``pipe:filename`` 4020 name pipe filename 4021 4022 ``COMn`` 4023 [Windows only] Use host serial port n 4024 4025 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]`` 4026 This implements UDP Net Console. When remote\_host or src\_ip 4027 are not specified they default to ``0.0.0.0``. When not using a 4028 specified src\_port a random port is automatically chosen. 4029 4030 If you just want a simple readonly console you can use 4031 ``netcat`` or ``nc``, by starting QEMU with: 4032 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time 4033 QEMU writes something to that port it will appear in the 4034 netconsole session. 4035 4036 If you plan to send characters back via netconsole or you want 4037 to stop and start QEMU a lot of times, you should have QEMU use 4038 the same source port each time by using something like ``-serial 4039 udp::4555@:4556`` to QEMU. Another approach is to use a patched 4040 version of netcat which can listen to a TCP port and send and 4041 receive characters via udp. If you have a patched version of 4042 netcat which activates telnet remote echo and single char 4043 transfer, then you can use the following options to set up a 4044 netcat redirector to allow telnet on port 5555 to access the 4045 QEMU port. 4046 4047 ``QEMU Options:`` 4048 -serial udp::4555@:4556 4049 4050 ``netcat options:`` 4051 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T 4052 4053 ``telnet options:`` 4054 localhost 5555 4055 4056 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]`` 4057 The TCP Net Console has two modes of operation. It can send the 4058 serial I/O to a location or wait for a connection from a 4059 location. By default the TCP Net Console is sent to host at the 4060 port. If you use the ``server=on`` option QEMU will wait for a client 4061 socket application to connect to the port before continuing, 4062 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off`` 4063 option disables the Nagle buffering algorithm. The ``reconnect=on`` 4064 option only applies if ``server=no`` is set, if the connection goes 4065 down it will attempt to reconnect at the given interval. If host 4066 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a 4067 time is accepted. You can use ``telnet=on`` to connect to the 4068 corresponding character device. 4069 4070 ``Example to send tcp console to 192.168.0.2 port 4444`` 4071 -serial tcp:192.168.0.2:4444 4072 4073 ``Example to listen and wait on port 4444 for connection`` 4074 -serial tcp::4444,server=on 4075 4076 ``Example to not wait and listen on ip 192.168.0.100 port 4444`` 4077 -serial tcp:192.168.0.100:4444,server=on,wait=off 4078 4079 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]`` 4080 The telnet protocol is used instead of raw tcp sockets. The 4081 options work the same as if you had specified ``-serial tcp``. 4082 The difference is that the port acts like a telnet server or 4083 client using telnet option negotiation. This will also allow you 4084 to send the MAGIC\_SYSRQ sequence if you use a telnet that 4085 supports sending the break sequence. Typically in unix telnet 4086 you do it with Control-] and then type "send break" followed by 4087 pressing the enter key. 4088 4089 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]`` 4090 The WebSocket protocol is used instead of raw tcp socket. The 4091 port acts as a WebSocket server. Client mode is not supported. 4092 4093 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]`` 4094 A unix domain socket is used instead of a tcp socket. The option 4095 works the same as if you had specified ``-serial tcp`` except 4096 the unix domain socket path is used for connections. 4097 4098 ``mon:dev_string`` 4099 This is a special option to allow the monitor to be multiplexed 4100 onto another serial port. The monitor is accessed with key 4101 sequence of Control-a and then pressing c. dev\_string should be 4102 any one of the serial devices specified above. An example to 4103 multiplex the monitor onto a telnet server listening on port 4104 4444 would be: 4105 4106 ``-serial mon:telnet::4444,server=on,wait=off`` 4107 4108 When the monitor is multiplexed to stdio in this way, Ctrl+C 4109 will not terminate QEMU any more but will be passed to the guest 4110 instead. 4111 4112 ``braille`` 4113 Braille device. This will use BrlAPI to display the braille 4114 output on a real or fake device. 4115 4116 ``msmouse`` 4117 Three button serial mouse. Configure the guest to use Microsoft 4118 protocol. 4119ERST 4120 4121DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \ 4122 "-parallel dev redirect the parallel port to char device 'dev'\n", 4123 QEMU_ARCH_ALL) 4124SRST 4125``-parallel dev`` 4126 Redirect the virtual parallel port to host device dev (same devices 4127 as the serial port). On Linux hosts, ``/dev/parportN`` can be used 4128 to use hardware devices connected on the corresponding host parallel 4129 port. 4130 4131 This option can be used several times to simulate up to 3 parallel 4132 ports. 4133 4134 Use ``-parallel none`` to disable all parallel ports. 4135ERST 4136 4137DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \ 4138 "-monitor dev redirect the monitor to char device 'dev'\n", 4139 QEMU_ARCH_ALL) 4140SRST 4141``-monitor dev`` 4142 Redirect the monitor to host device dev (same devices as the serial 4143 port). The default device is ``vc`` in graphical mode and ``stdio`` 4144 in non graphical mode. Use ``-monitor none`` to disable the default 4145 monitor. 4146ERST 4147DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \ 4148 "-qmp dev like -monitor but opens in 'control' mode\n", 4149 QEMU_ARCH_ALL) 4150SRST 4151``-qmp dev`` 4152 Like -monitor but opens in 'control' mode. 4153ERST 4154DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \ 4155 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n", 4156 QEMU_ARCH_ALL) 4157SRST 4158``-qmp-pretty dev`` 4159 Like -qmp but uses pretty JSON formatting. 4160ERST 4161 4162DEF("mon", HAS_ARG, QEMU_OPTION_mon, \ 4163 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL) 4164SRST 4165``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]`` 4166 Setup monitor on chardev name. ``mode=control`` configures 4167 a QMP monitor (a JSON RPC-style protocol) and it is not the 4168 same as HMP, the human monitor that has a "(qemu)" prompt. 4169 ``pretty`` is only valid when ``mode=control``, 4170 turning on JSON pretty printing to ease 4171 human reading and debugging. 4172ERST 4173 4174DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \ 4175 "-debugcon dev redirect the debug console to char device 'dev'\n", 4176 QEMU_ARCH_ALL) 4177SRST 4178``-debugcon dev`` 4179 Redirect the debug console to host device dev (same devices as the 4180 serial port). The debug console is an I/O port which is typically 4181 port 0xe9; writing to that I/O port sends output to this device. The 4182 default device is ``vc`` in graphical mode and ``stdio`` in non 4183 graphical mode. 4184ERST 4185 4186DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \ 4187 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL) 4188SRST 4189``-pidfile file`` 4190 Store the QEMU process PID in file. It is useful if you launch QEMU 4191 from a script. 4192ERST 4193 4194DEF("singlestep", 0, QEMU_OPTION_singlestep, \ 4195 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL) 4196SRST 4197``-singlestep`` 4198 Run the emulation in single step mode. 4199ERST 4200 4201DEF("preconfig", 0, QEMU_OPTION_preconfig, \ 4202 "--preconfig pause QEMU before machine is initialized (experimental)\n", 4203 QEMU_ARCH_ALL) 4204SRST 4205``--preconfig`` 4206 Pause QEMU for interactive configuration before the machine is 4207 created, which allows querying and configuring properties that will 4208 affect machine initialization. Use QMP command 'x-exit-preconfig' to 4209 exit the preconfig state and move to the next state (i.e. run guest 4210 if -S isn't used or pause the second time if -S is used). This 4211 option is experimental. 4212ERST 4213 4214DEF("S", 0, QEMU_OPTION_S, \ 4215 "-S freeze CPU at startup (use 'c' to start execution)\n", 4216 QEMU_ARCH_ALL) 4217SRST 4218``-S`` 4219 Do not start CPU at startup (you must type 'c' in the monitor). 4220ERST 4221 4222DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit, 4223 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n" 4224 " run qemu with overcommit hints\n" 4225 " mem-lock=on|off controls memory lock support (default: off)\n" 4226 " cpu-pm=on|off controls cpu power management (default: off)\n", 4227 QEMU_ARCH_ALL) 4228SRST 4229``-overcommit mem-lock=on|off`` 4230 \ 4231``-overcommit cpu-pm=on|off`` 4232 Run qemu with hints about host resource overcommit. The default is 4233 to assume that host overcommits all resources. 4234 4235 Locking qemu and guest memory can be enabled via ``mem-lock=on`` 4236 (disabled by default). This works when host memory is not 4237 overcommitted and reduces the worst-case latency for guest. 4238 4239 Guest ability to manage power state of host cpus (increasing latency 4240 for other processes on the same host cpu, but decreasing latency for 4241 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This 4242 works best when host CPU is not overcommitted. When used, host 4243 estimates of CPU cycle and power utilization will be incorrect, not 4244 taking into account guest idle time. 4245ERST 4246 4247DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \ 4248 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n" 4249 " the guest without waiting for gdb to connect; use -S too\n" 4250 " if you want it to not start execution.)\n", 4251 QEMU_ARCH_ALL) 4252SRST 4253``-gdb dev`` 4254 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter 4255 in the System Emulation Users Guide). Note that this option does not pause QEMU 4256 execution -- if you want QEMU to not start the guest until you 4257 connect with gdb and issue a ``continue`` command, you will need to 4258 also pass the ``-S`` option to QEMU. 4259 4260 The most usual configuration is to listen on a local TCP socket:: 4261 4262 -gdb tcp::3117 4263 4264 but you can specify other backends; UDP, pseudo TTY, or even stdio 4265 are all reasonable use cases. For example, a stdio connection 4266 allows you to start QEMU from within gdb and establish the 4267 connection via a pipe: 4268 4269 .. parsed-literal:: 4270 4271 (gdb) target remote | exec |qemu_system| -gdb stdio ... 4272ERST 4273 4274DEF("s", 0, QEMU_OPTION_s, \ 4275 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n", 4276 QEMU_ARCH_ALL) 4277SRST 4278``-s`` 4279 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234 4280 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide). 4281ERST 4282 4283DEF("d", HAS_ARG, QEMU_OPTION_d, \ 4284 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n", 4285 QEMU_ARCH_ALL) 4286SRST 4287``-d item1[,...]`` 4288 Enable logging of specified items. Use '-d help' for a list of log 4289 items. 4290ERST 4291 4292DEF("D", HAS_ARG, QEMU_OPTION_D, \ 4293 "-D logfile output log to logfile (default stderr)\n", 4294 QEMU_ARCH_ALL) 4295SRST 4296``-D logfile`` 4297 Output log in logfile instead of to stderr 4298ERST 4299 4300DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \ 4301 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n", 4302 QEMU_ARCH_ALL) 4303SRST 4304``-dfilter range1[,...]`` 4305 Filter debug output to that relevant to a range of target addresses. 4306 The filter spec can be either start+size, start-size or start..end 4307 where start end and size are the addresses and sizes required. For 4308 example: 4309 4310 :: 4311 4312 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000 4313 4314 Will dump output for any code in the 0x1000 sized block starting at 4315 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and 4316 another 0x1000 sized block starting at 0xffffffc00005f000. 4317ERST 4318 4319DEF("seed", HAS_ARG, QEMU_OPTION_seed, \ 4320 "-seed number seed the pseudo-random number generator\n", 4321 QEMU_ARCH_ALL) 4322SRST 4323``-seed number`` 4324 Force the guest to use a deterministic pseudo-random number 4325 generator, seeded with number. This does not affect crypto routines 4326 within the host. 4327ERST 4328 4329DEF("L", HAS_ARG, QEMU_OPTION_L, \ 4330 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n", 4331 QEMU_ARCH_ALL) 4332SRST 4333``-L path`` 4334 Set the directory for the BIOS, VGA BIOS and keymaps. 4335 4336 To list all the data directories, use ``-L help``. 4337ERST 4338 4339DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \ 4340 "-enable-kvm enable KVM full virtualization support\n", 4341 QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC | 4342 QEMU_ARCH_RISCV | QEMU_ARCH_S390X) 4343SRST 4344``-enable-kvm`` 4345 Enable KVM full virtualization support. This option is only 4346 available if KVM support is enabled when compiling. 4347ERST 4348 4349DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, 4350 "-xen-domid id specify xen guest domain id\n", 4351 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4352DEF("xen-attach", 0, QEMU_OPTION_xen_attach, 4353 "-xen-attach attach to existing xen domain\n" 4354 " libxl will use this when starting QEMU\n", 4355 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4356DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, 4357 "-xen-domid-restrict restrict set of available xen operations\n" 4358 " to specified domain id. (Does not affect\n" 4359 " xenpv machine type).\n", 4360 QEMU_ARCH_ARM | QEMU_ARCH_I386) 4361SRST 4362``-xen-domid id`` 4363 Specify xen guest domain id (XEN only). 4364 4365``-xen-attach`` 4366 Attach to existing xen domain. libxl will use this when starting 4367 QEMU (XEN only). Restrict set of available xen operations to 4368 specified domain id (XEN only). 4369ERST 4370 4371DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \ 4372 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL) 4373SRST 4374``-no-reboot`` 4375 Exit instead of rebooting. 4376ERST 4377 4378DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \ 4379 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL) 4380SRST 4381``-no-shutdown`` 4382 Don't exit QEMU on guest shutdown, but instead only stop the 4383 emulation. This allows for instance switching to monitor to commit 4384 changes to the disk image. 4385ERST 4386 4387DEF("action", HAS_ARG, QEMU_OPTION_action, 4388 "-action reboot=reset|shutdown\n" 4389 " action when guest reboots [default=reset]\n" 4390 "-action shutdown=poweroff|pause\n" 4391 " action when guest shuts down [default=poweroff]\n" 4392 "-action panic=pause|shutdown|exit-failure|none\n" 4393 " action when guest panics [default=shutdown]\n" 4394 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" 4395 " action when watchdog fires [default=reset]\n", 4396 QEMU_ARCH_ALL) 4397SRST 4398``-action event=action`` 4399 The action parameter serves to modify QEMU's default behavior when 4400 certain guest events occur. It provides a generic method for specifying the 4401 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown`` 4402 parameters. 4403 4404 Examples: 4405 4406 ``-action panic=none`` 4407 ``-action reboot=shutdown,shutdown=pause`` 4408 ``-device i6300esb -action watchdog=pause`` 4409 4410ERST 4411 4412DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \ 4413 "-loadvm [tag|id]\n" \ 4414 " start right away with a saved state (loadvm in monitor)\n", 4415 QEMU_ARCH_ALL) 4416SRST 4417``-loadvm file`` 4418 Start right away with a saved state (``loadvm`` in monitor) 4419ERST 4420 4421#ifndef _WIN32 4422DEF("daemonize", 0, QEMU_OPTION_daemonize, \ 4423 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL) 4424#endif 4425SRST 4426``-daemonize`` 4427 Daemonize the QEMU process after initialization. QEMU will not 4428 detach from standard IO until it is ready to receive connections on 4429 any of its devices. This option is a useful way for external 4430 programs to launch QEMU without having to cope with initialization 4431 race conditions. 4432ERST 4433 4434DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \ 4435 "-option-rom rom load a file, rom, into the option ROM space\n", 4436 QEMU_ARCH_ALL) 4437SRST 4438``-option-rom file`` 4439 Load the contents of file as an option ROM. This option is useful to 4440 load things like EtherBoot. 4441ERST 4442 4443DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \ 4444 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \ 4445 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n", 4446 QEMU_ARCH_ALL) 4447 4448SRST 4449``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]`` 4450 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at 4451 the current UTC or local time, respectively. ``localtime`` is 4452 required for correct date in MS-DOS or Windows. To start at a 4453 specific point in time, provide datetime in the format 4454 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC. 4455 4456 By default the RTC is driven by the host system time. This allows 4457 using of the RTC as accurate reference clock inside the guest, 4458 specifically if the host time is smoothly following an accurate 4459 external reference clock, e.g. via NTP. If you want to isolate the 4460 guest time from the host, you can set ``clock`` to ``rt`` instead, 4461 which provides a host monotonic clock if host support it. To even 4462 prevent the RTC from progressing during suspension, you can set 4463 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is 4464 recommended especially in icount mode in order to preserve 4465 determinism; however, note that in icount mode the speed of the 4466 virtual clock is variable and can in general differ from the host 4467 clock. 4468 4469 Enable ``driftfix`` (i386 targets only) if you experience time drift 4470 problems, specifically with Windows' ACPI HAL. This option will try 4471 to figure out how many timer interrupts were not processed by the 4472 Windows guest and will re-inject them. 4473ERST 4474 4475DEF("icount", HAS_ARG, QEMU_OPTION_icount, \ 4476 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \ 4477 " enable virtual instruction counter with 2^N clock ticks per\n" \ 4478 " instruction, enable aligning the host and virtual clocks\n" \ 4479 " or disable real time cpu sleeping, and optionally enable\n" \ 4480 " record-and-replay mode\n", QEMU_ARCH_ALL) 4481SRST 4482``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]`` 4483 Enable virtual instruction counter. The virtual cpu will execute one 4484 instruction every 2^N ns of virtual time. If ``auto`` is specified 4485 then the virtual cpu speed will be automatically adjusted to keep 4486 virtual time within a few seconds of real time. 4487 4488 Note that while this option can give deterministic behavior, it does 4489 not provide cycle accurate emulation. Modern CPUs contain 4490 superscalar out of order cores with complex cache hierarchies. The 4491 number of instructions executed often has little or no correlation 4492 with actual performance. 4493 4494 When the virtual cpu is sleeping, the virtual time will advance at 4495 default speed unless ``sleep=on`` is specified. With 4496 ``sleep=on``, the virtual time will jump to the next timer 4497 deadline instantly whenever the virtual cpu goes to sleep mode and 4498 will not advance if no timer is enabled. This behavior gives 4499 deterministic execution times from the guest point of view. 4500 The default if icount is enabled is ``sleep=off``. 4501 ``sleep=on`` cannot be used together with either ``shift=auto`` 4502 or ``align=on``. 4503 4504 ``align=on`` will activate the delay algorithm which will try to 4505 synchronise the host clock and the virtual clock. The goal is to 4506 have a guest running at the real frequency imposed by the shift 4507 option. Whenever the guest clock is behind the host clock and if 4508 ``align=on`` is specified then we print a message to the user to 4509 inform about the delay. Currently this option does not work when 4510 ``shift`` is ``auto``. Note: The sync algorithm will work for those 4511 shift values for which the guest clock runs ahead of the host clock. 4512 Typically this happens when the shift value is high (how high 4513 depends on the host machine). The default if icount is enabled 4514 is ``align=off``. 4515 4516 When the ``rr`` option is specified deterministic record/replay is 4517 enabled. The ``rrfile=`` option must also be provided to 4518 specify the path to the replay log. In record mode data is written 4519 to this file, and in replay mode it is read back. 4520 If the ``rrsnapshot`` option is given then it specifies a VM snapshot 4521 name. In record mode, a new VM snapshot with the given name is created 4522 at the start of execution recording. In replay mode this option 4523 specifies the snapshot name used to load the initial VM state. 4524ERST 4525 4526DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \ 4527 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \ 4528 " action when watchdog fires [default=reset]\n", 4529 QEMU_ARCH_ALL) 4530SRST 4531``-watchdog-action action`` 4532 The action controls what QEMU will do when the watchdog timer 4533 expires. The default is ``reset`` (forcefully reset the guest). 4534 Other possible actions are: ``shutdown`` (attempt to gracefully 4535 shutdown the guest), ``poweroff`` (forcefully poweroff the guest), 4536 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the 4537 guest), ``debug`` (print a debug message and continue), or ``none`` 4538 (do nothing). 4539 4540 Note that the ``shutdown`` action requires that the guest responds 4541 to ACPI signals, which it may not be able to do in the sort of 4542 situations where the watchdog would have expired, and thus 4543 ``-watchdog-action shutdown`` is not recommended for production use. 4544 4545 Examples: 4546 4547 ``-device i6300esb -watchdog-action pause`` 4548 4549ERST 4550 4551DEF("echr", HAS_ARG, QEMU_OPTION_echr, \ 4552 "-echr chr set terminal escape character instead of ctrl-a\n", 4553 QEMU_ARCH_ALL) 4554SRST 4555``-echr numeric_ascii_value`` 4556 Change the escape character used for switching to the monitor when 4557 using monitor and serial sharing. The default is ``0x01`` when using 4558 the ``-nographic`` option. ``0x01`` is equal to pressing 4559 ``Control-a``. You can select a different character from the ascii 4560 control keys where 1 through 26 map to Control-a through Control-z. 4561 For instance you could use the either of the following to change the 4562 escape character to Control-t. 4563 4564 ``-echr 0x14``; \ ``-echr 20`` 4565 4566ERST 4567 4568DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \ 4569 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \ 4570 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \ 4571 "-incoming unix:socketpath\n" \ 4572 " prepare for incoming migration, listen on\n" \ 4573 " specified protocol and socket address\n" \ 4574 "-incoming fd:fd\n" \ 4575 "-incoming exec:cmdline\n" \ 4576 " accept incoming migration on given file descriptor\n" \ 4577 " or from given external command\n" \ 4578 "-incoming defer\n" \ 4579 " wait for the URI to be specified via migrate_incoming\n", 4580 QEMU_ARCH_ALL) 4581SRST 4582``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]`` 4583 \ 4584``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]`` 4585 Prepare for incoming migration, listen on a given tcp port. 4586 4587``-incoming unix:socketpath`` 4588 Prepare for incoming migration, listen on a given unix socket. 4589 4590``-incoming fd:fd`` 4591 Accept incoming migration from a given filedescriptor. 4592 4593``-incoming exec:cmdline`` 4594 Accept incoming migration as an output from specified external 4595 command. 4596 4597``-incoming defer`` 4598 Wait for the URI to be specified via migrate\_incoming. The monitor 4599 can be used to change settings (such as migration parameters) prior 4600 to issuing the migrate\_incoming to allow the migration to begin. 4601ERST 4602 4603DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \ 4604 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL) 4605SRST 4606``-only-migratable`` 4607 Only allow migratable devices. Devices will not be allowed to enter 4608 an unmigratable state. 4609ERST 4610 4611DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \ 4612 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL) 4613SRST 4614``-nodefaults`` 4615 Don't create default devices. Normally, QEMU sets the default 4616 devices like serial port, parallel port, virtual console, monitor 4617 device, VGA adapter, floppy and CD-ROM drive and others. The 4618 ``-nodefaults`` option will disable all those default devices. 4619ERST 4620 4621#ifndef _WIN32 4622DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \ 4623 "-chroot dir chroot to dir just before starting the VM\n", 4624 QEMU_ARCH_ALL) 4625#endif 4626SRST 4627``-chroot dir`` 4628 Immediately before starting guest execution, chroot to the specified 4629 directory. Especially useful in combination with -runas. 4630ERST 4631 4632#ifndef _WIN32 4633DEF("runas", HAS_ARG, QEMU_OPTION_runas, \ 4634 "-runas user change to user id user just before starting the VM\n" \ 4635 " user can be numeric uid:gid instead\n", 4636 QEMU_ARCH_ALL) 4637#endif 4638SRST 4639``-runas user`` 4640 Immediately before starting guest execution, drop root privileges, 4641 switching to the specified user. 4642ERST 4643 4644DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env, 4645 "-prom-env variable=value\n" 4646 " set OpenBIOS nvram variables\n", 4647 QEMU_ARCH_PPC | QEMU_ARCH_SPARC) 4648SRST 4649``-prom-env variable=value`` 4650 Set OpenBIOS nvram variable to given value (PPC, SPARC only). 4651 4652 :: 4653 4654 qemu-system-sparc -prom-env 'auto-boot?=false' \ 4655 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single' 4656 4657 :: 4658 4659 qemu-system-ppc -prom-env 'auto-boot?=false' \ 4660 -prom-env 'boot-device=hd:2,\yaboot' \ 4661 -prom-env 'boot-args=conf=hd:2,\yaboot.conf' 4662ERST 4663DEF("semihosting", 0, QEMU_OPTION_semihosting, 4664 "-semihosting semihosting mode\n", 4665 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 4666 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) 4667SRST 4668``-semihosting`` 4669 Enable :ref:`Semihosting` mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only). 4670 4671 .. warning:: 4672 Note that this allows guest direct access to the host filesystem, so 4673 should only be used with a trusted guest OS. 4674 4675 See the -semihosting-config option documentation for further 4676 information about the facilities this enables. 4677ERST 4678DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config, 4679 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]\n" \ 4680 " semihosting configuration\n", 4681QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 4682QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) 4683SRST 4684``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]`` 4685 Enable and configure :ref:`Semihosting` (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V 4686 only). 4687 4688 .. warning:: 4689 Note that this allows guest direct access to the host filesystem, so 4690 should only be used with a trusted guest OS. 4691 4692 ``target=native|gdb|auto`` 4693 Defines where the semihosting calls will be addressed, to QEMU 4694 (``native``) or to GDB (``gdb``). The default is ``auto``, which 4695 means ``gdb`` during debug sessions and ``native`` otherwise. 4696 4697 ``chardev=str1`` 4698 Send the output to a chardev backend output for native or auto 4699 output when not in gdb 4700 4701 ``userspace=on|off`` 4702 Allows code running in guest userspace to access the semihosting 4703 interface. The default is that only privileged guest code can 4704 make semihosting calls. Note that setting ``userspace=on`` should 4705 only be used if all guest code is trusted (for example, in 4706 bare-metal test case code). 4707 4708 ``arg=str1,arg=str2,...`` 4709 Allows the user to pass input arguments, and can be used 4710 multiple times to build up a list. The old-style 4711 ``-kernel``/``-append`` method of passing a command line is 4712 still supported for backward compatibility. If both the 4713 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are 4714 specified, the former is passed to semihosting as it always 4715 takes precedence. 4716ERST 4717DEF("old-param", 0, QEMU_OPTION_old_param, 4718 "-old-param old param mode\n", QEMU_ARCH_ARM) 4719SRST 4720``-old-param`` 4721 Old param mode (ARM only). 4722ERST 4723 4724DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \ 4725 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \ 4726 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \ 4727 " Enable seccomp mode 2 system call filter (default 'off').\n" \ 4728 " use 'obsolete' to allow obsolete system calls that are provided\n" \ 4729 " by the kernel, but typically no longer used by modern\n" \ 4730 " C library implementations.\n" \ 4731 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \ 4732 " to elevate privileges using set*uid|gid system calls.\n" \ 4733 " The value 'children' will deny set*uid|gid system calls for\n" \ 4734 " main QEMU process but will allow forks and execves to run unprivileged\n" \ 4735 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \ 4736 " blocking *fork and execve\n" \ 4737 " use 'resourcecontrol' to disable process affinity and schedular priority\n", 4738 QEMU_ARCH_ALL) 4739SRST 4740``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]`` 4741 Enable Seccomp mode 2 system call filter. 'on' will enable syscall 4742 filtering and 'off' will disable it. The default is 'off'. 4743 4744 ``obsolete=string`` 4745 Enable Obsolete system calls 4746 4747 ``elevateprivileges=string`` 4748 Disable set\*uid\|gid system calls 4749 4750 ``spawn=string`` 4751 Disable \*fork and execve 4752 4753 ``resourcecontrol=string`` 4754 Disable process affinity and schedular priority 4755ERST 4756 4757DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig, 4758 "-readconfig <file>\n" 4759 " read config file\n", QEMU_ARCH_ALL) 4760SRST 4761``-readconfig file`` 4762 Read device configuration from file. This approach is useful when 4763 you want to spawn QEMU process with many command line options but 4764 you don't want to exceed the command line character limit. 4765ERST 4766 4767DEF("no-user-config", 0, QEMU_OPTION_nouserconfig, 4768 "-no-user-config\n" 4769 " do not load default user-provided config files at startup\n", 4770 QEMU_ARCH_ALL) 4771SRST 4772``-no-user-config`` 4773 The ``-no-user-config`` option makes QEMU not load any of the 4774 user-provided config files on sysconfdir. 4775ERST 4776 4777DEF("trace", HAS_ARG, QEMU_OPTION_trace, 4778 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n" 4779 " specify tracing options\n", 4780 QEMU_ARCH_ALL) 4781SRST 4782``-trace [[enable=]pattern][,events=file][,file=file]`` 4783 .. include:: ../qemu-option-trace.rst.inc 4784 4785ERST 4786DEF("plugin", HAS_ARG, QEMU_OPTION_plugin, 4787 "-plugin [file=]<file>[,<argname>=<argvalue>]\n" 4788 " load a plugin\n", 4789 QEMU_ARCH_ALL) 4790SRST 4791``-plugin file=file[,argname=argvalue]`` 4792 Load a plugin. 4793 4794 ``file=file`` 4795 Load the given plugin from a shared library file. 4796 4797 ``argname=argvalue`` 4798 Argument passed to the plugin. (Can be given multiple times.) 4799ERST 4800 4801HXCOMM Internal use 4802DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL) 4803DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL) 4804 4805#ifdef __linux__ 4806DEF("async-teardown", 0, QEMU_OPTION_asyncteardown, 4807 "-async-teardown enable asynchronous teardown\n", 4808 QEMU_ARCH_ALL) 4809#endif 4810SRST 4811``-async-teardown`` 4812 Enable asynchronous teardown. A new process called "cleanup/<QEMU_PID>" 4813 will be created at startup sharing the address space with the main qemu 4814 process, using clone. It will wait for the main qemu process to 4815 terminate completely, and then exit. 4816 This allows qemu to terminate very quickly even if the guest was 4817 huge, leaving the teardown of the address space to the cleanup 4818 process. Since the cleanup process shares the same cgroups as the 4819 main qemu process, accounting is performed correctly. This only 4820 works if the cleanup process is not forcefully killed with SIGKILL 4821 before the main qemu process has terminated completely. 4822ERST 4823 4824DEF("msg", HAS_ARG, QEMU_OPTION_msg, 4825 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n" 4826 " control error message format\n" 4827 " timestamp=on enables timestamps (default: off)\n" 4828 " guest-name=on enables guest name prefix but only if\n" 4829 " -name guest option is set (default: off)\n", 4830 QEMU_ARCH_ALL) 4831SRST 4832``-msg [timestamp[=on|off]][,guest-name[=on|off]]`` 4833 Control error message format. 4834 4835 ``timestamp=on|off`` 4836 Prefix messages with a timestamp. Default is off. 4837 4838 ``guest-name=on|off`` 4839 Prefix messages with guest name but only if -name guest option is set 4840 otherwise the option is ignored. Default is off. 4841ERST 4842 4843DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate, 4844 "-dump-vmstate <file>\n" 4845 " Output vmstate information in JSON format to file.\n" 4846 " Use the scripts/vmstate-static-checker.py file to\n" 4847 " check for possible regressions in migration code\n" 4848 " by comparing two such vmstate dumps.\n", 4849 QEMU_ARCH_ALL) 4850SRST 4851``-dump-vmstate file`` 4852 Dump json-encoded vmstate information for current machine type to 4853 file in file 4854ERST 4855 4856DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile, 4857 "-enable-sync-profile\n" 4858 " enable synchronization profiling\n", 4859 QEMU_ARCH_ALL) 4860SRST 4861``-enable-sync-profile`` 4862 Enable synchronization profiling. 4863ERST 4864 4865#if defined(CONFIG_TCG) && defined(CONFIG_LINUX) 4866DEF("perfmap", 0, QEMU_OPTION_perfmap, 4867 "-perfmap generate a /tmp/perf-${pid}.map file for perf\n", 4868 QEMU_ARCH_ALL) 4869SRST 4870``-perfmap`` 4871 Generate a map file for Linux perf tools that will allow basic profiling 4872 information to be broken down into basic blocks. 4873ERST 4874 4875DEF("jitdump", 0, QEMU_OPTION_jitdump, 4876 "-jitdump generate a jit-${pid}.dump file for perf\n", 4877 QEMU_ARCH_ALL) 4878SRST 4879``-jitdump`` 4880 Generate a dump file for Linux perf tools that maps basic blocks to symbol 4881 names, line numbers and JITted code. 4882ERST 4883#endif 4884 4885DEFHEADING() 4886 4887DEFHEADING(Generic object creation:) 4888 4889DEF("object", HAS_ARG, QEMU_OPTION_object, 4890 "-object TYPENAME[,PROP1=VALUE1,...]\n" 4891 " create a new object of type TYPENAME setting properties\n" 4892 " in the order they are specified. Note that the 'id'\n" 4893 " property must be set. These objects are placed in the\n" 4894 " '/objects' path.\n", 4895 QEMU_ARCH_ALL) 4896SRST 4897``-object typename[,prop1=value1,...]`` 4898 Create a new object of type typename setting properties in the order 4899 they are specified. Note that the 'id' property must be set. These 4900 objects are placed in the '/objects' path. 4901 4902 ``-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`` 4903 Creates a memory file backend object, which can be used to back 4904 the guest RAM with huge pages. 4905 4906 The ``id`` parameter is a unique ID that will be used to 4907 reference this memory region in other parameters, e.g. ``-numa``, 4908 ``-device nvdimm``, etc. 4909 4910 The ``size`` option provides the size of the memory region, and 4911 accepts common suffixes, e.g. ``500M``. 4912 4913 The ``mem-path`` provides the path to either a shared memory or 4914 huge page filesystem mount. 4915 4916 The ``share`` boolean option determines whether the memory 4917 region is marked as private to QEMU, or shared. The latter 4918 allows a co-operating external process to access the QEMU memory 4919 region. 4920 4921 The ``share`` is also required for pvrdma devices due to 4922 limitations in the RDMA API provided by Linux. 4923 4924 Setting share=on might affect the ability to configure NUMA 4925 bindings for the memory backend under some circumstances, see 4926 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel 4927 source tree for additional details. 4928 4929 Setting the ``discard-data`` boolean option to on indicates that 4930 file contents can be destroyed when QEMU exits, to avoid 4931 unnecessarily flushing data to the backing file. Note that 4932 ``discard-data`` is only an optimization, and QEMU might not 4933 discard file contents if it aborts unexpectedly or is terminated 4934 using SIGKILL. 4935 4936 The ``merge`` boolean option enables memory merge, also known as 4937 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider 4938 the pages for memory deduplication. 4939 4940 Setting the ``dump`` boolean option to off excludes the memory 4941 from core dumps. This feature is also known as MADV\_DONTDUMP. 4942 4943 The ``prealloc`` boolean option enables memory preallocation. 4944 4945 The ``host-nodes`` option binds the memory range to a list of 4946 NUMA host nodes. 4947 4948 The ``policy`` option sets the NUMA policy to one of the 4949 following values: 4950 4951 ``default`` 4952 default host policy 4953 4954 ``preferred`` 4955 prefer the given host node list for allocation 4956 4957 ``bind`` 4958 restrict memory allocation to the given host node list 4959 4960 ``interleave`` 4961 interleave memory allocations across the given host node 4962 list 4963 4964 The ``align`` option specifies the base address alignment when 4965 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg 4966 ``2M``. Some backend store specified by ``mem-path`` requires an 4967 alignment different than the default one used by QEMU, eg the 4968 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In 4969 such cases, users can specify the required alignment via this 4970 option. 4971 4972 The ``pmem`` option specifies whether the backing file specified 4973 by ``mem-path`` is in host persistent memory that can be 4974 accessed using the SNIA NVM programming model (e.g. Intel 4975 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary 4976 operations to guarantee the persistence of its own writes to 4977 ``mem-path`` (e.g. in vNVDIMM label emulation and live 4978 migration). Also, we will map the backend-file with MAP\_SYNC 4979 flag, which ensures the file metadata is in sync for 4980 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC 4981 requires support from both the host kernel (since Linux kernel 4982 4.15) and the filesystem of ``mem-path`` mounted with DAX 4983 option. 4984 4985 The ``readonly`` option specifies whether the backing file is opened 4986 read-only or read-write (default). 4987 4988 ``-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`` 4989 Creates a memory backend object, which can be used to back the 4990 guest RAM. Memory backend objects offer more control than the 4991 ``-m`` option that is traditionally used to define guest RAM. 4992 Please refer to ``memory-backend-file`` for a description of the 4993 options. 4994 4995 ``-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`` 4996 Creates an anonymous memory file backend object, which allows 4997 QEMU to share the memory with an external process (e.g. when 4998 using vhost-user). The memory is allocated with memfd and 4999 optional sealing. (Linux only) 5000 5001 The ``seal`` option creates a sealed-file, that will block 5002 further resizing the memory ('on' by default). 5003 5004 The ``hugetlb`` option specify the file to be created resides in 5005 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction 5006 with the ``hugetlb`` option, the ``hugetlbsize`` option specify 5007 the hugetlb page size on systems that support multiple hugetlb 5008 page sizes (it must be a power of 2 value supported by the 5009 system). 5010 5011 In some versions of Linux, the ``hugetlb`` option is 5012 incompatible with the ``seal`` option (requires at least Linux 5013 4.16). 5014 5015 Please refer to ``memory-backend-file`` for a description of the 5016 other options. 5017 5018 The ``share`` boolean option is on by default with memfd. 5019 5020 ``-object rng-builtin,id=id`` 5021 Creates a random number generator backend which obtains entropy 5022 from QEMU builtin functions. The ``id`` parameter is a unique ID 5023 that will be used to reference this entropy backend from the 5024 ``virtio-rng`` device. By default, the ``virtio-rng`` device 5025 uses this RNG backend. 5026 5027 ``-object rng-random,id=id,filename=/dev/random`` 5028 Creates a random number generator backend which obtains entropy 5029 from a device on the host. The ``id`` parameter is a unique ID 5030 that will be used to reference this entropy backend from the 5031 ``virtio-rng`` device. The ``filename`` parameter specifies 5032 which file to obtain entropy from and if omitted defaults to 5033 ``/dev/urandom``. 5034 5035 ``-object rng-egd,id=id,chardev=chardevid`` 5036 Creates a random number generator backend which obtains entropy 5037 from an external daemon running on the host. The ``id`` 5038 parameter is a unique ID that will be used to reference this 5039 entropy backend from the ``virtio-rng`` device. The ``chardev`` 5040 parameter is the unique ID of a character device backend that 5041 provides the connection to the RNG daemon. 5042 5043 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off`` 5044 Creates a TLS anonymous credentials object, which can be used to 5045 provide TLS support on network backends. The ``id`` parameter is 5046 a unique ID which network backends will use to access the 5047 credentials. The ``endpoint`` is either ``server`` or ``client`` 5048 depending on whether the QEMU network backend that uses the 5049 credentials will be acting as a client or as a server. If 5050 ``verify-peer`` is enabled (the default) then once the handshake 5051 is completed, the peer credentials will be verified, though this 5052 is a no-op for anonymous credentials. 5053 5054 The dir parameter tells QEMU where to find the credential files. 5055 For server endpoints, this directory may contain a file 5056 dh-params.pem providing diffie-hellman parameters to use for the 5057 TLS server. If the file is missing, QEMU will generate a set of 5058 DH parameters at startup. This is a computationally expensive 5059 operation that consumes random pool entropy, so it is 5060 recommended that a persistent set of parameters be generated 5061 upfront and saved. 5062 5063 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]`` 5064 Creates a TLS Pre-Shared Keys (PSK) credentials object, which 5065 can be used to provide TLS support on network backends. The 5066 ``id`` parameter is a unique ID which network backends will use 5067 to access the credentials. The ``endpoint`` is either ``server`` 5068 or ``client`` depending on whether the QEMU network backend that 5069 uses the credentials will be acting as a client or as a server. 5070 For clients only, ``username`` is the username which will be 5071 sent to the server. If omitted it defaults to "qemu". 5072 5073 The dir parameter tells QEMU where to find the keys file. It is 5074 called "dir/keys.psk" and contains "username:key" pairs. This 5075 file can most easily be created using the GnuTLS ``psktool`` 5076 program. 5077 5078 For server endpoints, dir may also contain a file dh-params.pem 5079 providing diffie-hellman parameters to use for the TLS server. 5080 If the file is missing, QEMU will generate a set of DH 5081 parameters at startup. This is a computationally expensive 5082 operation that consumes random pool entropy, so it is 5083 recommended that a persistent set of parameters be generated up 5084 front and saved. 5085 5086 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id`` 5087 Creates a TLS anonymous credentials object, which can be used to 5088 provide TLS support on network backends. The ``id`` parameter is 5089 a unique ID which network backends will use to access the 5090 credentials. The ``endpoint`` is either ``server`` or ``client`` 5091 depending on whether the QEMU network backend that uses the 5092 credentials will be acting as a client or as a server. If 5093 ``verify-peer`` is enabled (the default) then once the handshake 5094 is completed, the peer credentials will be verified. With x509 5095 certificates, this implies that the clients must be provided 5096 with valid client certificates too. 5097 5098 The dir parameter tells QEMU where to find the credential files. 5099 For server endpoints, this directory may contain a file 5100 dh-params.pem providing diffie-hellman parameters to use for the 5101 TLS server. If the file is missing, QEMU will generate a set of 5102 DH parameters at startup. This is a computationally expensive 5103 operation that consumes random pool entropy, so it is 5104 recommended that a persistent set of parameters be generated 5105 upfront and saved. 5106 5107 For x509 certificate credentials the directory will contain 5108 further files providing the x509 certificates. The certificates 5109 must be stored in PEM format, in filenames ca-cert.pem, 5110 ca-crl.pem (optional), server-cert.pem (only servers), 5111 server-key.pem (only servers), client-cert.pem (only clients), 5112 and client-key.pem (only clients). 5113 5114 For the server-key.pem and client-key.pem files which contain 5115 sensitive private keys, it is possible to use an encrypted 5116 version by providing the passwordid parameter. This provides the 5117 ID of a previously created ``secret`` object containing the 5118 password for decryption. 5119 5120 The priority parameter allows to override the global default 5121 priority used by gnutls. This can be useful if the system 5122 administrator needs to use a weaker set of crypto priorities for 5123 QEMU without potentially forcing the weakness onto all 5124 applications. Or conversely if one wants wants a stronger 5125 default for QEMU than for all other applications, they can do 5126 this through this parameter. Its format is a gnutls priority 5127 string as described at 5128 https://gnutls.org/manual/html_node/Priority-Strings.html. 5129 5130 ``-object tls-cipher-suites,id=id,priority=priority`` 5131 Creates a TLS cipher suites object, which can be used to control 5132 the TLS cipher/protocol algorithms that applications are permitted 5133 to use. 5134 5135 The ``id`` parameter is a unique ID which frontends will use to 5136 access the ordered list of permitted TLS cipher suites from the 5137 host. 5138 5139 The ``priority`` parameter allows to override the global default 5140 priority used by gnutls. This can be useful if the system 5141 administrator needs to use a weaker set of crypto priorities for 5142 QEMU without potentially forcing the weakness onto all 5143 applications. Or conversely if one wants wants a stronger 5144 default for QEMU than for all other applications, they can do 5145 this through this parameter. Its format is a gnutls priority 5146 string as described at 5147 https://gnutls.org/manual/html_node/Priority-Strings.html. 5148 5149 An example of use of this object is to control UEFI HTTPS Boot. 5150 The tls-cipher-suites object exposes the ordered list of permitted 5151 TLS cipher suites from the host side to the guest firmware, via 5152 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER 5153 objects. The firmware uses the IANA_TLS_CIPHER array for configuring 5154 guest-side TLS. 5155 5156 In the following example, the priority at which the host-side policy 5157 is retrieved is given by the ``priority`` property. 5158 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to 5159 refer to /etc/crypto-policies/back-ends/gnutls.config. 5160 5161 .. parsed-literal:: 5162 5163 # |qemu_system| \\ 5164 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\ 5165 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0 5166 5167 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]`` 5168 Interval t can't be 0, this filter batches the packet delivery: 5169 all packets arriving in a given interval on netdev netdevid are 5170 delayed until the end of the interval. Interval is in 5171 microseconds. ``status`` is optional that indicate whether the 5172 netfilter is on (enabled) or off (disabled), the default status 5173 for netfilter will be 'on'. 5174 5175 queue all\|rx\|tx is an option that can be applied to any 5176 netfilter. 5177 5178 ``all``: the filter is attached both to the receive and the 5179 transmit queue of the netdev (default). 5180 5181 ``rx``: the filter is attached to the receive queue of the 5182 netdev, where it will receive packets sent to the netdev. 5183 5184 ``tx``: the filter is attached to the transmit queue of the 5185 netdev, where it will receive packets sent by the netdev. 5186 5187 position head\|tail\|id=<id> is an option to specify where the 5188 filter should be inserted in the filter list. It can be applied 5189 to any netfilter. 5190 5191 ``head``: the filter is inserted at the head of the filter list, 5192 before any existing filters. 5193 5194 ``tail``: the filter is inserted at the tail of the filter list, 5195 behind any existing filters (default). 5196 5197 ``id=<id>``: the filter is inserted before or behind the filter 5198 specified by <id>, see the insert option below. 5199 5200 insert behind\|before is an option to specify where to insert 5201 the new filter relative to the one specified with 5202 position=id=<id>. It can be applied to any netfilter. 5203 5204 ``before``: insert before the specified filter. 5205 5206 ``behind``: insert behind the specified filter (default). 5207 5208 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5209 filter-mirror on netdev netdevid,mirror net packet to 5210 chardevchardevid, if it has the vnet\_hdr\_support flag, 5211 filter-mirror will mirror packet with vnet\_hdr\_len. 5212 5213 ``-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]`` 5214 filter-redirector on netdev netdevid,redirect filter's net 5215 packet to chardev chardevid,and redirect indev's packet to 5216 filter.if it has the vnet\_hdr\_support flag, filter-redirector 5217 will redirect packet with vnet\_hdr\_len. Create a 5218 filter-redirector we need to differ outdev id from indev id, id 5219 can not be the same. we can just use indev or outdev, but at 5220 least one of indev or outdev need to be specified. 5221 5222 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 5223 Filter-rewriter is a part of COLO project.It will rewrite tcp 5224 packet to secondary from primary to keep secondary tcp 5225 connection,and rewrite tcp packet to primary from secondary make 5226 tcp packet can be handled by client.if it has the 5227 vnet\_hdr\_support flag, we can parse packet with vnet header. 5228 5229 usage: colo secondary: -object 5230 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object 5231 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object 5232 filter-rewriter,id=rew0,netdev=hn0,queue=all 5233 5234 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]`` 5235 Dump the network traffic on netdev dev to the file specified by 5236 filename. At most len bytes (64k by default) per packet are 5237 stored. The file format is libpcap, so it can be analyzed with 5238 tools such as tcpdump or Wireshark. 5239 5240 ``-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}]`` 5241 Colo-compare gets packet from primary\_in chardevid and 5242 secondary\_in, then compare whether the payload of primary packet 5243 and secondary packet are the same. If same, it will output 5244 primary packet to out\_dev, else it will notify COLO-framework to do 5245 checkpoint and send primary packet to out\_dev. In order to 5246 improve efficiency, we need to put the task of comparison in 5247 another iothread. If it has the vnet\_hdr\_support flag, 5248 colo compare will send/recv packet with vnet\_hdr\_len. 5249 The compare\_timeout=@var{ms} determines the maximum time of the 5250 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms} 5251 is to set the period of scanning expired primary node network packets. 5252 The max\_queue\_size=@var{size} is to set the max compare queue 5253 size depend on user environment. 5254 If user want to use Xen COLO, need to add the notify\_dev to 5255 notify Xen colo-frame to do checkpoint. 5256 5257 COLO-compare must be used with the help of filter-mirror, 5258 filter-redirector and filter-rewriter. 5259 5260 :: 5261 5262 KVM COLO 5263 5264 primary: 5265 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 5266 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5267 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5268 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5269 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5270 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5271 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5272 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5273 -object iothread,id=iothread1 5274 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5275 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5276 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5277 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1 5278 5279 secondary: 5280 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 5281 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5282 -chardev socket,id=red0,host=3.3.3.3,port=9003 5283 -chardev socket,id=red1,host=3.3.3.3,port=9004 5284 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5285 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5286 5287 5288 Xen COLO 5289 5290 primary: 5291 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 5292 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5293 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5294 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5295 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5296 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5297 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5298 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5299 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off 5300 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5301 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5302 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5303 -object iothread,id=iothread1 5304 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1 5305 5306 secondary: 5307 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 5308 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5309 -chardev socket,id=red0,host=3.3.3.3,port=9003 5310 -chardev socket,id=red1,host=3.3.3.3,port=9004 5311 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5312 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5313 5314 If you want to know the detail of above command line, you can 5315 read the colo-compare git log. 5316 5317 ``-object cryptodev-backend-builtin,id=id[,queues=queues]`` 5318 Creates a cryptodev backend which executes crypto operations from 5319 the QEMU cipher APIs. The id parameter is a unique ID that will 5320 be used to reference this cryptodev backend from the 5321 ``virtio-crypto`` device. The queues parameter is optional, 5322 which specify the queue number of cryptodev backend, the default 5323 of queues is 1. 5324 5325 .. parsed-literal:: 5326 5327 # |qemu_system| \\ 5328 [...] \\ 5329 -object cryptodev-backend-builtin,id=cryptodev0 \\ 5330 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5331 [...] 5332 5333 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]`` 5334 Creates a vhost-user cryptodev backend, backed by a chardev 5335 chardevid. The id parameter is a unique ID that will be used to 5336 reference this cryptodev backend from the ``virtio-crypto`` 5337 device. The chardev should be a unix domain socket backed one. 5338 The vhost-user uses a specifically defined protocol to pass 5339 vhost ioctl replacement messages to an application on the other 5340 end of the socket. The queues parameter is optional, which 5341 specify the queue number of cryptodev backend for multiqueue 5342 vhost-user, the default of queues is 1. 5343 5344 .. parsed-literal:: 5345 5346 # |qemu_system| \\ 5347 [...] \\ 5348 -chardev socket,id=chardev0,path=/path/to/socket \\ 5349 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\ 5350 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5351 [...] 5352 5353 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]`` 5354 \ 5355 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]`` 5356 Defines a secret to store a password, encryption key, or some 5357 other sensitive data. The sensitive data can either be passed 5358 directly via the data parameter, or indirectly via the file 5359 parameter. Using the data parameter is insecure unless the 5360 sensitive data is encrypted. 5361 5362 The sensitive data can be provided in raw format (the default), 5363 or base64. When encoded as JSON, the raw format only supports 5364 valid UTF-8 characters, so base64 is recommended for sending 5365 binary data. QEMU will convert from which ever format is 5366 provided to the format it needs internally. eg, an RBD password 5367 can be provided in raw format, even though it will be base64 5368 encoded when passed onto the RBD sever. 5369 5370 For added protection, it is possible to encrypt the data 5371 associated with a secret using the AES-256-CBC cipher. Use of 5372 encryption is indicated by providing the keyid and iv 5373 parameters. The keyid parameter provides the ID of a previously 5374 defined secret that contains the AES-256 decryption key. This 5375 key should be 32-bytes long and be base64 encoded. The iv 5376 parameter provides the random initialization vector used for 5377 encryption of this particular secret and should be a base64 5378 encrypted string of the 16-byte IV. 5379 5380 The simplest (insecure) usage is to provide the secret inline 5381 5382 .. parsed-literal:: 5383 5384 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw 5385 5386 The simplest secure usage is to provide the secret via a file 5387 5388 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object 5389 secret,id=sec0,file=mypasswd.txt,format=raw 5390 5391 For greater security, AES-256-CBC should be used. To illustrate 5392 usage, consider the openssl command line tool which can encrypt 5393 the data. Note that when encrypting, the plaintext must be 5394 padded to the cipher block size (32 bytes) using the standard 5395 PKCS#5/6 compatible padding algorithm. 5396 5397 First a master key needs to be created in base64 encoding: 5398 5399 :: 5400 5401 # openssl rand -base64 32 > key.b64 5402 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"') 5403 5404 Each secret to be encrypted needs to have a random 5405 initialization vector generated. These do not need to be kept 5406 secret 5407 5408 :: 5409 5410 # openssl rand -base64 16 > iv.b64 5411 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"') 5412 5413 The secret to be defined can now be encrypted, in this case 5414 we're telling openssl to base64 encode the result, but it could 5415 be left as raw bytes if desired. 5416 5417 :: 5418 5419 # SECRET=$(printf "letmein" | 5420 openssl enc -aes-256-cbc -a -K $KEY -iv $IV) 5421 5422 When launching QEMU, create a master secret pointing to 5423 ``key.b64`` and specify that to be used to decrypt the user 5424 password. Pass the contents of ``iv.b64`` to the second secret 5425 5426 .. parsed-literal:: 5427 5428 # |qemu_system| \\ 5429 -object secret,id=secmaster0,format=base64,file=key.b64 \\ 5430 -object secret,id=sec0,keyid=secmaster0,format=base64,\\ 5431 data=$SECRET,iv=$(<iv.b64) 5432 5433 ``-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]`` 5434 Create a Secure Encrypted Virtualization (SEV) guest object, 5435 which can be used to provide the guest memory encryption support 5436 on AMD processors. 5437 5438 When memory encryption is enabled, one of the physical address 5439 bit (aka the C-bit) is utilized to mark if a memory page is 5440 protected. The ``cbitpos`` is used to provide the C-bit 5441 position. The C-bit position is Host family dependent hence user 5442 must provide this value. On EPYC, the value should be 47. 5443 5444 When memory encryption is enabled, we loose certain bits in 5445 physical address space. The ``reduced-phys-bits`` is used to 5446 provide the number of bits we loose in physical address space. 5447 Similar to C-bit, the value is Host family dependent. On EPYC, 5448 a guest will lose a maximum of 1 bit, so the value should be 1. 5449 5450 The ``sev-device`` provides the device file to use for 5451 communicating with the SEV firmware running inside AMD Secure 5452 Processor. The default device is '/dev/sev'. If hardware 5453 supports memory encryption then /dev/sev devices are created by 5454 CCP driver. 5455 5456 The ``policy`` provides the guest policy to be enforced by the 5457 SEV firmware and restrict what configuration and operational 5458 commands can be performed on this guest by the hypervisor. The 5459 policy should be provided by the guest owner and is bound to the 5460 guest and cannot be changed throughout the lifetime of the 5461 guest. The default is 0. 5462 5463 If guest ``policy`` allows sharing the key with another SEV 5464 guest then ``handle`` can be use to provide handle of the guest 5465 from which to share the key. 5466 5467 The ``dh-cert-file`` and ``session-file`` provides the guest 5468 owner's Public Diffie-Hillman key defined in SEV spec. The PDH 5469 and session parameters are used for establishing a cryptographic 5470 session with the guest owner to negotiate keys used for 5471 attestation. The file must be encoded in base64. 5472 5473 The ``kernel-hashes`` adds the hashes of given kernel/initrd/ 5474 cmdline to a designated guest firmware page for measured Linux 5475 boot with -kernel. The default is off. (Since 6.2) 5476 5477 e.g to launch a SEV guest 5478 5479 .. parsed-literal:: 5480 5481 # |qemu_system_x86| \\ 5482 ...... \\ 5483 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=1 \\ 5484 -machine ...,memory-encryption=sev0 \\ 5485 ..... 5486 5487 ``-object authz-simple,id=id,identity=string`` 5488 Create an authorization object that will control access to 5489 network services. 5490 5491 The ``identity`` parameter is identifies the user and its format 5492 depends on the network service that authorization object is 5493 associated with. For authorizing based on TLS x509 certificates, 5494 the identity must be the x509 distinguished name. Note that care 5495 must be taken to escape any commas in the distinguished name. 5496 5497 An example authorization object to validate a x509 distinguished 5498 name would look like: 5499 5500 .. parsed-literal:: 5501 5502 # |qemu_system| \\ 5503 ... \\ 5504 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\ 5505 ... 5506 5507 Note the use of quotes due to the x509 distinguished name 5508 containing whitespace, and escaping of ','. 5509 5510 ``-object authz-listfile,id=id,filename=path,refresh=on|off`` 5511 Create an authorization object that will control access to 5512 network services. 5513 5514 The ``filename`` parameter is the fully qualified path to a file 5515 containing the access control list rules in JSON format. 5516 5517 An example set of rules that match against SASL usernames might 5518 look like: 5519 5520 :: 5521 5522 { 5523 "rules": [ 5524 { "match": "fred", "policy": "allow", "format": "exact" }, 5525 { "match": "bob", "policy": "allow", "format": "exact" }, 5526 { "match": "danb", "policy": "deny", "format": "glob" }, 5527 { "match": "dan*", "policy": "allow", "format": "exact" }, 5528 ], 5529 "policy": "deny" 5530 } 5531 5532 When checking access the object will iterate over all the rules 5533 and the first rule to match will have its ``policy`` value 5534 returned as the result. If no rules match, then the default 5535 ``policy`` value is returned. 5536 5537 The rules can either be an exact string match, or they can use 5538 the simple UNIX glob pattern matching to allow wildcards to be 5539 used. 5540 5541 If ``refresh`` is set to true the file will be monitored and 5542 automatically reloaded whenever its content changes. 5543 5544 As with the ``authz-simple`` object, the format of the identity 5545 strings being matched depends on the network service, but is 5546 usually a TLS x509 distinguished name, or a SASL username. 5547 5548 An example authorization object to validate a SASL username 5549 would look like: 5550 5551 .. parsed-literal:: 5552 5553 # |qemu_system| \\ 5554 ... \\ 5555 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\ 5556 ... 5557 5558 ``-object authz-pam,id=id,service=string`` 5559 Create an authorization object that will control access to 5560 network services. 5561 5562 The ``service`` parameter provides the name of a PAM service to 5563 use for authorization. It requires that a file 5564 ``/etc/pam.d/service`` exist to provide the configuration for 5565 the ``account`` subsystem. 5566 5567 An example authorization object to validate a TLS x509 5568 distinguished name would look like: 5569 5570 .. parsed-literal:: 5571 5572 # |qemu_system| \\ 5573 ... \\ 5574 -object authz-pam,id=auth0,service=qemu-vnc \\ 5575 ... 5576 5577 There would then be a corresponding config file for PAM at 5578 ``/etc/pam.d/qemu-vnc`` that contains: 5579 5580 :: 5581 5582 account requisite pam_listfile.so item=user sense=allow \ 5583 file=/etc/qemu/vnc.allow 5584 5585 Finally the ``/etc/qemu/vnc.allow`` file would contain the list 5586 of x509 distinguished names that are permitted access 5587 5588 :: 5589 5590 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB 5591 5592 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch`` 5593 Creates a dedicated event loop thread that devices can be 5594 assigned to. This is known as an IOThread. By default device 5595 emulation happens in vCPU threads or the main event loop thread. 5596 This can become a scalability bottleneck. IOThreads allow device 5597 emulation and I/O to run on other host CPUs. 5598 5599 The ``id`` parameter is a unique ID that will be used to 5600 reference this IOThread from ``-device ...,iothread=id``. 5601 Multiple devices can be assigned to an IOThread. Note that not 5602 all devices support an ``iothread`` parameter. 5603 5604 The ``query-iothreads`` QMP command lists IOThreads and reports 5605 their thread IDs so that the user can configure host CPU 5606 pinning/affinity. 5607 5608 IOThreads use an adaptive polling algorithm to reduce event loop 5609 latency. Instead of entering a blocking system call to monitor 5610 file descriptors and then pay the cost of being woken up when an 5611 event occurs, the polling algorithm spins waiting for events for 5612 a short time. The algorithm's default parameters are suitable 5613 for many cases but can be adjusted based on knowledge of the 5614 workload and/or host device latency. 5615 5616 The ``poll-max-ns`` parameter is the maximum number of 5617 nanoseconds to busy wait for events. Polling can be disabled by 5618 setting this value to 0. 5619 5620 The ``poll-grow`` parameter is the multiplier used to increase 5621 the polling time when the algorithm detects it is missing events 5622 due to not polling long enough. 5623 5624 The ``poll-shrink`` parameter is the divisor used to decrease 5625 the polling time when the algorithm detects it is spending too 5626 long polling without encountering events. 5627 5628 The ``aio-max-batch`` parameter is the maximum number of requests 5629 in a batch for the AIO engine, 0 means that the engine will use 5630 its default. 5631 5632 The IOThread parameters can be modified at run-time using the 5633 ``qom-set`` command (where ``iothread1`` is the IOThread's 5634 ``id``): 5635 5636 :: 5637 5638 (qemu) qom-set /objects/iothread1 poll-max-ns 100000 5639ERST 5640 5641 5642HXCOMM This is the last statement. Insert new options before this line! 5643 5644#undef DEF 5645#undef DEFHEADING 5646#undef ARCHHEADING 5647
