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 QEMU_ARCH_ALL) 41SRST 42``-machine [type=]name[,prop=value[,...]]`` 43 Select the emulated machine by name. Use ``-machine help`` to list 44 available machines. 45 46 For architectures which aim to support live migration compatibility 47 across releases, each release will introduce a new versioned machine 48 type. For example, the 2.8.0 release introduced machine types 49 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures. 50 51 To allow live migration of guests from QEMU version 2.8.0, to QEMU 52 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8" 53 and "pc-q35-2.8" machines too. To allow users live migrating VMs to 54 skip multiple intermediate releases when upgrading, new releases of 55 QEMU will support machine types from many previous versions. 56 57 Supported machine properties are: 58 59 ``accel=accels1[:accels2[:...]]`` 60 This is used to enable an accelerator. Depending on the target 61 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available. 62 By default, tcg is used. If there is more than one accelerator 63 specified, the next one is used if the previous one fails to 64 initialize. 65 66 ``vmport=on|off|auto`` 67 Enables emulation of VMWare IO port, for vmmouse etc. auto says 68 to select the value based on accel. For accel=xen the default is 69 off otherwise the default is on. 70 71 ``dump-guest-core=on|off`` 72 Include guest memory in a core dump. The default is on. 73 74 ``mem-merge=on|off`` 75 Enables or disables memory merge support. This feature, when 76 supported by the host, de-duplicates identical memory pages 77 among VMs instances (enabled by default). 78 79 ``aes-key-wrap=on|off`` 80 Enables or disables AES key wrapping support on s390-ccw hosts. 81 This feature controls whether AES wrapping keys will be created 82 to allow execution of AES cryptographic functions. The default 83 is on. 84 85 ``dea-key-wrap=on|off`` 86 Enables or disables DEA key wrapping support on s390-ccw hosts. 87 This feature controls whether DEA wrapping keys will be created 88 to allow execution of DEA cryptographic functions. The default 89 is on. 90 91 ``nvdimm=on|off`` 92 Enables or disables NVDIMM support. The default is off. 93 94 ``memory-encryption=`` 95 Memory encryption object to use. The default is none. 96 97 ``hmat=on|off`` 98 Enables or disables ACPI Heterogeneous Memory Attribute Table 99 (HMAT) support. The default is off. 100 101 ``memory-backend='id'`` 102 An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options. 103 Allows to use a memory backend as main RAM. 104 105 For example: 106 :: 107 108 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on 109 -machine memory-backend=pc.ram 110 -m 512M 111 112 Migration compatibility note: 113 114 * as backend id one shall use value of 'default-ram-id', advertised by 115 machine type (available via ``query-machines`` QMP command), if migration 116 to/from old QEMU (<5.0) is expected. 117 * for machine types 4.0 and older, user shall 118 use ``x-use-canonical-path-for-ramblock-id=off`` backend option 119 if migration to/from old QEMU (<5.0) is expected. 120 121 For example: 122 :: 123 124 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off 125 -machine memory-backend=pc.ram 126 -m 512M 127ERST 128 129DEF("M", HAS_ARG, QEMU_OPTION_M, 130 " sgx-epc.0.memdev=memid\n", 131 QEMU_ARCH_ALL) 132 133SRST 134``sgx-epc.0.memdev=@var{memid}`` 135 Define an SGX EPC section. 136ERST 137 138DEF("cpu", HAS_ARG, QEMU_OPTION_cpu, 139 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL) 140SRST 141``-cpu model`` 142 Select CPU model (``-cpu help`` for list and additional feature 143 selection) 144ERST 145 146DEF("accel", HAS_ARG, QEMU_OPTION_accel, 147 "-accel [accel=]accelerator[,prop[=value][,...]]\n" 148 " select accelerator (kvm, xen, hax, hvf, nvmm, whpx or tcg; use 'help' for a list)\n" 149 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n" 150 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n" 151 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n" 152 " split-wx=on|off (enable TCG split w^x mapping)\n" 153 " tb-size=n (TCG translation block cache size)\n" 154 " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n" 155 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL) 156SRST 157``-accel name[,prop=value[,...]]`` 158 This is used to enable an accelerator. Depending on the target 159 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available. By 160 default, tcg is used. If there is more than one accelerator 161 specified, the next one is used if the previous one fails to 162 initialize. 163 164 ``igd-passthru=on|off`` 165 When Xen is in use, this option controls whether Intel 166 integrated graphics devices can be passed through to the guest 167 (default=off) 168 169 ``kernel-irqchip=on|off|split`` 170 Controls KVM in-kernel irqchip support. The default is full 171 acceleration of the interrupt controllers. On x86, split irqchip 172 reduces the kernel attack surface, at a performance cost for 173 non-MSI interrupts. Disabling the in-kernel irqchip completely 174 is not recommended except for debugging purposes. 175 176 ``kvm-shadow-mem=size`` 177 Defines the size of the KVM shadow MMU. 178 179 ``split-wx=on|off`` 180 Controls the use of split w^x mapping for the TCG code generation 181 buffer. Some operating systems require this to be enabled, and in 182 such a case this will default on. On other operating systems, this 183 will default off, but one may enable this for testing or debugging. 184 185 ``tb-size=n`` 186 Controls the size (in MiB) of the TCG translation block cache. 187 188 ``thread=single|multi`` 189 Controls number of TCG threads. When the TCG is multi-threaded 190 there will be one thread per vCPU therefore taking advantage of 191 additional host cores. The default is to enable multi-threading 192 where both the back-end and front-ends support it and no 193 incompatible TCG features have been enabled (e.g. 194 icount/replay). 195 196 ``dirty-ring-size=n`` 197 When the KVM accelerator is used, it controls the size of the per-vCPU 198 dirty page ring buffer (number of entries for each vCPU). It should 199 be a value that is power of two, and it should be 1024 or bigger (but 200 still less than the maximum value that the kernel supports). 4096 201 could be a good initial value if you have no idea which is the best. 202 Set this value to 0 to disable the feature. By default, this feature 203 is disabled (dirty-ring-size=0). When enabled, KVM will instead 204 record dirty pages in a bitmap. 205 206ERST 207 208DEF("smp", HAS_ARG, QEMU_OPTION_smp, 209 "-smp [[cpus=]n][,maxcpus=cpus][,sockets=sockets][,dies=dies][,cores=cores][,threads=threads]\n" 210 " set the number of CPUs to 'n' [default=1]\n" 211 " maxcpus= maximum number of total CPUs, including\n" 212 " offline CPUs for hotplug, etc\n" 213 " sockets= number of discrete sockets in the system\n" 214 " dies= number of CPU dies on one socket (for PC only)\n" 215 " cores= number of CPU cores on one socket (for PC, it's on one die)\n" 216 " threads= number of threads on one CPU core\n", 217 QEMU_ARCH_ALL) 218SRST 219``-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,cores=cores][,threads=threads]`` 220 Simulate a SMP system with '\ ``n``\ ' CPUs initially present on 221 the machine type board. On boards supporting CPU hotplug, the optional 222 '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be 223 added at runtime. If omitted the maximum number of CPUs will be 224 set to match the initial CPU count. Both parameters are subject to 225 an upper limit that is determined by the specific machine type chosen. 226 227 To control reporting of CPU topology information, the number of sockets, 228 dies per socket, cores per die, and threads per core can be specified. 229 The sum `` sockets * cores * dies * threads `` must be equal to the 230 maximum CPU count. CPU targets may only support a subset of the topology 231 parameters. Where a CPU target does not support use of a particular 232 topology parameter, its value should be assumed to be 1 for the purpose 233 of computing the CPU maximum count. 234 235 Either the initial CPU count, or at least one of the topology parameters 236 must be specified. Values for any omitted parameters will be computed 237 from those which are given. Historically preference was given to the 238 coarsest topology parameters when computing missing values (ie sockets 239 preferred over cores, which were preferred over threads), however, this 240 behaviour is considered liable to change. 241ERST 242 243DEF("numa", HAS_ARG, QEMU_OPTION_numa, 244 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 245 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 246 "-numa dist,src=source,dst=destination,val=distance\n" 247 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n" 248 "-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" 249 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n", 250 QEMU_ARCH_ALL) 251SRST 252``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 253 \ 254``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 255 \ 256``-numa dist,src=source,dst=destination,val=distance`` 257 \ 258``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]`` 259 \ 260``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=tpye[,latency=lat][,bandwidth=bw]`` 261 \ 262``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]`` 263 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA 264 distance from a source node to a destination node. Set the ACPI 265 Heterogeneous Memory Attributes for the given nodes. 266 267 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and 268 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a 269 contiguous range of CPU indexes (or a single VCPU if lastcpu is 270 omitted). A non-contiguous set of VCPUs can be represented by 271 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is 272 omitted on all nodes, VCPUs are automatically split between them. 273 274 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a 275 NUMA node: 276 277 :: 278 279 -numa node,cpus=0-2,cpus=5 280 281 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option 282 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to 283 assign CPU objects to a node using topology layout properties of 284 CPU. The set of properties is machine specific, and depends on used 285 machine type/'\ ``smp``\ ' options. It could be queried with 286 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ ' 287 property specifies node to which CPU object will be assigned, it's 288 required for node to be declared with '\ ``node``\ ' option before 289 it's used with '\ ``cpu``\ ' option. 290 291 For example: 292 293 :: 294 295 -M pc \ 296 -smp 1,sockets=2,maxcpus=2 \ 297 -numa node,nodeid=0 -numa node,nodeid=1 \ 298 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 299 300 Legacy '\ ``mem``\ ' assigns a given RAM amount to a node (not supported 301 for 5.1 and newer machine types). '\ ``memdev``\ ' assigns RAM from 302 a given memory backend device to a node. If '\ ``mem``\ ' and 303 '\ ``memdev``\ ' are omitted in all nodes, RAM is split equally between them. 304 305 306 '\ ``mem``\ ' and '\ ``memdev``\ ' are mutually exclusive. 307 Furthermore, if one node uses '\ ``memdev``\ ', all of them have to 308 use it. 309 310 '\ ``initiator``\ ' is an additional option that points to an 311 initiator NUMA node that has best performance (the lowest latency or 312 largest bandwidth) to this NUMA node. Note that this option can be 313 set only when the machine property 'hmat' is set to 'on'. 314 315 Following example creates a machine with 2 NUMA nodes, node 0 has 316 CPU. node 1 has only memory, and its initiator is node 0. Note that 317 because node 0 has CPU, by default the initiator of node 0 is itself 318 and must be itself. 319 320 :: 321 322 -machine hmat=on \ 323 -m 2G,slots=2,maxmem=4G \ 324 -object memory-backend-ram,size=1G,id=m0 \ 325 -object memory-backend-ram,size=1G,id=m1 \ 326 -numa node,nodeid=0,memdev=m0 \ 327 -numa node,nodeid=1,memdev=m1,initiator=0 \ 328 -smp 2,sockets=2,maxcpus=2 \ 329 -numa cpu,node-id=0,socket-id=0 \ 330 -numa cpu,node-id=0,socket-id=1 331 332 source and destination are NUMA node IDs. distance is the NUMA 333 distance from source to destination. The distance from a node to 334 itself is always 10. If any pair of nodes is given a distance, then 335 all pairs must be given distances. Although, when distances are only 336 given in one direction for each pair of nodes, then the distances in 337 the opposite directions are assumed to be the same. If, however, an 338 asymmetrical pair of distances is given for even one node pair, then 339 all node pairs must be provided distance values for both directions, 340 even when they are symmetrical. When a node is unreachable from 341 another node, set the pair's distance to 255. 342 343 Note that the -``numa`` option doesn't allocate any of the specified 344 resources, it just assigns existing resources to NUMA nodes. This 345 means that one still has to use the ``-m``, ``-smp`` options to 346 allocate RAM and VCPUs respectively. 347 348 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth 349 Information between initiator and target NUMA nodes in ACPI 350 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can 351 create memory requests, usually it has one or more processors. 352 Target NUMA node contains addressable memory. 353 354 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is 355 the memory hierarchy of the target NUMA node: if hierarchy is 356 'memory', the structure represents the memory performance; if 357 hierarchy is 'first-level\|second-level\|third-level', this 358 structure represents aggregated performance of memory side caches 359 for each domain. type of 'data-type' is type of data represented by 360 this structure instance: if 'hierarchy' is 'memory', 'data-type' is 361 'access\|read\|write' latency or 'access\|read\|write' bandwidth of 362 the target memory; if 'hierarchy' is 363 'first-level\|second-level\|third-level', 'data-type' is 364 'access\|read\|write' hit latency or 'access\|read\|write' hit 365 bandwidth of the target memory side cache. 366 367 lat is latency value in nanoseconds. bw is bandwidth value, the 368 possible value and units are NUM[M\|G\|T], mean that the bandwidth 369 value are NUM byte per second (or MB/s, GB/s or TB/s depending on 370 used suffix). Note that if latency or bandwidth value is 0, means 371 the corresponding latency or bandwidth information is not provided. 372 373 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory 374 belongs. size is the size of memory side cache in bytes. level is 375 the cache level described in this structure, note that the cache 376 level 0 should not be used with '\ ``hmat-cache``\ ' option. 377 associativity is the cache associativity, the possible value is 378 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy 379 is the write policy. line is the cache Line size in bytes. 380 381 For example, the following options describe 2 NUMA nodes. Node 0 has 382 2 cpus and a ram, node 1 has only a ram. The processors in node 0 383 access memory in node 0 with access-latency 5 nanoseconds, 384 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access 385 memory in NUMA node 1 with access-latency 10 nanoseconds, 386 access-bandwidth is 100 MB/s. And for memory side cache information, 387 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB, 388 policy is write-back, the cache Line size is 8 bytes: 389 390 :: 391 392 -machine hmat=on \ 393 -m 2G \ 394 -object memory-backend-ram,size=1G,id=m0 \ 395 -object memory-backend-ram,size=1G,id=m1 \ 396 -smp 2 \ 397 -numa node,nodeid=0,memdev=m0 \ 398 -numa node,nodeid=1,memdev=m1,initiator=0 \ 399 -numa cpu,node-id=0,socket-id=0 \ 400 -numa cpu,node-id=0,socket-id=1 \ 401 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \ 402 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \ 403 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \ 404 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \ 405 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \ 406 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8 407ERST 408 409DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd, 410 "-add-fd fd=fd,set=set[,opaque=opaque]\n" 411 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL) 412SRST 413``-add-fd fd=fd,set=set[,opaque=opaque]`` 414 Add a file descriptor to an fd set. Valid options are: 415 416 ``fd=fd`` 417 This option defines the file descriptor of which a duplicate is 418 added to fd set. The file descriptor cannot be stdin, stdout, or 419 stderr. 420 421 ``set=set`` 422 This option defines the ID of the fd set to add the file 423 descriptor to. 424 425 ``opaque=opaque`` 426 This option defines a free-form string that can be used to 427 describe fd. 428 429 You can open an image using pre-opened file descriptors from an fd 430 set: 431 432 .. parsed-literal:: 433 434 |qemu_system| \\ 435 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 436 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 437 -drive file=/dev/fdset/2,index=0,media=disk 438ERST 439 440DEF("set", HAS_ARG, QEMU_OPTION_set, 441 "-set group.id.arg=value\n" 442 " set <arg> parameter for item <id> of type <group>\n" 443 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL) 444SRST 445``-set group.id.arg=value`` 446 Set parameter arg for item id of type group 447ERST 448 449DEF("global", HAS_ARG, QEMU_OPTION_global, 450 "-global driver.property=value\n" 451 "-global driver=driver,property=property,value=value\n" 452 " set a global default for a driver property\n", 453 QEMU_ARCH_ALL) 454SRST 455``-global driver.prop=value`` 456 \ 457``-global driver=driver,property=property,value=value`` 458 Set default value of driver's property prop to value, e.g.: 459 460 .. parsed-literal:: 461 462 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img 463 464 In particular, you can use this to set driver properties for devices 465 which are created automatically by the machine model. To create a 466 device which is not created automatically and set properties on it, 467 use -``device``. 468 469 -global driver.prop=value is shorthand for -global 470 driver=driver,property=prop,value=value. The longhand syntax works 471 even when driver contains a dot. 472ERST 473 474DEF("boot", HAS_ARG, QEMU_OPTION_boot, 475 "-boot [order=drives][,once=drives][,menu=on|off]\n" 476 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n" 477 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n" 478 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n" 479 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n" 480 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n", 481 QEMU_ARCH_ALL) 482SRST 483``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]`` 484 Specify boot order drives as a string of drive letters. Valid drive 485 letters depend on the target architecture. The x86 PC uses: a, b 486 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p 487 (Etherboot from network adapter 1-4), hard disk boot is the default. 488 To apply a particular boot order only on the first startup, specify 489 it via ``once``. Note that the ``order`` or ``once`` parameter 490 should not be used together with the ``bootindex`` property of 491 devices, since the firmware implementations normally do not support 492 both at the same time. 493 494 Interactive boot menus/prompts can be enabled via ``menu=on`` as far 495 as firmware/BIOS supports them. The default is non-interactive boot. 496 497 A splash picture could be passed to bios, enabling user to show it 498 as logo, when option splash=sp\_name is given and menu=on, If 499 firmware/BIOS supports them. Currently Seabios for X86 system 500 support it. limitation: The splash file could be a jpeg file or a 501 BMP file in 24 BPP format(true color). The resolution should be 502 supported by the SVGA mode, so the recommended is 320x240, 640x480, 503 800x640. 504 505 A timeout could be passed to bios, guest will pause for rb\_timeout 506 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will 507 not reboot, qemu passes '-1' to bios by default. Currently Seabios 508 for X86 system support it. 509 510 Do strict boot via ``strict=on`` as far as firmware/BIOS supports 511 it. This only effects when boot priority is changed by bootindex 512 options. The default is non-strict boot. 513 514 .. parsed-literal:: 515 516 # try to boot from network first, then from hard disk 517 |qemu_system_x86| -boot order=nc 518 # boot from CD-ROM first, switch back to default order after reboot 519 |qemu_system_x86| -boot once=d 520 # boot with a splash picture for 5 seconds. 521 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000 522 523 Note: The legacy format '-boot drives' is still supported but its 524 use is discouraged as it may be removed from future versions. 525ERST 526 527DEF("m", HAS_ARG, QEMU_OPTION_m, 528 "-m [size=]megs[,slots=n,maxmem=size]\n" 529 " configure guest RAM\n" 530 " size: initial amount of guest memory\n" 531 " slots: number of hotplug slots (default: none)\n" 532 " maxmem: maximum amount of guest memory (default: none)\n" 533 "NOTE: Some architectures might enforce a specific granularity\n", 534 QEMU_ARCH_ALL) 535SRST 536``-m [size=]megs[,slots=n,maxmem=size]`` 537 Sets guest startup RAM size to megs megabytes. Default is 128 MiB. 538 Optionally, a suffix of "M" or "G" can be used to signify a value in 539 megabytes or gigabytes respectively. Optional pair slots, maxmem 540 could be used to set amount of hotpluggable memory slots and maximum 541 amount of memory. Note that maxmem must be aligned to the page size. 542 543 For example, the following command-line sets the guest startup RAM 544 size to 1GB, creates 3 slots to hotplug additional memory and sets 545 the maximum memory the guest can reach to 4GB: 546 547 .. parsed-literal:: 548 549 |qemu_system| -m 1G,slots=3,maxmem=4G 550 551 If slots and maxmem are not specified, memory hotplug won't be 552 enabled and the guest startup RAM will never increase. 553ERST 554 555DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath, 556 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL) 557SRST 558``-mem-path path`` 559 Allocate guest RAM from a temporarily created file in path. 560ERST 561 562DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc, 563 "-mem-prealloc preallocate guest memory (use with -mem-path)\n", 564 QEMU_ARCH_ALL) 565SRST 566``-mem-prealloc`` 567 Preallocate memory when using -mem-path. 568ERST 569 570DEF("k", HAS_ARG, QEMU_OPTION_k, 571 "-k language use keyboard layout (for example 'fr' for French)\n", 572 QEMU_ARCH_ALL) 573SRST 574``-k language`` 575 Use keyboard layout language (for example ``fr`` for French). This 576 option is only needed where it is not easy to get raw PC keycodes 577 (e.g. on Macs, with some X11 servers or with a VNC or curses 578 display). You don't normally need to use it on PC/Linux or 579 PC/Windows hosts. 580 581 The available layouts are: 582 583 :: 584 585 ar de-ch es fo fr-ca hu ja mk no pt-br sv 586 da en-gb et fr fr-ch is lt nl pl ru th 587 de en-us fi fr-be hr it lv nl-be pt sl tr 588 589 The default is ``en-us``. 590ERST 591 592 593HXCOMM Deprecated by -audiodev 594DEF("audio-help", 0, QEMU_OPTION_audio_help, 595 "-audio-help show -audiodev equivalent of the currently specified audio settings\n", 596 QEMU_ARCH_ALL) 597SRST 598``-audio-help`` 599 Will show the -audiodev equivalent of the currently specified 600 (deprecated) environment variables. 601ERST 602 603DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, 604 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n" 605 " specifies the audio backend to use\n" 606 " id= identifier of the backend\n" 607 " timer-period= timer period in microseconds\n" 608 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n" 609 " in|out.fixed-settings= use fixed settings for host audio\n" 610 " in|out.frequency= frequency to use with fixed settings\n" 611 " in|out.channels= number of channels to use with fixed settings\n" 612 " in|out.format= sample format to use with fixed settings\n" 613 " valid values: s8, s16, s32, u8, u16, u32, f32\n" 614 " in|out.voices= number of voices to use\n" 615 " in|out.buffer-length= length of buffer in microseconds\n" 616 "-audiodev none,id=id,[,prop[=value][,...]]\n" 617 " dummy driver that discards all output\n" 618#ifdef CONFIG_AUDIO_ALSA 619 "-audiodev alsa,id=id[,prop[=value][,...]]\n" 620 " in|out.dev= name of the audio device to use\n" 621 " in|out.period-length= length of period in microseconds\n" 622 " in|out.try-poll= attempt to use poll mode\n" 623 " threshold= threshold (in microseconds) when playback starts\n" 624#endif 625#ifdef CONFIG_AUDIO_COREAUDIO 626 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n" 627 " in|out.buffer-count= number of buffers\n" 628#endif 629#ifdef CONFIG_AUDIO_DSOUND 630 "-audiodev dsound,id=id[,prop[=value][,...]]\n" 631 " latency= add extra latency to playback in microseconds\n" 632#endif 633#ifdef CONFIG_AUDIO_OSS 634 "-audiodev oss,id=id[,prop[=value][,...]]\n" 635 " in|out.dev= path of the audio device to use\n" 636 " in|out.buffer-count= number of buffers\n" 637 " in|out.try-poll= attempt to use poll mode\n" 638 " try-mmap= try using memory mapped access\n" 639 " exclusive= open device in exclusive mode\n" 640 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n" 641#endif 642#ifdef CONFIG_AUDIO_PA 643 "-audiodev pa,id=id[,prop[=value][,...]]\n" 644 " server= PulseAudio server address\n" 645 " in|out.name= source/sink device name\n" 646 " in|out.latency= desired latency in microseconds\n" 647#endif 648#ifdef CONFIG_AUDIO_SDL 649 "-audiodev sdl,id=id[,prop[=value][,...]]\n" 650 " in|out.buffer-count= number of buffers\n" 651#endif 652#ifdef CONFIG_SPICE 653 "-audiodev spice,id=id[,prop[=value][,...]]\n" 654#endif 655 "-audiodev wav,id=id[,prop[=value][,...]]\n" 656 " path= path of wav file to record\n", 657 QEMU_ARCH_ALL) 658SRST 659``-audiodev [driver=]driver,id=id[,prop[=value][,...]]`` 660 Adds a new audio backend driver identified by id. There are global 661 and driver specific properties. Some values can be set differently 662 for input and output, they're marked with ``in|out.``. You can set 663 the input's property with ``in.prop`` and the output's property with 664 ``out.prop``. For example: 665 666 :: 667 668 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000 669 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified 670 671 NOTE: parameter validation is known to be incomplete, in many cases 672 specifying an invalid option causes QEMU to print an error message 673 and continue emulation without sound. 674 675 Valid global options are: 676 677 ``id=identifier`` 678 Identifies the audio backend. 679 680 ``timer-period=period`` 681 Sets the timer period used by the audio subsystem in 682 microseconds. Default is 10000 (10 ms). 683 684 ``in|out.mixing-engine=on|off`` 685 Use QEMU's mixing engine to mix all streams inside QEMU and 686 convert audio formats when not supported by the backend. When 687 off, fixed-settings must be off too. Note that disabling this 688 option means that the selected backend must support multiple 689 streams and the audio formats used by the virtual cards, 690 otherwise you'll get no sound. It's not recommended to disable 691 this option unless you want to use 5.1 or 7.1 audio, as mixing 692 engine only supports mono and stereo audio. Default is on. 693 694 ``in|out.fixed-settings=on|off`` 695 Use fixed settings for host audio. When off, it will change 696 based on how the guest opens the sound card. In this case you 697 must not specify frequency, channels or format. Default is on. 698 699 ``in|out.frequency=frequency`` 700 Specify the frequency to use when using fixed-settings. Default 701 is 44100Hz. 702 703 ``in|out.channels=channels`` 704 Specify the number of channels to use when using fixed-settings. 705 Default is 2 (stereo). 706 707 ``in|out.format=format`` 708 Specify the sample format to use when using fixed-settings. 709 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``, 710 ``u32``, ``f32``. Default is ``s16``. 711 712 ``in|out.voices=voices`` 713 Specify the number of voices to use. Default is 1. 714 715 ``in|out.buffer-length=usecs`` 716 Sets the size of the buffer in microseconds. 717 718``-audiodev none,id=id[,prop[=value][,...]]`` 719 Creates a dummy backend that discards all outputs. This backend has 720 no backend specific properties. 721 722``-audiodev alsa,id=id[,prop[=value][,...]]`` 723 Creates backend using the ALSA. This backend is only available on 724 Linux. 725 726 ALSA specific options are: 727 728 ``in|out.dev=device`` 729 Specify the ALSA device to use for input and/or output. Default 730 is ``default``. 731 732 ``in|out.period-length=usecs`` 733 Sets the period length in microseconds. 734 735 ``in|out.try-poll=on|off`` 736 Attempt to use poll mode with the device. Default is on. 737 738 ``threshold=threshold`` 739 Threshold (in microseconds) when playback starts. Default is 0. 740 741``-audiodev coreaudio,id=id[,prop[=value][,...]]`` 742 Creates a backend using Apple's Core Audio. This backend is only 743 available on Mac OS and only supports playback. 744 745 Core Audio specific options are: 746 747 ``in|out.buffer-count=count`` 748 Sets the count of the buffers. 749 750``-audiodev dsound,id=id[,prop[=value][,...]]`` 751 Creates a backend using Microsoft's DirectSound. This backend is 752 only available on Windows and only supports playback. 753 754 DirectSound specific options are: 755 756 ``latency=usecs`` 757 Add extra usecs microseconds latency to playback. Default is 758 10000 (10 ms). 759 760``-audiodev oss,id=id[,prop[=value][,...]]`` 761 Creates a backend using OSS. This backend is available on most 762 Unix-like systems. 763 764 OSS specific options are: 765 766 ``in|out.dev=device`` 767 Specify the file name of the OSS device to use. Default is 768 ``/dev/dsp``. 769 770 ``in|out.buffer-count=count`` 771 Sets the count of the buffers. 772 773 ``in|out.try-poll=on|of`` 774 Attempt to use poll mode with the device. Default is on. 775 776 ``try-mmap=on|off`` 777 Try using memory mapped device access. Default is off. 778 779 ``exclusive=on|off`` 780 Open the device in exclusive mode (vmix won't work in this 781 case). Default is off. 782 783 ``dsp-policy=policy`` 784 Sets the timing policy (between 0 and 10, where smaller number 785 means smaller latency but higher CPU usage). Use -1 to use 786 buffer sizes specified by ``buffer`` and ``buffer-count``. This 787 option is ignored if you do not have OSS 4. Default is 5. 788 789``-audiodev pa,id=id[,prop[=value][,...]]`` 790 Creates a backend using PulseAudio. This backend is available on 791 most systems. 792 793 PulseAudio specific options are: 794 795 ``server=server`` 796 Sets the PulseAudio server to connect to. 797 798 ``in|out.name=sink`` 799 Use the specified source/sink for recording/playback. 800 801 ``in|out.latency=usecs`` 802 Desired latency in microseconds. The PulseAudio server will try 803 to honor this value but actual latencies may be lower or higher. 804 805``-audiodev sdl,id=id[,prop[=value][,...]]`` 806 Creates a backend using SDL. This backend is available on most 807 systems, but you should use your platform's native backend if 808 possible. 809 810 SDL specific options are: 811 812 ``in|out.buffer-count=count`` 813 Sets the count of the buffers. 814 815``-audiodev spice,id=id[,prop[=value][,...]]`` 816 Creates a backend that sends audio through SPICE. This backend 817 requires ``-spice`` and automatically selected in that case, so 818 usually you can ignore this option. This backend has no backend 819 specific properties. 820 821``-audiodev wav,id=id[,prop[=value][,...]]`` 822 Creates a backend that writes audio to a WAV file. 823 824 Backend specific options are: 825 826 ``path=path`` 827 Write recorded audio into the specified file. Default is 828 ``qemu.wav``. 829ERST 830 831DEF("soundhw", HAS_ARG, QEMU_OPTION_soundhw, 832 "-soundhw c1,... enable audio support\n" 833 " and only specified sound cards (comma separated list)\n" 834 " use '-soundhw help' to get the list of supported cards\n" 835 " use '-soundhw all' to enable all of them\n", QEMU_ARCH_ALL) 836SRST 837``-soundhw card1[,card2,...] or -soundhw all`` 838 Enable audio and selected sound hardware. Use 'help' to print all 839 available sound hardware. For example: 840 841 .. parsed-literal:: 842 843 |qemu_system_x86| -soundhw sb16,adlib disk.img 844 |qemu_system_x86| -soundhw es1370 disk.img 845 |qemu_system_x86| -soundhw ac97 disk.img 846 |qemu_system_x86| -soundhw hda disk.img 847 |qemu_system_x86| -soundhw all disk.img 848 |qemu_system_x86| -soundhw help 849 850 Note that Linux's i810\_audio OSS kernel (for AC97) module might 851 require manually specifying clocking. 852 853 :: 854 855 modprobe i810_audio clocking=48000 856ERST 857 858DEF("device", HAS_ARG, QEMU_OPTION_device, 859 "-device driver[,prop[=value][,...]]\n" 860 " add device (based on driver)\n" 861 " prop=value,... sets driver properties\n" 862 " use '-device help' to print all possible drivers\n" 863 " use '-device driver,help' to print all possible properties\n", 864 QEMU_ARCH_ALL) 865SRST 866``-device driver[,prop[=value][,...]]`` 867 Add device driver. prop=value sets driver properties. Valid 868 properties depend on the driver. To get help on possible drivers and 869 properties, use ``-device help`` and ``-device driver,help``. 870 871 Some drivers are: 872 873``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]`` 874 Add an IPMI BMC. This is a simulation of a hardware management 875 interface processor that normally sits on a system. It provides a 876 watchdog and the ability to reset and power control the system. You 877 need to connect this to an IPMI interface to make it useful 878 879 The IPMI slave address to use for the BMC. The default is 0x20. This 880 address is the BMC's address on the I2C network of management 881 controllers. If you don't know what this means, it is safe to ignore 882 it. 883 884 ``id=id`` 885 The BMC id for interfaces to use this device. 886 887 ``slave_addr=val`` 888 Define slave address to use for the BMC. The default is 0x20. 889 890 ``sdrfile=file`` 891 file containing raw Sensor Data Records (SDR) data. The default 892 is none. 893 894 ``fruareasize=val`` 895 size of a Field Replaceable Unit (FRU) area. The default is 896 1024. 897 898 ``frudatafile=file`` 899 file containing raw Field Replaceable Unit (FRU) inventory data. 900 The default is none. 901 902 ``guid=uuid`` 903 value for the GUID for the BMC, in standard UUID format. If this 904 is set, get "Get GUID" command to the BMC will return it. 905 Otherwise "Get GUID" will return an error. 906 907``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]`` 908 Add a connection to an external IPMI BMC simulator. Instead of 909 locally emulating the BMC like the above item, instead connect to an 910 external entity that provides the IPMI services. 911 912 A connection is made to an external BMC simulator. If you do this, 913 it is strongly recommended that you use the "reconnect=" chardev 914 option to reconnect to the simulator if the connection is lost. Note 915 that if this is not used carefully, it can be a security issue, as 916 the interface has the ability to send resets, NMIs, and power off 917 the VM. It's best if QEMU makes a connection to an external 918 simulator running on a secure port on localhost, so neither the 919 simulator nor QEMU is exposed to any outside network. 920 921 See the "lanserv/README.vm" file in the OpenIPMI library for more 922 details on the external interface. 923 924``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]`` 925 Add a KCS IPMI interafce on the ISA bus. This also adds a 926 corresponding ACPI and SMBIOS entries, if appropriate. 927 928 ``bmc=id`` 929 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern 930 above. 931 932 ``ioport=val`` 933 Define the I/O address of the interface. The default is 0xca0 934 for KCS. 935 936 ``irq=val`` 937 Define the interrupt to use. The default is 5. To disable 938 interrupts, set this to 0. 939 940``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]`` 941 Like the KCS interface, but defines a BT interface. The default port 942 is 0xe4 and the default interrupt is 5. 943 944``-device pci-ipmi-kcs,bmc=id`` 945 Add a KCS IPMI interafce on the PCI bus. 946 947 ``bmc=id`` 948 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above. 949 950``-device pci-ipmi-bt,bmc=id`` 951 Like the KCS interface, but defines a BT interface on the PCI bus. 952 953``-device intel-iommu[,option=...]`` 954 This is only supported by ``-machine q35``, which will enable Intel VT-d 955 emulation within the guest. It supports below options: 956 957 ``intremap=on|off`` (default: auto) 958 This enables interrupt remapping feature. It's required to enable 959 complete x2apic. Currently it only supports kvm kernel-irqchip modes 960 ``off`` or ``split``, while full kernel-irqchip is not yet supported. 961 The default value is "auto", which will be decided by the mode of 962 kernel-irqchip. 963 964 ``caching-mode=on|off`` (default: off) 965 This enables caching mode for the VT-d emulated device. When 966 caching-mode is enabled, each guest DMA buffer mapping will generate an 967 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in 968 a synchronous way. It is required for ``-device vfio-pci`` to work 969 with the VT-d device, because host assigned devices requires to setup 970 the DMA mapping on the host before guest DMA starts. 971 972 ``device-iotlb=on|off`` (default: off) 973 This enables device-iotlb capability for the emulated VT-d device. So 974 far virtio/vhost should be the only real user for this parameter, 975 paired with ats=on configured for the device. 976 977 ``aw-bits=39|48`` (default: 39) 978 This decides the address width of IOVA address space. The address 979 space has 39 bits width for 3-level IOMMU page tables, and 48 bits for 980 4-level IOMMU page tables. 981 982 Please also refer to the wiki page for general scenarios of VT-d 983 emulation in QEMU: https://wiki.qemu.org/Features/VT-d. 984 985ERST 986 987DEF("name", HAS_ARG, QEMU_OPTION_name, 988 "-name string1[,process=string2][,debug-threads=on|off]\n" 989 " set the name of the guest\n" 990 " string1 sets the window title and string2 the process name\n" 991 " When debug-threads is enabled, individual threads are given a separate name\n" 992 " NOTE: The thread names are for debugging and not a stable API.\n", 993 QEMU_ARCH_ALL) 994SRST 995``-name name`` 996 Sets the name of the guest. This name will be displayed in the SDL 997 window caption. The name will also be used for the VNC server. Also 998 optionally set the top visible process name in Linux. Naming of 999 individual threads can also be enabled on Linux to aid debugging. 1000ERST 1001 1002DEF("uuid", HAS_ARG, QEMU_OPTION_uuid, 1003 "-uuid %08x-%04x-%04x-%04x-%012x\n" 1004 " specify machine UUID\n", QEMU_ARCH_ALL) 1005SRST 1006``-uuid uuid`` 1007 Set system UUID. 1008ERST 1009 1010DEFHEADING() 1011 1012DEFHEADING(Block device options:) 1013 1014DEF("fda", HAS_ARG, QEMU_OPTION_fda, 1015 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL) 1016DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL) 1017SRST 1018``-fda file`` 1019 \ 1020``-fdb file`` 1021 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in 1022 the System Emulation Users Guide). 1023ERST 1024 1025DEF("hda", HAS_ARG, QEMU_OPTION_hda, 1026 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL) 1027DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL) 1028DEF("hdc", HAS_ARG, QEMU_OPTION_hdc, 1029 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL) 1030DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL) 1031SRST 1032``-hda file`` 1033 \ 1034``-hdb file`` 1035 \ 1036``-hdc file`` 1037 \ 1038``-hdd file`` 1039 Use file as hard disk 0, 1, 2 or 3 image (see the :ref:`disk images` 1040 chapter in the System Emulation Users Guide). 1041ERST 1042 1043DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom, 1044 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n", 1045 QEMU_ARCH_ALL) 1046SRST 1047``-cdrom file`` 1048 Use file as CD-ROM image (you cannot use ``-hdc`` and ``-cdrom`` at 1049 the same time). You can use the host CD-ROM by using ``/dev/cdrom`` 1050 as filename. 1051ERST 1052 1053DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev, 1054 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n" 1055 " [,cache.direct=on|off][,cache.no-flush=on|off]\n" 1056 " [,read-only=on|off][,auto-read-only=on|off]\n" 1057 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n" 1058 " [,driver specific parameters...]\n" 1059 " configure a block backend\n", QEMU_ARCH_ALL) 1060SRST 1061``-blockdev option[,option[,option[,...]]]`` 1062 Define a new block driver node. Some of the options apply to all 1063 block drivers, other options are only accepted for a specific block 1064 driver. See below for a list of generic options and options for the 1065 most common block drivers. 1066 1067 Options that expect a reference to another node (e.g. ``file``) can 1068 be given in two ways. Either you specify the node name of an already 1069 existing node (file=node-name), or you define a new node inline, 1070 adding options for the referenced node after a dot 1071 (file.filename=path,file.aio=native). 1072 1073 A block driver node created with ``-blockdev`` can be used for a 1074 guest device by specifying its node name for the ``drive`` property 1075 in a ``-device`` argument that defines a block device. 1076 1077 ``Valid options for any block driver node:`` 1078 ``driver`` 1079 Specifies the block driver to use for the given node. 1080 1081 ``node-name`` 1082 This defines the name of the block driver node by which it 1083 will be referenced later. The name must be unique, i.e. it 1084 must not match the name of a different block driver node, or 1085 (if you use ``-drive`` as well) the ID of a drive. 1086 1087 If no node name is specified, it is automatically generated. 1088 The generated node name is not intended to be predictable 1089 and changes between QEMU invocations. For the top level, an 1090 explicit node name must be specified. 1091 1092 ``read-only`` 1093 Open the node read-only. Guest write attempts will fail. 1094 1095 Note that some block drivers support only read-only access, 1096 either generally or in certain configurations. In this case, 1097 the default value ``read-only=off`` does not work and the 1098 option must be specified explicitly. 1099 1100 ``auto-read-only`` 1101 If ``auto-read-only=on`` is set, QEMU may fall back to 1102 read-only usage even when ``read-only=off`` is requested, or 1103 even switch between modes as needed, e.g. depending on 1104 whether the image file is writable or whether a writing user 1105 is attached to the node. 1106 1107 ``force-share`` 1108 Override the image locking system of QEMU by forcing the 1109 node to utilize weaker shared access for permissions where 1110 it would normally request exclusive access. When there is 1111 the potential for multiple instances to have the same file 1112 open (whether this invocation of QEMU is the first or the 1113 second instance), both instances must permit shared access 1114 for the second instance to succeed at opening the file. 1115 1116 Enabling ``force-share=on`` requires ``read-only=on``. 1117 1118 ``cache.direct`` 1119 The host page cache can be avoided with ``cache.direct=on``. 1120 This will attempt to do disk IO directly to the guest's 1121 memory. QEMU may still perform an internal copy of the data. 1122 1123 ``cache.no-flush`` 1124 In case you don't care about data integrity over host 1125 failures, you can use ``cache.no-flush=on``. This option 1126 tells QEMU that it never needs to write any data to the disk 1127 but can instead keep things in cache. If anything goes 1128 wrong, like your host losing power, the disk storage getting 1129 disconnected accidentally, etc. your image will most 1130 probably be rendered unusable. 1131 1132 ``discard=discard`` 1133 discard is one of "ignore" (or "off") or "unmap" (or "on") 1134 and controls whether ``discard`` (also known as ``trim`` or 1135 ``unmap``) requests are ignored or passed to the filesystem. 1136 Some machine types may not support discard requests. 1137 1138 ``detect-zeroes=detect-zeroes`` 1139 detect-zeroes is "off", "on" or "unmap" and enables the 1140 automatic conversion of plain zero writes by the OS to 1141 driver specific optimized zero write commands. You may even 1142 choose "unmap" if discard is set to "unmap" to allow a zero 1143 write to be converted to an ``unmap`` operation. 1144 1145 ``Driver-specific options for file`` 1146 This is the protocol-level block driver for accessing regular 1147 files. 1148 1149 ``filename`` 1150 The path to the image file in the local filesystem 1151 1152 ``aio`` 1153 Specifies the AIO backend (threads/native/io_uring, 1154 default: threads) 1155 1156 ``locking`` 1157 Specifies whether the image file is protected with Linux OFD 1158 / POSIX locks. The default is to use the Linux Open File 1159 Descriptor API if available, otherwise no lock is applied. 1160 (auto/on/off, default: auto) 1161 1162 Example: 1163 1164 :: 1165 1166 -blockdev driver=file,node-name=disk,filename=disk.img 1167 1168 ``Driver-specific options for raw`` 1169 This is the image format block driver for raw images. It is 1170 usually stacked on top of a protocol level block driver such as 1171 ``file``. 1172 1173 ``file`` 1174 Reference to or definition of the data source block driver 1175 node (e.g. a ``file`` driver node) 1176 1177 Example 1: 1178 1179 :: 1180 1181 -blockdev driver=file,node-name=disk_file,filename=disk.img 1182 -blockdev driver=raw,node-name=disk,file=disk_file 1183 1184 Example 2: 1185 1186 :: 1187 1188 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img 1189 1190 ``Driver-specific options for qcow2`` 1191 This is the image format block driver for qcow2 images. It is 1192 usually stacked on top of a protocol level block driver such as 1193 ``file``. 1194 1195 ``file`` 1196 Reference to or definition of the data source block driver 1197 node (e.g. a ``file`` driver node) 1198 1199 ``backing`` 1200 Reference to or definition of the backing file block device 1201 (default is taken from the image file). It is allowed to 1202 pass ``null`` here in order to disable the default backing 1203 file. 1204 1205 ``lazy-refcounts`` 1206 Whether to enable the lazy refcounts feature (on/off; 1207 default is taken from the image file) 1208 1209 ``cache-size`` 1210 The maximum total size of the L2 table and refcount block 1211 caches in bytes (default: the sum of l2-cache-size and 1212 refcount-cache-size) 1213 1214 ``l2-cache-size`` 1215 The maximum size of the L2 table cache in bytes (default: if 1216 cache-size is not specified - 32M on Linux platforms, and 8M 1217 on non-Linux platforms; otherwise, as large as possible 1218 within the cache-size, while permitting the requested or the 1219 minimal refcount cache size) 1220 1221 ``refcount-cache-size`` 1222 The maximum size of the refcount block cache in bytes 1223 (default: 4 times the cluster size; or if cache-size is 1224 specified, the part of it which is not used for the L2 1225 cache) 1226 1227 ``cache-clean-interval`` 1228 Clean unused entries in the L2 and refcount caches. The 1229 interval is in seconds. The default value is 600 on 1230 supporting platforms, and 0 on other platforms. Setting it 1231 to 0 disables this feature. 1232 1233 ``pass-discard-request`` 1234 Whether discard requests to the qcow2 device should be 1235 forwarded to the data source (on/off; default: on if 1236 discard=unmap is specified, off otherwise) 1237 1238 ``pass-discard-snapshot`` 1239 Whether discard requests for the data source should be 1240 issued when a snapshot operation (e.g. deleting a snapshot) 1241 frees clusters in the qcow2 file (on/off; default: on) 1242 1243 ``pass-discard-other`` 1244 Whether discard requests for the data source should be 1245 issued on other occasions where a cluster gets freed 1246 (on/off; default: off) 1247 1248 ``overlap-check`` 1249 Which overlap checks to perform for writes to the image 1250 (none/constant/cached/all; default: cached). For details or 1251 finer granularity control refer to the QAPI documentation of 1252 ``blockdev-add``. 1253 1254 Example 1: 1255 1256 :: 1257 1258 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2 1259 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216 1260 1261 Example 2: 1262 1263 :: 1264 1265 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2 1266 1267 ``Driver-specific options for other drivers`` 1268 Please refer to the QAPI documentation of the ``blockdev-add`` 1269 QMP command. 1270ERST 1271 1272DEF("drive", HAS_ARG, QEMU_OPTION_drive, 1273 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n" 1274 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n" 1275 " [,snapshot=on|off][,rerror=ignore|stop|report]\n" 1276 " [,werror=ignore|stop|report|enospc][,id=name]\n" 1277 " [,aio=threads|native|io_uring]\n" 1278 " [,readonly=on|off][,copy-on-read=on|off]\n" 1279 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n" 1280 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n" 1281 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n" 1282 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n" 1283 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n" 1284 " [[,iops_size=is]]\n" 1285 " [[,group=g]]\n" 1286 " use 'file' as a drive image\n", QEMU_ARCH_ALL) 1287SRST 1288``-drive option[,option[,option[,...]]]`` 1289 Define a new drive. This includes creating a block driver node (the 1290 backend) as well as a guest device, and is mostly a shortcut for 1291 defining the corresponding ``-blockdev`` and ``-device`` options. 1292 1293 ``-drive`` accepts all options that are accepted by ``-blockdev``. 1294 In addition, it knows the following options: 1295 1296 ``file=file`` 1297 This option defines which disk image (see the :ref:`disk images` 1298 chapter in the System Emulation Users Guide) to use with this drive. 1299 If the filename contains comma, you must double it (for instance, 1300 "file=my,,file" to use file "my,file"). 1301 1302 Special files such as iSCSI devices can be specified using 1303 protocol specific URLs. See the section for "Device URL Syntax" 1304 for more information. 1305 1306 ``if=interface`` 1307 This option defines on which type on interface the drive is 1308 connected. Available types are: ide, scsi, sd, mtd, floppy, 1309 pflash, virtio, none. 1310 1311 ``bus=bus,unit=unit`` 1312 These options define where is connected the drive by defining 1313 the bus number and the unit id. 1314 1315 ``index=index`` 1316 This option defines where is connected the drive by using an 1317 index in the list of available connectors of a given interface 1318 type. 1319 1320 ``media=media`` 1321 This option defines the type of the media: disk or cdrom. 1322 1323 ``snapshot=snapshot`` 1324 snapshot is "on" or "off" and controls snapshot mode for the 1325 given drive (see ``-snapshot``). 1326 1327 ``cache=cache`` 1328 cache is "none", "writeback", "unsafe", "directsync" or 1329 "writethrough" and controls how the host cache is used to access 1330 block data. This is a shortcut that sets the ``cache.direct`` 1331 and ``cache.no-flush`` options (as in ``-blockdev``), and 1332 additionally ``cache.writeback``, which provides a default for 1333 the ``write-cache`` option of block guest devices (as in 1334 ``-device``). The modes correspond to the following settings: 1335 1336 ============= =============== ============ ============== 1337 \ cache.writeback cache.direct cache.no-flush 1338 ============= =============== ============ ============== 1339 writeback on off off 1340 none on on off 1341 writethrough off off off 1342 directsync off on off 1343 unsafe on off on 1344 ============= =============== ============ ============== 1345 1346 The default mode is ``cache=writeback``. 1347 1348 ``aio=aio`` 1349 aio is "threads", "native", or "io_uring" and selects between pthread 1350 based disk I/O, native Linux AIO, or Linux io_uring API. 1351 1352 ``format=format`` 1353 Specify which disk format will be used rather than detecting the 1354 format. Can be used to specify format=raw to avoid interpreting 1355 an untrusted format header. 1356 1357 ``werror=action,rerror=action`` 1358 Specify which action to take on write and read errors. Valid 1359 actions are: "ignore" (ignore the error and try to continue), 1360 "stop" (pause QEMU), "report" (report the error to the guest), 1361 "enospc" (pause QEMU only if the host disk is full; report the 1362 error to the guest otherwise). The default setting is 1363 ``werror=enospc`` and ``rerror=report``. 1364 1365 ``copy-on-read=copy-on-read`` 1366 copy-on-read is "on" or "off" and enables whether to copy read 1367 backing file sectors into the image file. 1368 1369 ``bps=b,bps_rd=r,bps_wr=w`` 1370 Specify bandwidth throttling limits in bytes per second, either 1371 for all request types or for reads or writes only. Small values 1372 can lead to timeouts or hangs inside the guest. A safe minimum 1373 for disks is 2 MB/s. 1374 1375 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm`` 1376 Specify bursts in bytes per second, either for all request types 1377 or for reads or writes only. Bursts allow the guest I/O to spike 1378 above the limit temporarily. 1379 1380 ``iops=i,iops_rd=r,iops_wr=w`` 1381 Specify request rate limits in requests per second, either for 1382 all request types or for reads or writes only. 1383 1384 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm`` 1385 Specify bursts in requests per second, either for all request 1386 types or for reads or writes only. Bursts allow the guest I/O to 1387 spike above the limit temporarily. 1388 1389 ``iops_size=is`` 1390 Let every is bytes of a request count as a new request for iops 1391 throttling purposes. Use this option to prevent guests from 1392 circumventing iops limits by sending fewer but larger requests. 1393 1394 ``group=g`` 1395 Join a throttling quota group with given name g. All drives that 1396 are members of the same group are accounted for together. Use 1397 this option to prevent guests from circumventing throttling 1398 limits by using many small disks instead of a single larger 1399 disk. 1400 1401 By default, the ``cache.writeback=on`` mode is used. It will report 1402 data writes as completed as soon as the data is present in the host 1403 page cache. This is safe as long as your guest OS makes sure to 1404 correctly flush disk caches where needed. If your guest OS does not 1405 handle volatile disk write caches correctly and your host crashes or 1406 loses power, then the guest may experience data corruption. 1407 1408 For such guests, you should consider using ``cache.writeback=off``. 1409 This means that the host page cache will be used to read and write 1410 data, but write notification will be sent to the guest only after 1411 QEMU has made sure to flush each write to the disk. Be aware that 1412 this has a major impact on performance. 1413 1414 When using the ``-snapshot`` option, unsafe caching is always used. 1415 1416 Copy-on-read avoids accessing the same backing file sectors 1417 repeatedly and is useful when the backing file is over a slow 1418 network. By default copy-on-read is off. 1419 1420 Instead of ``-cdrom`` you can use: 1421 1422 .. parsed-literal:: 1423 1424 |qemu_system| -drive file=file,index=2,media=cdrom 1425 1426 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use: 1427 1428 .. parsed-literal:: 1429 1430 |qemu_system| -drive file=file,index=0,media=disk 1431 |qemu_system| -drive file=file,index=1,media=disk 1432 |qemu_system| -drive file=file,index=2,media=disk 1433 |qemu_system| -drive file=file,index=3,media=disk 1434 1435 You can open an image using pre-opened file descriptors from an fd 1436 set: 1437 1438 .. parsed-literal:: 1439 1440 |qemu_system| \\ 1441 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 1442 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 1443 -drive file=/dev/fdset/2,index=0,media=disk 1444 1445 You can connect a CDROM to the slave of ide0: 1446 1447 .. parsed-literal:: 1448 1449 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom 1450 1451 If you don't specify the "file=" argument, you define an empty 1452 drive: 1453 1454 .. parsed-literal:: 1455 1456 |qemu_system_x86| -drive if=ide,index=1,media=cdrom 1457 1458 Instead of ``-fda``, ``-fdb``, you can use: 1459 1460 .. parsed-literal:: 1461 1462 |qemu_system_x86| -drive file=file,index=0,if=floppy 1463 |qemu_system_x86| -drive file=file,index=1,if=floppy 1464 1465 By default, interface is "ide" and index is automatically 1466 incremented: 1467 1468 .. parsed-literal:: 1469 1470 |qemu_system_x86| -drive file=a -drive file=b" 1471 1472 is interpreted like: 1473 1474 .. parsed-literal:: 1475 1476 |qemu_system_x86| -hda a -hdb b 1477ERST 1478 1479DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock, 1480 "-mtdblock file use 'file' as on-board Flash memory image\n", 1481 QEMU_ARCH_ALL) 1482SRST 1483``-mtdblock file`` 1484 Use file as on-board Flash memory image. 1485ERST 1486 1487DEF("sd", HAS_ARG, QEMU_OPTION_sd, 1488 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL) 1489SRST 1490``-sd file`` 1491 Use file as SecureDigital card image. 1492ERST 1493 1494DEF("pflash", HAS_ARG, QEMU_OPTION_pflash, 1495 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL) 1496SRST 1497``-pflash file`` 1498 Use file as a parallel flash image. 1499ERST 1500 1501DEF("snapshot", 0, QEMU_OPTION_snapshot, 1502 "-snapshot write to temporary files instead of disk image files\n", 1503 QEMU_ARCH_ALL) 1504SRST 1505``-snapshot`` 1506 Write to temporary files instead of disk image files. In this case, 1507 the raw disk image you use is not written back. You can however 1508 force the write back by pressing C-a s (see the :ref:`disk images` 1509 chapter in the System Emulation Users Guide). 1510ERST 1511 1512DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev, 1513 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1514 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n" 1515 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n" 1516 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n" 1517 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n" 1518 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n" 1519 " [[,throttling.iops-size=is]]\n" 1520 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n" 1521 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n" 1522 "-fsdev synth,id=id\n", 1523 QEMU_ARCH_ALL) 1524 1525SRST 1526``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]`` 1527 \ 1528``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]`` 1529 \ 1530``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]`` 1531 \ 1532``-fsdev synth,id=id[,readonly=on]`` 1533 Define a new file system device. Valid options are: 1534 1535 ``local`` 1536 Accesses to the filesystem are done by QEMU. 1537 1538 ``proxy`` 1539 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1540 1541 ``synth`` 1542 Synthetic filesystem, only used by QTests. 1543 1544 ``id=id`` 1545 Specifies identifier for this device. 1546 1547 ``path=path`` 1548 Specifies the export path for the file system device. Files 1549 under this path will be available to the 9p client on the guest. 1550 1551 ``security_model=security_model`` 1552 Specifies the security model to be used for this export path. 1553 Supported security models are "passthrough", "mapped-xattr", 1554 "mapped-file" and "none". In "passthrough" security model, files 1555 are stored using the same credentials as they are created on the 1556 guest. This requires QEMU to run as root. In "mapped-xattr" 1557 security model, some of the file attributes like uid, gid, mode 1558 bits and link target are stored as file attributes. For 1559 "mapped-file" these attributes are stored in the hidden 1560 .virtfs\_metadata directory. Directories exported by this 1561 security model cannot interact with other unix tools. "none" 1562 security model is same as passthrough except the sever won't 1563 report failures if it fails to set file attributes like 1564 ownership. Security model is mandatory only for local fsdriver. 1565 Other fsdrivers (like proxy) don't take security model as a 1566 parameter. 1567 1568 ``writeout=writeout`` 1569 This is an optional argument. The only supported value is 1570 "immediate". This means that host page cache will be used to 1571 read and write data but write notification will be sent to the 1572 guest only when the data has been reported as written by the 1573 storage subsystem. 1574 1575 ``readonly=on`` 1576 Enables exporting 9p share as a readonly mount for guests. By 1577 default read-write access is given. 1578 1579 ``socket=socket`` 1580 Enables proxy filesystem driver to use passed socket file for 1581 communicating with virtfs-proxy-helper(1). 1582 1583 ``sock_fd=sock_fd`` 1584 Enables proxy filesystem driver to use passed socket descriptor 1585 for communicating with virtfs-proxy-helper(1). Usually a helper 1586 like libvirt will create socketpair and pass one of the fds as 1587 sock\_fd. 1588 1589 ``fmode=fmode`` 1590 Specifies the default mode for newly created files on the host. 1591 Works only with security models "mapped-xattr" and 1592 "mapped-file". 1593 1594 ``dmode=dmode`` 1595 Specifies the default mode for newly created directories on the 1596 host. Works only with security models "mapped-xattr" and 1597 "mapped-file". 1598 1599 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w`` 1600 Specify bandwidth throttling limits in bytes per second, either 1601 for all request types or for reads or writes only. 1602 1603 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm`` 1604 Specify bursts in bytes per second, either for all request types 1605 or for reads or writes only. Bursts allow the guest I/O to spike 1606 above the limit temporarily. 1607 1608 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w`` 1609 Specify request rate limits in requests per second, either for 1610 all request types or for reads or writes only. 1611 1612 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm`` 1613 Specify bursts in requests per second, either for all request 1614 types or for reads or writes only. Bursts allow the guest I/O to 1615 spike above the limit temporarily. 1616 1617 ``throttling.iops-size=is`` 1618 Let every is bytes of a request count as a new request for iops 1619 throttling purposes. 1620 1621 -fsdev option is used along with -device driver "virtio-9p-...". 1622 1623``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag`` 1624 Options for virtio-9p-... driver are: 1625 1626 ``type`` 1627 Specifies the variant to be used. Supported values are "pci", 1628 "ccw" or "device", depending on the machine type. 1629 1630 ``fsdev=id`` 1631 Specifies the id value specified along with -fsdev option. 1632 1633 ``mount_tag=mount_tag`` 1634 Specifies the tag name to be used by the guest to mount this 1635 export point. 1636ERST 1637 1638DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs, 1639 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1640 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n" 1641 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n" 1642 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n" 1643 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n", 1644 QEMU_ARCH_ALL) 1645 1646SRST 1647``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]`` 1648 \ 1649``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1650 \ 1651``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1652 \ 1653``-virtfs synth,mount_tag=mount_tag`` 1654 Define a new virtual filesystem device and expose it to the guest using 1655 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain 1656 directory on host is made directly accessible by guest as a pass-through 1657 file system by using the 9P network protocol for communication between 1658 host and guests, if desired even accessible, shared by several guests 1659 simultaniously. 1660 1661 Note that ``-virtfs`` is actually just a convenience shortcut for its 1662 generalized form ``-fsdev -device virtio-9p-pci``. 1663 1664 The general form of pass-through file system options are: 1665 1666 ``local`` 1667 Accesses to the filesystem are done by QEMU. 1668 1669 ``proxy`` 1670 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1671 1672 ``synth`` 1673 Synthetic filesystem, only used by QTests. 1674 1675 ``id=id`` 1676 Specifies identifier for the filesystem device 1677 1678 ``path=path`` 1679 Specifies the export path for the file system device. Files 1680 under this path will be available to the 9p client on the guest. 1681 1682 ``security_model=security_model`` 1683 Specifies the security model to be used for this export path. 1684 Supported security models are "passthrough", "mapped-xattr", 1685 "mapped-file" and "none". In "passthrough" security model, files 1686 are stored using the same credentials as they are created on the 1687 guest. This requires QEMU to run as root. In "mapped-xattr" 1688 security model, some of the file attributes like uid, gid, mode 1689 bits and link target are stored as file attributes. For 1690 "mapped-file" these attributes are stored in the hidden 1691 .virtfs\_metadata directory. Directories exported by this 1692 security model cannot interact with other unix tools. "none" 1693 security model is same as passthrough except the sever won't 1694 report failures if it fails to set file attributes like 1695 ownership. Security model is mandatory only for local fsdriver. 1696 Other fsdrivers (like proxy) don't take security model as a 1697 parameter. 1698 1699 ``writeout=writeout`` 1700 This is an optional argument. The only supported value is 1701 "immediate". This means that host page cache will be used to 1702 read and write data but write notification will be sent to the 1703 guest only when the data has been reported as written by the 1704 storage subsystem. 1705 1706 ``readonly=on`` 1707 Enables exporting 9p share as a readonly mount for guests. By 1708 default read-write access is given. 1709 1710 ``socket=socket`` 1711 Enables proxy filesystem driver to use passed socket file for 1712 communicating with virtfs-proxy-helper(1). Usually a helper like 1713 libvirt will create socketpair and pass one of the fds as 1714 sock\_fd. 1715 1716 ``sock_fd`` 1717 Enables proxy filesystem driver to use passed 'sock\_fd' as the 1718 socket descriptor for interfacing with virtfs-proxy-helper(1). 1719 1720 ``fmode=fmode`` 1721 Specifies the default mode for newly created files on the host. 1722 Works only with security models "mapped-xattr" and 1723 "mapped-file". 1724 1725 ``dmode=dmode`` 1726 Specifies the default mode for newly created directories on the 1727 host. Works only with security models "mapped-xattr" and 1728 "mapped-file". 1729 1730 ``mount_tag=mount_tag`` 1731 Specifies the tag name to be used by the guest to mount this 1732 export point. 1733 1734 ``multidevs=multidevs`` 1735 Specifies how to deal with multiple devices being shared with a 1736 9p export. Supported behaviours are either "remap", "forbid" or 1737 "warn". The latter is the default behaviour on which virtfs 9p 1738 expects only one device to be shared with the same export, and 1739 if more than one device is shared and accessed via the same 9p 1740 export then only a warning message is logged (once) by qemu on 1741 host side. In order to avoid file ID collisions on guest you 1742 should either create a separate virtfs export for each device to 1743 be shared with guests (recommended way) or you might use "remap" 1744 instead which allows you to share multiple devices with only one 1745 export instead, which is achieved by remapping the original 1746 inode numbers from host to guest in a way that would prevent 1747 such collisions. Remapping inodes in such use cases is required 1748 because the original device IDs from host are never passed and 1749 exposed on guest. Instead all files of an export shared with 1750 virtfs always share the same device id on guest. So two files 1751 with identical inode numbers but from actually different devices 1752 on host would otherwise cause a file ID collision and hence 1753 potential misbehaviours on guest. "forbid" on the other hand 1754 assumes like "warn" that only one device is shared by the same 1755 export, however it will not only log a warning message but also 1756 deny access to additional devices on guest. Note though that 1757 "forbid" does currently not block all possible file access 1758 operations (e.g. readdir() would still return entries from other 1759 devices). 1760ERST 1761 1762DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi, 1763 "-iscsi [user=user][,password=password]\n" 1764 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE\n" 1765 " [,initiator-name=initiator-iqn][,id=target-iqn]\n" 1766 " [,timeout=timeout]\n" 1767 " iSCSI session parameters\n", QEMU_ARCH_ALL) 1768 1769SRST 1770``-iscsi`` 1771 Configure iSCSI session parameters. 1772ERST 1773 1774DEFHEADING() 1775 1776DEFHEADING(USB convenience options:) 1777 1778DEF("usb", 0, QEMU_OPTION_usb, 1779 "-usb enable on-board USB host controller (if not enabled by default)\n", 1780 QEMU_ARCH_ALL) 1781SRST 1782``-usb`` 1783 Enable USB emulation on machine types with an on-board USB host 1784 controller (if not enabled by default). Note that on-board USB host 1785 controllers may not support USB 3.0. In this case 1786 ``-device qemu-xhci`` can be used instead on machines with PCI. 1787ERST 1788 1789DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice, 1790 "-usbdevice name add the host or guest USB device 'name'\n", 1791 QEMU_ARCH_ALL) 1792SRST 1793``-usbdevice devname`` 1794 Add the USB device devname, and enable an on-board USB controller 1795 if possible and necessary (just like it can be done via 1796 ``-machine usb=on``). Note that this option is mainly intended for 1797 the user's convenience only. More fine-grained control can be 1798 achieved by selecting a USB host controller (if necessary) and the 1799 desired USB device via the ``-device`` option instead. For example, 1800 instead of using ``-usbdevice mouse`` it is possible to use 1801 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse 1802 to a USB 3.0 controller instead (at least on machines that support 1803 PCI and do not have an USB controller enabled by default yet). 1804 For more details, see the chapter about 1805 :ref:`Connecting USB devices` in the System Emulation Users Guide. 1806 Possible devices for devname are: 1807 1808 ``braille`` 1809 Braille device. This will use BrlAPI to display the braille 1810 output on a real or fake device (i.e. it also creates a 1811 corresponding ``braille`` chardev automatically beside the 1812 ``usb-braille`` USB device). 1813 1814 ``keyboard`` 1815 Standard USB keyboard. Will override the PS/2 keyboard (if present). 1816 1817 ``mouse`` 1818 Virtual Mouse. This will override the PS/2 mouse emulation when 1819 activated. 1820 1821 ``tablet`` 1822 Pointer device that uses absolute coordinates (like a 1823 touchscreen). This means QEMU is able to report the mouse 1824 position without having to grab the mouse. Also overrides the 1825 PS/2 mouse emulation when activated. 1826 1827 ``wacom-tablet`` 1828 Wacom PenPartner USB tablet. 1829 1830 1831ERST 1832 1833DEFHEADING() 1834 1835DEFHEADING(Display options:) 1836 1837DEF("display", HAS_ARG, QEMU_OPTION_display, 1838#if defined(CONFIG_SPICE) 1839 "-display spice-app[,gl=on|off]\n" 1840#endif 1841#if defined(CONFIG_SDL) 1842 "-display sdl[,alt_grab=on|off][,ctrl_grab=on|off][,gl=on|core|es|off]\n" 1843 " [,grab-mod=<mod>][,show-cursor=on|off][,window-close=on|off]\n" 1844#endif 1845#if defined(CONFIG_GTK) 1846 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n" 1847 " [,show-cursor=on|off][,window-close=on|off]\n" 1848#endif 1849#if defined(CONFIG_VNC) 1850 "-display vnc=<display>[,<optargs>]\n" 1851#endif 1852#if defined(CONFIG_CURSES) 1853 "-display curses[,charset=<encoding>]\n" 1854#endif 1855#if defined(CONFIG_OPENGL) 1856 "-display egl-headless[,rendernode=<file>]\n" 1857#endif 1858 "-display none\n" 1859 " select display backend type\n" 1860 " The default display is equivalent to\n " 1861#if defined(CONFIG_GTK) 1862 "\"-display gtk\"\n" 1863#elif defined(CONFIG_SDL) 1864 "\"-display sdl\"\n" 1865#elif defined(CONFIG_COCOA) 1866 "\"-display cocoa\"\n" 1867#elif defined(CONFIG_VNC) 1868 "\"-vnc localhost:0,to=99,id=default\"\n" 1869#else 1870 "\"-display none\"\n" 1871#endif 1872 , QEMU_ARCH_ALL) 1873SRST 1874``-display type`` 1875 Select type of display to use. This option is a replacement for the 1876 old style -sdl/-curses/... options. Use ``-display help`` to list 1877 the available display types. Valid values for type are 1878 1879 ``spice-app[,gl=on|off]`` 1880 Start QEMU as a Spice server and launch the default Spice client 1881 application. The Spice server will redirect the serial consoles 1882 and QEMU monitors. (Since 4.0) 1883 1884 ``sdl`` 1885 Display video output via SDL (usually in a separate graphics 1886 window; see the SDL documentation for other possibilities). 1887 Valid parameters are: 1888 1889 ``grab-mod=<mods>`` : Used to select the modifier keys for toggling 1890 the mouse grabbing in conjunction with the "g" key. `<mods>` can be 1891 either `lshift-lctrl-lalt` or `rctrl`. 1892 1893 ``alt_grab=on|off`` : Use Control+Alt+Shift-g to toggle mouse grabbing. 1894 This parameter is deprecated - use ``grab-mod`` instead. 1895 1896 ``ctrl_grab=on|off`` : Use Right-Control-g to toggle mouse grabbing. 1897 This parameter is deprecated - use ``grab-mod`` instead. 1898 1899 ``gl=on|off|core|es`` : Use OpenGL for displaying 1900 1901 ``show-cursor=on|off`` : Force showing the mouse cursor 1902 1903 ``window-close=on|off`` : Allow to quit qemu with window close button 1904 1905 ``gtk`` 1906 Display video output in a GTK window. This interface provides 1907 drop-down menus and other UI elements to configure and control 1908 the VM during runtime. Valid parameters are: 1909 1910 ``full-screen=on|off`` : Start in fullscreen mode 1911 1912 ``gl=on|off`` : Use OpenGL for displaying 1913 1914 ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover 1915 1916 ``show-cursor=on|off`` : Force showing the mouse cursor 1917 1918 ``window-close=on|off`` : Allow to quit qemu with window close button 1919 1920 ``curses[,charset=<encoding>]`` 1921 Display video output via curses. For graphics device models 1922 which support a text mode, QEMU can display this output using a 1923 curses/ncurses interface. Nothing is displayed when the graphics 1924 device is in graphical mode or if the graphics device does not 1925 support a text mode. Generally only the VGA device models 1926 support text mode. The font charset used by the guest can be 1927 specified with the ``charset`` option, for example 1928 ``charset=CP850`` for IBM CP850 encoding. The default is 1929 ``CP437``. 1930 1931 ``egl-headless[,rendernode=<file>]`` 1932 Offload all OpenGL operations to a local DRI device. For any 1933 graphical display, this display needs to be paired with either 1934 VNC or SPICE displays. 1935 1936 ``vnc=<display>`` 1937 Start a VNC server on display <display> 1938 1939 ``none`` 1940 Do not display video output. The guest will still see an 1941 emulated graphics card, but its output will not be displayed to 1942 the QEMU user. This option differs from the -nographic option in 1943 that it only affects what is done with video output; -nographic 1944 also changes the destination of the serial and parallel port 1945 data. 1946ERST 1947 1948DEF("nographic", 0, QEMU_OPTION_nographic, 1949 "-nographic disable graphical output and redirect serial I/Os to console\n", 1950 QEMU_ARCH_ALL) 1951SRST 1952``-nographic`` 1953 Normally, if QEMU is compiled with graphical window support, it 1954 displays output such as guest graphics, guest console, and the QEMU 1955 monitor in a window. With this option, you can totally disable 1956 graphical output so that QEMU is a simple command line application. 1957 The emulated serial port is redirected on the console and muxed with 1958 the monitor (unless redirected elsewhere explicitly). Therefore, you 1959 can still use QEMU to debug a Linux kernel with a serial console. 1960 Use C-a h for help on switching between the console and monitor. 1961ERST 1962 1963DEF("curses", 0, QEMU_OPTION_curses, 1964 "-curses shorthand for -display curses\n", 1965 QEMU_ARCH_ALL) 1966SRST 1967``-curses`` 1968 Normally, if QEMU is compiled with graphical window support, it 1969 displays output such as guest graphics, guest console, and the QEMU 1970 monitor in a window. With this option, QEMU can display the VGA 1971 output when in text mode using a curses/ncurses interface. Nothing 1972 is displayed in graphical mode. 1973ERST 1974 1975DEF("alt-grab", 0, QEMU_OPTION_alt_grab, 1976 "-alt-grab use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt)\n", 1977 QEMU_ARCH_ALL) 1978SRST 1979``-alt-grab`` 1980 Use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt). Note that 1981 this also affects the special keys (for fullscreen, monitor-mode 1982 switching, etc). This option is deprecated - please use 1983 ``-display sdl,grab-mod=lshift-lctrl-lalt`` instead. 1984ERST 1985 1986DEF("ctrl-grab", 0, QEMU_OPTION_ctrl_grab, 1987 "-ctrl-grab use Right-Ctrl to grab mouse (instead of Ctrl-Alt)\n", 1988 QEMU_ARCH_ALL) 1989SRST 1990``-ctrl-grab`` 1991 Use Right-Ctrl to grab mouse (instead of Ctrl-Alt). Note that this 1992 also affects the special keys (for fullscreen, monitor-mode 1993 switching, etc). This option is deprecated - please use 1994 ``-display sdl,grab-mod=rctrl`` instead. 1995ERST 1996 1997DEF("no-quit", 0, QEMU_OPTION_no_quit, 1998 "-no-quit disable SDL/GTK window close capability (deprecated)\n", QEMU_ARCH_ALL) 1999SRST 2000``-no-quit`` 2001 Disable window close capability (SDL and GTK only). This option is 2002 deprecated, please use ``-display ...,window-close=off`` instead. 2003ERST 2004 2005DEF("sdl", 0, QEMU_OPTION_sdl, 2006 "-sdl shorthand for -display sdl\n", QEMU_ARCH_ALL) 2007SRST 2008``-sdl`` 2009 Enable SDL. 2010ERST 2011 2012DEF("spice", HAS_ARG, QEMU_OPTION_spice, 2013 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n" 2014 " [,x509-key-file=<file>][,x509-key-password=<file>]\n" 2015 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n" 2016 " [,x509-dh-key-file=<file>][,addr=addr]\n" 2017 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n" 2018 " [,tls-ciphers=<list>]\n" 2019 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n" 2020 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n" 2021 " [,sasl=on|off][,disable-ticketing=on|off]\n" 2022 " [,password=<string>][,password-secret=<secret-id>]\n" 2023 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n" 2024 " [,jpeg-wan-compression=[auto|never|always]]\n" 2025 " [,zlib-glz-wan-compression=[auto|never|always]]\n" 2026 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n" 2027 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n" 2028 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n" 2029 " [,gl=[on|off]][,rendernode=<file>]\n" 2030 " enable spice\n" 2031 " at least one of {port, tls-port} is mandatory\n", 2032 QEMU_ARCH_ALL) 2033SRST 2034``-spice option[,option[,...]]`` 2035 Enable the spice remote desktop protocol. Valid options are 2036 2037 ``port=<nr>`` 2038 Set the TCP port spice is listening on for plaintext channels. 2039 2040 ``addr=<addr>`` 2041 Set the IP address spice is listening on. Default is any 2042 address. 2043 2044 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off`` 2045 Force using the specified IP version. 2046 2047 ``password=<string>`` 2048 Set the password you need to authenticate. 2049 2050 This option is deprecated and insecure because it leaves the 2051 password visible in the process listing. Use ``password-secret`` 2052 instead. 2053 2054 ``password-secret=<secret-id>`` 2055 Set the ID of the ``secret`` object containing the password 2056 you need to authenticate. 2057 2058 ``sasl=on|off`` 2059 Require that the client use SASL to authenticate with the spice. 2060 The exact choice of authentication method used is controlled 2061 from the system / user's SASL configuration file for the 'qemu' 2062 service. This is typically found in /etc/sasl2/qemu.conf. If 2063 running QEMU as an unprivileged user, an environment variable 2064 SASL\_CONF\_PATH can be used to make it search alternate 2065 locations for the service config. While some SASL auth methods 2066 can also provide data encryption (eg GSSAPI), it is recommended 2067 that SASL always be combined with the 'tls' and 'x509' settings 2068 to enable use of SSL and server certificates. This ensures a 2069 data encryption preventing compromise of authentication 2070 credentials. 2071 2072 ``disable-ticketing=on|off`` 2073 Allow client connects without authentication. 2074 2075 ``disable-copy-paste=on|off`` 2076 Disable copy paste between the client and the guest. 2077 2078 ``disable-agent-file-xfer=on|off`` 2079 Disable spice-vdagent based file-xfer between the client and the 2080 guest. 2081 2082 ``tls-port=<nr>`` 2083 Set the TCP port spice is listening on for encrypted channels. 2084 2085 ``x509-dir=<dir>`` 2086 Set the x509 file directory. Expects same filenames as -vnc 2087 $display,x509=$dir 2088 2089 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>`` 2090 The x509 file names can also be configured individually. 2091 2092 ``tls-ciphers=<list>`` 2093 Specify which ciphers to use. 2094 2095 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]`` 2096 Force specific channel to be used with or without TLS 2097 encryption. The options can be specified multiple times to 2098 configure multiple channels. The special name "default" can be 2099 used to set the default mode. For channels which are not 2100 explicitly forced into one mode the spice client is allowed to 2101 pick tls/plaintext as he pleases. 2102 2103 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]`` 2104 Configure image compression (lossless). Default is auto\_glz. 2105 2106 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]`` 2107 Configure wan image compression (lossy for slow links). Default 2108 is auto. 2109 2110 ``streaming-video=[off|all|filter]`` 2111 Configure video stream detection. Default is off. 2112 2113 ``agent-mouse=[on|off]`` 2114 Enable/disable passing mouse events via vdagent. Default is on. 2115 2116 ``playback-compression=[on|off]`` 2117 Enable/disable audio stream compression (using celt 0.5.1). 2118 Default is on. 2119 2120 ``seamless-migration=[on|off]`` 2121 Enable/disable spice seamless migration. Default is off. 2122 2123 ``gl=[on|off]`` 2124 Enable/disable OpenGL context. Default is off. 2125 2126 ``rendernode=<file>`` 2127 DRM render node for OpenGL rendering. If not specified, it will 2128 pick the first available. (Since 2.9) 2129ERST 2130 2131DEF("portrait", 0, QEMU_OPTION_portrait, 2132 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n", 2133 QEMU_ARCH_ALL) 2134SRST 2135``-portrait`` 2136 Rotate graphical output 90 deg left (only PXA LCD). 2137ERST 2138 2139DEF("rotate", HAS_ARG, QEMU_OPTION_rotate, 2140 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n", 2141 QEMU_ARCH_ALL) 2142SRST 2143``-rotate deg`` 2144 Rotate graphical output some deg left (only PXA LCD). 2145ERST 2146 2147DEF("vga", HAS_ARG, QEMU_OPTION_vga, 2148 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n" 2149 " select video card type\n", QEMU_ARCH_ALL) 2150SRST 2151``-vga type`` 2152 Select type of VGA card to emulate. Valid values for type are 2153 2154 ``cirrus`` 2155 Cirrus Logic GD5446 Video card. All Windows versions starting 2156 from Windows 95 should recognize and use this graphic card. For 2157 optimal performances, use 16 bit color depth in the guest and 2158 the host OS. (This card was the default before QEMU 2.2) 2159 2160 ``std`` 2161 Standard VGA card with Bochs VBE extensions. If your guest OS 2162 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if 2163 you want to use high resolution modes (>= 1280x1024x16) then you 2164 should use this option. (This card is the default since QEMU 2165 2.2) 2166 2167 ``vmware`` 2168 VMWare SVGA-II compatible adapter. Use it if you have 2169 sufficiently recent XFree86/XOrg server or Windows guest with a 2170 driver for this card. 2171 2172 ``qxl`` 2173 QXL paravirtual graphic card. It is VGA compatible (including 2174 VESA 2.0 VBE support). Works best with qxl guest drivers 2175 installed though. Recommended choice when using the spice 2176 protocol. 2177 2178 ``tcx`` 2179 (sun4m only) Sun TCX framebuffer. This is the default 2180 framebuffer for sun4m machines and offers both 8-bit and 24-bit 2181 colour depths at a fixed resolution of 1024x768. 2182 2183 ``cg3`` 2184 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit 2185 framebuffer for sun4m machines available in both 1024x768 2186 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people 2187 wishing to run older Solaris versions. 2188 2189 ``virtio`` 2190 Virtio VGA card. 2191 2192 ``none`` 2193 Disable VGA card. 2194ERST 2195 2196DEF("full-screen", 0, QEMU_OPTION_full_screen, 2197 "-full-screen start in full screen\n", QEMU_ARCH_ALL) 2198SRST 2199``-full-screen`` 2200 Start in full screen. 2201ERST 2202 2203DEF("g", HAS_ARG, QEMU_OPTION_g , 2204 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n", 2205 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K) 2206SRST 2207``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]`` 2208 Set the initial graphical resolution and depth (PPC, SPARC only). 2209 2210 For PPC the default is 800x600x32. 2211 2212 For SPARC with the TCX graphics device, the default is 1024x768x8 2213 with the option of 1024x768x24. For cgthree, the default is 2214 1024x768x8 with the option of 1152x900x8 for people who wish to use 2215 OBP. 2216ERST 2217 2218DEF("vnc", HAS_ARG, QEMU_OPTION_vnc , 2219 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL) 2220SRST 2221``-vnc display[,option[,option[,...]]]`` 2222 Normally, if QEMU is compiled with graphical window support, it 2223 displays output such as guest graphics, guest console, and the QEMU 2224 monitor in a window. With this option, you can have QEMU listen on 2225 VNC display display and redirect the VGA display over the VNC 2226 session. It is very useful to enable the usb tablet device when 2227 using this option (option ``-device usb-tablet``). When using the 2228 VNC display, you must use the ``-k`` parameter to set the keyboard 2229 layout if you are not using en-us. Valid syntax for the display is 2230 2231 ``to=L`` 2232 With this option, QEMU will try next available VNC displays, 2233 until the number L, if the origianlly defined "-vnc display" is 2234 not available, e.g. port 5900+display is already used by another 2235 application. By default, to=0. 2236 2237 ``host:d`` 2238 TCP connections will only be allowed from host on display d. By 2239 convention the TCP port is 5900+d. Optionally, host can be 2240 omitted in which case the server will accept connections from 2241 any host. 2242 2243 ``unix:path`` 2244 Connections will be allowed over UNIX domain sockets where path 2245 is the location of a unix socket to listen for connections on. 2246 2247 ``none`` 2248 VNC is initialized but not started. The monitor ``change`` 2249 command can be used to later start the VNC server. 2250 2251 Following the display value there may be one or more option flags 2252 separated by commas. Valid options are 2253 2254 ``reverse=on|off`` 2255 Connect to a listening VNC client via a "reverse" connection. 2256 The client is specified by the display. For reverse network 2257 connections (host:d,``reverse``), the d argument is a TCP port 2258 number, not a display number. 2259 2260 ``websocket=on|off`` 2261 Opens an additional TCP listening port dedicated to VNC 2262 Websocket connections. If a bare websocket option is given, the 2263 Websocket port is 5700+display. An alternative port can be 2264 specified with the syntax ``websocket``\ =port. 2265 2266 If host is specified connections will only be allowed from this 2267 host. It is possible to control the websocket listen address 2268 independently, using the syntax ``websocket``\ =host:port. 2269 2270 If no TLS credentials are provided, the websocket connection 2271 runs in unencrypted mode. If TLS credentials are provided, the 2272 websocket connection requires encrypted client connections. 2273 2274 ``password=on|off`` 2275 Require that password based authentication is used for client 2276 connections. 2277 2278 The password must be set separately using the ``set_password`` 2279 command in the :ref:`QEMU monitor`. The 2280 syntax to change your password is: 2281 ``set_password <protocol> <password>`` where <protocol> could be 2282 either "vnc" or "spice". 2283 2284 If you would like to change <protocol> password expiration, you 2285 should use ``expire_password <protocol> <expiration-time>`` 2286 where expiration time could be one of the following options: 2287 now, never, +seconds or UNIX time of expiration, e.g. +60 to 2288 make password expire in 60 seconds, or 1335196800 to make 2289 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for 2290 this date and time). 2291 2292 You can also use keywords "now" or "never" for the expiration 2293 time to allow <protocol> password to expire immediately or never 2294 expire. 2295 2296 ``password-secret=<secret-id>`` 2297 Require that password based authentication is used for client 2298 connections, using the password provided by the ``secret`` 2299 object identified by ``secret-id``. 2300 2301 ``tls-creds=ID`` 2302 Provides the ID of a set of TLS credentials to use to secure the 2303 VNC server. They will apply to both the normal VNC server socket 2304 and the websocket socket (if enabled). Setting TLS credentials 2305 will cause the VNC server socket to enable the VeNCrypt auth 2306 mechanism. The credentials should have been previously created 2307 using the ``-object tls-creds`` argument. 2308 2309 ``tls-authz=ID`` 2310 Provides the ID of the QAuthZ authorization object against which 2311 the client's x509 distinguished name will validated. This object 2312 is only resolved at time of use, so can be deleted and recreated 2313 on the fly while the VNC server is active. If missing, it will 2314 default to denying access. 2315 2316 ``sasl=on|off`` 2317 Require that the client use SASL to authenticate with the VNC 2318 server. The exact choice of authentication method used is 2319 controlled from the system / user's SASL configuration file for 2320 the 'qemu' service. This is typically found in 2321 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user, 2322 an environment variable SASL\_CONF\_PATH can be used to make it 2323 search alternate locations for the service config. While some 2324 SASL auth methods can also provide data encryption (eg GSSAPI), 2325 it is recommended that SASL always be combined with the 'tls' 2326 and 'x509' settings to enable use of SSL and server 2327 certificates. This ensures a data encryption preventing 2328 compromise of authentication credentials. See the 2329 :ref:`VNC security` section in the System Emulation Users Guide 2330 for details on using SASL authentication. 2331 2332 ``sasl-authz=ID`` 2333 Provides the ID of the QAuthZ authorization object against which 2334 the client's SASL username will validated. This object is only 2335 resolved at time of use, so can be deleted and recreated on the 2336 fly while the VNC server is active. If missing, it will default 2337 to denying access. 2338 2339 ``acl=on|off`` 2340 Legacy method for enabling authorization of clients against the 2341 x509 distinguished name and SASL username. It results in the 2342 creation of two ``authz-list`` objects with IDs of 2343 ``vnc.username`` and ``vnc.x509dname``. The rules for these 2344 objects must be configured with the HMP ACL commands. 2345 2346 This option is deprecated and should no longer be used. The new 2347 ``sasl-authz`` and ``tls-authz`` options are a replacement. 2348 2349 ``lossy=on|off`` 2350 Enable lossy compression methods (gradient, JPEG, ...). If this 2351 option is set, VNC client may receive lossy framebuffer updates 2352 depending on its encoding settings. Enabling this option can 2353 save a lot of bandwidth at the expense of quality. 2354 2355 ``non-adaptive=on|off`` 2356 Disable adaptive encodings. Adaptive encodings are enabled by 2357 default. An adaptive encoding will try to detect frequently 2358 updated screen regions, and send updates in these regions using 2359 a lossy encoding (like JPEG). This can be really helpful to save 2360 bandwidth when playing videos. Disabling adaptive encodings 2361 restores the original static behavior of encodings like Tight. 2362 2363 ``share=[allow-exclusive|force-shared|ignore]`` 2364 Set display sharing policy. 'allow-exclusive' allows clients to 2365 ask for exclusive access. As suggested by the rfb spec this is 2366 implemented by dropping other connections. Connecting multiple 2367 clients in parallel requires all clients asking for a shared 2368 session (vncviewer: -shared switch). This is the default. 2369 'force-shared' disables exclusive client access. Useful for 2370 shared desktop sessions, where you don't want someone forgetting 2371 specify -shared disconnect everybody else. 'ignore' completely 2372 ignores the shared flag and allows everybody connect 2373 unconditionally. Doesn't conform to the rfb spec but is 2374 traditional QEMU behavior. 2375 2376 ``key-delay-ms`` 2377 Set keyboard delay, for key down and key up events, in 2378 milliseconds. Default is 10. Keyboards are low-bandwidth 2379 devices, so this slowdown can help the device and guest to keep 2380 up and not lose events in case events are arriving in bulk. 2381 Possible causes for the latter are flaky network connections, or 2382 scripts for automated testing. 2383 2384 ``audiodev=audiodev`` 2385 Use the specified audiodev when the VNC client requests audio 2386 transmission. When not using an -audiodev argument, this option 2387 must be omitted, otherwise is must be present and specify a 2388 valid audiodev. 2389 2390 ``power-control=on|off`` 2391 Permit the remote client to issue shutdown, reboot or reset power 2392 control requests. 2393ERST 2394 2395ARCHHEADING(, QEMU_ARCH_I386) 2396 2397ARCHHEADING(i386 target only:, QEMU_ARCH_I386) 2398 2399DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack, 2400 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n", 2401 QEMU_ARCH_I386) 2402SRST 2403``-win2k-hack`` 2404 Use it when installing Windows 2000 to avoid a disk full bug. After 2405 Windows 2000 is installed, you no longer need this option (this 2406 option slows down the IDE transfers). 2407ERST 2408 2409DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk, 2410 "-no-fd-bootchk disable boot signature checking for floppy disks\n", 2411 QEMU_ARCH_I386) 2412SRST 2413``-no-fd-bootchk`` 2414 Disable boot signature checking for floppy disks in BIOS. May be 2415 needed to boot from old floppy disks. 2416ERST 2417 2418DEF("no-acpi", 0, QEMU_OPTION_no_acpi, 2419 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2420SRST 2421``-no-acpi`` 2422 Disable ACPI (Advanced Configuration and Power Interface) support. 2423 Use it if your guest OS complains about ACPI problems (PC target 2424 machine only). 2425ERST 2426 2427DEF("no-hpet", 0, QEMU_OPTION_no_hpet, 2428 "-no-hpet disable HPET\n", QEMU_ARCH_I386) 2429SRST 2430``-no-hpet`` 2431 Disable HPET support. 2432ERST 2433 2434DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable, 2435 "-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" 2436 " ACPI table description\n", QEMU_ARCH_I386) 2437SRST 2438``-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]...]`` 2439 Add ACPI table with specified header fields and context from 2440 specified files. For file=, take whole ACPI table from the specified 2441 files, including all ACPI headers (possible overridden by other 2442 options). For data=, only data portion of the table is used, all 2443 header information is specified in the command line. If a SLIC table 2444 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id 2445 fields will override the same in the RSDT and the FADT (a.k.a. 2446 FACP), in order to ensure the field matches required by the 2447 Microsoft SLIC spec and the ACPI spec. 2448ERST 2449 2450DEF("smbios", HAS_ARG, QEMU_OPTION_smbios, 2451 "-smbios file=binary\n" 2452 " load SMBIOS entry from binary file\n" 2453 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n" 2454 " [,uefi=on|off]\n" 2455 " specify SMBIOS type 0 fields\n" 2456 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2457 " [,uuid=uuid][,sku=str][,family=str]\n" 2458 " specify SMBIOS type 1 fields\n" 2459 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2460 " [,asset=str][,location=str]\n" 2461 " specify SMBIOS type 2 fields\n" 2462 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n" 2463 " [,sku=str]\n" 2464 " specify SMBIOS type 3 fields\n" 2465 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n" 2466 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n" 2467 " specify SMBIOS type 4 fields\n" 2468 "-smbios type=11[,value=str][,path=filename]\n" 2469 " specify SMBIOS type 11 fields\n" 2470 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n" 2471 " [,asset=str][,part=str][,speed=%d]\n" 2472 " specify SMBIOS type 17 fields\n" 2473 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n" 2474 " specify SMBIOS type 41 fields\n", 2475 QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2476SRST 2477``-smbios file=binary`` 2478 Load SMBIOS entry from binary file. 2479 2480``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]`` 2481 Specify SMBIOS type 0 fields 2482 2483``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]`` 2484 Specify SMBIOS type 1 fields 2485 2486``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]`` 2487 Specify SMBIOS type 2 fields 2488 2489``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]`` 2490 Specify SMBIOS type 3 fields 2491 2492``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str]`` 2493 Specify SMBIOS type 4 fields 2494 2495``-smbios type=11[,value=str][,path=filename]`` 2496 Specify SMBIOS type 11 fields 2497 2498 This argument can be repeated multiple times, and values are added in the order they are parsed. 2499 Applications intending to use OEM strings data are encouraged to use their application name as 2500 a prefix for the value string. This facilitates passing information for multiple applications 2501 concurrently. 2502 2503 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax 2504 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes. 2505 2506 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to 2507 the SMBIOS table in the order in which they appear. 2508 2509 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535 2510 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the 2511 guest. Instead it should be used as a indicator to inform the guest where to locate the real 2512 data set, for example, by specifying the serial ID of a block device. 2513 2514 An example passing three strings is 2515 2516 .. parsed-literal:: 2517 2518 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\ 2519 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\ 2520 path=/some/file/with/oemstringsdata.txt 2521 2522 In the guest OS this is visible with the ``dmidecode`` command 2523 2524 .. parsed-literal:: 2525 2526 $ dmidecode -t 11 2527 Handle 0x0E00, DMI type 11, 5 bytes 2528 OEM Strings 2529 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/ 2530 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os 2531 String 3: myapp:some extra data 2532 2533 2534``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]`` 2535 Specify SMBIOS type 17 fields 2536 2537``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]`` 2538 Specify SMBIOS type 41 fields 2539 2540 This argument can be repeated multiple times. Its main use is to allow network interfaces be created 2541 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface 2542 position on the PCI bus. 2543 2544 Here is an example of use: 2545 2546 .. parsed-literal:: 2547 2548 -netdev user,id=internet \\ 2549 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\ 2550 -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev 2551 2552 In the guest OS, the device should then appear as ``eno1``: 2553 2554 ..parsed-literal:: 2555 2556 $ ip -brief l 2557 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP> 2558 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP> 2559 2560 Currently, the PCI device has to be attached to the root bus. 2561 2562ERST 2563 2564DEFHEADING() 2565 2566DEFHEADING(Network options:) 2567 2568DEF("netdev", HAS_ARG, QEMU_OPTION_netdev, 2569#ifdef CONFIG_SLIRP 2570 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n" 2571 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n" 2572 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n" 2573 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n" 2574 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]" 2575#ifndef _WIN32 2576 "[,smb=dir[,smbserver=addr]]\n" 2577#endif 2578 " configure a user mode network backend with ID 'str',\n" 2579 " its DHCP server and optional services\n" 2580#endif 2581#ifdef _WIN32 2582 "-netdev tap,id=str,ifname=name\n" 2583 " configure a host TAP network backend with ID 'str'\n" 2584#else 2585 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n" 2586 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n" 2587 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n" 2588 " [,poll-us=n]\n" 2589 " configure a host TAP network backend with ID 'str'\n" 2590 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2591 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n" 2592 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n" 2593 " to deconfigure it\n" 2594 " use '[down]script=no' to disable script execution\n" 2595 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n" 2596 " configure it\n" 2597 " use 'fd=h' to connect to an already opened TAP interface\n" 2598 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n" 2599 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n" 2600 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n" 2601 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n" 2602 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n" 2603 " use vhost=on to enable experimental in kernel accelerator\n" 2604 " (only has effect for virtio guests which use MSIX)\n" 2605 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n" 2606 " use 'vhostfd=h' to connect to an already opened vhost net device\n" 2607 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n" 2608 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n" 2609 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n" 2610 " spent on busy polling for vhost net\n" 2611 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n" 2612 " configure a host TAP network backend with ID 'str' that is\n" 2613 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2614 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n" 2615#endif 2616#ifdef __linux__ 2617 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n" 2618 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n" 2619 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n" 2620 " [,rxcookie=rxcookie][,offset=offset]\n" 2621 " configure a network backend with ID 'str' connected to\n" 2622 " an Ethernet over L2TPv3 pseudowire.\n" 2623 " Linux kernel 3.3+ as well as most routers can talk\n" 2624 " L2TPv3. This transport allows connecting a VM to a VM,\n" 2625 " VM to a router and even VM to Host. It is a nearly-universal\n" 2626 " standard (RFC3931). Note - this implementation uses static\n" 2627 " pre-configured tunnels (same as the Linux kernel).\n" 2628 " use 'src=' to specify source address\n" 2629 " use 'dst=' to specify destination address\n" 2630 " use 'udp=on' to specify udp encapsulation\n" 2631 " use 'srcport=' to specify source udp port\n" 2632 " use 'dstport=' to specify destination udp port\n" 2633 " use 'ipv6=on' to force v6\n" 2634 " L2TPv3 uses cookies to prevent misconfiguration as\n" 2635 " well as a weak security measure\n" 2636 " use 'rxcookie=0x012345678' to specify a rxcookie\n" 2637 " use 'txcookie=0x012345678' to specify a txcookie\n" 2638 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n" 2639 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n" 2640 " use 'pincounter=on' to work around broken counter handling in peer\n" 2641 " use 'offset=X' to add an extra offset between header and data\n" 2642#endif 2643 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n" 2644 " configure a network backend to connect to another network\n" 2645 " using a socket connection\n" 2646 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n" 2647 " configure a network backend to connect to a multicast maddr and port\n" 2648 " use 'localaddr=addr' to specify the host address to send packets from\n" 2649 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n" 2650 " configure a network backend to connect to another network\n" 2651 " using an UDP tunnel\n" 2652#ifdef CONFIG_VDE 2653 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n" 2654 " configure a network backend to connect to port 'n' of a vde switch\n" 2655 " running on host and listening for incoming connections on 'socketpath'.\n" 2656 " Use group 'groupname' and mode 'octalmode' to change default\n" 2657 " ownership and permissions for communication port.\n" 2658#endif 2659#ifdef CONFIG_NETMAP 2660 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n" 2661 " attach to the existing netmap-enabled network interface 'name', or to a\n" 2662 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n" 2663 " netmap device, defaults to '/dev/netmap')\n" 2664#endif 2665#ifdef CONFIG_POSIX 2666 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n" 2667 " configure a vhost-user network, backed by a chardev 'dev'\n" 2668#endif 2669#ifdef __linux__ 2670 "-netdev vhost-vdpa,id=str,vhostdev=/path/to/dev\n" 2671 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n" 2672#endif 2673 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n" 2674 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL) 2675DEF("nic", HAS_ARG, QEMU_OPTION_nic, 2676 "-nic [tap|bridge|" 2677#ifdef CONFIG_SLIRP 2678 "user|" 2679#endif 2680#ifdef __linux__ 2681 "l2tpv3|" 2682#endif 2683#ifdef CONFIG_VDE 2684 "vde|" 2685#endif 2686#ifdef CONFIG_NETMAP 2687 "netmap|" 2688#endif 2689#ifdef CONFIG_POSIX 2690 "vhost-user|" 2691#endif 2692 "socket][,option][,...][mac=macaddr]\n" 2693 " initialize an on-board / default host NIC (using MAC address\n" 2694 " macaddr) and connect it to the given host network backend\n" 2695 "-nic none use it alone to have zero network devices (the default is to\n" 2696 " provided a 'user' network connection)\n", 2697 QEMU_ARCH_ALL) 2698DEF("net", HAS_ARG, QEMU_OPTION_net, 2699 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n" 2700 " configure or create an on-board (or machine default) NIC and\n" 2701 " connect it to hub 0 (please use -nic unless you need a hub)\n" 2702 "-net [" 2703#ifdef CONFIG_SLIRP 2704 "user|" 2705#endif 2706 "tap|" 2707 "bridge|" 2708#ifdef CONFIG_VDE 2709 "vde|" 2710#endif 2711#ifdef CONFIG_NETMAP 2712 "netmap|" 2713#endif 2714 "socket][,option][,option][,...]\n" 2715 " old way to initialize a host network interface\n" 2716 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL) 2717SRST 2718``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]`` 2719 This option is a shortcut for configuring both the on-board 2720 (default) guest NIC hardware and the host network backend in one go. 2721 The host backend options are the same as with the corresponding 2722 ``-netdev`` options below. The guest NIC model can be set with 2723 ``model=modelname``. Use ``model=help`` to list the available device 2724 types. The hardware MAC address can be set with ``mac=macaddr``. 2725 2726 The following two example do exactly the same, to show how ``-nic`` 2727 can be used to shorten the command line length: 2728 2729 .. parsed-literal:: 2730 2731 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32 2732 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32 2733 2734``-nic none`` 2735 Indicate that no network devices should be configured. It is used to 2736 override the default configuration (default NIC with "user" host 2737 network backend) which is activated if no other networking options 2738 are provided. 2739 2740``-netdev user,id=id[,option][,option][,...]`` 2741 Configure user mode host network backend which requires no 2742 administrator privilege to run. Valid options are: 2743 2744 ``id=id`` 2745 Assign symbolic name for use in monitor commands. 2746 2747 ``ipv4=on|off and ipv6=on|off`` 2748 Specify that either IPv4 or IPv6 must be enabled. If neither is 2749 specified both protocols are enabled. 2750 2751 ``net=addr[/mask]`` 2752 Set IP network address the guest will see. Optionally specify 2753 the netmask, either in the form a.b.c.d or as number of valid 2754 top-most bits. Default is 10.0.2.0/24. 2755 2756 ``host=addr`` 2757 Specify the guest-visible address of the host. Default is the 2758 2nd IP in the guest network, i.e. x.x.x.2. 2759 2760 ``ipv6-net=addr[/int]`` 2761 Set IPv6 network address the guest will see (default is 2762 fec0::/64). The network prefix is given in the usual hexadecimal 2763 IPv6 address notation. The prefix size is optional, and is given 2764 as the number of valid top-most bits (default is 64). 2765 2766 ``ipv6-host=addr`` 2767 Specify the guest-visible IPv6 address of the host. Default is 2768 the 2nd IPv6 in the guest network, i.e. xxxx::2. 2769 2770 ``restrict=on|off`` 2771 If this option is enabled, the guest will be isolated, i.e. it 2772 will not be able to contact the host and no guest IP packets 2773 will be routed over the host to the outside. This option does 2774 not affect any explicitly set forwarding rules. 2775 2776 ``hostname=name`` 2777 Specifies the client hostname reported by the built-in DHCP 2778 server. 2779 2780 ``dhcpstart=addr`` 2781 Specify the first of the 16 IPs the built-in DHCP server can 2782 assign. Default is the 15th to 31st IP in the guest network, 2783 i.e. x.x.x.15 to x.x.x.31. 2784 2785 ``dns=addr`` 2786 Specify the guest-visible address of the virtual nameserver. The 2787 address must be different from the host address. Default is the 2788 3rd IP in the guest network, i.e. x.x.x.3. 2789 2790 ``ipv6-dns=addr`` 2791 Specify the guest-visible address of the IPv6 virtual 2792 nameserver. The address must be different from the host address. 2793 Default is the 3rd IP in the guest network, i.e. xxxx::3. 2794 2795 ``dnssearch=domain`` 2796 Provides an entry for the domain-search list sent by the 2797 built-in DHCP server. More than one domain suffix can be 2798 transmitted by specifying this option multiple times. If 2799 supported, this will cause the guest to automatically try to 2800 append the given domain suffix(es) in case a domain name can not 2801 be resolved. 2802 2803 Example: 2804 2805 .. parsed-literal:: 2806 2807 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org 2808 2809 ``domainname=domain`` 2810 Specifies the client domain name reported by the built-in DHCP 2811 server. 2812 2813 ``tftp=dir`` 2814 When using the user mode network stack, activate a built-in TFTP 2815 server. The files in dir will be exposed as the root of a TFTP 2816 server. The TFTP client on the guest must be configured in 2817 binary mode (use the command ``bin`` of the Unix TFTP client). 2818 2819 ``tftp-server-name=name`` 2820 In BOOTP reply, broadcast name as the "TFTP server name" 2821 (RFC2132 option 66). This can be used to advise the guest to 2822 load boot files or configurations from a different server than 2823 the host address. 2824 2825 ``bootfile=file`` 2826 When using the user mode network stack, broadcast file as the 2827 BOOTP filename. In conjunction with ``tftp``, this can be used 2828 to network boot a guest from a local directory. 2829 2830 Example (using pxelinux): 2831 2832 .. parsed-literal:: 2833 2834 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\ 2835 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0 2836 2837 ``smb=dir[,smbserver=addr]`` 2838 When using the user mode network stack, activate a built-in SMB 2839 server so that Windows OSes can access to the host files in 2840 ``dir`` transparently. The IP address of the SMB server can be 2841 set to addr. By default the 4th IP in the guest network is used, 2842 i.e. x.x.x.4. 2843 2844 In the guest Windows OS, the line: 2845 2846 :: 2847 2848 10.0.2.4 smbserver 2849 2850 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows 2851 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows 2852 NT/2000). 2853 2854 Then ``dir`` can be accessed in ``\\smbserver\qemu``. 2855 2856 Note that a SAMBA server must be installed on the host OS. 2857 2858 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport`` 2859 Redirect incoming TCP or UDP connections to the host port 2860 hostport to the guest IP address guestaddr on guest port 2861 guestport. If guestaddr is not specified, its value is x.x.x.15 2862 (default first address given by the built-in DHCP server). By 2863 specifying hostaddr, the rule can be bound to a specific host 2864 interface. If no connection type is set, TCP is used. This 2865 option can be given multiple times. 2866 2867 For example, to redirect host X11 connection from screen 1 to 2868 guest screen 0, use the following: 2869 2870 .. parsed-literal:: 2871 2872 # on the host 2873 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000 2874 # this host xterm should open in the guest X11 server 2875 xterm -display :1 2876 2877 To redirect telnet connections from host port 5555 to telnet 2878 port on the guest, use the following: 2879 2880 .. parsed-literal:: 2881 2882 # on the host 2883 |qemu_system| -nic user,hostfwd=tcp::5555-:23 2884 telnet localhost 5555 2885 2886 Then when you use on the host ``telnet localhost 5555``, you 2887 connect to the guest telnet server. 2888 2889 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command`` 2890 Forward guest TCP connections to the IP address server on port 2891 port to the character device dev or to a program executed by 2892 cmd:command which gets spawned for each connection. This option 2893 can be given multiple times. 2894 2895 You can either use a chardev directly and have that one used 2896 throughout QEMU's lifetime, like in the following example: 2897 2898 .. parsed-literal:: 2899 2900 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever 2901 # the guest accesses it 2902 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321 2903 2904 Or you can execute a command on every TCP connection established 2905 by the guest, so that QEMU behaves similar to an inetd process 2906 for that virtual server: 2907 2908 .. parsed-literal:: 2909 2910 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234 2911 # and connect the TCP stream to its stdin/stdout 2912 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321' 2913 2914``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]`` 2915 Configure a host TAP network backend with ID id. 2916 2917 Use the network script file to configure it and the network script 2918 dfile to deconfigure it. If name is not provided, the OS 2919 automatically provides one. The default network configure script is 2920 ``/etc/qemu-ifup`` and the default network deconfigure script is 2921 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to 2922 disable script execution. 2923 2924 If running QEMU as an unprivileged user, use the network helper 2925 to configure the TAP interface and attach it to the bridge. 2926 The default network helper executable is 2927 ``/path/to/qemu-bridge-helper`` and the default bridge device is 2928 ``br0``. 2929 2930 ``fd``\ =h can be used to specify the handle of an already opened 2931 host TAP interface. 2932 2933 Examples: 2934 2935 .. parsed-literal:: 2936 2937 #launch a QEMU instance with the default network script 2938 |qemu_system| linux.img -nic tap 2939 2940 .. parsed-literal:: 2941 2942 #launch a QEMU instance with two NICs, each one connected 2943 #to a TAP device 2944 |qemu_system| linux.img \\ 2945 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\ 2946 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1 2947 2948 .. parsed-literal:: 2949 2950 #launch a QEMU instance with the default network helper to 2951 #connect a TAP device to bridge br0 2952 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 2953 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper" 2954 2955``-netdev bridge,id=id[,br=bridge][,helper=helper]`` 2956 Connect a host TAP network interface to a host bridge device. 2957 2958 Use the network helper helper to configure the TAP interface and 2959 attach it to the bridge. The default network helper executable is 2960 ``/path/to/qemu-bridge-helper`` and the default bridge device is 2961 ``br0``. 2962 2963 Examples: 2964 2965 .. parsed-literal:: 2966 2967 #launch a QEMU instance with the default network helper to 2968 #connect a TAP device to bridge br0 2969 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1 2970 2971 .. parsed-literal:: 2972 2973 #launch a QEMU instance with the default network helper to 2974 #connect a TAP device to bridge qemubr0 2975 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1 2976 2977``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]`` 2978 This host network backend can be used to connect the guest's network 2979 to another QEMU virtual machine using a TCP socket connection. If 2980 ``listen`` is specified, QEMU waits for incoming connections on port 2981 (host is optional). ``connect`` is used to connect to another QEMU 2982 instance using the ``listen`` option. ``fd``\ =h specifies an 2983 already opened TCP socket. 2984 2985 Example: 2986 2987 .. parsed-literal:: 2988 2989 # launch a first QEMU instance 2990 |qemu_system| linux.img \\ 2991 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 2992 -netdev socket,id=n1,listen=:1234 2993 # connect the network of this instance to the network of the first instance 2994 |qemu_system| linux.img \\ 2995 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 2996 -netdev socket,id=n2,connect=127.0.0.1:1234 2997 2998``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]`` 2999 Configure a socket host network backend to share the guest's network 3000 traffic with another QEMU virtual machines using a UDP multicast 3001 socket, effectively making a bus for every QEMU with same multicast 3002 address maddr and port. NOTES: 3003 3004 1. Several QEMU can be running on different hosts and share same bus 3005 (assuming correct multicast setup for these hosts). 3006 3007 2. mcast support is compatible with User Mode Linux (argument 3008 ``ethN=mcast``), see http://user-mode-linux.sf.net. 3009 3010 3. Use ``fd=h`` to specify an already opened UDP multicast socket. 3011 3012 Example: 3013 3014 .. parsed-literal:: 3015 3016 # launch one QEMU instance 3017 |qemu_system| linux.img \\ 3018 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3019 -netdev socket,id=n1,mcast=230.0.0.1:1234 3020 # launch another QEMU instance on same "bus" 3021 |qemu_system| linux.img \\ 3022 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 3023 -netdev socket,id=n2,mcast=230.0.0.1:1234 3024 # launch yet another QEMU instance on same "bus" 3025 |qemu_system| linux.img \\ 3026 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\ 3027 -netdev socket,id=n3,mcast=230.0.0.1:1234 3028 3029 Example (User Mode Linux compat.): 3030 3031 .. parsed-literal:: 3032 3033 # launch QEMU instance (note mcast address selected is UML's default) 3034 |qemu_system| linux.img \\ 3035 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3036 -netdev socket,id=n1,mcast=239.192.168.1:1102 3037 # launch UML 3038 /path/to/linux ubd0=/path/to/root_fs eth0=mcast 3039 3040 Example (send packets from host's 1.2.3.4): 3041 3042 .. parsed-literal:: 3043 3044 |qemu_system| linux.img \\ 3045 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 3046 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4 3047 3048``-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]`` 3049 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931) 3050 is a popular protocol to transport Ethernet (and other Layer 2) data 3051 frames between two systems. It is present in routers, firewalls and 3052 the Linux kernel (from version 3.3 onwards). 3053 3054 This transport allows a VM to communicate to another VM, router or 3055 firewall directly. 3056 3057 ``src=srcaddr`` 3058 source address (mandatory) 3059 3060 ``dst=dstaddr`` 3061 destination address (mandatory) 3062 3063 ``udp`` 3064 select udp encapsulation (default is ip). 3065 3066 ``srcport=srcport`` 3067 source udp port. 3068 3069 ``dstport=dstport`` 3070 destination udp port. 3071 3072 ``ipv6`` 3073 force v6, otherwise defaults to v4. 3074 3075 ``rxcookie=rxcookie``; \ ``txcookie=txcookie`` 3076 Cookies are a weak form of security in the l2tpv3 specification. 3077 Their function is mostly to prevent misconfiguration. By default 3078 they are 32 bit. 3079 3080 ``cookie64`` 3081 Set cookie size to 64 bit instead of the default 32 3082 3083 ``counter=off`` 3084 Force a 'cut-down' L2TPv3 with no counter as in 3085 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00 3086 3087 ``pincounter=on`` 3088 Work around broken counter handling in peer. This may also help 3089 on networks which have packet reorder. 3090 3091 ``offset=offset`` 3092 Add an extra offset between header and data 3093 3094 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to 3095 the bridge br-lan on the remote Linux host 1.2.3.4: 3096 3097 .. parsed-literal:: 3098 3099 # Setup tunnel on linux host using raw ip as encapsulation 3100 # on 1.2.3.4 3101 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\ 3102 encap udp udp_sport 16384 udp_dport 16384 3103 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\ 3104 0xFFFFFFFF peer_session_id 0xFFFFFFFF 3105 ifconfig vmtunnel0 mtu 1500 3106 ifconfig vmtunnel0 up 3107 brctl addif br-lan vmtunnel0 3108 3109 3110 # on 4.3.2.1 3111 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter 3112 3113 |qemu_system| linux.img -device e1000,netdev=n1 \\ 3114 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter 3115 3116``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]`` 3117 Configure VDE backend to connect to PORT n of a vde switch running 3118 on host and listening for incoming connections on socketpath. Use 3119 GROUP groupname and MODE octalmode to change default ownership and 3120 permissions for communication port. This option is only available if 3121 QEMU has been compiled with vde support enabled. 3122 3123 Example: 3124 3125 .. parsed-literal:: 3126 3127 # launch vde switch 3128 vde_switch -F -sock /tmp/myswitch 3129 # launch QEMU instance 3130 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch 3131 3132``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]`` 3133 Establish a vhost-user netdev, backed by a chardev id. The chardev 3134 should be a unix domain socket backed one. The vhost-user uses a 3135 specifically defined protocol to pass vhost ioctl replacement 3136 messages to an application on the other end of the socket. On 3137 non-MSIX guests, the feature can be forced with vhostforce. Use 3138 'queues=n' to specify the number of queues to be created for 3139 multiqueue vhost-user. 3140 3141 Example: 3142 3143 :: 3144 3145 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \ 3146 -numa node,memdev=mem \ 3147 -chardev socket,id=chr0,path=/path/to/socket \ 3148 -netdev type=vhost-user,id=net0,chardev=chr0 \ 3149 -device virtio-net-pci,netdev=net0 3150 3151``-netdev vhost-vdpa,vhostdev=/path/to/dev`` 3152 Establish a vhost-vdpa netdev. 3153 3154 vDPA device is a device that uses a datapath which complies with 3155 the virtio specifications with a vendor specific control path. 3156 vDPA devices can be both physically located on the hardware or 3157 emulated by software. 3158 3159``-netdev hubport,id=id,hubid=hubid[,netdev=nd]`` 3160 Create a hub port on the emulated hub with ID hubid. 3161 3162 The hubport netdev lets you connect a NIC to a QEMU emulated hub 3163 instead of a single netdev. Alternatively, you can also connect the 3164 hubport to another netdev with ID nd by using the ``netdev=nd`` 3165 option. 3166 3167``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]`` 3168 Legacy option to configure or create an on-board (or machine 3169 default) Network Interface Card(NIC) and connect it either to the 3170 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd. 3171 If model is omitted, then the default NIC model associated with the 3172 machine type is used. Note that the default NIC model may change in 3173 future QEMU releases, so it is highly recommended to always specify 3174 a model. Optionally, the MAC address can be changed to mac, the 3175 device address set to addr (PCI cards only), and a name can be 3176 assigned for use in monitor commands. Optionally, for PCI cards, you 3177 can specify the number v of MSI-X vectors that the card should have; 3178 this option currently only affects virtio cards; set v = 0 to 3179 disable MSI-X. If no ``-net`` option is specified, a single NIC is 3180 created. QEMU can emulate several different models of network card. 3181 Use ``-net nic,model=help`` for a list of available devices for your 3182 target. 3183 3184``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]`` 3185 Configure a host network backend (with the options corresponding to 3186 the same ``-netdev`` option) and connect it to the emulated hub 0 3187 (the default hub). Use name to specify the name of the hub port. 3188ERST 3189 3190DEFHEADING() 3191 3192DEFHEADING(Character device options:) 3193 3194DEF("chardev", HAS_ARG, QEMU_OPTION_chardev, 3195 "-chardev help\n" 3196 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3197 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off][,reconnect=seconds]\n" 3198 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n" 3199 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n" 3200 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n" 3201 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n" 3202 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n" 3203 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n" 3204 " [,logfile=PATH][,logappend=on|off]\n" 3205 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3206 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n" 3207 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3208 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n" 3209 "-chardev file,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3210 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3211#ifdef _WIN32 3212 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3213 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3214#else 3215 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3216 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n" 3217#endif 3218#ifdef CONFIG_BRLAPI 3219 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3220#endif 3221#if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \ 3222 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) 3223 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3224 "-chardev tty,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3225#endif 3226#if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__) 3227 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3228 "-chardev parport,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3229#endif 3230#if defined(CONFIG_SPICE) 3231 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3232 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3233#endif 3234 , QEMU_ARCH_ALL 3235) 3236 3237SRST 3238The general form of a character device option is: 3239 3240``-chardev backend,id=id[,mux=on|off][,options]`` 3241 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``, 3242 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``, 3243 ``pty``, ``stdio``, ``braille``, ``tty``, ``parallel``, ``parport``, 3244 ``spicevmc``, ``spiceport``. The specific backend will determine the 3245 applicable options. 3246 3247 Use ``-chardev help`` to print all available chardev backend types. 3248 3249 All devices must have an id, which can be any string up to 127 3250 characters long. It is used to uniquely identify this device in 3251 other command line directives. 3252 3253 A character device may be used in multiplexing mode by multiple 3254 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is 3255 a "1:N" device, and here the "1" end is your specified chardev 3256 backend, and the "N" end is the various parts of QEMU that can talk 3257 to a chardev. If you create a chardev with ``id=myid`` and 3258 ``mux=on``, QEMU will create a multiplexer with your specified ID, 3259 and you can then configure multiple front ends to use that chardev 3260 ID for their input/output. Up to four different front ends can be 3261 connected to a single multiplexed chardev. (Without multiplexing 3262 enabled, a chardev can only be used by a single front end.) For 3263 instance you could use this to allow a single stdio chardev to be 3264 used by two serial ports and the QEMU monitor: 3265 3266 :: 3267 3268 -chardev stdio,mux=on,id=char0 \ 3269 -mon chardev=char0,mode=readline \ 3270 -serial chardev:char0 \ 3271 -serial chardev:char0 3272 3273 You can have more than one multiplexer in a system configuration; 3274 for instance you could have a TCP port multiplexed between UART 0 3275 and UART 1, and stdio multiplexed between the QEMU monitor and a 3276 parallel port: 3277 3278 :: 3279 3280 -chardev stdio,mux=on,id=char0 \ 3281 -mon chardev=char0,mode=readline \ 3282 -parallel chardev:char0 \ 3283 -chardev tcp,...,mux=on,id=char1 \ 3284 -serial chardev:char1 \ 3285 -serial chardev:char1 3286 3287 When you're using a multiplexed character device, some escape 3288 sequences are interpreted in the input. See the chapter about 3289 :ref:`keys in the character backend multiplexer` in the 3290 System Emulation Users Guide for more details. 3291 3292 Note that some other command line options may implicitly create 3293 multiplexed character backends; for instance ``-serial mon:stdio`` 3294 creates a multiplexed stdio backend connected to the serial port and 3295 the QEMU monitor, and ``-nographic`` also multiplexes the console 3296 and the monitor to stdio. 3297 3298 There is currently no support for multiplexing in the other 3299 direction (where a single QEMU front end takes input and output from 3300 multiple chardevs). 3301 3302 Every backend supports the ``logfile`` option, which supplies the 3303 path to a file to record all data transmitted via the backend. The 3304 ``logappend`` option controls whether the log file will be truncated 3305 or appended to when opened. 3306 3307The available backends are: 3308 3309``-chardev null,id=id`` 3310 A void device. This device will not emit any data, and will drop any 3311 data it receives. The null backend does not take any options. 3312 3313``-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]`` 3314 Create a two-way stream socket, which can be either a TCP or a unix 3315 socket. A unix socket will be created if ``path`` is specified. 3316 Behaviour is undefined if TCP options are specified for a unix 3317 socket. 3318 3319 ``server=on|off`` specifies that the socket shall be a listening socket. 3320 3321 ``wait=on|off`` specifies that QEMU should not block waiting for a client 3322 to connect to a listening socket. 3323 3324 ``telnet=on|off`` specifies that traffic on the socket should interpret 3325 telnet escape sequences. 3326 3327 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for 3328 communication. 3329 3330 ``reconnect`` sets the timeout for reconnecting on non-server 3331 sockets when the remote end goes away. qemu will delay this many 3332 seconds and then attempt to reconnect. Zero disables reconnecting, 3333 and is the default. 3334 3335 ``tls-creds`` requests enablement of the TLS protocol for 3336 encryption, and specifies the id of the TLS credentials to use for 3337 the handshake. The credentials must be previously created with the 3338 ``-object tls-creds`` argument. 3339 3340 ``tls-auth`` provides the ID of the QAuthZ authorization object 3341 against which the client's x509 distinguished name will be 3342 validated. This object is only resolved at time of use, so can be 3343 deleted and recreated on the fly while the chardev server is active. 3344 If missing, it will default to denying access. 3345 3346 TCP and unix socket options are given below: 3347 3348 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]`` 3349 ``host`` for a listening socket specifies the local address to 3350 be bound. For a connecting socket species the remote host to 3351 connect to. ``host`` is optional for listening sockets. If not 3352 specified it defaults to ``0.0.0.0``. 3353 3354 ``port`` for a listening socket specifies the local port to be 3355 bound. For a connecting socket specifies the port on the remote 3356 host to connect to. ``port`` can be given as either a port 3357 number or a service name. ``port`` is required. 3358 3359 ``to`` is only relevant to listening sockets. If it is 3360 specified, and ``port`` cannot be bound, QEMU will attempt to 3361 bind to subsequent ports up to and including ``to`` until it 3362 succeeds. ``to`` must be specified as a port number. 3363 3364 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 3365 or IPv6 must be used. If neither is specified the socket may 3366 use either protocol. 3367 3368 ``nodelay=on|off`` disables the Nagle algorithm. 3369 3370 ``unix options: path=path[,abstract=on|off][,tight=on|off]`` 3371 ``path`` specifies the local path of the unix socket. ``path`` 3372 is required. 3373 ``abstract=on|off`` specifies the use of the abstract socket namespace, 3374 rather than the filesystem. Optional, defaults to false. 3375 ``tight=on|off`` sets the socket length of abstract sockets to their minimum, 3376 rather than the full sun_path length. Optional, defaults to true. 3377 3378``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]`` 3379 Sends all traffic from the guest to a remote host over UDP. 3380 3381 ``host`` specifies the remote host to connect to. If not specified 3382 it defaults to ``localhost``. 3383 3384 ``port`` specifies the port on the remote host to connect to. 3385 ``port`` is required. 3386 3387 ``localaddr`` specifies the local address to bind to. If not 3388 specified it defaults to ``0.0.0.0``. 3389 3390 ``localport`` specifies the local port to bind to. If not specified 3391 any available local port will be used. 3392 3393 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used. 3394 If neither is specified the device may use either protocol. 3395 3396``-chardev msmouse,id=id`` 3397 Forward QEMU's emulated msmouse events to the guest. ``msmouse`` 3398 does not take any options. 3399 3400``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]`` 3401 Connect to a QEMU text console. ``vc`` may optionally be given a 3402 specific size. 3403 3404 ``width`` and ``height`` specify the width and height respectively 3405 of the console, in pixels. 3406 3407 ``cols`` and ``rows`` specify that the console be sized to fit a 3408 text console with the given dimensions. 3409 3410``-chardev ringbuf,id=id[,size=size]`` 3411 Create a ring buffer with fixed size ``size``. size must be a power 3412 of two and defaults to ``64K``. 3413 3414``-chardev file,id=id,path=path`` 3415 Log all traffic received from the guest to a file. 3416 3417 ``path`` specifies the path of the file to be opened. This file will 3418 be created if it does not already exist, and overwritten if it does. 3419 ``path`` is required. 3420 3421``-chardev pipe,id=id,path=path`` 3422 Create a two-way connection to the guest. The behaviour differs 3423 slightly between Windows hosts and other hosts: 3424 3425 On Windows, a single duplex pipe will be created at 3426 ``\\.pipe\path``. 3427 3428 On other hosts, 2 pipes will be created called ``path.in`` and 3429 ``path.out``. Data written to ``path.in`` will be received by the 3430 guest. Data written by the guest can be read from ``path.out``. QEMU 3431 will not create these fifos, and requires them to be present. 3432 3433 ``path`` forms part of the pipe path as described above. ``path`` is 3434 required. 3435 3436``-chardev console,id=id`` 3437 Send traffic from the guest to QEMU's standard output. ``console`` 3438 does not take any options. 3439 3440 ``console`` is only available on Windows hosts. 3441 3442``-chardev serial,id=id,path=path`` 3443 Send traffic from the guest to a serial device on the host. 3444 3445 On Unix hosts serial will actually accept any tty device, not only 3446 serial lines. 3447 3448 ``path`` specifies the name of the serial device to open. 3449 3450``-chardev pty,id=id`` 3451 Create a new pseudo-terminal on the host and connect to it. ``pty`` 3452 does not take any options. 3453 3454 ``pty`` is not available on Windows hosts. 3455 3456``-chardev stdio,id=id[,signal=on|off]`` 3457 Connect to standard input and standard output of the QEMU process. 3458 3459 ``signal`` controls if signals are enabled on the terminal, that 3460 includes exiting QEMU with the key sequence Control-c. This option 3461 is enabled by default, use ``signal=off`` to disable it. 3462 3463``-chardev braille,id=id`` 3464 Connect to a local BrlAPI server. ``braille`` does not take any 3465 options. 3466 3467``-chardev tty,id=id,path=path`` 3468 ``tty`` is only available on Linux, Sun, FreeBSD, NetBSD, OpenBSD 3469 and DragonFlyBSD hosts. It is an alias for ``serial``. 3470 3471 ``path`` specifies the path to the tty. ``path`` is required. 3472 3473``-chardev parallel,id=id,path=path`` 3474 \ 3475``-chardev parport,id=id,path=path`` 3476 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD 3477 hosts. 3478 3479 Connect to a local parallel port. 3480 3481 ``path`` specifies the path to the parallel port device. ``path`` is 3482 required. 3483 3484``-chardev spicevmc,id=id,debug=debug,name=name`` 3485 ``spicevmc`` is only available when spice support is built in. 3486 3487 ``debug`` debug level for spicevmc 3488 3489 ``name`` name of spice channel to connect to 3490 3491 Connect to a spice virtual machine channel, such as vdiport. 3492 3493``-chardev spiceport,id=id,debug=debug,name=name`` 3494 ``spiceport`` is only available when spice support is built in. 3495 3496 ``debug`` debug level for spicevmc 3497 3498 ``name`` name of spice port to connect to 3499 3500 Connect to a spice port, allowing a Spice client to handle the 3501 traffic identified by a name (preferably a fqdn). 3502ERST 3503 3504DEFHEADING() 3505 3506#ifdef CONFIG_TPM 3507DEFHEADING(TPM device options:) 3508 3509DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \ 3510 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n" 3511 " use path to provide path to a character device; default is /dev/tpm0\n" 3512 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n" 3513 " not provided it will be searched for in /sys/class/misc/tpm?/device\n" 3514 "-tpmdev emulator,id=id,chardev=dev\n" 3515 " configure the TPM device using chardev backend\n", 3516 QEMU_ARCH_ALL) 3517SRST 3518The general form of a TPM device option is: 3519 3520``-tpmdev backend,id=id[,options]`` 3521 The specific backend type will determine the applicable options. The 3522 ``-tpmdev`` option creates the TPM backend and requires a 3523 ``-device`` option that specifies the TPM frontend interface model. 3524 3525 Use ``-tpmdev help`` to print all available TPM backend types. 3526 3527The available backends are: 3528 3529``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path`` 3530 (Linux-host only) Enable access to the host's TPM using the 3531 passthrough driver. 3532 3533 ``path`` specifies the path to the host's TPM device, i.e., on a 3534 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by 3535 default ``/dev/tpm0`` is used. 3536 3537 ``cancel-path`` specifies the path to the host TPM device's sysfs 3538 entry allowing for cancellation of an ongoing TPM command. 3539 ``cancel-path`` is optional and by default QEMU will search for the 3540 sysfs entry to use. 3541 3542 Some notes about using the host's TPM with the passthrough driver: 3543 3544 The TPM device accessed by the passthrough driver must not be used 3545 by any other application on the host. 3546 3547 Since the host's firmware (BIOS/UEFI) has already initialized the 3548 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize 3549 the TPM again and may therefore not show a TPM-specific menu that 3550 would otherwise allow the user to configure the TPM, e.g., allow the 3551 user to enable/disable or activate/deactivate the TPM. Further, if 3552 TPM ownership is released from within a VM then the host's TPM will 3553 get disabled and deactivated. To enable and activate the TPM again 3554 afterwards, the host has to be rebooted and the user is required to 3555 enter the firmware's menu to enable and activate the TPM. If the TPM 3556 is left disabled and/or deactivated most TPM commands will fail. 3557 3558 To create a passthrough TPM use the following two options: 3559 3560 :: 3561 3562 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0 3563 3564 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by 3565 ``tpmdev=tpm0`` in the device option. 3566 3567``-tpmdev emulator,id=id,chardev=dev`` 3568 (Linux-host only) Enable access to a TPM emulator using Unix domain 3569 socket based chardev backend. 3570 3571 ``chardev`` specifies the unique ID of a character device backend 3572 that provides connection to the software TPM server. 3573 3574 To create a TPM emulator backend device with chardev socket backend: 3575 3576 :: 3577 3578 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0 3579ERST 3580 3581DEFHEADING() 3582 3583#endif 3584 3585DEFHEADING(Linux/Multiboot boot specific:) 3586SRST 3587When using these options, you can use a given Linux or Multiboot kernel 3588without installing it in the disk image. It can be useful for easier 3589testing of various kernels. 3590 3591 3592ERST 3593 3594DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \ 3595 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL) 3596SRST 3597``-kernel bzImage`` 3598 Use bzImage as kernel image. The kernel can be either a Linux kernel 3599 or in multiboot format. 3600ERST 3601 3602DEF("append", HAS_ARG, QEMU_OPTION_append, \ 3603 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL) 3604SRST 3605``-append cmdline`` 3606 Use cmdline as kernel command line 3607ERST 3608 3609DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \ 3610 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL) 3611SRST 3612``-initrd file`` 3613 Use file as initial ram disk. 3614 3615``-initrd "file1 arg=foo,file2"`` 3616 This syntax is only available with multiboot. 3617 3618 Use file1 and file2 as modules and pass arg=foo as parameter to the 3619 first module. 3620ERST 3621 3622DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \ 3623 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL) 3624SRST 3625``-dtb file`` 3626 Use file as a device tree binary (dtb) image and pass it to the 3627 kernel on boot. 3628ERST 3629 3630DEFHEADING() 3631 3632DEFHEADING(Debug/Expert options:) 3633 3634DEF("compat", HAS_ARG, QEMU_OPTION_compat, 3635 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n" 3636 " Policy for handling deprecated management interfaces\n", 3637 QEMU_ARCH_ALL) 3638SRST 3639``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]`` 3640 Set policy for handling deprecated management interfaces (experimental): 3641 3642 ``deprecated-input=accept`` (default) 3643 Accept deprecated commands and arguments 3644 ``deprecated-input=reject`` 3645 Reject deprecated commands and arguments 3646 ``deprecated-input=crash`` 3647 Crash on deprecated commands and arguments 3648 ``deprecated-output=accept`` (default) 3649 Emit deprecated command results and events 3650 ``deprecated-output=hide`` 3651 Suppress deprecated command results and events 3652 3653 Limitation: covers only syntactic aspects of QMP. 3654ERST 3655 3656DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg, 3657 "-fw_cfg [name=]<name>,file=<file>\n" 3658 " add named fw_cfg entry with contents from file\n" 3659 "-fw_cfg [name=]<name>,string=<str>\n" 3660 " add named fw_cfg entry with contents from string\n", 3661 QEMU_ARCH_ALL) 3662SRST 3663``-fw_cfg [name=]name,file=file`` 3664 Add named fw\_cfg entry with contents from file file. 3665 3666``-fw_cfg [name=]name,string=str`` 3667 Add named fw\_cfg entry with contents from string str. 3668 3669 The terminating NUL character of the contents of str will not be 3670 included as part of the fw\_cfg item data. To insert contents with 3671 embedded NUL characters, you have to use the file parameter. 3672 3673 The fw\_cfg entries are passed by QEMU through to the guest. 3674 3675 Example: 3676 3677 :: 3678 3679 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin 3680 3681 creates an fw\_cfg entry named opt/com.mycompany/blob with contents 3682 from ./my\_blob.bin. 3683ERST 3684 3685DEF("serial", HAS_ARG, QEMU_OPTION_serial, \ 3686 "-serial dev redirect the serial port to char device 'dev'\n", 3687 QEMU_ARCH_ALL) 3688SRST 3689``-serial dev`` 3690 Redirect the virtual serial port to host character device dev. The 3691 default device is ``vc`` in graphical mode and ``stdio`` in non 3692 graphical mode. 3693 3694 This option can be used several times to simulate up to 4 serial 3695 ports. 3696 3697 Use ``-serial none`` to disable all serial ports. 3698 3699 Available character devices are: 3700 3701 ``vc[:WxH]`` 3702 Virtual console. Optionally, a width and height can be given in 3703 pixel with 3704 3705 :: 3706 3707 vc:800x600 3708 3709 It is also possible to specify width or height in characters: 3710 3711 :: 3712 3713 vc:80Cx24C 3714 3715 ``pty`` 3716 [Linux only] Pseudo TTY (a new PTY is automatically allocated) 3717 3718 ``none`` 3719 No device is allocated. 3720 3721 ``null`` 3722 void device 3723 3724 ``chardev:id`` 3725 Use a named character device defined with the ``-chardev`` 3726 option. 3727 3728 ``/dev/XXX`` 3729 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial 3730 port parameters are set according to the emulated ones. 3731 3732 ``/dev/parportN`` 3733 [Linux only, parallel port only] Use host parallel port N. 3734 Currently SPP and EPP parallel port features can be used. 3735 3736 ``file:filename`` 3737 Write output to filename. No character can be read. 3738 3739 ``stdio`` 3740 [Unix only] standard input/output 3741 3742 ``pipe:filename`` 3743 name pipe filename 3744 3745 ``COMn`` 3746 [Windows only] Use host serial port n 3747 3748 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]`` 3749 This implements UDP Net Console. When remote\_host or src\_ip 3750 are not specified they default to ``0.0.0.0``. When not using a 3751 specified src\_port a random port is automatically chosen. 3752 3753 If you just want a simple readonly console you can use 3754 ``netcat`` or ``nc``, by starting QEMU with: 3755 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time 3756 QEMU writes something to that port it will appear in the 3757 netconsole session. 3758 3759 If you plan to send characters back via netconsole or you want 3760 to stop and start QEMU a lot of times, you should have QEMU use 3761 the same source port each time by using something like ``-serial 3762 udp::4555@:4556`` to QEMU. Another approach is to use a patched 3763 version of netcat which can listen to a TCP port and send and 3764 receive characters via udp. If you have a patched version of 3765 netcat which activates telnet remote echo and single char 3766 transfer, then you can use the following options to set up a 3767 netcat redirector to allow telnet on port 5555 to access the 3768 QEMU port. 3769 3770 ``QEMU Options:`` 3771 -serial udp::4555@:4556 3772 3773 ``netcat options:`` 3774 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T 3775 3776 ``telnet options:`` 3777 localhost 5555 3778 3779 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]`` 3780 The TCP Net Console has two modes of operation. It can send the 3781 serial I/O to a location or wait for a connection from a 3782 location. By default the TCP Net Console is sent to host at the 3783 port. If you use the ``server=on`` option QEMU will wait for a client 3784 socket application to connect to the port before continuing, 3785 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off`` 3786 option disables the Nagle buffering algorithm. The ``reconnect=on`` 3787 option only applies if ``server=no`` is set, if the connection goes 3788 down it will attempt to reconnect at the given interval. If host 3789 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a 3790 time is accepted. You can use ``telnet=on`` to connect to the 3791 corresponding character device. 3792 3793 ``Example to send tcp console to 192.168.0.2 port 4444`` 3794 -serial tcp:192.168.0.2:4444 3795 3796 ``Example to listen and wait on port 4444 for connection`` 3797 -serial tcp::4444,server=on 3798 3799 ``Example to not wait and listen on ip 192.168.0.100 port 4444`` 3800 -serial tcp:192.168.0.100:4444,server=on,wait=off 3801 3802 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]`` 3803 The telnet protocol is used instead of raw tcp sockets. The 3804 options work the same as if you had specified ``-serial tcp``. 3805 The difference is that the port acts like a telnet server or 3806 client using telnet option negotiation. This will also allow you 3807 to send the MAGIC\_SYSRQ sequence if you use a telnet that 3808 supports sending the break sequence. Typically in unix telnet 3809 you do it with Control-] and then type "send break" followed by 3810 pressing the enter key. 3811 3812 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]`` 3813 The WebSocket protocol is used instead of raw tcp socket. The 3814 port acts as a WebSocket server. Client mode is not supported. 3815 3816 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]`` 3817 A unix domain socket is used instead of a tcp socket. The option 3818 works the same as if you had specified ``-serial tcp`` except 3819 the unix domain socket path is used for connections. 3820 3821 ``mon:dev_string`` 3822 This is a special option to allow the monitor to be multiplexed 3823 onto another serial port. The monitor is accessed with key 3824 sequence of Control-a and then pressing c. dev\_string should be 3825 any one of the serial devices specified above. An example to 3826 multiplex the monitor onto a telnet server listening on port 3827 4444 would be: 3828 3829 ``-serial mon:telnet::4444,server=on,wait=off`` 3830 3831 When the monitor is multiplexed to stdio in this way, Ctrl+C 3832 will not terminate QEMU any more but will be passed to the guest 3833 instead. 3834 3835 ``braille`` 3836 Braille device. This will use BrlAPI to display the braille 3837 output on a real or fake device. 3838 3839 ``msmouse`` 3840 Three button serial mouse. Configure the guest to use Microsoft 3841 protocol. 3842ERST 3843 3844DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \ 3845 "-parallel dev redirect the parallel port to char device 'dev'\n", 3846 QEMU_ARCH_ALL) 3847SRST 3848``-parallel dev`` 3849 Redirect the virtual parallel port to host device dev (same devices 3850 as the serial port). On Linux hosts, ``/dev/parportN`` can be used 3851 to use hardware devices connected on the corresponding host parallel 3852 port. 3853 3854 This option can be used several times to simulate up to 3 parallel 3855 ports. 3856 3857 Use ``-parallel none`` to disable all parallel ports. 3858ERST 3859 3860DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \ 3861 "-monitor dev redirect the monitor to char device 'dev'\n", 3862 QEMU_ARCH_ALL) 3863SRST 3864``-monitor dev`` 3865 Redirect the monitor to host device dev (same devices as the serial 3866 port). The default device is ``vc`` in graphical mode and ``stdio`` 3867 in non graphical mode. Use ``-monitor none`` to disable the default 3868 monitor. 3869ERST 3870DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \ 3871 "-qmp dev like -monitor but opens in 'control' mode\n", 3872 QEMU_ARCH_ALL) 3873SRST 3874``-qmp dev`` 3875 Like -monitor but opens in 'control' mode. 3876ERST 3877DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \ 3878 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n", 3879 QEMU_ARCH_ALL) 3880SRST 3881``-qmp-pretty dev`` 3882 Like -qmp but uses pretty JSON formatting. 3883ERST 3884 3885DEF("mon", HAS_ARG, QEMU_OPTION_mon, \ 3886 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL) 3887SRST 3888``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]`` 3889 Setup monitor on chardev name. ``mode=control`` configures 3890 a QMP monitor (a JSON RPC-style protocol) and it is not the 3891 same as HMP, the human monitor that has a "(qemu)" prompt. 3892 ``pretty`` is only valid when ``mode=control``, 3893 turning on JSON pretty printing to ease 3894 human reading and debugging. 3895ERST 3896 3897DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \ 3898 "-debugcon dev redirect the debug console to char device 'dev'\n", 3899 QEMU_ARCH_ALL) 3900SRST 3901``-debugcon dev`` 3902 Redirect the debug console to host device dev (same devices as the 3903 serial port). The debug console is an I/O port which is typically 3904 port 0xe9; writing to that I/O port sends output to this device. The 3905 default device is ``vc`` in graphical mode and ``stdio`` in non 3906 graphical mode. 3907ERST 3908 3909DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \ 3910 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL) 3911SRST 3912``-pidfile file`` 3913 Store the QEMU process PID in file. It is useful if you launch QEMU 3914 from a script. 3915ERST 3916 3917DEF("singlestep", 0, QEMU_OPTION_singlestep, \ 3918 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL) 3919SRST 3920``-singlestep`` 3921 Run the emulation in single step mode. 3922ERST 3923 3924DEF("preconfig", 0, QEMU_OPTION_preconfig, \ 3925 "--preconfig pause QEMU before machine is initialized (experimental)\n", 3926 QEMU_ARCH_ALL) 3927SRST 3928``--preconfig`` 3929 Pause QEMU for interactive configuration before the machine is 3930 created, which allows querying and configuring properties that will 3931 affect machine initialization. Use QMP command 'x-exit-preconfig' to 3932 exit the preconfig state and move to the next state (i.e. run guest 3933 if -S isn't used or pause the second time if -S is used). This 3934 option is experimental. 3935ERST 3936 3937DEF("S", 0, QEMU_OPTION_S, \ 3938 "-S freeze CPU at startup (use 'c' to start execution)\n", 3939 QEMU_ARCH_ALL) 3940SRST 3941``-S`` 3942 Do not start CPU at startup (you must type 'c' in the monitor). 3943ERST 3944 3945DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit, 3946 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n" 3947 " run qemu with overcommit hints\n" 3948 " mem-lock=on|off controls memory lock support (default: off)\n" 3949 " cpu-pm=on|off controls cpu power management (default: off)\n", 3950 QEMU_ARCH_ALL) 3951SRST 3952``-overcommit mem-lock=on|off`` 3953 \ 3954``-overcommit cpu-pm=on|off`` 3955 Run qemu with hints about host resource overcommit. The default is 3956 to assume that host overcommits all resources. 3957 3958 Locking qemu and guest memory can be enabled via ``mem-lock=on`` 3959 (disabled by default). This works when host memory is not 3960 overcommitted and reduces the worst-case latency for guest. 3961 3962 Guest ability to manage power state of host cpus (increasing latency 3963 for other processes on the same host cpu, but decreasing latency for 3964 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This 3965 works best when host CPU is not overcommitted. When used, host 3966 estimates of CPU cycle and power utilization will be incorrect, not 3967 taking into account guest idle time. 3968ERST 3969 3970DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \ 3971 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n" 3972 " the guest without waiting for gdb to connect; use -S too\n" 3973 " if you want it to not start execution.)\n", 3974 QEMU_ARCH_ALL) 3975SRST 3976``-gdb dev`` 3977 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter 3978 in the System Emulation Users Guide). Note that this option does not pause QEMU 3979 execution -- if you want QEMU to not start the guest until you 3980 connect with gdb and issue a ``continue`` command, you will need to 3981 also pass the ``-S`` option to QEMU. 3982 3983 The most usual configuration is to listen on a local TCP socket:: 3984 3985 -gdb tcp::3117 3986 3987 but you can specify other backends; UDP, pseudo TTY, or even stdio 3988 are all reasonable use cases. For example, a stdio connection 3989 allows you to start QEMU from within gdb and establish the 3990 connection via a pipe: 3991 3992 .. parsed-literal:: 3993 3994 (gdb) target remote | exec |qemu_system| -gdb stdio ... 3995ERST 3996 3997DEF("s", 0, QEMU_OPTION_s, \ 3998 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n", 3999 QEMU_ARCH_ALL) 4000SRST 4001``-s`` 4002 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234 4003 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide). 4004ERST 4005 4006DEF("d", HAS_ARG, QEMU_OPTION_d, \ 4007 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n", 4008 QEMU_ARCH_ALL) 4009SRST 4010``-d item1[,...]`` 4011 Enable logging of specified items. Use '-d help' for a list of log 4012 items. 4013ERST 4014 4015DEF("D", HAS_ARG, QEMU_OPTION_D, \ 4016 "-D logfile output log to logfile (default stderr)\n", 4017 QEMU_ARCH_ALL) 4018SRST 4019``-D logfile`` 4020 Output log in logfile instead of to stderr 4021ERST 4022 4023DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \ 4024 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n", 4025 QEMU_ARCH_ALL) 4026SRST 4027``-dfilter range1[,...]`` 4028 Filter debug output to that relevant to a range of target addresses. 4029 The filter spec can be either start+size, start-size or start..end 4030 where start end and size are the addresses and sizes required. For 4031 example: 4032 4033 :: 4034 4035 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000 4036 4037 Will dump output for any code in the 0x1000 sized block starting at 4038 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and 4039 another 0x1000 sized block starting at 0xffffffc00005f000. 4040ERST 4041 4042DEF("seed", HAS_ARG, QEMU_OPTION_seed, \ 4043 "-seed number seed the pseudo-random number generator\n", 4044 QEMU_ARCH_ALL) 4045SRST 4046``-seed number`` 4047 Force the guest to use a deterministic pseudo-random number 4048 generator, seeded with number. This does not affect crypto routines 4049 within the host. 4050ERST 4051 4052DEF("L", HAS_ARG, QEMU_OPTION_L, \ 4053 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n", 4054 QEMU_ARCH_ALL) 4055SRST 4056``-L path`` 4057 Set the directory for the BIOS, VGA BIOS and keymaps. 4058 4059 To list all the data directories, use ``-L help``. 4060ERST 4061 4062DEF("bios", HAS_ARG, QEMU_OPTION_bios, \ 4063 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL) 4064SRST 4065``-bios file`` 4066 Set the filename for the BIOS. 4067ERST 4068 4069DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \ 4070 "-enable-kvm enable KVM full virtualization support\n", QEMU_ARCH_ALL) 4071SRST 4072``-enable-kvm`` 4073 Enable KVM full virtualization support. This option is only 4074 available if KVM support is enabled when compiling. 4075ERST 4076 4077DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, 4078 "-xen-domid id specify xen guest domain id\n", QEMU_ARCH_ALL) 4079DEF("xen-attach", 0, QEMU_OPTION_xen_attach, 4080 "-xen-attach attach to existing xen domain\n" 4081 " libxl will use this when starting QEMU\n", 4082 QEMU_ARCH_ALL) 4083DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, 4084 "-xen-domid-restrict restrict set of available xen operations\n" 4085 " to specified domain id. (Does not affect\n" 4086 " xenpv machine type).\n", 4087 QEMU_ARCH_ALL) 4088SRST 4089``-xen-domid id`` 4090 Specify xen guest domain id (XEN only). 4091 4092``-xen-attach`` 4093 Attach to existing xen domain. libxl will use this when starting 4094 QEMU (XEN only). Restrict set of available xen operations to 4095 specified domain id (XEN only). 4096ERST 4097 4098DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \ 4099 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL) 4100SRST 4101``-no-reboot`` 4102 Exit instead of rebooting. 4103ERST 4104 4105DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \ 4106 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL) 4107SRST 4108``-no-shutdown`` 4109 Don't exit QEMU on guest shutdown, but instead only stop the 4110 emulation. This allows for instance switching to monitor to commit 4111 changes to the disk image. 4112ERST 4113 4114DEF("action", HAS_ARG, QEMU_OPTION_action, 4115 "-action reboot=reset|shutdown\n" 4116 " action when guest reboots [default=reset]\n" 4117 "-action shutdown=poweroff|pause\n" 4118 " action when guest shuts down [default=poweroff]\n" 4119 "-action panic=pause|shutdown|none\n" 4120 " action when guest panics [default=shutdown]\n" 4121 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" 4122 " action when watchdog fires [default=reset]\n", 4123 QEMU_ARCH_ALL) 4124SRST 4125``-action event=action`` 4126 The action parameter serves to modify QEMU's default behavior when 4127 certain guest events occur. It provides a generic method for specifying the 4128 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown`` 4129 parameters. 4130 4131 Examples: 4132 4133 ``-action panic=none`` 4134 ``-action reboot=shutdown,shutdown=pause`` 4135 ``-watchdog i6300esb -action watchdog=pause`` 4136 4137ERST 4138 4139DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \ 4140 "-loadvm [tag|id]\n" \ 4141 " start right away with a saved state (loadvm in monitor)\n", 4142 QEMU_ARCH_ALL) 4143SRST 4144``-loadvm file`` 4145 Start right away with a saved state (``loadvm`` in monitor) 4146ERST 4147 4148#ifndef _WIN32 4149DEF("daemonize", 0, QEMU_OPTION_daemonize, \ 4150 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL) 4151#endif 4152SRST 4153``-daemonize`` 4154 Daemonize the QEMU process after initialization. QEMU will not 4155 detach from standard IO until it is ready to receive connections on 4156 any of its devices. This option is a useful way for external 4157 programs to launch QEMU without having to cope with initialization 4158 race conditions. 4159ERST 4160 4161DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \ 4162 "-option-rom rom load a file, rom, into the option ROM space\n", 4163 QEMU_ARCH_ALL) 4164SRST 4165``-option-rom file`` 4166 Load the contents of file as an option ROM. This option is useful to 4167 load things like EtherBoot. 4168ERST 4169 4170DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \ 4171 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \ 4172 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n", 4173 QEMU_ARCH_ALL) 4174 4175SRST 4176``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]`` 4177 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at 4178 the current UTC or local time, respectively. ``localtime`` is 4179 required for correct date in MS-DOS or Windows. To start at a 4180 specific point in time, provide datetime in the format 4181 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC. 4182 4183 By default the RTC is driven by the host system time. This allows 4184 using of the RTC as accurate reference clock inside the guest, 4185 specifically if the host time is smoothly following an accurate 4186 external reference clock, e.g. via NTP. If you want to isolate the 4187 guest time from the host, you can set ``clock`` to ``rt`` instead, 4188 which provides a host monotonic clock if host support it. To even 4189 prevent the RTC from progressing during suspension, you can set 4190 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is 4191 recommended especially in icount mode in order to preserve 4192 determinism; however, note that in icount mode the speed of the 4193 virtual clock is variable and can in general differ from the host 4194 clock. 4195 4196 Enable ``driftfix`` (i386 targets only) if you experience time drift 4197 problems, specifically with Windows' ACPI HAL. This option will try 4198 to figure out how many timer interrupts were not processed by the 4199 Windows guest and will re-inject them. 4200ERST 4201 4202DEF("icount", HAS_ARG, QEMU_OPTION_icount, \ 4203 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \ 4204 " enable virtual instruction counter with 2^N clock ticks per\n" \ 4205 " instruction, enable aligning the host and virtual clocks\n" \ 4206 " or disable real time cpu sleeping, and optionally enable\n" \ 4207 " record-and-replay mode\n", QEMU_ARCH_ALL) 4208SRST 4209``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]`` 4210 Enable virtual instruction counter. The virtual cpu will execute one 4211 instruction every 2^N ns of virtual time. If ``auto`` is specified 4212 then the virtual cpu speed will be automatically adjusted to keep 4213 virtual time within a few seconds of real time. 4214 4215 Note that while this option can give deterministic behavior, it does 4216 not provide cycle accurate emulation. Modern CPUs contain 4217 superscalar out of order cores with complex cache hierarchies. The 4218 number of instructions executed often has little or no correlation 4219 with actual performance. 4220 4221 When the virtual cpu is sleeping, the virtual time will advance at 4222 default speed unless ``sleep=on`` is specified. With 4223 ``sleep=on``, the virtual time will jump to the next timer 4224 deadline instantly whenever the virtual cpu goes to sleep mode and 4225 will not advance if no timer is enabled. This behavior gives 4226 deterministic execution times from the guest point of view. 4227 The default if icount is enabled is ``sleep=off``. 4228 ``sleep=on`` cannot be used together with either ``shift=auto`` 4229 or ``align=on``. 4230 4231 ``align=on`` will activate the delay algorithm which will try to 4232 synchronise the host clock and the virtual clock. The goal is to 4233 have a guest running at the real frequency imposed by the shift 4234 option. Whenever the guest clock is behind the host clock and if 4235 ``align=on`` is specified then we print a message to the user to 4236 inform about the delay. Currently this option does not work when 4237 ``shift`` is ``auto``. Note: The sync algorithm will work for those 4238 shift values for which the guest clock runs ahead of the host clock. 4239 Typically this happens when the shift value is high (how high 4240 depends on the host machine). The default if icount is enabled 4241 is ``align=off``. 4242 4243 When the ``rr`` option is specified deterministic record/replay is 4244 enabled. The ``rrfile=`` option must also be provided to 4245 specify the path to the replay log. In record mode data is written 4246 to this file, and in replay mode it is read back. 4247 If the ``rrsnapshot`` option is given then it specifies a VM snapshot 4248 name. In record mode, a new VM snapshot with the given name is created 4249 at the start of execution recording. In replay mode this option 4250 specifies the snapshot name used to load the initial VM state. 4251ERST 4252 4253DEF("watchdog", HAS_ARG, QEMU_OPTION_watchdog, \ 4254 "-watchdog model\n" \ 4255 " enable virtual hardware watchdog [default=none]\n", 4256 QEMU_ARCH_ALL) 4257SRST 4258``-watchdog model`` 4259 Create a virtual hardware watchdog device. Once enabled (by a guest 4260 action), the watchdog must be periodically polled by an agent inside 4261 the guest or else the guest will be restarted. Choose a model for 4262 which your guest has drivers. 4263 4264 The model is the model of hardware watchdog to emulate. Use 4265 ``-watchdog help`` to list available hardware models. Only one 4266 watchdog can be enabled for a guest. 4267 4268 The following models may be available: 4269 4270 ``ib700`` 4271 iBASE 700 is a very simple ISA watchdog with a single timer. 4272 4273 ``i6300esb`` 4274 Intel 6300ESB I/O controller hub is a much more featureful 4275 PCI-based dual-timer watchdog. 4276 4277 ``diag288`` 4278 A virtual watchdog for s390x backed by the diagnose 288 4279 hypercall (currently KVM only). 4280ERST 4281 4282DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \ 4283 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \ 4284 " action when watchdog fires [default=reset]\n", 4285 QEMU_ARCH_ALL) 4286SRST 4287``-watchdog-action action`` 4288 The action controls what QEMU will do when the watchdog timer 4289 expires. The default is ``reset`` (forcefully reset the guest). 4290 Other possible actions are: ``shutdown`` (attempt to gracefully 4291 shutdown the guest), ``poweroff`` (forcefully poweroff the guest), 4292 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the 4293 guest), ``debug`` (print a debug message and continue), or ``none`` 4294 (do nothing). 4295 4296 Note that the ``shutdown`` action requires that the guest responds 4297 to ACPI signals, which it may not be able to do in the sort of 4298 situations where the watchdog would have expired, and thus 4299 ``-watchdog-action shutdown`` is not recommended for production use. 4300 4301 Examples: 4302 4303 ``-watchdog i6300esb -watchdog-action pause``; \ ``-watchdog ib700`` 4304 4305ERST 4306 4307DEF("echr", HAS_ARG, QEMU_OPTION_echr, \ 4308 "-echr chr set terminal escape character instead of ctrl-a\n", 4309 QEMU_ARCH_ALL) 4310SRST 4311``-echr numeric_ascii_value`` 4312 Change the escape character used for switching to the monitor when 4313 using monitor and serial sharing. The default is ``0x01`` when using 4314 the ``-nographic`` option. ``0x01`` is equal to pressing 4315 ``Control-a``. You can select a different character from the ascii 4316 control keys where 1 through 26 map to Control-a through Control-z. 4317 For instance you could use the either of the following to change the 4318 escape character to Control-t. 4319 4320 ``-echr 0x14``; \ ``-echr 20`` 4321 4322ERST 4323 4324DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \ 4325 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \ 4326 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \ 4327 "-incoming unix:socketpath\n" \ 4328 " prepare for incoming migration, listen on\n" \ 4329 " specified protocol and socket address\n" \ 4330 "-incoming fd:fd\n" \ 4331 "-incoming exec:cmdline\n" \ 4332 " accept incoming migration on given file descriptor\n" \ 4333 " or from given external command\n" \ 4334 "-incoming defer\n" \ 4335 " wait for the URI to be specified via migrate_incoming\n", 4336 QEMU_ARCH_ALL) 4337SRST 4338``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]`` 4339 \ 4340``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]`` 4341 Prepare for incoming migration, listen on a given tcp port. 4342 4343``-incoming unix:socketpath`` 4344 Prepare for incoming migration, listen on a given unix socket. 4345 4346``-incoming fd:fd`` 4347 Accept incoming migration from a given filedescriptor. 4348 4349``-incoming exec:cmdline`` 4350 Accept incoming migration as an output from specified external 4351 command. 4352 4353``-incoming defer`` 4354 Wait for the URI to be specified via migrate\_incoming. The monitor 4355 can be used to change settings (such as migration parameters) prior 4356 to issuing the migrate\_incoming to allow the migration to begin. 4357ERST 4358 4359DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \ 4360 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL) 4361SRST 4362``-only-migratable`` 4363 Only allow migratable devices. Devices will not be allowed to enter 4364 an unmigratable state. 4365ERST 4366 4367DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \ 4368 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL) 4369SRST 4370``-nodefaults`` 4371 Don't create default devices. Normally, QEMU sets the default 4372 devices like serial port, parallel port, virtual console, monitor 4373 device, VGA adapter, floppy and CD-ROM drive and others. The 4374 ``-nodefaults`` option will disable all those default devices. 4375ERST 4376 4377#ifndef _WIN32 4378DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \ 4379 "-chroot dir chroot to dir just before starting the VM\n", 4380 QEMU_ARCH_ALL) 4381#endif 4382SRST 4383``-chroot dir`` 4384 Immediately before starting guest execution, chroot to the specified 4385 directory. Especially useful in combination with -runas. 4386ERST 4387 4388#ifndef _WIN32 4389DEF("runas", HAS_ARG, QEMU_OPTION_runas, \ 4390 "-runas user change to user id user just before starting the VM\n" \ 4391 " user can be numeric uid:gid instead\n", 4392 QEMU_ARCH_ALL) 4393#endif 4394SRST 4395``-runas user`` 4396 Immediately before starting guest execution, drop root privileges, 4397 switching to the specified user. 4398ERST 4399 4400DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env, 4401 "-prom-env variable=value\n" 4402 " set OpenBIOS nvram variables\n", 4403 QEMU_ARCH_PPC | QEMU_ARCH_SPARC) 4404SRST 4405``-prom-env variable=value`` 4406 Set OpenBIOS nvram variable to given value (PPC, SPARC only). 4407 4408 :: 4409 4410 qemu-system-sparc -prom-env 'auto-boot?=false' \ 4411 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single' 4412 4413 :: 4414 4415 qemu-system-ppc -prom-env 'auto-boot?=false' \ 4416 -prom-env 'boot-device=hd:2,\yaboot' \ 4417 -prom-env 'boot-args=conf=hd:2,\yaboot.conf' 4418ERST 4419DEF("semihosting", 0, QEMU_OPTION_semihosting, 4420 "-semihosting semihosting mode\n", 4421 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 4422 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) 4423SRST 4424``-semihosting`` 4425 Enable semihosting mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only). 4426 4427 Note that this allows guest direct access to the host filesystem, so 4428 should only be used with a trusted guest OS. 4429 4430 See the -semihosting-config option documentation for further 4431 information about the facilities this enables. 4432ERST 4433DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config, 4434 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]\n" \ 4435 " semihosting configuration\n", 4436QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | 4437QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) 4438SRST 4439``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]`` 4440 Enable and configure semihosting (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V 4441 only). 4442 4443 Note that this allows guest direct access to the host filesystem, so 4444 should only be used with a trusted guest OS. 4445 4446 On Arm this implements the standard semihosting API, version 2.0. 4447 4448 On M68K this implements the "ColdFire GDB" interface used by 4449 libgloss. 4450 4451 Xtensa semihosting provides basic file IO calls, such as 4452 open/read/write/seek/select. Tensilica baremetal libc for ISS and 4453 linux platform "sim" use this interface. 4454 4455 On RISC-V this implements the standard semihosting API, version 0.2. 4456 4457 ``target=native|gdb|auto`` 4458 Defines where the semihosting calls will be addressed, to QEMU 4459 (``native``) or to GDB (``gdb``). The default is ``auto``, which 4460 means ``gdb`` during debug sessions and ``native`` otherwise. 4461 4462 ``chardev=str1`` 4463 Send the output to a chardev backend output for native or auto 4464 output when not in gdb 4465 4466 ``arg=str1,arg=str2,...`` 4467 Allows the user to pass input arguments, and can be used 4468 multiple times to build up a list. The old-style 4469 ``-kernel``/``-append`` method of passing a command line is 4470 still supported for backward compatibility. If both the 4471 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are 4472 specified, the former is passed to semihosting as it always 4473 takes precedence. 4474ERST 4475DEF("old-param", 0, QEMU_OPTION_old_param, 4476 "-old-param old param mode\n", QEMU_ARCH_ARM) 4477SRST 4478``-old-param`` 4479 Old param mode (ARM only). 4480ERST 4481 4482DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \ 4483 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \ 4484 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \ 4485 " Enable seccomp mode 2 system call filter (default 'off').\n" \ 4486 " use 'obsolete' to allow obsolete system calls that are provided\n" \ 4487 " by the kernel, but typically no longer used by modern\n" \ 4488 " C library implementations.\n" \ 4489 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \ 4490 " to elevate privileges using set*uid|gid system calls.\n" \ 4491 " The value 'children' will deny set*uid|gid system calls for\n" \ 4492 " main QEMU process but will allow forks and execves to run unprivileged\n" \ 4493 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \ 4494 " blocking *fork and execve\n" \ 4495 " use 'resourcecontrol' to disable process affinity and schedular priority\n", 4496 QEMU_ARCH_ALL) 4497SRST 4498``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]`` 4499 Enable Seccomp mode 2 system call filter. 'on' will enable syscall 4500 filtering and 'off' will disable it. The default is 'off'. 4501 4502 ``obsolete=string`` 4503 Enable Obsolete system calls 4504 4505 ``elevateprivileges=string`` 4506 Disable set\*uid\|gid system calls 4507 4508 ``spawn=string`` 4509 Disable \*fork and execve 4510 4511 ``resourcecontrol=string`` 4512 Disable process affinity and schedular priority 4513ERST 4514 4515DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig, 4516 "-readconfig <file>\n", QEMU_ARCH_ALL) 4517SRST 4518``-readconfig file`` 4519 Read device configuration from file. This approach is useful when 4520 you want to spawn QEMU process with many command line options but 4521 you don't want to exceed the command line character limit. 4522ERST 4523DEF("writeconfig", HAS_ARG, QEMU_OPTION_writeconfig, 4524 "-writeconfig <file>\n" 4525 " read/write config file (deprecated)\n", QEMU_ARCH_ALL) 4526SRST 4527ERST 4528 4529DEF("no-user-config", 0, QEMU_OPTION_nouserconfig, 4530 "-no-user-config\n" 4531 " do not load default user-provided config files at startup\n", 4532 QEMU_ARCH_ALL) 4533SRST 4534``-no-user-config`` 4535 The ``-no-user-config`` option makes QEMU not load any of the 4536 user-provided config files on sysconfdir. 4537ERST 4538 4539DEF("trace", HAS_ARG, QEMU_OPTION_trace, 4540 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n" 4541 " specify tracing options\n", 4542 QEMU_ARCH_ALL) 4543SRST 4544``-trace [[enable=]pattern][,events=file][,file=file]`` 4545 .. include:: ../qemu-option-trace.rst.inc 4546 4547ERST 4548DEF("plugin", HAS_ARG, QEMU_OPTION_plugin, 4549 "-plugin [file=]<file>[,<argname>=<argvalue>]\n" 4550 " load a plugin\n", 4551 QEMU_ARCH_ALL) 4552SRST 4553``-plugin file=file[,argname=argvalue]`` 4554 Load a plugin. 4555 4556 ``file=file`` 4557 Load the given plugin from a shared library file. 4558 4559 ``argname=argvalue`` 4560 Argument passed to the plugin. (Can be given multiple times.) 4561ERST 4562 4563HXCOMM Internal use 4564DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL) 4565DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL) 4566 4567#ifdef __linux__ 4568DEF("enable-fips", 0, QEMU_OPTION_enablefips, 4569 "-enable-fips enable FIPS 140-2 compliance\n", 4570 QEMU_ARCH_ALL) 4571#endif 4572SRST 4573``-enable-fips`` 4574 Enable FIPS 140-2 compliance mode. 4575ERST 4576 4577DEF("msg", HAS_ARG, QEMU_OPTION_msg, 4578 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n" 4579 " control error message format\n" 4580 " timestamp=on enables timestamps (default: off)\n" 4581 " guest-name=on enables guest name prefix but only if\n" 4582 " -name guest option is set (default: off)\n", 4583 QEMU_ARCH_ALL) 4584SRST 4585``-msg [timestamp[=on|off]][,guest-name[=on|off]]`` 4586 Control error message format. 4587 4588 ``timestamp=on|off`` 4589 Prefix messages with a timestamp. Default is off. 4590 4591 ``guest-name=on|off`` 4592 Prefix messages with guest name but only if -name guest option is set 4593 otherwise the option is ignored. Default is off. 4594ERST 4595 4596DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate, 4597 "-dump-vmstate <file>\n" 4598 " Output vmstate information in JSON format to file.\n" 4599 " Use the scripts/vmstate-static-checker.py file to\n" 4600 " check for possible regressions in migration code\n" 4601 " by comparing two such vmstate dumps.\n", 4602 QEMU_ARCH_ALL) 4603SRST 4604``-dump-vmstate file`` 4605 Dump json-encoded vmstate information for current machine type to 4606 file in file 4607ERST 4608 4609DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile, 4610 "-enable-sync-profile\n" 4611 " enable synchronization profiling\n", 4612 QEMU_ARCH_ALL) 4613SRST 4614``-enable-sync-profile`` 4615 Enable synchronization profiling. 4616ERST 4617 4618DEFHEADING() 4619 4620DEFHEADING(Generic object creation:) 4621 4622DEF("object", HAS_ARG, QEMU_OPTION_object, 4623 "-object TYPENAME[,PROP1=VALUE1,...]\n" 4624 " create a new object of type TYPENAME setting properties\n" 4625 " in the order they are specified. Note that the 'id'\n" 4626 " property must be set. These objects are placed in the\n" 4627 " '/objects' path.\n", 4628 QEMU_ARCH_ALL) 4629SRST 4630``-object typename[,prop1=value1,...]`` 4631 Create a new object of type typename setting properties in the order 4632 they are specified. Note that the 'id' property must be set. These 4633 objects are placed in the '/objects' path. 4634 4635 ``-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`` 4636 Creates a memory file backend object, which can be used to back 4637 the guest RAM with huge pages. 4638 4639 The ``id`` parameter is a unique ID that will be used to 4640 reference this memory region in other parameters, e.g. ``-numa``, 4641 ``-device nvdimm``, etc. 4642 4643 The ``size`` option provides the size of the memory region, and 4644 accepts common suffixes, e.g. ``500M``. 4645 4646 The ``mem-path`` provides the path to either a shared memory or 4647 huge page filesystem mount. 4648 4649 The ``share`` boolean option determines whether the memory 4650 region is marked as private to QEMU, or shared. The latter 4651 allows a co-operating external process to access the QEMU memory 4652 region. 4653 4654 The ``share`` is also required for pvrdma devices due to 4655 limitations in the RDMA API provided by Linux. 4656 4657 Setting share=on might affect the ability to configure NUMA 4658 bindings for the memory backend under some circumstances, see 4659 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel 4660 source tree for additional details. 4661 4662 Setting the ``discard-data`` boolean option to on indicates that 4663 file contents can be destroyed when QEMU exits, to avoid 4664 unnecessarily flushing data to the backing file. Note that 4665 ``discard-data`` is only an optimization, and QEMU might not 4666 discard file contents if it aborts unexpectedly or is terminated 4667 using SIGKILL. 4668 4669 The ``merge`` boolean option enables memory merge, also known as 4670 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider 4671 the pages for memory deduplication. 4672 4673 Setting the ``dump`` boolean option to off excludes the memory 4674 from core dumps. This feature is also known as MADV\_DONTDUMP. 4675 4676 The ``prealloc`` boolean option enables memory preallocation. 4677 4678 The ``host-nodes`` option binds the memory range to a list of 4679 NUMA host nodes. 4680 4681 The ``policy`` option sets the NUMA policy to one of the 4682 following values: 4683 4684 ``default`` 4685 default host policy 4686 4687 ``preferred`` 4688 prefer the given host node list for allocation 4689 4690 ``bind`` 4691 restrict memory allocation to the given host node list 4692 4693 ``interleave`` 4694 interleave memory allocations across the given host node 4695 list 4696 4697 The ``align`` option specifies the base address alignment when 4698 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg 4699 ``2M``. Some backend store specified by ``mem-path`` requires an 4700 alignment different than the default one used by QEMU, eg the 4701 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In 4702 such cases, users can specify the required alignment via this 4703 option. 4704 4705 The ``pmem`` option specifies whether the backing file specified 4706 by ``mem-path`` is in host persistent memory that can be 4707 accessed using the SNIA NVM programming model (e.g. Intel 4708 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary 4709 operations to guarantee the persistence of its own writes to 4710 ``mem-path`` (e.g. in vNVDIMM label emulation and live 4711 migration). Also, we will map the backend-file with MAP\_SYNC 4712 flag, which ensures the file metadata is in sync for 4713 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC 4714 requires support from both the host kernel (since Linux kernel 4715 4.15) and the filesystem of ``mem-path`` mounted with DAX 4716 option. 4717 4718 The ``readonly`` option specifies whether the backing file is opened 4719 read-only or read-write (default). 4720 4721 ``-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`` 4722 Creates a memory backend object, which can be used to back the 4723 guest RAM. Memory backend objects offer more control than the 4724 ``-m`` option that is traditionally used to define guest RAM. 4725 Please refer to ``memory-backend-file`` for a description of the 4726 options. 4727 4728 ``-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`` 4729 Creates an anonymous memory file backend object, which allows 4730 QEMU to share the memory with an external process (e.g. when 4731 using vhost-user). The memory is allocated with memfd and 4732 optional sealing. (Linux only) 4733 4734 The ``seal`` option creates a sealed-file, that will block 4735 further resizing the memory ('on' by default). 4736 4737 The ``hugetlb`` option specify the file to be created resides in 4738 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction 4739 with the ``hugetlb`` option, the ``hugetlbsize`` option specify 4740 the hugetlb page size on systems that support multiple hugetlb 4741 page sizes (it must be a power of 2 value supported by the 4742 system). 4743 4744 In some versions of Linux, the ``hugetlb`` option is 4745 incompatible with the ``seal`` option (requires at least Linux 4746 4.16). 4747 4748 Please refer to ``memory-backend-file`` for a description of the 4749 other options. 4750 4751 The ``share`` boolean option is on by default with memfd. 4752 4753 ``-object rng-builtin,id=id`` 4754 Creates a random number generator backend which obtains entropy 4755 from QEMU builtin functions. The ``id`` parameter is a unique ID 4756 that will be used to reference this entropy backend from the 4757 ``virtio-rng`` device. By default, the ``virtio-rng`` device 4758 uses this RNG backend. 4759 4760 ``-object rng-random,id=id,filename=/dev/random`` 4761 Creates a random number generator backend which obtains entropy 4762 from a device on the host. The ``id`` parameter is a unique ID 4763 that will be used to reference this entropy backend from the 4764 ``virtio-rng`` device. The ``filename`` parameter specifies 4765 which file to obtain entropy from and if omitted defaults to 4766 ``/dev/urandom``. 4767 4768 ``-object rng-egd,id=id,chardev=chardevid`` 4769 Creates a random number generator backend which obtains entropy 4770 from an external daemon running on the host. The ``id`` 4771 parameter is a unique ID that will be used to reference this 4772 entropy backend from the ``virtio-rng`` device. The ``chardev`` 4773 parameter is the unique ID of a character device backend that 4774 provides the connection to the RNG daemon. 4775 4776 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off`` 4777 Creates a TLS anonymous credentials object, which can be used to 4778 provide TLS support on network backends. The ``id`` parameter is 4779 a unique ID which network backends will use to access the 4780 credentials. The ``endpoint`` is either ``server`` or ``client`` 4781 depending on whether the QEMU network backend that uses the 4782 credentials will be acting as a client or as a server. If 4783 ``verify-peer`` is enabled (the default) then once the handshake 4784 is completed, the peer credentials will be verified, though this 4785 is a no-op for anonymous credentials. 4786 4787 The dir parameter tells QEMU where to find the credential files. 4788 For server endpoints, this directory may contain a file 4789 dh-params.pem providing diffie-hellman parameters to use for the 4790 TLS server. If the file is missing, QEMU will generate a set of 4791 DH parameters at startup. This is a computationally expensive 4792 operation that consumes random pool entropy, so it is 4793 recommended that a persistent set of parameters be generated 4794 upfront and saved. 4795 4796 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]`` 4797 Creates a TLS Pre-Shared Keys (PSK) credentials object, which 4798 can be used to provide TLS support on network backends. The 4799 ``id`` parameter is a unique ID which network backends will use 4800 to access the credentials. The ``endpoint`` is either ``server`` 4801 or ``client`` depending on whether the QEMU network backend that 4802 uses the credentials will be acting as a client or as a server. 4803 For clients only, ``username`` is the username which will be 4804 sent to the server. If omitted it defaults to "qemu". 4805 4806 The dir parameter tells QEMU where to find the keys file. It is 4807 called "dir/keys.psk" and contains "username:key" pairs. This 4808 file can most easily be created using the GnuTLS ``psktool`` 4809 program. 4810 4811 For server endpoints, dir may also contain a file dh-params.pem 4812 providing diffie-hellman parameters to use for the TLS server. 4813 If the file is missing, QEMU will generate a set of DH 4814 parameters at startup. This is a computationally expensive 4815 operation that consumes random pool entropy, so it is 4816 recommended that a persistent set of parameters be generated up 4817 front and saved. 4818 4819 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id`` 4820 Creates a TLS anonymous credentials object, which can be used to 4821 provide TLS support on network backends. The ``id`` parameter is 4822 a unique ID which network backends will use to access the 4823 credentials. The ``endpoint`` is either ``server`` or ``client`` 4824 depending on whether the QEMU network backend that uses the 4825 credentials will be acting as a client or as a server. If 4826 ``verify-peer`` is enabled (the default) then once the handshake 4827 is completed, the peer credentials will be verified. With x509 4828 certificates, this implies that the clients must be provided 4829 with valid client certificates too. 4830 4831 The dir parameter tells QEMU where to find the credential files. 4832 For server endpoints, this directory may contain a file 4833 dh-params.pem providing diffie-hellman parameters to use for the 4834 TLS server. If the file is missing, QEMU will generate a set of 4835 DH parameters at startup. This is a computationally expensive 4836 operation that consumes random pool entropy, so it is 4837 recommended that a persistent set of parameters be generated 4838 upfront and saved. 4839 4840 For x509 certificate credentials the directory will contain 4841 further files providing the x509 certificates. The certificates 4842 must be stored in PEM format, in filenames ca-cert.pem, 4843 ca-crl.pem (optional), server-cert.pem (only servers), 4844 server-key.pem (only servers), client-cert.pem (only clients), 4845 and client-key.pem (only clients). 4846 4847 For the server-key.pem and client-key.pem files which contain 4848 sensitive private keys, it is possible to use an encrypted 4849 version by providing the passwordid parameter. This provides the 4850 ID of a previously created ``secret`` object containing the 4851 password for decryption. 4852 4853 The priority parameter allows to override the global default 4854 priority used by gnutls. This can be useful if the system 4855 administrator needs to use a weaker set of crypto priorities for 4856 QEMU without potentially forcing the weakness onto all 4857 applications. Or conversely if one wants wants a stronger 4858 default for QEMU than for all other applications, they can do 4859 this through this parameter. Its format is a gnutls priority 4860 string as described at 4861 https://gnutls.org/manual/html_node/Priority-Strings.html. 4862 4863 ``-object tls-cipher-suites,id=id,priority=priority`` 4864 Creates a TLS cipher suites object, which can be used to control 4865 the TLS cipher/protocol algorithms that applications are permitted 4866 to use. 4867 4868 The ``id`` parameter is a unique ID which frontends will use to 4869 access the ordered list of permitted TLS cipher suites from the 4870 host. 4871 4872 The ``priority`` parameter allows to override the global default 4873 priority used by gnutls. This can be useful if the system 4874 administrator needs to use a weaker set of crypto priorities for 4875 QEMU without potentially forcing the weakness onto all 4876 applications. Or conversely if one wants wants a stronger 4877 default for QEMU than for all other applications, they can do 4878 this through this parameter. Its format is a gnutls priority 4879 string as described at 4880 https://gnutls.org/manual/html_node/Priority-Strings.html. 4881 4882 An example of use of this object is to control UEFI HTTPS Boot. 4883 The tls-cipher-suites object exposes the ordered list of permitted 4884 TLS cipher suites from the host side to the guest firmware, via 4885 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER 4886 objects. The firmware uses the IANA_TLS_CIPHER array for configuring 4887 guest-side TLS. 4888 4889 In the following example, the priority at which the host-side policy 4890 is retrieved is given by the ``priority`` property. 4891 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to 4892 refer to /etc/crypto-policies/back-ends/gnutls.config. 4893 4894 .. parsed-literal:: 4895 4896 # |qemu_system| \\ 4897 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\ 4898 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0 4899 4900 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]`` 4901 Interval t can't be 0, this filter batches the packet delivery: 4902 all packets arriving in a given interval on netdev netdevid are 4903 delayed until the end of the interval. Interval is in 4904 microseconds. ``status`` is optional that indicate whether the 4905 netfilter is on (enabled) or off (disabled), the default status 4906 for netfilter will be 'on'. 4907 4908 queue all\|rx\|tx is an option that can be applied to any 4909 netfilter. 4910 4911 ``all``: the filter is attached both to the receive and the 4912 transmit queue of the netdev (default). 4913 4914 ``rx``: the filter is attached to the receive queue of the 4915 netdev, where it will receive packets sent to the netdev. 4916 4917 ``tx``: the filter is attached to the transmit queue of the 4918 netdev, where it will receive packets sent by the netdev. 4919 4920 position head\|tail\|id=<id> is an option to specify where the 4921 filter should be inserted in the filter list. It can be applied 4922 to any netfilter. 4923 4924 ``head``: the filter is inserted at the head of the filter list, 4925 before any existing filters. 4926 4927 ``tail``: the filter is inserted at the tail of the filter list, 4928 behind any existing filters (default). 4929 4930 ``id=<id>``: the filter is inserted before or behind the filter 4931 specified by <id>, see the insert option below. 4932 4933 insert behind\|before is an option to specify where to insert 4934 the new filter relative to the one specified with 4935 position=id=<id>. It can be applied to any netfilter. 4936 4937 ``before``: insert before the specified filter. 4938 4939 ``behind``: insert behind the specified filter (default). 4940 4941 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 4942 filter-mirror on netdev netdevid,mirror net packet to 4943 chardevchardevid, if it has the vnet\_hdr\_support flag, 4944 filter-mirror will mirror packet with vnet\_hdr\_len. 4945 4946 ``-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]`` 4947 filter-redirector on netdev netdevid,redirect filter's net 4948 packet to chardev chardevid,and redirect indev's packet to 4949 filter.if it has the vnet\_hdr\_support flag, filter-redirector 4950 will redirect packet with vnet\_hdr\_len. Create a 4951 filter-redirector we need to differ outdev id from indev id, id 4952 can not be the same. we can just use indev or outdev, but at 4953 least one of indev or outdev need to be specified. 4954 4955 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 4956 Filter-rewriter is a part of COLO project.It will rewrite tcp 4957 packet to secondary from primary to keep secondary tcp 4958 connection,and rewrite tcp packet to primary from secondary make 4959 tcp packet can be handled by client.if it has the 4960 vnet\_hdr\_support flag, we can parse packet with vnet header. 4961 4962 usage: colo secondary: -object 4963 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object 4964 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object 4965 filter-rewriter,id=rew0,netdev=hn0,queue=all 4966 4967 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]`` 4968 Dump the network traffic on netdev dev to the file specified by 4969 filename. At most len bytes (64k by default) per packet are 4970 stored. The file format is libpcap, so it can be analyzed with 4971 tools such as tcpdump or Wireshark. 4972 4973 ``-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}]`` 4974 Colo-compare gets packet from primary\_in chardevid and 4975 secondary\_in, then compare whether the payload of primary packet 4976 and secondary packet are the same. If same, it will output 4977 primary packet to out\_dev, else it will notify COLO-framework to do 4978 checkpoint and send primary packet to out\_dev. In order to 4979 improve efficiency, we need to put the task of comparison in 4980 another iothread. If it has the vnet\_hdr\_support flag, 4981 colo compare will send/recv packet with vnet\_hdr\_len. 4982 The compare\_timeout=@var{ms} determines the maximum time of the 4983 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms} 4984 is to set the period of scanning expired primary node network packets. 4985 The max\_queue\_size=@var{size} is to set the max compare queue 4986 size depend on user environment. 4987 If user want to use Xen COLO, need to add the notify\_dev to 4988 notify Xen colo-frame to do checkpoint. 4989 4990 COLO-compare must be used with the help of filter-mirror, 4991 filter-redirector and filter-rewriter. 4992 4993 :: 4994 4995 KVM COLO 4996 4997 primary: 4998 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 4999 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5000 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5001 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5002 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5003 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5004 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5005 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5006 -object iothread,id=iothread1 5007 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5008 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5009 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5010 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1 5011 5012 secondary: 5013 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 5014 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5015 -chardev socket,id=red0,host=3.3.3.3,port=9003 5016 -chardev socket,id=red1,host=3.3.3.3,port=9004 5017 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5018 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5019 5020 5021 Xen COLO 5022 5023 primary: 5024 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 5025 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 5026 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off 5027 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off 5028 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off 5029 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 5030 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off 5031 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 5032 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off 5033 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 5034 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 5035 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 5036 -object iothread,id=iothread1 5037 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1 5038 5039 secondary: 5040 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 5041 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 5042 -chardev socket,id=red0,host=3.3.3.3,port=9003 5043 -chardev socket,id=red1,host=3.3.3.3,port=9004 5044 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 5045 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 5046 5047 If you want to know the detail of above command line, you can 5048 read the colo-compare git log. 5049 5050 ``-object cryptodev-backend-builtin,id=id[,queues=queues]`` 5051 Creates a cryptodev backend which executes crypto opreation from 5052 the QEMU cipher APIS. The id parameter is a unique ID that will 5053 be used to reference this cryptodev backend from the 5054 ``virtio-crypto`` device. The queues parameter is optional, 5055 which specify the queue number of cryptodev backend, the default 5056 of queues is 1. 5057 5058 .. parsed-literal:: 5059 5060 # |qemu_system| \\ 5061 [...] \\ 5062 -object cryptodev-backend-builtin,id=cryptodev0 \\ 5063 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5064 [...] 5065 5066 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]`` 5067 Creates a vhost-user cryptodev backend, backed by a chardev 5068 chardevid. The id parameter is a unique ID that will be used to 5069 reference this cryptodev backend from the ``virtio-crypto`` 5070 device. The chardev should be a unix domain socket backed one. 5071 The vhost-user uses a specifically defined protocol to pass 5072 vhost ioctl replacement messages to an application on the other 5073 end of the socket. The queues parameter is optional, which 5074 specify the queue number of cryptodev backend for multiqueue 5075 vhost-user, the default of queues is 1. 5076 5077 .. parsed-literal:: 5078 5079 # |qemu_system| \\ 5080 [...] \\ 5081 -chardev socket,id=chardev0,path=/path/to/socket \\ 5082 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\ 5083 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 5084 [...] 5085 5086 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]`` 5087 \ 5088 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]`` 5089 Defines a secret to store a password, encryption key, or some 5090 other sensitive data. The sensitive data can either be passed 5091 directly via the data parameter, or indirectly via the file 5092 parameter. Using the data parameter is insecure unless the 5093 sensitive data is encrypted. 5094 5095 The sensitive data can be provided in raw format (the default), 5096 or base64. When encoded as JSON, the raw format only supports 5097 valid UTF-8 characters, so base64 is recommended for sending 5098 binary data. QEMU will convert from which ever format is 5099 provided to the format it needs internally. eg, an RBD password 5100 can be provided in raw format, even though it will be base64 5101 encoded when passed onto the RBD sever. 5102 5103 For added protection, it is possible to encrypt the data 5104 associated with a secret using the AES-256-CBC cipher. Use of 5105 encryption is indicated by providing the keyid and iv 5106 parameters. The keyid parameter provides the ID of a previously 5107 defined secret that contains the AES-256 decryption key. This 5108 key should be 32-bytes long and be base64 encoded. The iv 5109 parameter provides the random initialization vector used for 5110 encryption of this particular secret and should be a base64 5111 encrypted string of the 16-byte IV. 5112 5113 The simplest (insecure) usage is to provide the secret inline 5114 5115 .. parsed-literal:: 5116 5117 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw 5118 5119 The simplest secure usage is to provide the secret via a file 5120 5121 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object 5122 secret,id=sec0,file=mypasswd.txt,format=raw 5123 5124 For greater security, AES-256-CBC should be used. To illustrate 5125 usage, consider the openssl command line tool which can encrypt 5126 the data. Note that when encrypting, the plaintext must be 5127 padded to the cipher block size (32 bytes) using the standard 5128 PKCS#5/6 compatible padding algorithm. 5129 5130 First a master key needs to be created in base64 encoding: 5131 5132 :: 5133 5134 # openssl rand -base64 32 > key.b64 5135 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"') 5136 5137 Each secret to be encrypted needs to have a random 5138 initialization vector generated. These do not need to be kept 5139 secret 5140 5141 :: 5142 5143 # openssl rand -base64 16 > iv.b64 5144 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"') 5145 5146 The secret to be defined can now be encrypted, in this case 5147 we're telling openssl to base64 encode the result, but it could 5148 be left as raw bytes if desired. 5149 5150 :: 5151 5152 # SECRET=$(printf "letmein" | 5153 openssl enc -aes-256-cbc -a -K $KEY -iv $IV) 5154 5155 When launching QEMU, create a master secret pointing to 5156 ``key.b64`` and specify that to be used to decrypt the user 5157 password. Pass the contents of ``iv.b64`` to the second secret 5158 5159 .. parsed-literal:: 5160 5161 # |qemu_system| \\ 5162 -object secret,id=secmaster0,format=base64,file=key.b64 \\ 5163 -object secret,id=sec0,keyid=secmaster0,format=base64,\\ 5164 data=$SECRET,iv=$(<iv.b64) 5165 5166 ``-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]`` 5167 Create a Secure Encrypted Virtualization (SEV) guest object, 5168 which can be used to provide the guest memory encryption support 5169 on AMD processors. 5170 5171 When memory encryption is enabled, one of the physical address 5172 bit (aka the C-bit) is utilized to mark if a memory page is 5173 protected. The ``cbitpos`` is used to provide the C-bit 5174 position. The C-bit position is Host family dependent hence user 5175 must provide this value. On EPYC, the value should be 47. 5176 5177 When memory encryption is enabled, we loose certain bits in 5178 physical address space. The ``reduced-phys-bits`` is used to 5179 provide the number of bits we loose in physical address space. 5180 Similar to C-bit, the value is Host family dependent. On EPYC, 5181 the value should be 5. 5182 5183 The ``sev-device`` provides the device file to use for 5184 communicating with the SEV firmware running inside AMD Secure 5185 Processor. The default device is '/dev/sev'. If hardware 5186 supports memory encryption then /dev/sev devices are created by 5187 CCP driver. 5188 5189 The ``policy`` provides the guest policy to be enforced by the 5190 SEV firmware and restrict what configuration and operational 5191 commands can be performed on this guest by the hypervisor. The 5192 policy should be provided by the guest owner and is bound to the 5193 guest and cannot be changed throughout the lifetime of the 5194 guest. The default is 0. 5195 5196 If guest ``policy`` allows sharing the key with another SEV 5197 guest then ``handle`` can be use to provide handle of the guest 5198 from which to share the key. 5199 5200 The ``dh-cert-file`` and ``session-file`` provides the guest 5201 owner's Public Diffie-Hillman key defined in SEV spec. The PDH 5202 and session parameters are used for establishing a cryptographic 5203 session with the guest owner to negotiate keys used for 5204 attestation. The file must be encoded in base64. 5205 5206 e.g to launch a SEV guest 5207 5208 .. parsed-literal:: 5209 5210 # |qemu_system_x86| \\ 5211 ...... \\ 5212 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \\ 5213 -machine ...,memory-encryption=sev0 \\ 5214 ..... 5215 5216 ``-object authz-simple,id=id,identity=string`` 5217 Create an authorization object that will control access to 5218 network services. 5219 5220 The ``identity`` parameter is identifies the user and its format 5221 depends on the network service that authorization object is 5222 associated with. For authorizing based on TLS x509 certificates, 5223 the identity must be the x509 distinguished name. Note that care 5224 must be taken to escape any commas in the distinguished name. 5225 5226 An example authorization object to validate a x509 distinguished 5227 name would look like: 5228 5229 .. parsed-literal:: 5230 5231 # |qemu_system| \\ 5232 ... \\ 5233 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\ 5234 ... 5235 5236 Note the use of quotes due to the x509 distinguished name 5237 containing whitespace, and escaping of ','. 5238 5239 ``-object authz-listfile,id=id,filename=path,refresh=on|off`` 5240 Create an authorization object that will control access to 5241 network services. 5242 5243 The ``filename`` parameter is the fully qualified path to a file 5244 containing the access control list rules in JSON format. 5245 5246 An example set of rules that match against SASL usernames might 5247 look like: 5248 5249 :: 5250 5251 { 5252 "rules": [ 5253 { "match": "fred", "policy": "allow", "format": "exact" }, 5254 { "match": "bob", "policy": "allow", "format": "exact" }, 5255 { "match": "danb", "policy": "deny", "format": "glob" }, 5256 { "match": "dan*", "policy": "allow", "format": "exact" }, 5257 ], 5258 "policy": "deny" 5259 } 5260 5261 When checking access the object will iterate over all the rules 5262 and the first rule to match will have its ``policy`` value 5263 returned as the result. If no rules match, then the default 5264 ``policy`` value is returned. 5265 5266 The rules can either be an exact string match, or they can use 5267 the simple UNIX glob pattern matching to allow wildcards to be 5268 used. 5269 5270 If ``refresh`` is set to true the file will be monitored and 5271 automatically reloaded whenever its content changes. 5272 5273 As with the ``authz-simple`` object, the format of the identity 5274 strings being matched depends on the network service, but is 5275 usually a TLS x509 distinguished name, or a SASL username. 5276 5277 An example authorization object to validate a SASL username 5278 would look like: 5279 5280 .. parsed-literal:: 5281 5282 # |qemu_system| \\ 5283 ... \\ 5284 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\ 5285 ... 5286 5287 ``-object authz-pam,id=id,service=string`` 5288 Create an authorization object that will control access to 5289 network services. 5290 5291 The ``service`` parameter provides the name of a PAM service to 5292 use for authorization. It requires that a file 5293 ``/etc/pam.d/service`` exist to provide the configuration for 5294 the ``account`` subsystem. 5295 5296 An example authorization object to validate a TLS x509 5297 distinguished name would look like: 5298 5299 .. parsed-literal:: 5300 5301 # |qemu_system| \\ 5302 ... \\ 5303 -object authz-pam,id=auth0,service=qemu-vnc \\ 5304 ... 5305 5306 There would then be a corresponding config file for PAM at 5307 ``/etc/pam.d/qemu-vnc`` that contains: 5308 5309 :: 5310 5311 account requisite pam_listfile.so item=user sense=allow \ 5312 file=/etc/qemu/vnc.allow 5313 5314 Finally the ``/etc/qemu/vnc.allow`` file would contain the list 5315 of x509 distingished names that are permitted access 5316 5317 :: 5318 5319 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB 5320 5321 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch`` 5322 Creates a dedicated event loop thread that devices can be 5323 assigned to. This is known as an IOThread. By default device 5324 emulation happens in vCPU threads or the main event loop thread. 5325 This can become a scalability bottleneck. IOThreads allow device 5326 emulation and I/O to run on other host CPUs. 5327 5328 The ``id`` parameter is a unique ID that will be used to 5329 reference this IOThread from ``-device ...,iothread=id``. 5330 Multiple devices can be assigned to an IOThread. Note that not 5331 all devices support an ``iothread`` parameter. 5332 5333 The ``query-iothreads`` QMP command lists IOThreads and reports 5334 their thread IDs so that the user can configure host CPU 5335 pinning/affinity. 5336 5337 IOThreads use an adaptive polling algorithm to reduce event loop 5338 latency. Instead of entering a blocking system call to monitor 5339 file descriptors and then pay the cost of being woken up when an 5340 event occurs, the polling algorithm spins waiting for events for 5341 a short time. The algorithm's default parameters are suitable 5342 for many cases but can be adjusted based on knowledge of the 5343 workload and/or host device latency. 5344 5345 The ``poll-max-ns`` parameter is the maximum number of 5346 nanoseconds to busy wait for events. Polling can be disabled by 5347 setting this value to 0. 5348 5349 The ``poll-grow`` parameter is the multiplier used to increase 5350 the polling time when the algorithm detects it is missing events 5351 due to not polling long enough. 5352 5353 The ``poll-shrink`` parameter is the divisor used to decrease 5354 the polling time when the algorithm detects it is spending too 5355 long polling without encountering events. 5356 5357 The ``aio-max-batch`` parameter is the maximum number of requests 5358 in a batch for the AIO engine, 0 means that the engine will use 5359 its default. 5360 5361 The IOThread parameters can be modified at run-time using the 5362 ``qom-set`` command (where ``iothread1`` is the IOThread's 5363 ``id``): 5364 5365 :: 5366 5367 (qemu) qom-set /objects/iothread1 poll-max-ns 100000 5368ERST 5369 5370 5371HXCOMM This is the last statement. Insert new options before this line! 5372 5373#undef DEF 5374#undef DEFHEADING 5375#undef ARCHHEADING 5376
