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, 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 QEMU_ARCH_ALL) 40SRST 41``-machine [type=]name[,prop=value[,...]]`` 42 Select the emulated machine by name. Use ``-machine help`` to list 43 available machines. 44 45 For architectures which aim to support live migration compatibility 46 across releases, each release will introduce a new versioned machine 47 type. For example, the 2.8.0 release introduced machine types 48 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures. 49 50 To allow live migration of guests from QEMU version 2.8.0, to QEMU 51 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8" 52 and "pc-q35-2.8" machines too. To allow users live migrating VMs to 53 skip multiple intermediate releases when upgrading, new releases of 54 QEMU will support machine types from many previous versions. 55 56 Supported machine properties are: 57 58 ``accel=accels1[:accels2[:...]]`` 59 This is used to enable an accelerator. Depending on the target 60 architecture, kvm, xen, hax, hvf, whpx or tcg can be available. 61 By default, tcg is used. If there is more than one accelerator 62 specified, the next one is used if the previous one fails to 63 initialize. 64 65 ``vmport=on|off|auto`` 66 Enables emulation of VMWare IO port, for vmmouse etc. auto says 67 to select the value based on accel. For accel=xen the default is 68 off otherwise the default is on. 69 70 ``dump-guest-core=on|off`` 71 Include guest memory in a core dump. The default is on. 72 73 ``mem-merge=on|off`` 74 Enables or disables memory merge support. This feature, when 75 supported by the host, de-duplicates identical memory pages 76 among VMs instances (enabled by default). 77 78 ``aes-key-wrap=on|off`` 79 Enables or disables AES key wrapping support on s390-ccw hosts. 80 This feature controls whether AES wrapping keys will be created 81 to allow execution of AES cryptographic functions. The default 82 is on. 83 84 ``dea-key-wrap=on|off`` 85 Enables or disables DEA key wrapping support on s390-ccw hosts. 86 This feature controls whether DEA wrapping keys will be created 87 to allow execution of DEA cryptographic functions. The default 88 is on. 89 90 ``nvdimm=on|off`` 91 Enables or disables NVDIMM support. The default is off. 92 93 ``memory-encryption=`` 94 Memory encryption object to use. The default is none. 95 96 ``hmat=on|off`` 97 Enables or disables ACPI Heterogeneous Memory Attribute Table 98 (HMAT) support. The default is off. 99ERST 100 101HXCOMM Deprecated by -machine 102DEF("M", HAS_ARG, QEMU_OPTION_M, "", QEMU_ARCH_ALL) 103 104DEF("cpu", HAS_ARG, QEMU_OPTION_cpu, 105 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL) 106SRST 107``-cpu model`` 108 Select CPU model (``-cpu help`` for list and additional feature 109 selection) 110ERST 111 112DEF("accel", HAS_ARG, QEMU_OPTION_accel, 113 "-accel [accel=]accelerator[,prop[=value][,...]]\n" 114 " select accelerator (kvm, xen, hax, hvf, whpx or tcg; use 'help' for a list)\n" 115 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n" 116 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n" 117 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n" 118 " tb-size=n (TCG translation block cache size)\n" 119 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL) 120SRST 121``-accel name[,prop=value[,...]]`` 122 This is used to enable an accelerator. Depending on the target 123 architecture, kvm, xen, hax, hvf, whpx or tcg can be available. By 124 default, tcg is used. If there is more than one accelerator 125 specified, the next one is used if the previous one fails to 126 initialize. 127 128 ``igd-passthru=on|off`` 129 When Xen is in use, this option controls whether Intel 130 integrated graphics devices can be passed through to the guest 131 (default=off) 132 133 ``kernel-irqchip=on|off|split`` 134 Controls KVM in-kernel irqchip support. The default is full 135 acceleration of the interrupt controllers. On x86, split irqchip 136 reduces the kernel attack surface, at a performance cost for 137 non-MSI interrupts. Disabling the in-kernel irqchip completely 138 is not recommended except for debugging purposes. 139 140 ``kvm-shadow-mem=size`` 141 Defines the size of the KVM shadow MMU. 142 143 ``tb-size=n`` 144 Controls the size (in MiB) of the TCG translation block cache. 145 146 ``thread=single|multi`` 147 Controls number of TCG threads. When the TCG is multi-threaded 148 there will be one thread per vCPU therefor taking advantage of 149 additional host cores. The default is to enable multi-threading 150 where both the back-end and front-ends support it and no 151 incompatible TCG features have been enabled (e.g. 152 icount/replay). 153ERST 154 155DEF("smp", HAS_ARG, QEMU_OPTION_smp, 156 "-smp [cpus=]n[,maxcpus=cpus][,cores=cores][,threads=threads][,dies=dies][,sockets=sockets]\n" 157 " set the number of CPUs to 'n' [default=1]\n" 158 " maxcpus= maximum number of total cpus, including\n" 159 " offline CPUs for hotplug, etc\n" 160 " cores= number of CPU cores on one socket (for PC, it's on one die)\n" 161 " threads= number of threads on one CPU core\n" 162 " dies= number of CPU dies on one socket (for PC only)\n" 163 " sockets= number of discrete sockets in the system\n", 164 QEMU_ARCH_ALL) 165SRST 166``-smp [cpus=]n[,cores=cores][,threads=threads][,dies=dies][,sockets=sockets][,maxcpus=maxcpus]`` 167 Simulate an SMP system with n CPUs. On the PC target, up to 255 CPUs 168 are supported. On Sparc32 target, Linux limits the number of usable 169 CPUs to 4. For the PC target, the number of cores per die, the 170 number of threads per cores, the number of dies per packages and the 171 total number of sockets can be specified. Missing values will be 172 computed. If any on the three values is given, the total number of 173 CPUs n can be omitted. maxcpus specifies the maximum number of 174 hotpluggable CPUs. 175ERST 176 177DEF("numa", HAS_ARG, QEMU_OPTION_numa, 178 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 179 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" 180 "-numa dist,src=source,dst=destination,val=distance\n" 181 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n" 182 "-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" 183 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n", 184 QEMU_ARCH_ALL) 185SRST 186``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 187 \ 188``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` 189 \ 190``-numa dist,src=source,dst=destination,val=distance`` 191 \ 192``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]`` 193 \ 194``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=tpye[,latency=lat][,bandwidth=bw]`` 195 \ 196``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]`` 197 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA 198 distance from a source node to a destination node. Set the ACPI 199 Heterogeneous Memory Attributes for the given nodes. 200 201 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and 202 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a 203 contiguous range of CPU indexes (or a single VCPU if lastcpu is 204 omitted). A non-contiguous set of VCPUs can be represented by 205 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is 206 omitted on all nodes, VCPUs are automatically split between them. 207 208 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a 209 NUMA node: 210 211 :: 212 213 -numa node,cpus=0-2,cpus=5 214 215 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option 216 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to 217 assign CPU objects to a node using topology layout properties of 218 CPU. The set of properties is machine specific, and depends on used 219 machine type/'\ ``smp``\ ' options. It could be queried with 220 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ ' 221 property specifies node to which CPU object will be assigned, it's 222 required for node to be declared with '\ ``node``\ ' option before 223 it's used with '\ ``cpu``\ ' option. 224 225 For example: 226 227 :: 228 229 -M pc \ 230 -smp 1,sockets=2,maxcpus=2 \ 231 -numa node,nodeid=0 -numa node,nodeid=1 \ 232 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 233 234 Legacy '\ ``mem``\ ' assigns a given RAM amount to a node (not supported 235 for 5.1 and newer machine types). '\ ``memdev``\ ' assigns RAM from 236 a given memory backend device to a node. If '\ ``mem``\ ' and 237 '\ ``memdev``\ ' are omitted in all nodes, RAM is split equally between them. 238 239 240 '\ ``mem``\ ' and '\ ``memdev``\ ' are mutually exclusive. 241 Furthermore, if one node uses '\ ``memdev``\ ', all of them have to 242 use it. 243 244 '\ ``initiator``\ ' is an additional option that points to an 245 initiator NUMA node that has best performance (the lowest latency or 246 largest bandwidth) to this NUMA node. Note that this option can be 247 set only when the machine property 'hmat' is set to 'on'. 248 249 Following example creates a machine with 2 NUMA nodes, node 0 has 250 CPU. node 1 has only memory, and its initiator is node 0. Note that 251 because node 0 has CPU, by default the initiator of node 0 is itself 252 and must be itself. 253 254 :: 255 256 -machine hmat=on \ 257 -m 2G,slots=2,maxmem=4G \ 258 -object memory-backend-ram,size=1G,id=m0 \ 259 -object memory-backend-ram,size=1G,id=m1 \ 260 -numa node,nodeid=0,memdev=m0 \ 261 -numa node,nodeid=1,memdev=m1,initiator=0 \ 262 -smp 2,sockets=2,maxcpus=2 \ 263 -numa cpu,node-id=0,socket-id=0 \ 264 -numa cpu,node-id=0,socket-id=1 265 266 source and destination are NUMA node IDs. distance is the NUMA 267 distance from source to destination. The distance from a node to 268 itself is always 10. If any pair of nodes is given a distance, then 269 all pairs must be given distances. Although, when distances are only 270 given in one direction for each pair of nodes, then the distances in 271 the opposite directions are assumed to be the same. If, however, an 272 asymmetrical pair of distances is given for even one node pair, then 273 all node pairs must be provided distance values for both directions, 274 even when they are symmetrical. When a node is unreachable from 275 another node, set the pair's distance to 255. 276 277 Note that the -``numa`` option doesn't allocate any of the specified 278 resources, it just assigns existing resources to NUMA nodes. This 279 means that one still has to use the ``-m``, ``-smp`` options to 280 allocate RAM and VCPUs respectively. 281 282 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth 283 Information between initiator and target NUMA nodes in ACPI 284 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can 285 create memory requests, usually it has one or more processors. 286 Target NUMA node contains addressable memory. 287 288 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is 289 the memory hierarchy of the target NUMA node: if hierarchy is 290 'memory', the structure represents the memory performance; if 291 hierarchy is 'first-level\|second-level\|third-level', this 292 structure represents aggregated performance of memory side caches 293 for each domain. type of 'data-type' is type of data represented by 294 this structure instance: if 'hierarchy' is 'memory', 'data-type' is 295 'access\|read\|write' latency or 'access\|read\|write' bandwidth of 296 the target memory; if 'hierarchy' is 297 'first-level\|second-level\|third-level', 'data-type' is 298 'access\|read\|write' hit latency or 'access\|read\|write' hit 299 bandwidth of the target memory side cache. 300 301 lat is latency value in nanoseconds. bw is bandwidth value, the 302 possible value and units are NUM[M\|G\|T], mean that the bandwidth 303 value are NUM byte per second (or MB/s, GB/s or TB/s depending on 304 used suffix). Note that if latency or bandwidth value is 0, means 305 the corresponding latency or bandwidth information is not provided. 306 307 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory 308 belongs. size is the size of memory side cache in bytes. level is 309 the cache level described in this structure, note that the cache 310 level 0 should not be used with '\ ``hmat-cache``\ ' option. 311 associativity is the cache associativity, the possible value is 312 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy 313 is the write policy. line is the cache Line size in bytes. 314 315 For example, the following options describe 2 NUMA nodes. Node 0 has 316 2 cpus and a ram, node 1 has only a ram. The processors in node 0 317 access memory in node 0 with access-latency 5 nanoseconds, 318 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access 319 memory in NUMA node 1 with access-latency 10 nanoseconds, 320 access-bandwidth is 100 MB/s. And for memory side cache information, 321 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB, 322 policy is write-back, the cache Line size is 8 bytes: 323 324 :: 325 326 -machine hmat=on \ 327 -m 2G \ 328 -object memory-backend-ram,size=1G,id=m0 \ 329 -object memory-backend-ram,size=1G,id=m1 \ 330 -smp 2 \ 331 -numa node,nodeid=0,memdev=m0 \ 332 -numa node,nodeid=1,memdev=m1,initiator=0 \ 333 -numa cpu,node-id=0,socket-id=0 \ 334 -numa cpu,node-id=0,socket-id=1 \ 335 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \ 336 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \ 337 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \ 338 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \ 339 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \ 340 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8 341ERST 342 343DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd, 344 "-add-fd fd=fd,set=set[,opaque=opaque]\n" 345 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL) 346SRST 347``-add-fd fd=fd,set=set[,opaque=opaque]`` 348 Add a file descriptor to an fd set. Valid options are: 349 350 ``fd=fd`` 351 This option defines the file descriptor of which a duplicate is 352 added to fd set. The file descriptor cannot be stdin, stdout, or 353 stderr. 354 355 ``set=set`` 356 This option defines the ID of the fd set to add the file 357 descriptor to. 358 359 ``opaque=opaque`` 360 This option defines a free-form string that can be used to 361 describe fd. 362 363 You can open an image using pre-opened file descriptors from an fd 364 set: 365 366 .. parsed-literal:: 367 368 |qemu_system| \\ 369 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 370 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 371 -drive file=/dev/fdset/2,index=0,media=disk 372ERST 373 374DEF("set", HAS_ARG, QEMU_OPTION_set, 375 "-set group.id.arg=value\n" 376 " set <arg> parameter for item <id> of type <group>\n" 377 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL) 378SRST 379``-set group.id.arg=value`` 380 Set parameter arg for item id of type group 381ERST 382 383DEF("global", HAS_ARG, QEMU_OPTION_global, 384 "-global driver.property=value\n" 385 "-global driver=driver,property=property,value=value\n" 386 " set a global default for a driver property\n", 387 QEMU_ARCH_ALL) 388SRST 389``-global driver.prop=value`` 390 \ 391``-global driver=driver,property=property,value=value`` 392 Set default value of driver's property prop to value, e.g.: 393 394 .. parsed-literal:: 395 396 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img 397 398 In particular, you can use this to set driver properties for devices 399 which are created automatically by the machine model. To create a 400 device which is not created automatically and set properties on it, 401 use -``device``. 402 403 -global driver.prop=value is shorthand for -global 404 driver=driver,property=prop,value=value. The longhand syntax works 405 even when driver contains a dot. 406ERST 407 408DEF("boot", HAS_ARG, QEMU_OPTION_boot, 409 "-boot [order=drives][,once=drives][,menu=on|off]\n" 410 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n" 411 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n" 412 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n" 413 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n" 414 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n", 415 QEMU_ARCH_ALL) 416SRST 417``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]`` 418 Specify boot order drives as a string of drive letters. Valid drive 419 letters depend on the target architecture. The x86 PC uses: a, b 420 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p 421 (Etherboot from network adapter 1-4), hard disk boot is the default. 422 To apply a particular boot order only on the first startup, specify 423 it via ``once``. Note that the ``order`` or ``once`` parameter 424 should not be used together with the ``bootindex`` property of 425 devices, since the firmware implementations normally do not support 426 both at the same time. 427 428 Interactive boot menus/prompts can be enabled via ``menu=on`` as far 429 as firmware/BIOS supports them. The default is non-interactive boot. 430 431 A splash picture could be passed to bios, enabling user to show it 432 as logo, when option splash=sp\_name is given and menu=on, If 433 firmware/BIOS supports them. Currently Seabios for X86 system 434 support it. limitation: The splash file could be a jpeg file or a 435 BMP file in 24 BPP format(true color). The resolution should be 436 supported by the SVGA mode, so the recommended is 320x240, 640x480, 437 800x640. 438 439 A timeout could be passed to bios, guest will pause for rb\_timeout 440 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will 441 not reboot, qemu passes '-1' to bios by default. Currently Seabios 442 for X86 system support it. 443 444 Do strict boot via ``strict=on`` as far as firmware/BIOS supports 445 it. This only effects when boot priority is changed by bootindex 446 options. The default is non-strict boot. 447 448 .. parsed-literal:: 449 450 # try to boot from network first, then from hard disk 451 |qemu_system_x86| -boot order=nc 452 # boot from CD-ROM first, switch back to default order after reboot 453 |qemu_system_x86| -boot once=d 454 # boot with a splash picture for 5 seconds. 455 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000 456 457 Note: The legacy format '-boot drives' is still supported but its 458 use is discouraged as it may be removed from future versions. 459ERST 460 461DEF("m", HAS_ARG, QEMU_OPTION_m, 462 "-m [size=]megs[,slots=n,maxmem=size]\n" 463 " configure guest RAM\n" 464 " size: initial amount of guest memory\n" 465 " slots: number of hotplug slots (default: none)\n" 466 " maxmem: maximum amount of guest memory (default: none)\n" 467 "NOTE: Some architectures might enforce a specific granularity\n", 468 QEMU_ARCH_ALL) 469SRST 470``-m [size=]megs[,slots=n,maxmem=size]`` 471 Sets guest startup RAM size to megs megabytes. Default is 128 MiB. 472 Optionally, a suffix of "M" or "G" can be used to signify a value in 473 megabytes or gigabytes respectively. Optional pair slots, maxmem 474 could be used to set amount of hotpluggable memory slots and maximum 475 amount of memory. Note that maxmem must be aligned to the page size. 476 477 For example, the following command-line sets the guest startup RAM 478 size to 1GB, creates 3 slots to hotplug additional memory and sets 479 the maximum memory the guest can reach to 4GB: 480 481 .. parsed-literal:: 482 483 |qemu_system| -m 1G,slots=3,maxmem=4G 484 485 If slots and maxmem are not specified, memory hotplug won't be 486 enabled and the guest startup RAM will never increase. 487ERST 488 489DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath, 490 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL) 491SRST 492``-mem-path path`` 493 Allocate guest RAM from a temporarily created file in path. 494ERST 495 496DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc, 497 "-mem-prealloc preallocate guest memory (use with -mem-path)\n", 498 QEMU_ARCH_ALL) 499SRST 500``-mem-prealloc`` 501 Preallocate memory when using -mem-path. 502ERST 503 504DEF("k", HAS_ARG, QEMU_OPTION_k, 505 "-k language use keyboard layout (for example 'fr' for French)\n", 506 QEMU_ARCH_ALL) 507SRST 508``-k language`` 509 Use keyboard layout language (for example ``fr`` for French). This 510 option is only needed where it is not easy to get raw PC keycodes 511 (e.g. on Macs, with some X11 servers or with a VNC or curses 512 display). You don't normally need to use it on PC/Linux or 513 PC/Windows hosts. 514 515 The available layouts are: 516 517 :: 518 519 ar de-ch es fo fr-ca hu ja mk no pt-br sv 520 da en-gb et fr fr-ch is lt nl pl ru th 521 de en-us fi fr-be hr it lv nl-be pt sl tr 522 523 The default is ``en-us``. 524ERST 525 526 527HXCOMM Deprecated by -audiodev 528DEF("audio-help", 0, QEMU_OPTION_audio_help, 529 "-audio-help show -audiodev equivalent of the currently specified audio settings\n", 530 QEMU_ARCH_ALL) 531SRST 532``-audio-help`` 533 Will show the -audiodev equivalent of the currently specified 534 (deprecated) environment variables. 535ERST 536 537DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, 538 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n" 539 " specifies the audio backend to use\n" 540 " id= identifier of the backend\n" 541 " timer-period= timer period in microseconds\n" 542 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n" 543 " in|out.fixed-settings= use fixed settings for host audio\n" 544 " in|out.frequency= frequency to use with fixed settings\n" 545 " in|out.channels= number of channels to use with fixed settings\n" 546 " in|out.format= sample format to use with fixed settings\n" 547 " valid values: s8, s16, s32, u8, u16, u32, f32\n" 548 " in|out.voices= number of voices to use\n" 549 " in|out.buffer-length= length of buffer in microseconds\n" 550 "-audiodev none,id=id,[,prop[=value][,...]]\n" 551 " dummy driver that discards all output\n" 552#ifdef CONFIG_AUDIO_ALSA 553 "-audiodev alsa,id=id[,prop[=value][,...]]\n" 554 " in|out.dev= name of the audio device to use\n" 555 " in|out.period-length= length of period in microseconds\n" 556 " in|out.try-poll= attempt to use poll mode\n" 557 " threshold= threshold (in microseconds) when playback starts\n" 558#endif 559#ifdef CONFIG_AUDIO_COREAUDIO 560 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n" 561 " in|out.buffer-count= number of buffers\n" 562#endif 563#ifdef CONFIG_AUDIO_DSOUND 564 "-audiodev dsound,id=id[,prop[=value][,...]]\n" 565 " latency= add extra latency to playback in microseconds\n" 566#endif 567#ifdef CONFIG_AUDIO_OSS 568 "-audiodev oss,id=id[,prop[=value][,...]]\n" 569 " in|out.dev= path of the audio device to use\n" 570 " in|out.buffer-count= number of buffers\n" 571 " in|out.try-poll= attempt to use poll mode\n" 572 " try-mmap= try using memory mapped access\n" 573 " exclusive= open device in exclusive mode\n" 574 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n" 575#endif 576#ifdef CONFIG_AUDIO_PA 577 "-audiodev pa,id=id[,prop[=value][,...]]\n" 578 " server= PulseAudio server address\n" 579 " in|out.name= source/sink device name\n" 580 " in|out.latency= desired latency in microseconds\n" 581#endif 582#ifdef CONFIG_AUDIO_SDL 583 "-audiodev sdl,id=id[,prop[=value][,...]]\n" 584#endif 585#ifdef CONFIG_SPICE 586 "-audiodev spice,id=id[,prop[=value][,...]]\n" 587#endif 588 "-audiodev wav,id=id[,prop[=value][,...]]\n" 589 " path= path of wav file to record\n", 590 QEMU_ARCH_ALL) 591SRST 592``-audiodev [driver=]driver,id=id[,prop[=value][,...]]`` 593 Adds a new audio backend driver identified by id. There are global 594 and driver specific properties. Some values can be set differently 595 for input and output, they're marked with ``in|out.``. You can set 596 the input's property with ``in.prop`` and the output's property with 597 ``out.prop``. For example: 598 599 :: 600 601 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000 602 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified 603 604 NOTE: parameter validation is known to be incomplete, in many cases 605 specifying an invalid option causes QEMU to print an error message 606 and continue emulation without sound. 607 608 Valid global options are: 609 610 ``id=identifier`` 611 Identifies the audio backend. 612 613 ``timer-period=period`` 614 Sets the timer period used by the audio subsystem in 615 microseconds. Default is 10000 (10 ms). 616 617 ``in|out.mixing-engine=on|off`` 618 Use QEMU's mixing engine to mix all streams inside QEMU and 619 convert audio formats when not supported by the backend. When 620 off, fixed-settings must be off too. Note that disabling this 621 option means that the selected backend must support multiple 622 streams and the audio formats used by the virtual cards, 623 otherwise you'll get no sound. It's not recommended to disable 624 this option unless you want to use 5.1 or 7.1 audio, as mixing 625 engine only supports mono and stereo audio. Default is on. 626 627 ``in|out.fixed-settings=on|off`` 628 Use fixed settings for host audio. When off, it will change 629 based on how the guest opens the sound card. In this case you 630 must not specify frequency, channels or format. Default is on. 631 632 ``in|out.frequency=frequency`` 633 Specify the frequency to use when using fixed-settings. Default 634 is 44100Hz. 635 636 ``in|out.channels=channels`` 637 Specify the number of channels to use when using fixed-settings. 638 Default is 2 (stereo). 639 640 ``in|out.format=format`` 641 Specify the sample format to use when using fixed-settings. 642 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``, 643 ``u32``, ``f32``. Default is ``s16``. 644 645 ``in|out.voices=voices`` 646 Specify the number of voices to use. Default is 1. 647 648 ``in|out.buffer-length=usecs`` 649 Sets the size of the buffer in microseconds. 650 651``-audiodev none,id=id[,prop[=value][,...]]`` 652 Creates a dummy backend that discards all outputs. This backend has 653 no backend specific properties. 654 655``-audiodev alsa,id=id[,prop[=value][,...]]`` 656 Creates backend using the ALSA. This backend is only available on 657 Linux. 658 659 ALSA specific options are: 660 661 ``in|out.dev=device`` 662 Specify the ALSA device to use for input and/or output. Default 663 is ``default``. 664 665 ``in|out.period-length=usecs`` 666 Sets the period length in microseconds. 667 668 ``in|out.try-poll=on|off`` 669 Attempt to use poll mode with the device. Default is on. 670 671 ``threshold=threshold`` 672 Threshold (in microseconds) when playback starts. Default is 0. 673 674``-audiodev coreaudio,id=id[,prop[=value][,...]]`` 675 Creates a backend using Apple's Core Audio. This backend is only 676 available on Mac OS and only supports playback. 677 678 Core Audio specific options are: 679 680 ``in|out.buffer-count=count`` 681 Sets the count of the buffers. 682 683``-audiodev dsound,id=id[,prop[=value][,...]]`` 684 Creates a backend using Microsoft's DirectSound. This backend is 685 only available on Windows and only supports playback. 686 687 DirectSound specific options are: 688 689 ``latency=usecs`` 690 Add extra usecs microseconds latency to playback. Default is 691 10000 (10 ms). 692 693``-audiodev oss,id=id[,prop[=value][,...]]`` 694 Creates a backend using OSS. This backend is available on most 695 Unix-like systems. 696 697 OSS specific options are: 698 699 ``in|out.dev=device`` 700 Specify the file name of the OSS device to use. Default is 701 ``/dev/dsp``. 702 703 ``in|out.buffer-count=count`` 704 Sets the count of the buffers. 705 706 ``in|out.try-poll=on|of`` 707 Attempt to use poll mode with the device. Default is on. 708 709 ``try-mmap=on|off`` 710 Try using memory mapped device access. Default is off. 711 712 ``exclusive=on|off`` 713 Open the device in exclusive mode (vmix won't work in this 714 case). Default is off. 715 716 ``dsp-policy=policy`` 717 Sets the timing policy (between 0 and 10, where smaller number 718 means smaller latency but higher CPU usage). Use -1 to use 719 buffer sizes specified by ``buffer`` and ``buffer-count``. This 720 option is ignored if you do not have OSS 4. Default is 5. 721 722``-audiodev pa,id=id[,prop[=value][,...]]`` 723 Creates a backend using PulseAudio. This backend is available on 724 most systems. 725 726 PulseAudio specific options are: 727 728 ``server=server`` 729 Sets the PulseAudio server to connect to. 730 731 ``in|out.name=sink`` 732 Use the specified source/sink for recording/playback. 733 734 ``in|out.latency=usecs`` 735 Desired latency in microseconds. The PulseAudio server will try 736 to honor this value but actual latencies may be lower or higher. 737 738``-audiodev sdl,id=id[,prop[=value][,...]]`` 739 Creates a backend using SDL. This backend is available on most 740 systems, but you should use your platform's native backend if 741 possible. This backend has no backend specific properties. 742 743``-audiodev spice,id=id[,prop[=value][,...]]`` 744 Creates a backend that sends audio through SPICE. This backend 745 requires ``-spice`` and automatically selected in that case, so 746 usually you can ignore this option. This backend has no backend 747 specific properties. 748 749``-audiodev wav,id=id[,prop[=value][,...]]`` 750 Creates a backend that writes audio to a WAV file. 751 752 Backend specific options are: 753 754 ``path=path`` 755 Write recorded audio into the specified file. Default is 756 ``qemu.wav``. 757ERST 758 759DEF("soundhw", HAS_ARG, QEMU_OPTION_soundhw, 760 "-soundhw c1,... enable audio support\n" 761 " and only specified sound cards (comma separated list)\n" 762 " use '-soundhw help' to get the list of supported cards\n" 763 " use '-soundhw all' to enable all of them\n", QEMU_ARCH_ALL) 764SRST 765``-soundhw card1[,card2,...] or -soundhw all`` 766 Enable audio and selected sound hardware. Use 'help' to print all 767 available sound hardware. For example: 768 769 .. parsed-literal:: 770 771 |qemu_system_x86| -soundhw sb16,adlib disk.img 772 |qemu_system_x86| -soundhw es1370 disk.img 773 |qemu_system_x86| -soundhw ac97 disk.img 774 |qemu_system_x86| -soundhw hda disk.img 775 |qemu_system_x86| -soundhw all disk.img 776 |qemu_system_x86| -soundhw help 777 778 Note that Linux's i810\_audio OSS kernel (for AC97) module might 779 require manually specifying clocking. 780 781 :: 782 783 modprobe i810_audio clocking=48000 784ERST 785 786DEF("device", HAS_ARG, QEMU_OPTION_device, 787 "-device driver[,prop[=value][,...]]\n" 788 " add device (based on driver)\n" 789 " prop=value,... sets driver properties\n" 790 " use '-device help' to print all possible drivers\n" 791 " use '-device driver,help' to print all possible properties\n", 792 QEMU_ARCH_ALL) 793SRST 794``-device driver[,prop[=value][,...]]`` 795 Add device driver. prop=value sets driver properties. Valid 796 properties depend on the driver. To get help on possible drivers and 797 properties, use ``-device help`` and ``-device driver,help``. 798 799 Some drivers are: 800 801``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]`` 802 Add an IPMI BMC. This is a simulation of a hardware management 803 interface processor that normally sits on a system. It provides a 804 watchdog and the ability to reset and power control the system. You 805 need to connect this to an IPMI interface to make it useful 806 807 The IPMI slave address to use for the BMC. The default is 0x20. This 808 address is the BMC's address on the I2C network of management 809 controllers. If you don't know what this means, it is safe to ignore 810 it. 811 812 ``id=id`` 813 The BMC id for interfaces to use this device. 814 815 ``slave_addr=val`` 816 Define slave address to use for the BMC. The default is 0x20. 817 818 ``sdrfile=file`` 819 file containing raw Sensor Data Records (SDR) data. The default 820 is none. 821 822 ``fruareasize=val`` 823 size of a Field Replaceable Unit (FRU) area. The default is 824 1024. 825 826 ``frudatafile=file`` 827 file containing raw Field Replaceable Unit (FRU) inventory data. 828 The default is none. 829 830 ``guid=uuid`` 831 value for the GUID for the BMC, in standard UUID format. If this 832 is set, get "Get GUID" command to the BMC will return it. 833 Otherwise "Get GUID" will return an error. 834 835``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]`` 836 Add a connection to an external IPMI BMC simulator. Instead of 837 locally emulating the BMC like the above item, instead connect to an 838 external entity that provides the IPMI services. 839 840 A connection is made to an external BMC simulator. If you do this, 841 it is strongly recommended that you use the "reconnect=" chardev 842 option to reconnect to the simulator if the connection is lost. Note 843 that if this is not used carefully, it can be a security issue, as 844 the interface has the ability to send resets, NMIs, and power off 845 the VM. It's best if QEMU makes a connection to an external 846 simulator running on a secure port on localhost, so neither the 847 simulator nor QEMU is exposed to any outside network. 848 849 See the "lanserv/README.vm" file in the OpenIPMI library for more 850 details on the external interface. 851 852``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]`` 853 Add a KCS IPMI interafce on the ISA bus. This also adds a 854 corresponding ACPI and SMBIOS entries, if appropriate. 855 856 ``bmc=id`` 857 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern 858 above. 859 860 ``ioport=val`` 861 Define the I/O address of the interface. The default is 0xca0 862 for KCS. 863 864 ``irq=val`` 865 Define the interrupt to use. The default is 5. To disable 866 interrupts, set this to 0. 867 868``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]`` 869 Like the KCS interface, but defines a BT interface. The default port 870 is 0xe4 and the default interrupt is 5. 871 872``-device pci-ipmi-kcs,bmc=id`` 873 Add a KCS IPMI interafce on the PCI bus. 874 875 ``bmc=id`` 876 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above. 877 878``-device pci-ipmi-bt,bmc=id`` 879 Like the KCS interface, but defines a BT interface on the PCI bus. 880ERST 881 882DEF("name", HAS_ARG, QEMU_OPTION_name, 883 "-name string1[,process=string2][,debug-threads=on|off]\n" 884 " set the name of the guest\n" 885 " string1 sets the window title and string2 the process name\n" 886 " When debug-threads is enabled, individual threads are given a separate name\n" 887 " NOTE: The thread names are for debugging and not a stable API.\n", 888 QEMU_ARCH_ALL) 889SRST 890``-name name`` 891 Sets the name of the guest. This name will be displayed in the SDL 892 window caption. The name will also be used for the VNC server. Also 893 optionally set the top visible process name in Linux. Naming of 894 individual threads can also be enabled on Linux to aid debugging. 895ERST 896 897DEF("uuid", HAS_ARG, QEMU_OPTION_uuid, 898 "-uuid %08x-%04x-%04x-%04x-%012x\n" 899 " specify machine UUID\n", QEMU_ARCH_ALL) 900SRST 901``-uuid uuid`` 902 Set system UUID. 903ERST 904 905DEFHEADING() 906 907DEFHEADING(Block device options:) 908 909DEF("fda", HAS_ARG, QEMU_OPTION_fda, 910 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL) 911DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL) 912SRST 913``-fda file`` 914 \ 915``-fdb file`` 916 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in 917 the System Emulation Users Guide). 918ERST 919 920DEF("hda", HAS_ARG, QEMU_OPTION_hda, 921 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL) 922DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL) 923DEF("hdc", HAS_ARG, QEMU_OPTION_hdc, 924 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL) 925DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL) 926SRST 927``-hda file`` 928 \ 929``-hdb file`` 930 \ 931``-hdc file`` 932 \ 933``-hdd file`` 934 Use file as hard disk 0, 1, 2 or 3 image (see the :ref:`disk images` 935 chapter in the System Emulation Users Guide). 936ERST 937 938DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom, 939 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n", 940 QEMU_ARCH_ALL) 941SRST 942``-cdrom file`` 943 Use file as CD-ROM image (you cannot use ``-hdc`` and ``-cdrom`` at 944 the same time). You can use the host CD-ROM by using ``/dev/cdrom`` 945 as filename. 946ERST 947 948DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev, 949 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n" 950 " [,cache.direct=on|off][,cache.no-flush=on|off]\n" 951 " [,read-only=on|off][,auto-read-only=on|off]\n" 952 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n" 953 " [,driver specific parameters...]\n" 954 " configure a block backend\n", QEMU_ARCH_ALL) 955SRST 956``-blockdev option[,option[,option[,...]]]`` 957 Define a new block driver node. Some of the options apply to all 958 block drivers, other options are only accepted for a specific block 959 driver. See below for a list of generic options and options for the 960 most common block drivers. 961 962 Options that expect a reference to another node (e.g. ``file``) can 963 be given in two ways. Either you specify the node name of an already 964 existing node (file=node-name), or you define a new node inline, 965 adding options for the referenced node after a dot 966 (file.filename=path,file.aio=native). 967 968 A block driver node created with ``-blockdev`` can be used for a 969 guest device by specifying its node name for the ``drive`` property 970 in a ``-device`` argument that defines a block device. 971 972 ``Valid options for any block driver node:`` 973 ``driver`` 974 Specifies the block driver to use for the given node. 975 976 ``node-name`` 977 This defines the name of the block driver node by which it 978 will be referenced later. The name must be unique, i.e. it 979 must not match the name of a different block driver node, or 980 (if you use ``-drive`` as well) the ID of a drive. 981 982 If no node name is specified, it is automatically generated. 983 The generated node name is not intended to be predictable 984 and changes between QEMU invocations. For the top level, an 985 explicit node name must be specified. 986 987 ``read-only`` 988 Open the node read-only. Guest write attempts will fail. 989 990 Note that some block drivers support only read-only access, 991 either generally or in certain configurations. In this case, 992 the default value ``read-only=off`` does not work and the 993 option must be specified explicitly. 994 995 ``auto-read-only`` 996 If ``auto-read-only=on`` is set, QEMU may fall back to 997 read-only usage even when ``read-only=off`` is requested, or 998 even switch between modes as needed, e.g. depending on 999 whether the image file is writable or whether a writing user 1000 is attached to the node. 1001 1002 ``force-share`` 1003 Override the image locking system of QEMU by forcing the 1004 node to utilize weaker shared access for permissions where 1005 it would normally request exclusive access. When there is 1006 the potential for multiple instances to have the same file 1007 open (whether this invocation of QEMU is the first or the 1008 second instance), both instances must permit shared access 1009 for the second instance to succeed at opening the file. 1010 1011 Enabling ``force-share=on`` requires ``read-only=on``. 1012 1013 ``cache.direct`` 1014 The host page cache can be avoided with ``cache.direct=on``. 1015 This will attempt to do disk IO directly to the guest's 1016 memory. QEMU may still perform an internal copy of the data. 1017 1018 ``cache.no-flush`` 1019 In case you don't care about data integrity over host 1020 failures, you can use ``cache.no-flush=on``. This option 1021 tells QEMU that it never needs to write any data to the disk 1022 but can instead keep things in cache. If anything goes 1023 wrong, like your host losing power, the disk storage getting 1024 disconnected accidentally, etc. your image will most 1025 probably be rendered unusable. 1026 1027 ``discard=discard`` 1028 discard is one of "ignore" (or "off") or "unmap" (or "on") 1029 and controls whether ``discard`` (also known as ``trim`` or 1030 ``unmap``) requests are ignored or passed to the filesystem. 1031 Some machine types may not support discard requests. 1032 1033 ``detect-zeroes=detect-zeroes`` 1034 detect-zeroes is "off", "on" or "unmap" and enables the 1035 automatic conversion of plain zero writes by the OS to 1036 driver specific optimized zero write commands. You may even 1037 choose "unmap" if discard is set to "unmap" to allow a zero 1038 write to be converted to an ``unmap`` operation. 1039 1040 ``Driver-specific options for file`` 1041 This is the protocol-level block driver for accessing regular 1042 files. 1043 1044 ``filename`` 1045 The path to the image file in the local filesystem 1046 1047 ``aio`` 1048 Specifies the AIO backend (threads/native/io_uring, 1049 default: threads) 1050 1051 ``locking`` 1052 Specifies whether the image file is protected with Linux OFD 1053 / POSIX locks. The default is to use the Linux Open File 1054 Descriptor API if available, otherwise no lock is applied. 1055 (auto/on/off, default: auto) 1056 1057 Example: 1058 1059 :: 1060 1061 -blockdev driver=file,node-name=disk,filename=disk.img 1062 1063 ``Driver-specific options for raw`` 1064 This is the image format block driver for raw images. It is 1065 usually stacked on top of a protocol level block driver such as 1066 ``file``. 1067 1068 ``file`` 1069 Reference to or definition of the data source block driver 1070 node (e.g. a ``file`` driver node) 1071 1072 Example 1: 1073 1074 :: 1075 1076 -blockdev driver=file,node-name=disk_file,filename=disk.img 1077 -blockdev driver=raw,node-name=disk,file=disk_file 1078 1079 Example 2: 1080 1081 :: 1082 1083 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img 1084 1085 ``Driver-specific options for qcow2`` 1086 This is the image format block driver for qcow2 images. It is 1087 usually stacked on top of a protocol level block driver such as 1088 ``file``. 1089 1090 ``file`` 1091 Reference to or definition of the data source block driver 1092 node (e.g. a ``file`` driver node) 1093 1094 ``backing`` 1095 Reference to or definition of the backing file block device 1096 (default is taken from the image file). It is allowed to 1097 pass ``null`` here in order to disable the default backing 1098 file. 1099 1100 ``lazy-refcounts`` 1101 Whether to enable the lazy refcounts feature (on/off; 1102 default is taken from the image file) 1103 1104 ``cache-size`` 1105 The maximum total size of the L2 table and refcount block 1106 caches in bytes (default: the sum of l2-cache-size and 1107 refcount-cache-size) 1108 1109 ``l2-cache-size`` 1110 The maximum size of the L2 table cache in bytes (default: if 1111 cache-size is not specified - 32M on Linux platforms, and 8M 1112 on non-Linux platforms; otherwise, as large as possible 1113 within the cache-size, while permitting the requested or the 1114 minimal refcount cache size) 1115 1116 ``refcount-cache-size`` 1117 The maximum size of the refcount block cache in bytes 1118 (default: 4 times the cluster size; or if cache-size is 1119 specified, the part of it which is not used for the L2 1120 cache) 1121 1122 ``cache-clean-interval`` 1123 Clean unused entries in the L2 and refcount caches. The 1124 interval is in seconds. The default value is 600 on 1125 supporting platforms, and 0 on other platforms. Setting it 1126 to 0 disables this feature. 1127 1128 ``pass-discard-request`` 1129 Whether discard requests to the qcow2 device should be 1130 forwarded to the data source (on/off; default: on if 1131 discard=unmap is specified, off otherwise) 1132 1133 ``pass-discard-snapshot`` 1134 Whether discard requests for the data source should be 1135 issued when a snapshot operation (e.g. deleting a snapshot) 1136 frees clusters in the qcow2 file (on/off; default: on) 1137 1138 ``pass-discard-other`` 1139 Whether discard requests for the data source should be 1140 issued on other occasions where a cluster gets freed 1141 (on/off; default: off) 1142 1143 ``overlap-check`` 1144 Which overlap checks to perform for writes to the image 1145 (none/constant/cached/all; default: cached). For details or 1146 finer granularity control refer to the QAPI documentation of 1147 ``blockdev-add``. 1148 1149 Example 1: 1150 1151 :: 1152 1153 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2 1154 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216 1155 1156 Example 2: 1157 1158 :: 1159 1160 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2 1161 1162 ``Driver-specific options for other drivers`` 1163 Please refer to the QAPI documentation of the ``blockdev-add`` 1164 QMP command. 1165ERST 1166 1167DEF("drive", HAS_ARG, QEMU_OPTION_drive, 1168 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n" 1169 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n" 1170 " [,snapshot=on|off][,rerror=ignore|stop|report]\n" 1171 " [,werror=ignore|stop|report|enospc][,id=name]\n" 1172 " [,aio=threads|native|io_uring]\n" 1173 " [,readonly=on|off][,copy-on-read=on|off]\n" 1174 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n" 1175 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n" 1176 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n" 1177 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n" 1178 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n" 1179 " [[,iops_size=is]]\n" 1180 " [[,group=g]]\n" 1181 " use 'file' as a drive image\n", QEMU_ARCH_ALL) 1182SRST 1183``-drive option[,option[,option[,...]]]`` 1184 Define a new drive. This includes creating a block driver node (the 1185 backend) as well as a guest device, and is mostly a shortcut for 1186 defining the corresponding ``-blockdev`` and ``-device`` options. 1187 1188 ``-drive`` accepts all options that are accepted by ``-blockdev``. 1189 In addition, it knows the following options: 1190 1191 ``file=file`` 1192 This option defines which disk image (see the :ref:`disk images` 1193 chapter in the System Emulation Users Guide) to use with this drive. 1194 If the filename contains comma, you must double it (for instance, 1195 "file=my,,file" to use file "my,file"). 1196 1197 Special files such as iSCSI devices can be specified using 1198 protocol specific URLs. See the section for "Device URL Syntax" 1199 for more information. 1200 1201 ``if=interface`` 1202 This option defines on which type on interface the drive is 1203 connected. Available types are: ide, scsi, sd, mtd, floppy, 1204 pflash, virtio, none. 1205 1206 ``bus=bus,unit=unit`` 1207 These options define where is connected the drive by defining 1208 the bus number and the unit id. 1209 1210 ``index=index`` 1211 This option defines where is connected the drive by using an 1212 index in the list of available connectors of a given interface 1213 type. 1214 1215 ``media=media`` 1216 This option defines the type of the media: disk or cdrom. 1217 1218 ``snapshot=snapshot`` 1219 snapshot is "on" or "off" and controls snapshot mode for the 1220 given drive (see ``-snapshot``). 1221 1222 ``cache=cache`` 1223 cache is "none", "writeback", "unsafe", "directsync" or 1224 "writethrough" and controls how the host cache is used to access 1225 block data. This is a shortcut that sets the ``cache.direct`` 1226 and ``cache.no-flush`` options (as in ``-blockdev``), and 1227 additionally ``cache.writeback``, which provides a default for 1228 the ``write-cache`` option of block guest devices (as in 1229 ``-device``). The modes correspond to the following settings: 1230 1231 ============= =============== ============ ============== 1232 \ cache.writeback cache.direct cache.no-flush 1233 ============= =============== ============ ============== 1234 writeback on off off 1235 none on on off 1236 writethrough off off off 1237 directsync off on off 1238 unsafe on off on 1239 ============= =============== ============ ============== 1240 1241 The default mode is ``cache=writeback``. 1242 1243 ``aio=aio`` 1244 aio is "threads", "native", or "io_uring" and selects between pthread 1245 based disk I/O, native Linux AIO, or Linux io_uring API. 1246 1247 ``format=format`` 1248 Specify which disk format will be used rather than detecting the 1249 format. Can be used to specify format=raw to avoid interpreting 1250 an untrusted format header. 1251 1252 ``werror=action,rerror=action`` 1253 Specify which action to take on write and read errors. Valid 1254 actions are: "ignore" (ignore the error and try to continue), 1255 "stop" (pause QEMU), "report" (report the error to the guest), 1256 "enospc" (pause QEMU only if the host disk is full; report the 1257 error to the guest otherwise). The default setting is 1258 ``werror=enospc`` and ``rerror=report``. 1259 1260 ``copy-on-read=copy-on-read`` 1261 copy-on-read is "on" or "off" and enables whether to copy read 1262 backing file sectors into the image file. 1263 1264 ``bps=b,bps_rd=r,bps_wr=w`` 1265 Specify bandwidth throttling limits in bytes per second, either 1266 for all request types or for reads or writes only. Small values 1267 can lead to timeouts or hangs inside the guest. A safe minimum 1268 for disks is 2 MB/s. 1269 1270 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm`` 1271 Specify bursts in bytes per second, either for all request types 1272 or for reads or writes only. Bursts allow the guest I/O to spike 1273 above the limit temporarily. 1274 1275 ``iops=i,iops_rd=r,iops_wr=w`` 1276 Specify request rate limits in requests per second, either for 1277 all request types or for reads or writes only. 1278 1279 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm`` 1280 Specify bursts in requests per second, either for all request 1281 types or for reads or writes only. Bursts allow the guest I/O to 1282 spike above the limit temporarily. 1283 1284 ``iops_size=is`` 1285 Let every is bytes of a request count as a new request for iops 1286 throttling purposes. Use this option to prevent guests from 1287 circumventing iops limits by sending fewer but larger requests. 1288 1289 ``group=g`` 1290 Join a throttling quota group with given name g. All drives that 1291 are members of the same group are accounted for together. Use 1292 this option to prevent guests from circumventing throttling 1293 limits by using many small disks instead of a single larger 1294 disk. 1295 1296 By default, the ``cache.writeback=on`` mode is used. It will report 1297 data writes as completed as soon as the data is present in the host 1298 page cache. This is safe as long as your guest OS makes sure to 1299 correctly flush disk caches where needed. If your guest OS does not 1300 handle volatile disk write caches correctly and your host crashes or 1301 loses power, then the guest may experience data corruption. 1302 1303 For such guests, you should consider using ``cache.writeback=off``. 1304 This means that the host page cache will be used to read and write 1305 data, but write notification will be sent to the guest only after 1306 QEMU has made sure to flush each write to the disk. Be aware that 1307 this has a major impact on performance. 1308 1309 When using the ``-snapshot`` option, unsafe caching is always used. 1310 1311 Copy-on-read avoids accessing the same backing file sectors 1312 repeatedly and is useful when the backing file is over a slow 1313 network. By default copy-on-read is off. 1314 1315 Instead of ``-cdrom`` you can use: 1316 1317 .. parsed-literal:: 1318 1319 |qemu_system| -drive file=file,index=2,media=cdrom 1320 1321 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use: 1322 1323 .. parsed-literal:: 1324 1325 |qemu_system| -drive file=file,index=0,media=disk 1326 |qemu_system| -drive file=file,index=1,media=disk 1327 |qemu_system| -drive file=file,index=2,media=disk 1328 |qemu_system| -drive file=file,index=3,media=disk 1329 1330 You can open an image using pre-opened file descriptors from an fd 1331 set: 1332 1333 .. parsed-literal:: 1334 1335 |qemu_system| \\ 1336 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ 1337 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ 1338 -drive file=/dev/fdset/2,index=0,media=disk 1339 1340 You can connect a CDROM to the slave of ide0: 1341 1342 .. parsed-literal:: 1343 1344 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom 1345 1346 If you don't specify the "file=" argument, you define an empty 1347 drive: 1348 1349 .. parsed-literal:: 1350 1351 |qemu_system_x86| -drive if=ide,index=1,media=cdrom 1352 1353 Instead of ``-fda``, ``-fdb``, you can use: 1354 1355 .. parsed-literal:: 1356 1357 |qemu_system_x86| -drive file=file,index=0,if=floppy 1358 |qemu_system_x86| -drive file=file,index=1,if=floppy 1359 1360 By default, interface is "ide" and index is automatically 1361 incremented: 1362 1363 .. parsed-literal:: 1364 1365 |qemu_system_x86| -drive file=a -drive file=b" 1366 1367 is interpreted like: 1368 1369 .. parsed-literal:: 1370 1371 |qemu_system_x86| -hda a -hdb b 1372ERST 1373 1374DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock, 1375 "-mtdblock file use 'file' as on-board Flash memory image\n", 1376 QEMU_ARCH_ALL) 1377SRST 1378``-mtdblock file`` 1379 Use file as on-board Flash memory image. 1380ERST 1381 1382DEF("sd", HAS_ARG, QEMU_OPTION_sd, 1383 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL) 1384SRST 1385``-sd file`` 1386 Use file as SecureDigital card image. 1387ERST 1388 1389DEF("pflash", HAS_ARG, QEMU_OPTION_pflash, 1390 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL) 1391SRST 1392``-pflash file`` 1393 Use file as a parallel flash image. 1394ERST 1395 1396DEF("snapshot", 0, QEMU_OPTION_snapshot, 1397 "-snapshot write to temporary files instead of disk image files\n", 1398 QEMU_ARCH_ALL) 1399SRST 1400``-snapshot`` 1401 Write to temporary files instead of disk image files. In this case, 1402 the raw disk image you use is not written back. You can however 1403 force the write back by pressing C-a s (see the :ref:`disk images` 1404 chapter in the System Emulation Users Guide). 1405ERST 1406 1407DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev, 1408 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1409 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n" 1410 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n" 1411 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n" 1412 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n" 1413 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n" 1414 " [[,throttling.iops-size=is]]\n" 1415 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n" 1416 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n" 1417 "-fsdev synth,id=id\n", 1418 QEMU_ARCH_ALL) 1419 1420SRST 1421``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]`` 1422 \ 1423``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]`` 1424 \ 1425``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]`` 1426 \ 1427``-fsdev synth,id=id[,readonly=on]`` 1428 Define a new file system device. Valid options are: 1429 1430 ``local`` 1431 Accesses to the filesystem are done by QEMU. 1432 1433 ``proxy`` 1434 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1435 1436 ``synth`` 1437 Synthetic filesystem, only used by QTests. 1438 1439 ``id=id`` 1440 Specifies identifier for this device. 1441 1442 ``path=path`` 1443 Specifies the export path for the file system device. Files 1444 under this path will be available to the 9p client on the guest. 1445 1446 ``security_model=security_model`` 1447 Specifies the security model to be used for this export path. 1448 Supported security models are "passthrough", "mapped-xattr", 1449 "mapped-file" and "none". In "passthrough" security model, files 1450 are stored using the same credentials as they are created on the 1451 guest. This requires QEMU to run as root. In "mapped-xattr" 1452 security model, some of the file attributes like uid, gid, mode 1453 bits and link target are stored as file attributes. For 1454 "mapped-file" these attributes are stored in the hidden 1455 .virtfs\_metadata directory. Directories exported by this 1456 security model cannot interact with other unix tools. "none" 1457 security model is same as passthrough except the sever won't 1458 report failures if it fails to set file attributes like 1459 ownership. Security model is mandatory only for local fsdriver. 1460 Other fsdrivers (like proxy) don't take security model as a 1461 parameter. 1462 1463 ``writeout=writeout`` 1464 This is an optional argument. The only supported value is 1465 "immediate". This means that host page cache will be used to 1466 read and write data but write notification will be sent to the 1467 guest only when the data has been reported as written by the 1468 storage subsystem. 1469 1470 ``readonly=on`` 1471 Enables exporting 9p share as a readonly mount for guests. By 1472 default read-write access is given. 1473 1474 ``socket=socket`` 1475 Enables proxy filesystem driver to use passed socket file for 1476 communicating with virtfs-proxy-helper(1). 1477 1478 ``sock_fd=sock_fd`` 1479 Enables proxy filesystem driver to use passed socket descriptor 1480 for communicating with virtfs-proxy-helper(1). Usually a helper 1481 like libvirt will create socketpair and pass one of the fds as 1482 sock\_fd. 1483 1484 ``fmode=fmode`` 1485 Specifies the default mode for newly created files on the host. 1486 Works only with security models "mapped-xattr" and 1487 "mapped-file". 1488 1489 ``dmode=dmode`` 1490 Specifies the default mode for newly created directories on the 1491 host. Works only with security models "mapped-xattr" and 1492 "mapped-file". 1493 1494 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w`` 1495 Specify bandwidth throttling limits in bytes per second, either 1496 for all request types or for reads or writes only. 1497 1498 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm`` 1499 Specify bursts in bytes per second, either for all request types 1500 or for reads or writes only. Bursts allow the guest I/O to spike 1501 above the limit temporarily. 1502 1503 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w`` 1504 Specify request rate limits in requests per second, either for 1505 all request types or for reads or writes only. 1506 1507 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm`` 1508 Specify bursts in requests per second, either for all request 1509 types or for reads or writes only. Bursts allow the guest I/O to 1510 spike above the limit temporarily. 1511 1512 ``throttling.iops-size=is`` 1513 Let every is bytes of a request count as a new request for iops 1514 throttling purposes. 1515 1516 -fsdev option is used along with -device driver "virtio-9p-...". 1517 1518``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag`` 1519 Options for virtio-9p-... driver are: 1520 1521 ``type`` 1522 Specifies the variant to be used. Supported values are "pci", 1523 "ccw" or "device", depending on the machine type. 1524 1525 ``fsdev=id`` 1526 Specifies the id value specified along with -fsdev option. 1527 1528 ``mount_tag=mount_tag`` 1529 Specifies the tag name to be used by the guest to mount this 1530 export point. 1531ERST 1532 1533DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs, 1534 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1535 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n" 1536 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n" 1537 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n" 1538 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n", 1539 QEMU_ARCH_ALL) 1540 1541SRST 1542``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]`` 1543 \ 1544``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1545 \ 1546``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` 1547 \ 1548``-virtfs synth,mount_tag=mount_tag`` 1549 Define a new virtual filesystem device and expose it to the guest using 1550 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain 1551 directory on host is made directly accessible by guest as a pass-through 1552 file system by using the 9P network protocol for communication between 1553 host and guests, if desired even accessible, shared by several guests 1554 simultaniously. 1555 1556 Note that ``-virtfs`` is actually just a convenience shortcut for its 1557 generalized form ``-fsdev -device virtio-9p-pci``. 1558 1559 The general form of pass-through file system options are: 1560 1561 ``local`` 1562 Accesses to the filesystem are done by QEMU. 1563 1564 ``proxy`` 1565 Accesses to the filesystem are done by virtfs-proxy-helper(1). 1566 1567 ``synth`` 1568 Synthetic filesystem, only used by QTests. 1569 1570 ``id=id`` 1571 Specifies identifier for the filesystem device 1572 1573 ``path=path`` 1574 Specifies the export path for the file system device. Files 1575 under this path will be available to the 9p client on the guest. 1576 1577 ``security_model=security_model`` 1578 Specifies the security model to be used for this export path. 1579 Supported security models are "passthrough", "mapped-xattr", 1580 "mapped-file" and "none". In "passthrough" security model, files 1581 are stored using the same credentials as they are created on the 1582 guest. This requires QEMU to run as root. In "mapped-xattr" 1583 security model, some of the file attributes like uid, gid, mode 1584 bits and link target are stored as file attributes. For 1585 "mapped-file" these attributes are stored in the hidden 1586 .virtfs\_metadata directory. Directories exported by this 1587 security model cannot interact with other unix tools. "none" 1588 security model is same as passthrough except the sever won't 1589 report failures if it fails to set file attributes like 1590 ownership. Security model is mandatory only for local fsdriver. 1591 Other fsdrivers (like proxy) don't take security model as a 1592 parameter. 1593 1594 ``writeout=writeout`` 1595 This is an optional argument. The only supported value is 1596 "immediate". This means that host page cache will be used to 1597 read and write data but write notification will be sent to the 1598 guest only when the data has been reported as written by the 1599 storage subsystem. 1600 1601 ``readonly=on`` 1602 Enables exporting 9p share as a readonly mount for guests. By 1603 default read-write access is given. 1604 1605 ``socket=socket`` 1606 Enables proxy filesystem driver to use passed socket file for 1607 communicating with virtfs-proxy-helper(1). Usually a helper like 1608 libvirt will create socketpair and pass one of the fds as 1609 sock\_fd. 1610 1611 ``sock_fd`` 1612 Enables proxy filesystem driver to use passed 'sock\_fd' as the 1613 socket descriptor for interfacing with virtfs-proxy-helper(1). 1614 1615 ``fmode=fmode`` 1616 Specifies the default mode for newly created files on the host. 1617 Works only with security models "mapped-xattr" and 1618 "mapped-file". 1619 1620 ``dmode=dmode`` 1621 Specifies the default mode for newly created directories on the 1622 host. Works only with security models "mapped-xattr" and 1623 "mapped-file". 1624 1625 ``mount_tag=mount_tag`` 1626 Specifies the tag name to be used by the guest to mount this 1627 export point. 1628 1629 ``multidevs=multidevs`` 1630 Specifies how to deal with multiple devices being shared with a 1631 9p export. Supported behaviours are either "remap", "forbid" or 1632 "warn". The latter is the default behaviour on which virtfs 9p 1633 expects only one device to be shared with the same export, and 1634 if more than one device is shared and accessed via the same 9p 1635 export then only a warning message is logged (once) by qemu on 1636 host side. In order to avoid file ID collisions on guest you 1637 should either create a separate virtfs export for each device to 1638 be shared with guests (recommended way) or you might use "remap" 1639 instead which allows you to share multiple devices with only one 1640 export instead, which is achieved by remapping the original 1641 inode numbers from host to guest in a way that would prevent 1642 such collisions. Remapping inodes in such use cases is required 1643 because the original device IDs from host are never passed and 1644 exposed on guest. Instead all files of an export shared with 1645 virtfs always share the same device id on guest. So two files 1646 with identical inode numbers but from actually different devices 1647 on host would otherwise cause a file ID collision and hence 1648 potential misbehaviours on guest. "forbid" on the other hand 1649 assumes like "warn" that only one device is shared by the same 1650 export, however it will not only log a warning message but also 1651 deny access to additional devices on guest. Note though that 1652 "forbid" does currently not block all possible file access 1653 operations (e.g. readdir() would still return entries from other 1654 devices). 1655ERST 1656 1657DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi, 1658 "-iscsi [user=user][,password=password]\n" 1659 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE\n" 1660 " [,initiator-name=initiator-iqn][,id=target-iqn]\n" 1661 " [,timeout=timeout]\n" 1662 " iSCSI session parameters\n", QEMU_ARCH_ALL) 1663 1664SRST 1665``-iscsi`` 1666 Configure iSCSI session parameters. 1667ERST 1668 1669DEFHEADING() 1670 1671DEFHEADING(USB options:) 1672 1673DEF("usb", 0, QEMU_OPTION_usb, 1674 "-usb enable on-board USB host controller (if not enabled by default)\n", 1675 QEMU_ARCH_ALL) 1676SRST 1677``-usb`` 1678 Enable USB emulation on machine types with an on-board USB host 1679 controller (if not enabled by default). Note that on-board USB host 1680 controllers may not support USB 3.0. In this case 1681 ``-device qemu-xhci`` can be used instead on machines with PCI. 1682ERST 1683 1684DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice, 1685 "-usbdevice name add the host or guest USB device 'name'\n", 1686 QEMU_ARCH_ALL) 1687SRST 1688``-usbdevice devname`` 1689 Add the USB device devname. Note that this option is deprecated, 1690 please use ``-device usb-...`` instead. See the chapter about 1691 :ref:`Connecting USB devices` in the System Emulation Users Guide. 1692 1693 ``mouse`` 1694 Virtual Mouse. This will override the PS/2 mouse emulation when 1695 activated. 1696 1697 ``tablet`` 1698 Pointer device that uses absolute coordinates (like a 1699 touchscreen). This means QEMU is able to report the mouse 1700 position without having to grab the mouse. Also overrides the 1701 PS/2 mouse emulation when activated. 1702 1703 ``braille`` 1704 Braille device. This will use BrlAPI to display the braille 1705 output on a real or fake device. 1706ERST 1707 1708DEFHEADING() 1709 1710DEFHEADING(Display options:) 1711 1712DEF("display", HAS_ARG, QEMU_OPTION_display, 1713#if defined(CONFIG_SPICE) 1714 "-display spice-app[,gl=on|off]\n" 1715#endif 1716#if defined(CONFIG_SDL) 1717 "-display sdl[,alt_grab=on|off][,ctrl_grab=on|off]\n" 1718 " [,window_close=on|off][,gl=on|core|es|off]\n" 1719#endif 1720#if defined(CONFIG_GTK) 1721 "-display gtk[,grab_on_hover=on|off][,gl=on|off]|\n" 1722#endif 1723#if defined(CONFIG_VNC) 1724 "-display vnc=<display>[,<optargs>]\n" 1725#endif 1726#if defined(CONFIG_CURSES) 1727 "-display curses[,charset=<encoding>]\n" 1728#endif 1729#if defined(CONFIG_OPENGL) 1730 "-display egl-headless[,rendernode=<file>]\n" 1731#endif 1732 "-display none\n" 1733 " select display backend type\n" 1734 " The default display is equivalent to\n " 1735#if defined(CONFIG_GTK) 1736 "\"-display gtk\"\n" 1737#elif defined(CONFIG_SDL) 1738 "\"-display sdl\"\n" 1739#elif defined(CONFIG_COCOA) 1740 "\"-display cocoa\"\n" 1741#elif defined(CONFIG_VNC) 1742 "\"-vnc localhost:0,to=99,id=default\"\n" 1743#else 1744 "\"-display none\"\n" 1745#endif 1746 , QEMU_ARCH_ALL) 1747SRST 1748``-display type`` 1749 Select type of display to use. This option is a replacement for the 1750 old style -sdl/-curses/... options. Use ``-display help`` to list 1751 the available display types. Valid values for type are 1752 1753 ``sdl`` 1754 Display video output via SDL (usually in a separate graphics 1755 window; see the SDL documentation for other possibilities). 1756 1757 ``curses`` 1758 Display video output via curses. For graphics device models 1759 which support a text mode, QEMU can display this output using a 1760 curses/ncurses interface. Nothing is displayed when the graphics 1761 device is in graphical mode or if the graphics device does not 1762 support a text mode. Generally only the VGA device models 1763 support text mode. The font charset used by the guest can be 1764 specified with the ``charset`` option, for example 1765 ``charset=CP850`` for IBM CP850 encoding. The default is 1766 ``CP437``. 1767 1768 ``none`` 1769 Do not display video output. The guest will still see an 1770 emulated graphics card, but its output will not be displayed to 1771 the QEMU user. This option differs from the -nographic option in 1772 that it only affects what is done with video output; -nographic 1773 also changes the destination of the serial and parallel port 1774 data. 1775 1776 ``gtk`` 1777 Display video output in a GTK window. This interface provides 1778 drop-down menus and other UI elements to configure and control 1779 the VM during runtime. 1780 1781 ``vnc`` 1782 Start a VNC server on display <arg> 1783 1784 ``egl-headless`` 1785 Offload all OpenGL operations to a local DRI device. For any 1786 graphical display, this display needs to be paired with either 1787 VNC or SPICE displays. 1788 1789 ``spice-app`` 1790 Start QEMU as a Spice server and launch the default Spice client 1791 application. The Spice server will redirect the serial consoles 1792 and QEMU monitors. (Since 4.0) 1793ERST 1794 1795DEF("nographic", 0, QEMU_OPTION_nographic, 1796 "-nographic disable graphical output and redirect serial I/Os to console\n", 1797 QEMU_ARCH_ALL) 1798SRST 1799``-nographic`` 1800 Normally, if QEMU is compiled with graphical window support, it 1801 displays output such as guest graphics, guest console, and the QEMU 1802 monitor in a window. With this option, you can totally disable 1803 graphical output so that QEMU is a simple command line application. 1804 The emulated serial port is redirected on the console and muxed with 1805 the monitor (unless redirected elsewhere explicitly). Therefore, you 1806 can still use QEMU to debug a Linux kernel with a serial console. 1807 Use C-a h for help on switching between the console and monitor. 1808ERST 1809 1810DEF("curses", 0, QEMU_OPTION_curses, 1811 "-curses shorthand for -display curses\n", 1812 QEMU_ARCH_ALL) 1813SRST 1814``-curses`` 1815 Normally, if QEMU is compiled with graphical window support, it 1816 displays output such as guest graphics, guest console, and the QEMU 1817 monitor in a window. With this option, QEMU can display the VGA 1818 output when in text mode using a curses/ncurses interface. Nothing 1819 is displayed in graphical mode. 1820ERST 1821 1822DEF("alt-grab", 0, QEMU_OPTION_alt_grab, 1823 "-alt-grab use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt)\n", 1824 QEMU_ARCH_ALL) 1825SRST 1826``-alt-grab`` 1827 Use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt). Note that 1828 this also affects the special keys (for fullscreen, monitor-mode 1829 switching, etc). 1830ERST 1831 1832DEF("ctrl-grab", 0, QEMU_OPTION_ctrl_grab, 1833 "-ctrl-grab use Right-Ctrl to grab mouse (instead of Ctrl-Alt)\n", 1834 QEMU_ARCH_ALL) 1835SRST 1836``-ctrl-grab`` 1837 Use Right-Ctrl to grab mouse (instead of Ctrl-Alt). Note that this 1838 also affects the special keys (for fullscreen, monitor-mode 1839 switching, etc). 1840ERST 1841 1842DEF("no-quit", 0, QEMU_OPTION_no_quit, 1843 "-no-quit disable SDL window close capability\n", QEMU_ARCH_ALL) 1844SRST 1845``-no-quit`` 1846 Disable SDL window close capability. 1847ERST 1848 1849DEF("sdl", 0, QEMU_OPTION_sdl, 1850 "-sdl shorthand for -display sdl\n", QEMU_ARCH_ALL) 1851SRST 1852``-sdl`` 1853 Enable SDL. 1854ERST 1855 1856DEF("spice", HAS_ARG, QEMU_OPTION_spice, 1857 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n" 1858 " [,x509-key-file=<file>][,x509-key-password=<file>]\n" 1859 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n" 1860 " [,x509-dh-key-file=<file>][,addr=addr][,ipv4|ipv6|unix]\n" 1861 " [,tls-ciphers=<list>]\n" 1862 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n" 1863 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n" 1864 " [,sasl][,password=<secret>][,disable-ticketing]\n" 1865 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n" 1866 " [,jpeg-wan-compression=[auto|never|always]]\n" 1867 " [,zlib-glz-wan-compression=[auto|never|always]]\n" 1868 " [,streaming-video=[off|all|filter]][,disable-copy-paste]\n" 1869 " [,disable-agent-file-xfer][,agent-mouse=[on|off]]\n" 1870 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n" 1871 " [,gl=[on|off]][,rendernode=<file>]\n" 1872 " enable spice\n" 1873 " at least one of {port, tls-port} is mandatory\n", 1874 QEMU_ARCH_ALL) 1875SRST 1876``-spice option[,option[,...]]`` 1877 Enable the spice remote desktop protocol. Valid options are 1878 1879 ``port=<nr>`` 1880 Set the TCP port spice is listening on for plaintext channels. 1881 1882 ``addr=<addr>`` 1883 Set the IP address spice is listening on. Default is any 1884 address. 1885 1886 ``ipv4``; \ ``ipv6``; \ ``unix`` 1887 Force using the specified IP version. 1888 1889 ``password=<secret>`` 1890 Set the password you need to authenticate. 1891 1892 ``sasl`` 1893 Require that the client use SASL to authenticate with the spice. 1894 The exact choice of authentication method used is controlled 1895 from the system / user's SASL configuration file for the 'qemu' 1896 service. This is typically found in /etc/sasl2/qemu.conf. If 1897 running QEMU as an unprivileged user, an environment variable 1898 SASL\_CONF\_PATH can be used to make it search alternate 1899 locations for the service config. While some SASL auth methods 1900 can also provide data encryption (eg GSSAPI), it is recommended 1901 that SASL always be combined with the 'tls' and 'x509' settings 1902 to enable use of SSL and server certificates. This ensures a 1903 data encryption preventing compromise of authentication 1904 credentials. 1905 1906 ``disable-ticketing`` 1907 Allow client connects without authentication. 1908 1909 ``disable-copy-paste`` 1910 Disable copy paste between the client and the guest. 1911 1912 ``disable-agent-file-xfer`` 1913 Disable spice-vdagent based file-xfer between the client and the 1914 guest. 1915 1916 ``tls-port=<nr>`` 1917 Set the TCP port spice is listening on for encrypted channels. 1918 1919 ``x509-dir=<dir>`` 1920 Set the x509 file directory. Expects same filenames as -vnc 1921 $display,x509=$dir 1922 1923 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>`` 1924 The x509 file names can also be configured individually. 1925 1926 ``tls-ciphers=<list>`` 1927 Specify which ciphers to use. 1928 1929 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]`` 1930 Force specific channel to be used with or without TLS 1931 encryption. The options can be specified multiple times to 1932 configure multiple channels. The special name "default" can be 1933 used to set the default mode. For channels which are not 1934 explicitly forced into one mode the spice client is allowed to 1935 pick tls/plaintext as he pleases. 1936 1937 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]`` 1938 Configure image compression (lossless). Default is auto\_glz. 1939 1940 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]`` 1941 Configure wan image compression (lossy for slow links). Default 1942 is auto. 1943 1944 ``streaming-video=[off|all|filter]`` 1945 Configure video stream detection. Default is off. 1946 1947 ``agent-mouse=[on|off]`` 1948 Enable/disable passing mouse events via vdagent. Default is on. 1949 1950 ``playback-compression=[on|off]`` 1951 Enable/disable audio stream compression (using celt 0.5.1). 1952 Default is on. 1953 1954 ``seamless-migration=[on|off]`` 1955 Enable/disable spice seamless migration. Default is off. 1956 1957 ``gl=[on|off]`` 1958 Enable/disable OpenGL context. Default is off. 1959 1960 ``rendernode=<file>`` 1961 DRM render node for OpenGL rendering. If not specified, it will 1962 pick the first available. (Since 2.9) 1963ERST 1964 1965DEF("portrait", 0, QEMU_OPTION_portrait, 1966 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n", 1967 QEMU_ARCH_ALL) 1968SRST 1969``-portrait`` 1970 Rotate graphical output 90 deg left (only PXA LCD). 1971ERST 1972 1973DEF("rotate", HAS_ARG, QEMU_OPTION_rotate, 1974 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n", 1975 QEMU_ARCH_ALL) 1976SRST 1977``-rotate deg`` 1978 Rotate graphical output some deg left (only PXA LCD). 1979ERST 1980 1981DEF("vga", HAS_ARG, QEMU_OPTION_vga, 1982 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n" 1983 " select video card type\n", QEMU_ARCH_ALL) 1984SRST 1985``-vga type`` 1986 Select type of VGA card to emulate. Valid values for type are 1987 1988 ``cirrus`` 1989 Cirrus Logic GD5446 Video card. All Windows versions starting 1990 from Windows 95 should recognize and use this graphic card. For 1991 optimal performances, use 16 bit color depth in the guest and 1992 the host OS. (This card was the default before QEMU 2.2) 1993 1994 ``std`` 1995 Standard VGA card with Bochs VBE extensions. If your guest OS 1996 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if 1997 you want to use high resolution modes (>= 1280x1024x16) then you 1998 should use this option. (This card is the default since QEMU 1999 2.2) 2000 2001 ``vmware`` 2002 VMWare SVGA-II compatible adapter. Use it if you have 2003 sufficiently recent XFree86/XOrg server or Windows guest with a 2004 driver for this card. 2005 2006 ``qxl`` 2007 QXL paravirtual graphic card. It is VGA compatible (including 2008 VESA 2.0 VBE support). Works best with qxl guest drivers 2009 installed though. Recommended choice when using the spice 2010 protocol. 2011 2012 ``tcx`` 2013 (sun4m only) Sun TCX framebuffer. This is the default 2014 framebuffer for sun4m machines and offers both 8-bit and 24-bit 2015 colour depths at a fixed resolution of 1024x768. 2016 2017 ``cg3`` 2018 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit 2019 framebuffer for sun4m machines available in both 1024x768 2020 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people 2021 wishing to run older Solaris versions. 2022 2023 ``virtio`` 2024 Virtio VGA card. 2025 2026 ``none`` 2027 Disable VGA card. 2028ERST 2029 2030DEF("full-screen", 0, QEMU_OPTION_full_screen, 2031 "-full-screen start in full screen\n", QEMU_ARCH_ALL) 2032SRST 2033``-full-screen`` 2034 Start in full screen. 2035ERST 2036 2037DEF("g", HAS_ARG, QEMU_OPTION_g , 2038 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n", 2039 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K) 2040SRST 2041``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]`` 2042 Set the initial graphical resolution and depth (PPC, SPARC only). 2043 2044 For PPC the default is 800x600x32. 2045 2046 For SPARC with the TCX graphics device, the default is 1024x768x8 2047 with the option of 1024x768x24. For cgthree, the default is 2048 1024x768x8 with the option of 1152x900x8 for people who wish to use 2049 OBP. 2050ERST 2051 2052DEF("vnc", HAS_ARG, QEMU_OPTION_vnc , 2053 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL) 2054SRST 2055``-vnc display[,option[,option[,...]]]`` 2056 Normally, if QEMU is compiled with graphical window support, it 2057 displays output such as guest graphics, guest console, and the QEMU 2058 monitor in a window. With this option, you can have QEMU listen on 2059 VNC display display and redirect the VGA display over the VNC 2060 session. It is very useful to enable the usb tablet device when 2061 using this option (option ``-device usb-tablet``). When using the 2062 VNC display, you must use the ``-k`` parameter to set the keyboard 2063 layout if you are not using en-us. Valid syntax for the display is 2064 2065 ``to=L`` 2066 With this option, QEMU will try next available VNC displays, 2067 until the number L, if the origianlly defined "-vnc display" is 2068 not available, e.g. port 5900+display is already used by another 2069 application. By default, to=0. 2070 2071 ``host:d`` 2072 TCP connections will only be allowed from host on display d. By 2073 convention the TCP port is 5900+d. Optionally, host can be 2074 omitted in which case the server will accept connections from 2075 any host. 2076 2077 ``unix:path`` 2078 Connections will be allowed over UNIX domain sockets where path 2079 is the location of a unix socket to listen for connections on. 2080 2081 ``none`` 2082 VNC is initialized but not started. The monitor ``change`` 2083 command can be used to later start the VNC server. 2084 2085 Following the display value there may be one or more option flags 2086 separated by commas. Valid options are 2087 2088 ``reverse`` 2089 Connect to a listening VNC client via a "reverse" connection. 2090 The client is specified by the display. For reverse network 2091 connections (host:d,``reverse``), the d argument is a TCP port 2092 number, not a display number. 2093 2094 ``websocket`` 2095 Opens an additional TCP listening port dedicated to VNC 2096 Websocket connections. If a bare websocket option is given, the 2097 Websocket port is 5700+display. An alternative port can be 2098 specified with the syntax ``websocket``\ =port. 2099 2100 If host is specified connections will only be allowed from this 2101 host. It is possible to control the websocket listen address 2102 independently, using the syntax ``websocket``\ =host:port. 2103 2104 If no TLS credentials are provided, the websocket connection 2105 runs in unencrypted mode. If TLS credentials are provided, the 2106 websocket connection requires encrypted client connections. 2107 2108 ``password`` 2109 Require that password based authentication is used for client 2110 connections. 2111 2112 The password must be set separately using the ``set_password`` 2113 command in the :ref:`QEMU monitor`. The 2114 syntax to change your password is: 2115 ``set_password <protocol> <password>`` where <protocol> could be 2116 either "vnc" or "spice". 2117 2118 If you would like to change <protocol> password expiration, you 2119 should use ``expire_password <protocol> <expiration-time>`` 2120 where expiration time could be one of the following options: 2121 now, never, +seconds or UNIX time of expiration, e.g. +60 to 2122 make password expire in 60 seconds, or 1335196800 to make 2123 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for 2124 this date and time). 2125 2126 You can also use keywords "now" or "never" for the expiration 2127 time to allow <protocol> password to expire immediately or never 2128 expire. 2129 2130 ``tls-creds=ID`` 2131 Provides the ID of a set of TLS credentials to use to secure the 2132 VNC server. They will apply to both the normal VNC server socket 2133 and the websocket socket (if enabled). Setting TLS credentials 2134 will cause the VNC server socket to enable the VeNCrypt auth 2135 mechanism. The credentials should have been previously created 2136 using the ``-object tls-creds`` argument. 2137 2138 ``tls-authz=ID`` 2139 Provides the ID of the QAuthZ authorization object against which 2140 the client's x509 distinguished name will validated. This object 2141 is only resolved at time of use, so can be deleted and recreated 2142 on the fly while the VNC server is active. If missing, it will 2143 default to denying access. 2144 2145 ``sasl`` 2146 Require that the client use SASL to authenticate with the VNC 2147 server. The exact choice of authentication method used is 2148 controlled from the system / user's SASL configuration file for 2149 the 'qemu' service. This is typically found in 2150 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user, 2151 an environment variable SASL\_CONF\_PATH can be used to make it 2152 search alternate locations for the service config. While some 2153 SASL auth methods can also provide data encryption (eg GSSAPI), 2154 it is recommended that SASL always be combined with the 'tls' 2155 and 'x509' settings to enable use of SSL and server 2156 certificates. This ensures a data encryption preventing 2157 compromise of authentication credentials. See the 2158 :ref:`VNC security` section in the System Emulation Users Guide 2159 for details on using SASL authentication. 2160 2161 ``sasl-authz=ID`` 2162 Provides the ID of the QAuthZ authorization object against which 2163 the client's SASL username will validated. This object is only 2164 resolved at time of use, so can be deleted and recreated on the 2165 fly while the VNC server is active. If missing, it will default 2166 to denying access. 2167 2168 ``acl`` 2169 Legacy method for enabling authorization of clients against the 2170 x509 distinguished name and SASL username. It results in the 2171 creation of two ``authz-list`` objects with IDs of 2172 ``vnc.username`` and ``vnc.x509dname``. The rules for these 2173 objects must be configured with the HMP ACL commands. 2174 2175 This option is deprecated and should no longer be used. The new 2176 ``sasl-authz`` and ``tls-authz`` options are a replacement. 2177 2178 ``lossy`` 2179 Enable lossy compression methods (gradient, JPEG, ...). If this 2180 option is set, VNC client may receive lossy framebuffer updates 2181 depending on its encoding settings. Enabling this option can 2182 save a lot of bandwidth at the expense of quality. 2183 2184 ``non-adaptive`` 2185 Disable adaptive encodings. Adaptive encodings are enabled by 2186 default. An adaptive encoding will try to detect frequently 2187 updated screen regions, and send updates in these regions using 2188 a lossy encoding (like JPEG). This can be really helpful to save 2189 bandwidth when playing videos. Disabling adaptive encodings 2190 restores the original static behavior of encodings like Tight. 2191 2192 ``share=[allow-exclusive|force-shared|ignore]`` 2193 Set display sharing policy. 'allow-exclusive' allows clients to 2194 ask for exclusive access. As suggested by the rfb spec this is 2195 implemented by dropping other connections. Connecting multiple 2196 clients in parallel requires all clients asking for a shared 2197 session (vncviewer: -shared switch). This is the default. 2198 'force-shared' disables exclusive client access. Useful for 2199 shared desktop sessions, where you don't want someone forgetting 2200 specify -shared disconnect everybody else. 'ignore' completely 2201 ignores the shared flag and allows everybody connect 2202 unconditionally. Doesn't conform to the rfb spec but is 2203 traditional QEMU behavior. 2204 2205 ``key-delay-ms`` 2206 Set keyboard delay, for key down and key up events, in 2207 milliseconds. Default is 10. Keyboards are low-bandwidth 2208 devices, so this slowdown can help the device and guest to keep 2209 up and not lose events in case events are arriving in bulk. 2210 Possible causes for the latter are flaky network connections, or 2211 scripts for automated testing. 2212 2213 ``audiodev=audiodev`` 2214 Use the specified audiodev when the VNC client requests audio 2215 transmission. When not using an -audiodev argument, this option 2216 must be omitted, otherwise is must be present and specify a 2217 valid audiodev. 2218ERST 2219 2220ARCHHEADING(, QEMU_ARCH_I386) 2221 2222ARCHHEADING(i386 target only:, QEMU_ARCH_I386) 2223 2224DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack, 2225 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n", 2226 QEMU_ARCH_I386) 2227SRST 2228``-win2k-hack`` 2229 Use it when installing Windows 2000 to avoid a disk full bug. After 2230 Windows 2000 is installed, you no longer need this option (this 2231 option slows down the IDE transfers). 2232ERST 2233 2234DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk, 2235 "-no-fd-bootchk disable boot signature checking for floppy disks\n", 2236 QEMU_ARCH_I386) 2237SRST 2238``-no-fd-bootchk`` 2239 Disable boot signature checking for floppy disks in BIOS. May be 2240 needed to boot from old floppy disks. 2241ERST 2242 2243DEF("no-acpi", 0, QEMU_OPTION_no_acpi, 2244 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2245SRST 2246``-no-acpi`` 2247 Disable ACPI (Advanced Configuration and Power Interface) support. 2248 Use it if your guest OS complains about ACPI problems (PC target 2249 machine only). 2250ERST 2251 2252DEF("no-hpet", 0, QEMU_OPTION_no_hpet, 2253 "-no-hpet disable HPET\n", QEMU_ARCH_I386) 2254SRST 2255``-no-hpet`` 2256 Disable HPET support. 2257ERST 2258 2259DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable, 2260 "-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" 2261 " ACPI table description\n", QEMU_ARCH_I386) 2262SRST 2263``-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]...]`` 2264 Add ACPI table with specified header fields and context from 2265 specified files. For file=, take whole ACPI table from the specified 2266 files, including all ACPI headers (possible overridden by other 2267 options). For data=, only data portion of the table is used, all 2268 header information is specified in the command line. If a SLIC table 2269 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id 2270 fields will override the same in the RSDT and the FADT (a.k.a. 2271 FACP), in order to ensure the field matches required by the 2272 Microsoft SLIC spec and the ACPI spec. 2273ERST 2274 2275DEF("smbios", HAS_ARG, QEMU_OPTION_smbios, 2276 "-smbios file=binary\n" 2277 " load SMBIOS entry from binary file\n" 2278 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n" 2279 " [,uefi=on|off]\n" 2280 " specify SMBIOS type 0 fields\n" 2281 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2282 " [,uuid=uuid][,sku=str][,family=str]\n" 2283 " specify SMBIOS type 1 fields\n" 2284 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2285 " [,asset=str][,location=str]\n" 2286 " specify SMBIOS type 2 fields\n" 2287 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n" 2288 " [,sku=str]\n" 2289 " specify SMBIOS type 3 fields\n" 2290 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n" 2291 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n" 2292 " specify SMBIOS type 4 fields\n" 2293 "-smbios type=11[,value=str][,path=filename]\n" 2294 " specify SMBIOS type 11 fields\n" 2295 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n" 2296 " [,asset=str][,part=str][,speed=%d]\n" 2297 " specify SMBIOS type 17 fields\n", 2298 QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2299SRST 2300``-smbios file=binary`` 2301 Load SMBIOS entry from binary file. 2302 2303``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]`` 2304 Specify SMBIOS type 0 fields 2305 2306``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]`` 2307 Specify SMBIOS type 1 fields 2308 2309``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]`` 2310 Specify SMBIOS type 2 fields 2311 2312``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]`` 2313 Specify SMBIOS type 3 fields 2314 2315``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str]`` 2316 Specify SMBIOS type 4 fields 2317 2318``-smbios type=11[,value=str][,path=filename]`` 2319 Specify SMBIOS type 11 fields 2320 2321 This argument can be repeated multiple times, and values are added in the order they are parsed. 2322 Applications intending to use OEM strings data are encouraged to use their application name as 2323 a prefix for the value string. This facilitates passing information for multiple applications 2324 concurrently. 2325 2326 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax 2327 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes. 2328 2329 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to 2330 the SMBIOS table in the order in which they appear. 2331 2332 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535 2333 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the 2334 guest. Instead it should be used as a indicator to inform the guest where to locate the real 2335 data set, for example, by specifying the serial ID of a block device. 2336 2337 An example passing three strings is 2338 2339 .. parsed-literal:: 2340 2341 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\ 2342 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\ 2343 path=/some/file/with/oemstringsdata.txt 2344 2345 In the guest OS this is visible with the ``dmidecode`` command 2346 2347 .. parsed-literal:: 2348 2349 $ dmidecode -t 11 2350 Handle 0x0E00, DMI type 11, 5 bytes 2351 OEM Strings 2352 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/ 2353 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os 2354 String 3: myapp:some extra data 2355 2356 2357``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]`` 2358 Specify SMBIOS type 17 fields 2359ERST 2360 2361DEFHEADING() 2362 2363DEFHEADING(Network options:) 2364 2365DEF("netdev", HAS_ARG, QEMU_OPTION_netdev, 2366#ifdef CONFIG_SLIRP 2367 "-netdev user,id=str[,ipv4[=on|off]][,net=addr[/mask]][,host=addr]\n" 2368 " [,ipv6[=on|off]][,ipv6-net=addr[/int]][,ipv6-host=addr]\n" 2369 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n" 2370 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n" 2371 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]" 2372#ifndef _WIN32 2373 "[,smb=dir[,smbserver=addr]]\n" 2374#endif 2375 " configure a user mode network backend with ID 'str',\n" 2376 " its DHCP server and optional services\n" 2377#endif 2378#ifdef _WIN32 2379 "-netdev tap,id=str,ifname=name\n" 2380 " configure a host TAP network backend with ID 'str'\n" 2381#else 2382 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n" 2383 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n" 2384 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n" 2385 " [,poll-us=n]\n" 2386 " configure a host TAP network backend with ID 'str'\n" 2387 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2388 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n" 2389 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n" 2390 " to deconfigure it\n" 2391 " use '[down]script=no' to disable script execution\n" 2392 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n" 2393 " configure it\n" 2394 " use 'fd=h' to connect to an already opened TAP interface\n" 2395 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n" 2396 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n" 2397 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n" 2398 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n" 2399 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n" 2400 " use vhost=on to enable experimental in kernel accelerator\n" 2401 " (only has effect for virtio guests which use MSIX)\n" 2402 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n" 2403 " use 'vhostfd=h' to connect to an already opened vhost net device\n" 2404 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n" 2405 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n" 2406 " use 'poll-us=n' to speciy the maximum number of microseconds that could be\n" 2407 " spent on busy polling for vhost net\n" 2408 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n" 2409 " configure a host TAP network backend with ID 'str' that is\n" 2410 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2411 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n" 2412#endif 2413#ifdef __linux__ 2414 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n" 2415 " [,rxsession=rxsession],txsession=txsession[,ipv6=on/off][,udp=on/off]\n" 2416 " [,cookie64=on/off][,counter][,pincounter][,txcookie=txcookie]\n" 2417 " [,rxcookie=rxcookie][,offset=offset]\n" 2418 " configure a network backend with ID 'str' connected to\n" 2419 " an Ethernet over L2TPv3 pseudowire.\n" 2420 " Linux kernel 3.3+ as well as most routers can talk\n" 2421 " L2TPv3. This transport allows connecting a VM to a VM,\n" 2422 " VM to a router and even VM to Host. It is a nearly-universal\n" 2423 " standard (RFC3931). Note - this implementation uses static\n" 2424 " pre-configured tunnels (same as the Linux kernel).\n" 2425 " use 'src=' to specify source address\n" 2426 " use 'dst=' to specify destination address\n" 2427 " use 'udp=on' to specify udp encapsulation\n" 2428 " use 'srcport=' to specify source udp port\n" 2429 " use 'dstport=' to specify destination udp port\n" 2430 " use 'ipv6=on' to force v6\n" 2431 " L2TPv3 uses cookies to prevent misconfiguration as\n" 2432 " well as a weak security measure\n" 2433 " use 'rxcookie=0x012345678' to specify a rxcookie\n" 2434 " use 'txcookie=0x012345678' to specify a txcookie\n" 2435 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n" 2436 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n" 2437 " use 'pincounter=on' to work around broken counter handling in peer\n" 2438 " use 'offset=X' to add an extra offset between header and data\n" 2439#endif 2440 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n" 2441 " configure a network backend to connect to another network\n" 2442 " using a socket connection\n" 2443 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n" 2444 " configure a network backend to connect to a multicast maddr and port\n" 2445 " use 'localaddr=addr' to specify the host address to send packets from\n" 2446 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n" 2447 " configure a network backend to connect to another network\n" 2448 " using an UDP tunnel\n" 2449#ifdef CONFIG_VDE 2450 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n" 2451 " configure a network backend to connect to port 'n' of a vde switch\n" 2452 " running on host and listening for incoming connections on 'socketpath'.\n" 2453 " Use group 'groupname' and mode 'octalmode' to change default\n" 2454 " ownership and permissions for communication port.\n" 2455#endif 2456#ifdef CONFIG_NETMAP 2457 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n" 2458 " attach to the existing netmap-enabled network interface 'name', or to a\n" 2459 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n" 2460 " netmap device, defaults to '/dev/netmap')\n" 2461#endif 2462#ifdef CONFIG_POSIX 2463 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n" 2464 " configure a vhost-user network, backed by a chardev 'dev'\n" 2465#endif 2466#ifdef __linux__ 2467 "-netdev vhost-vdpa,id=str,vhostdev=/path/to/dev\n" 2468 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n" 2469#endif 2470 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n" 2471 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL) 2472DEF("nic", HAS_ARG, QEMU_OPTION_nic, 2473 "-nic [tap|bridge|" 2474#ifdef CONFIG_SLIRP 2475 "user|" 2476#endif 2477#ifdef __linux__ 2478 "l2tpv3|" 2479#endif 2480#ifdef CONFIG_VDE 2481 "vde|" 2482#endif 2483#ifdef CONFIG_NETMAP 2484 "netmap|" 2485#endif 2486#ifdef CONFIG_POSIX 2487 "vhost-user|" 2488#endif 2489 "socket][,option][,...][mac=macaddr]\n" 2490 " initialize an on-board / default host NIC (using MAC address\n" 2491 " macaddr) and connect it to the given host network backend\n" 2492 "-nic none use it alone to have zero network devices (the default is to\n" 2493 " provided a 'user' network connection)\n", 2494 QEMU_ARCH_ALL) 2495DEF("net", HAS_ARG, QEMU_OPTION_net, 2496 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n" 2497 " configure or create an on-board (or machine default) NIC and\n" 2498 " connect it to hub 0 (please use -nic unless you need a hub)\n" 2499 "-net [" 2500#ifdef CONFIG_SLIRP 2501 "user|" 2502#endif 2503 "tap|" 2504 "bridge|" 2505#ifdef CONFIG_VDE 2506 "vde|" 2507#endif 2508#ifdef CONFIG_NETMAP 2509 "netmap|" 2510#endif 2511 "socket][,option][,option][,...]\n" 2512 " old way to initialize a host network interface\n" 2513 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL) 2514SRST 2515``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]`` 2516 This option is a shortcut for configuring both the on-board 2517 (default) guest NIC hardware and the host network backend in one go. 2518 The host backend options are the same as with the corresponding 2519 ``-netdev`` options below. The guest NIC model can be set with 2520 ``model=modelname``. Use ``model=help`` to list the available device 2521 types. The hardware MAC address can be set with ``mac=macaddr``. 2522 2523 The following two example do exactly the same, to show how ``-nic`` 2524 can be used to shorten the command line length: 2525 2526 .. parsed-literal:: 2527 2528 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32 2529 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32 2530 2531``-nic none`` 2532 Indicate that no network devices should be configured. It is used to 2533 override the default configuration (default NIC with "user" host 2534 network backend) which is activated if no other networking options 2535 are provided. 2536 2537``-netdev user,id=id[,option][,option][,...]`` 2538 Configure user mode host network backend which requires no 2539 administrator privilege to run. Valid options are: 2540 2541 ``id=id`` 2542 Assign symbolic name for use in monitor commands. 2543 2544 ``ipv4=on|off and ipv6=on|off`` 2545 Specify that either IPv4 or IPv6 must be enabled. If neither is 2546 specified both protocols are enabled. 2547 2548 ``net=addr[/mask]`` 2549 Set IP network address the guest will see. Optionally specify 2550 the netmask, either in the form a.b.c.d or as number of valid 2551 top-most bits. Default is 10.0.2.0/24. 2552 2553 ``host=addr`` 2554 Specify the guest-visible address of the host. Default is the 2555 2nd IP in the guest network, i.e. x.x.x.2. 2556 2557 ``ipv6-net=addr[/int]`` 2558 Set IPv6 network address the guest will see (default is 2559 fec0::/64). The network prefix is given in the usual hexadecimal 2560 IPv6 address notation. The prefix size is optional, and is given 2561 as the number of valid top-most bits (default is 64). 2562 2563 ``ipv6-host=addr`` 2564 Specify the guest-visible IPv6 address of the host. Default is 2565 the 2nd IPv6 in the guest network, i.e. xxxx::2. 2566 2567 ``restrict=on|off`` 2568 If this option is enabled, the guest will be isolated, i.e. it 2569 will not be able to contact the host and no guest IP packets 2570 will be routed over the host to the outside. This option does 2571 not affect any explicitly set forwarding rules. 2572 2573 ``hostname=name`` 2574 Specifies the client hostname reported by the built-in DHCP 2575 server. 2576 2577 ``dhcpstart=addr`` 2578 Specify the first of the 16 IPs the built-in DHCP server can 2579 assign. Default is the 15th to 31st IP in the guest network, 2580 i.e. x.x.x.15 to x.x.x.31. 2581 2582 ``dns=addr`` 2583 Specify the guest-visible address of the virtual nameserver. The 2584 address must be different from the host address. Default is the 2585 3rd IP in the guest network, i.e. x.x.x.3. 2586 2587 ``ipv6-dns=addr`` 2588 Specify the guest-visible address of the IPv6 virtual 2589 nameserver. The address must be different from the host address. 2590 Default is the 3rd IP in the guest network, i.e. xxxx::3. 2591 2592 ``dnssearch=domain`` 2593 Provides an entry for the domain-search list sent by the 2594 built-in DHCP server. More than one domain suffix can be 2595 transmitted by specifying this option multiple times. If 2596 supported, this will cause the guest to automatically try to 2597 append the given domain suffix(es) in case a domain name can not 2598 be resolved. 2599 2600 Example: 2601 2602 .. parsed-literal:: 2603 2604 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org 2605 2606 ``domainname=domain`` 2607 Specifies the client domain name reported by the built-in DHCP 2608 server. 2609 2610 ``tftp=dir`` 2611 When using the user mode network stack, activate a built-in TFTP 2612 server. The files in dir will be exposed as the root of a TFTP 2613 server. The TFTP client on the guest must be configured in 2614 binary mode (use the command ``bin`` of the Unix TFTP client). 2615 2616 ``tftp-server-name=name`` 2617 In BOOTP reply, broadcast name as the "TFTP server name" 2618 (RFC2132 option 66). This can be used to advise the guest to 2619 load boot files or configurations from a different server than 2620 the host address. 2621 2622 ``bootfile=file`` 2623 When using the user mode network stack, broadcast file as the 2624 BOOTP filename. In conjunction with ``tftp``, this can be used 2625 to network boot a guest from a local directory. 2626 2627 Example (using pxelinux): 2628 2629 .. parsed-literal:: 2630 2631 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\ 2632 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0 2633 2634 ``smb=dir[,smbserver=addr]`` 2635 When using the user mode network stack, activate a built-in SMB 2636 server so that Windows OSes can access to the host files in 2637 ``dir`` transparently. The IP address of the SMB server can be 2638 set to addr. By default the 4th IP in the guest network is used, 2639 i.e. x.x.x.4. 2640 2641 In the guest Windows OS, the line: 2642 2643 :: 2644 2645 10.0.2.4 smbserver 2646 2647 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows 2648 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows 2649 NT/2000). 2650 2651 Then ``dir`` can be accessed in ``\\smbserver\qemu``. 2652 2653 Note that a SAMBA server must be installed on the host OS. 2654 2655 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport`` 2656 Redirect incoming TCP or UDP connections to the host port 2657 hostport to the guest IP address guestaddr on guest port 2658 guestport. If guestaddr is not specified, its value is x.x.x.15 2659 (default first address given by the built-in DHCP server). By 2660 specifying hostaddr, the rule can be bound to a specific host 2661 interface. If no connection type is set, TCP is used. This 2662 option can be given multiple times. 2663 2664 For example, to redirect host X11 connection from screen 1 to 2665 guest screen 0, use the following: 2666 2667 .. parsed-literal:: 2668 2669 # on the host 2670 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000 2671 # this host xterm should open in the guest X11 server 2672 xterm -display :1 2673 2674 To redirect telnet connections from host port 5555 to telnet 2675 port on the guest, use the following: 2676 2677 .. parsed-literal:: 2678 2679 # on the host 2680 |qemu_system| -nic user,hostfwd=tcp::5555-:23 2681 telnet localhost 5555 2682 2683 Then when you use on the host ``telnet localhost 5555``, you 2684 connect to the guest telnet server. 2685 2686 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command`` 2687 Forward guest TCP connections to the IP address server on port 2688 port to the character device dev or to a program executed by 2689 cmd:command which gets spawned for each connection. This option 2690 can be given multiple times. 2691 2692 You can either use a chardev directly and have that one used 2693 throughout QEMU's lifetime, like in the following example: 2694 2695 .. parsed-literal:: 2696 2697 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever 2698 # the guest accesses it 2699 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321 2700 2701 Or you can execute a command on every TCP connection established 2702 by the guest, so that QEMU behaves similar to an inetd process 2703 for that virtual server: 2704 2705 .. parsed-literal:: 2706 2707 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234 2708 # and connect the TCP stream to its stdin/stdout 2709 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321' 2710 2711``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]`` 2712 Configure a host TAP network backend with ID id. 2713 2714 Use the network script file to configure it and the network script 2715 dfile to deconfigure it. If name is not provided, the OS 2716 automatically provides one. The default network configure script is 2717 ``/etc/qemu-ifup`` and the default network deconfigure script is 2718 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to 2719 disable script execution. 2720 2721 If running QEMU as an unprivileged user, use the network helper 2722 to configure the TAP interface and attach it to the bridge. 2723 The default network helper executable is 2724 ``/path/to/qemu-bridge-helper`` and the default bridge device is 2725 ``br0``. 2726 2727 ``fd``\ =h can be used to specify the handle of an already opened 2728 host TAP interface. 2729 2730 Examples: 2731 2732 .. parsed-literal:: 2733 2734 #launch a QEMU instance with the default network script 2735 |qemu_system| linux.img -nic tap 2736 2737 .. parsed-literal:: 2738 2739 #launch a QEMU instance with two NICs, each one connected 2740 #to a TAP device 2741 |qemu_system| linux.img \\ 2742 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\ 2743 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1 2744 2745 .. parsed-literal:: 2746 2747 #launch a QEMU instance with the default network helper to 2748 #connect a TAP device to bridge br0 2749 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ 2750 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper" 2751 2752``-netdev bridge,id=id[,br=bridge][,helper=helper]`` 2753 Connect a host TAP network interface to a host bridge device. 2754 2755 Use the network helper helper to configure the TAP interface and 2756 attach it to the bridge. The default network helper executable is 2757 ``/path/to/qemu-bridge-helper`` and the default bridge device is 2758 ``br0``. 2759 2760 Examples: 2761 2762 .. parsed-literal:: 2763 2764 #launch a QEMU instance with the default network helper to 2765 #connect a TAP device to bridge br0 2766 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1 2767 2768 .. parsed-literal:: 2769 2770 #launch a QEMU instance with the default network helper to 2771 #connect a TAP device to bridge qemubr0 2772 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1 2773 2774``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]`` 2775 This host network backend can be used to connect the guest's network 2776 to another QEMU virtual machine using a TCP socket connection. If 2777 ``listen`` is specified, QEMU waits for incoming connections on port 2778 (host is optional). ``connect`` is used to connect to another QEMU 2779 instance using the ``listen`` option. ``fd``\ =h specifies an 2780 already opened TCP socket. 2781 2782 Example: 2783 2784 .. parsed-literal:: 2785 2786 # launch a first QEMU instance 2787 |qemu_system| linux.img \\ 2788 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 2789 -netdev socket,id=n1,listen=:1234 2790 # connect the network of this instance to the network of the first instance 2791 |qemu_system| linux.img \\ 2792 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 2793 -netdev socket,id=n2,connect=127.0.0.1:1234 2794 2795``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]`` 2796 Configure a socket host network backend to share the guest's network 2797 traffic with another QEMU virtual machines using a UDP multicast 2798 socket, effectively making a bus for every QEMU with same multicast 2799 address maddr and port. NOTES: 2800 2801 1. Several QEMU can be running on different hosts and share same bus 2802 (assuming correct multicast setup for these hosts). 2803 2804 2. mcast support is compatible with User Mode Linux (argument 2805 ``ethN=mcast``), see http://user-mode-linux.sf.net. 2806 2807 3. Use ``fd=h`` to specify an already opened UDP multicast socket. 2808 2809 Example: 2810 2811 .. parsed-literal:: 2812 2813 # launch one QEMU instance 2814 |qemu_system| linux.img \\ 2815 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 2816 -netdev socket,id=n1,mcast=230.0.0.1:1234 2817 # launch another QEMU instance on same "bus" 2818 |qemu_system| linux.img \\ 2819 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ 2820 -netdev socket,id=n2,mcast=230.0.0.1:1234 2821 # launch yet another QEMU instance on same "bus" 2822 |qemu_system| linux.img \\ 2823 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\ 2824 -netdev socket,id=n3,mcast=230.0.0.1:1234 2825 2826 Example (User Mode Linux compat.): 2827 2828 .. parsed-literal:: 2829 2830 # launch QEMU instance (note mcast address selected is UML's default) 2831 |qemu_system| linux.img \\ 2832 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 2833 -netdev socket,id=n1,mcast=239.192.168.1:1102 2834 # launch UML 2835 /path/to/linux ubd0=/path/to/root_fs eth0=mcast 2836 2837 Example (send packets from host's 1.2.3.4): 2838 2839 .. parsed-literal:: 2840 2841 |qemu_system| linux.img \\ 2842 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ 2843 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4 2844 2845``-netdev l2tpv3,id=id,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport],txsession=txsession[,rxsession=rxsession][,ipv6][,udp][,cookie64][,counter][,pincounter][,txcookie=txcookie][,rxcookie=rxcookie][,offset=offset]`` 2846 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931) 2847 is a popular protocol to transport Ethernet (and other Layer 2) data 2848 frames between two systems. It is present in routers, firewalls and 2849 the Linux kernel (from version 3.3 onwards). 2850 2851 This transport allows a VM to communicate to another VM, router or 2852 firewall directly. 2853 2854 ``src=srcaddr`` 2855 source address (mandatory) 2856 2857 ``dst=dstaddr`` 2858 destination address (mandatory) 2859 2860 ``udp`` 2861 select udp encapsulation (default is ip). 2862 2863 ``srcport=srcport`` 2864 source udp port. 2865 2866 ``dstport=dstport`` 2867 destination udp port. 2868 2869 ``ipv6`` 2870 force v6, otherwise defaults to v4. 2871 2872 ``rxcookie=rxcookie``; \ ``txcookie=txcookie`` 2873 Cookies are a weak form of security in the l2tpv3 specification. 2874 Their function is mostly to prevent misconfiguration. By default 2875 they are 32 bit. 2876 2877 ``cookie64`` 2878 Set cookie size to 64 bit instead of the default 32 2879 2880 ``counter=off`` 2881 Force a 'cut-down' L2TPv3 with no counter as in 2882 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00 2883 2884 ``pincounter=on`` 2885 Work around broken counter handling in peer. This may also help 2886 on networks which have packet reorder. 2887 2888 ``offset=offset`` 2889 Add an extra offset between header and data 2890 2891 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to 2892 the bridge br-lan on the remote Linux host 1.2.3.4: 2893 2894 .. parsed-literal:: 2895 2896 # Setup tunnel on linux host using raw ip as encapsulation 2897 # on 1.2.3.4 2898 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\ 2899 encap udp udp_sport 16384 udp_dport 16384 2900 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\ 2901 0xFFFFFFFF peer_session_id 0xFFFFFFFF 2902 ifconfig vmtunnel0 mtu 1500 2903 ifconfig vmtunnel0 up 2904 brctl addif br-lan vmtunnel0 2905 2906 2907 # on 4.3.2.1 2908 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter 2909 2910 |qemu_system| linux.img -device e1000,netdev=n1 \\ 2911 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter 2912 2913``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]`` 2914 Configure VDE backend to connect to PORT n of a vde switch running 2915 on host and listening for incoming connections on socketpath. Use 2916 GROUP groupname and MODE octalmode to change default ownership and 2917 permissions for communication port. This option is only available if 2918 QEMU has been compiled with vde support enabled. 2919 2920 Example: 2921 2922 .. parsed-literal:: 2923 2924 # launch vde switch 2925 vde_switch -F -sock /tmp/myswitch 2926 # launch QEMU instance 2927 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch 2928 2929``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]`` 2930 Establish a vhost-user netdev, backed by a chardev id. The chardev 2931 should be a unix domain socket backed one. The vhost-user uses a 2932 specifically defined protocol to pass vhost ioctl replacement 2933 messages to an application on the other end of the socket. On 2934 non-MSIX guests, the feature can be forced with vhostforce. Use 2935 'queues=n' to specify the number of queues to be created for 2936 multiqueue vhost-user. 2937 2938 Example: 2939 2940 :: 2941 2942 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \ 2943 -numa node,memdev=mem \ 2944 -chardev socket,id=chr0,path=/path/to/socket \ 2945 -netdev type=vhost-user,id=net0,chardev=chr0 \ 2946 -device virtio-net-pci,netdev=net0 2947 2948``-netdev vhost-vdpa,vhostdev=/path/to/dev`` 2949 Establish a vhost-vdpa netdev. 2950 2951 vDPA device is a device that uses a datapath which complies with 2952 the virtio specifications with a vendor specific control path. 2953 vDPA devices can be both physically located on the hardware or 2954 emulated by software. 2955 2956``-netdev hubport,id=id,hubid=hubid[,netdev=nd]`` 2957 Create a hub port on the emulated hub with ID hubid. 2958 2959 The hubport netdev lets you connect a NIC to a QEMU emulated hub 2960 instead of a single netdev. Alternatively, you can also connect the 2961 hubport to another netdev with ID nd by using the ``netdev=nd`` 2962 option. 2963 2964``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]`` 2965 Legacy option to configure or create an on-board (or machine 2966 default) Network Interface Card(NIC) and connect it either to the 2967 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd. 2968 If model is omitted, then the default NIC model associated with the 2969 machine type is used. Note that the default NIC model may change in 2970 future QEMU releases, so it is highly recommended to always specify 2971 a model. Optionally, the MAC address can be changed to mac, the 2972 device address set to addr (PCI cards only), and a name can be 2973 assigned for use in monitor commands. Optionally, for PCI cards, you 2974 can specify the number v of MSI-X vectors that the card should have; 2975 this option currently only affects virtio cards; set v = 0 to 2976 disable MSI-X. If no ``-net`` option is specified, a single NIC is 2977 created. QEMU can emulate several different models of network card. 2978 Use ``-net nic,model=help`` for a list of available devices for your 2979 target. 2980 2981``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]`` 2982 Configure a host network backend (with the options corresponding to 2983 the same ``-netdev`` option) and connect it to the emulated hub 0 2984 (the default hub). Use name to specify the name of the hub port. 2985ERST 2986 2987DEFHEADING() 2988 2989DEFHEADING(Character device options:) 2990 2991DEF("chardev", HAS_ARG, QEMU_OPTION_chardev, 2992 "-chardev help\n" 2993 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2994 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4][,ipv6][,nodelay][,reconnect=seconds]\n" 2995 " [,server][,nowait][,telnet][,websocket][,reconnect=seconds][,mux=on|off]\n" 2996 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n" 2997 "-chardev socket,id=id,path=path[,server][,nowait][,telnet][,websocket][,reconnect=seconds]\n" 2998 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n" 2999 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n" 3000 " [,localport=localport][,ipv4][,ipv6][,mux=on|off]\n" 3001 " [,logfile=PATH][,logappend=on|off]\n" 3002 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3003 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n" 3004 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3005 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n" 3006 "-chardev file,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3007 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3008#ifdef _WIN32 3009 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3010 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3011#else 3012 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3013 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n" 3014#endif 3015#ifdef CONFIG_BRLAPI 3016 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3017#endif 3018#if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \ 3019 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) 3020 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3021 "-chardev tty,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3022#endif 3023#if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__) 3024 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3025 "-chardev parport,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 3026#endif 3027#if defined(CONFIG_SPICE) 3028 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3029 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 3030#endif 3031 , QEMU_ARCH_ALL 3032) 3033 3034SRST 3035The general form of a character device option is: 3036 3037``-chardev backend,id=id[,mux=on|off][,options]`` 3038 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``, 3039 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``, 3040 ``pty``, ``stdio``, ``braille``, ``tty``, ``parallel``, ``parport``, 3041 ``spicevmc``, ``spiceport``. The specific backend will determine the 3042 applicable options. 3043 3044 Use ``-chardev help`` to print all available chardev backend types. 3045 3046 All devices must have an id, which can be any string up to 127 3047 characters long. It is used to uniquely identify this device in 3048 other command line directives. 3049 3050 A character device may be used in multiplexing mode by multiple 3051 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is 3052 a "1:N" device, and here the "1" end is your specified chardev 3053 backend, and the "N" end is the various parts of QEMU that can talk 3054 to a chardev. If you create a chardev with ``id=myid`` and 3055 ``mux=on``, QEMU will create a multiplexer with your specified ID, 3056 and you can then configure multiple front ends to use that chardev 3057 ID for their input/output. Up to four different front ends can be 3058 connected to a single multiplexed chardev. (Without multiplexing 3059 enabled, a chardev can only be used by a single front end.) For 3060 instance you could use this to allow a single stdio chardev to be 3061 used by two serial ports and the QEMU monitor: 3062 3063 :: 3064 3065 -chardev stdio,mux=on,id=char0 \ 3066 -mon chardev=char0,mode=readline \ 3067 -serial chardev:char0 \ 3068 -serial chardev:char0 3069 3070 You can have more than one multiplexer in a system configuration; 3071 for instance you could have a TCP port multiplexed between UART 0 3072 and UART 1, and stdio multiplexed between the QEMU monitor and a 3073 parallel port: 3074 3075 :: 3076 3077 -chardev stdio,mux=on,id=char0 \ 3078 -mon chardev=char0,mode=readline \ 3079 -parallel chardev:char0 \ 3080 -chardev tcp,...,mux=on,id=char1 \ 3081 -serial chardev:char1 \ 3082 -serial chardev:char1 3083 3084 When you're using a multiplexed character device, some escape 3085 sequences are interpreted in the input. See the chapter about 3086 :ref:`keys in the character backend multiplexer` in the 3087 System Emulation Users Guide for more details. 3088 3089 Note that some other command line options may implicitly create 3090 multiplexed character backends; for instance ``-serial mon:stdio`` 3091 creates a multiplexed stdio backend connected to the serial port and 3092 the QEMU monitor, and ``-nographic`` also multiplexes the console 3093 and the monitor to stdio. 3094 3095 There is currently no support for multiplexing in the other 3096 direction (where a single QEMU front end takes input and output from 3097 multiple chardevs). 3098 3099 Every backend supports the ``logfile`` option, which supplies the 3100 path to a file to record all data transmitted via the backend. The 3101 ``logappend`` option controls whether the log file will be truncated 3102 or appended to when opened. 3103 3104The available backends are: 3105 3106``-chardev null,id=id`` 3107 A void device. This device will not emit any data, and will drop any 3108 data it receives. The null backend does not take any options. 3109 3110``-chardev socket,id=id[,TCP options or unix options][,server][,nowait][,telnet][,websocket][,reconnect=seconds][,tls-creds=id][,tls-authz=id]`` 3111 Create a two-way stream socket, which can be either a TCP or a unix 3112 socket. A unix socket will be created if ``path`` is specified. 3113 Behaviour is undefined if TCP options are specified for a unix 3114 socket. 3115 3116 ``server`` specifies that the socket shall be a listening socket. 3117 3118 ``nowait`` specifies that QEMU should not block waiting for a client 3119 to connect to a listening socket. 3120 3121 ``telnet`` specifies that traffic on the socket should interpret 3122 telnet escape sequences. 3123 3124 ``websocket`` specifies that the socket uses WebSocket protocol for 3125 communication. 3126 3127 ``reconnect`` sets the timeout for reconnecting on non-server 3128 sockets when the remote end goes away. qemu will delay this many 3129 seconds and then attempt to reconnect. Zero disables reconnecting, 3130 and is the default. 3131 3132 ``tls-creds`` requests enablement of the TLS protocol for 3133 encryption, and specifies the id of the TLS credentials to use for 3134 the handshake. The credentials must be previously created with the 3135 ``-object tls-creds`` argument. 3136 3137 ``tls-auth`` provides the ID of the QAuthZ authorization object 3138 against which the client's x509 distinguished name will be 3139 validated. This object is only resolved at time of use, so can be 3140 deleted and recreated on the fly while the chardev server is active. 3141 If missing, it will default to denying access. 3142 3143 TCP and unix socket options are given below: 3144 3145 ``TCP options: port=port[,host=host][,to=to][,ipv4][,ipv6][,nodelay]`` 3146 ``host`` for a listening socket specifies the local address to 3147 be bound. For a connecting socket species the remote host to 3148 connect to. ``host`` is optional for listening sockets. If not 3149 specified it defaults to ``0.0.0.0``. 3150 3151 ``port`` for a listening socket specifies the local port to be 3152 bound. For a connecting socket specifies the port on the remote 3153 host to connect to. ``port`` can be given as either a port 3154 number or a service name. ``port`` is required. 3155 3156 ``to`` is only relevant to listening sockets. If it is 3157 specified, and ``port`` cannot be bound, QEMU will attempt to 3158 bind to subsequent ports up to and including ``to`` until it 3159 succeeds. ``to`` must be specified as a port number. 3160 3161 ``ipv4`` and ``ipv6`` specify that either IPv4 or IPv6 must be 3162 used. If neither is specified the socket may use either 3163 protocol. 3164 3165 ``nodelay`` disables the Nagle algorithm. 3166 3167 ``unix options: path=path[,abstract=on|off][,tight=on|off]`` 3168 ``path`` specifies the local path of the unix socket. ``path`` 3169 is required. 3170 ``abstract`` specifies the use of the abstract socket namespace, 3171 rather than the filesystem. Optional, defaults to false. 3172 ``tight`` sets the socket length of abstract sockets to their minimum, 3173 rather than the full sun_path length. Optional, defaults to true. 3174 3175``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4][,ipv6]`` 3176 Sends all traffic from the guest to a remote host over UDP. 3177 3178 ``host`` specifies the remote host to connect to. If not specified 3179 it defaults to ``localhost``. 3180 3181 ``port`` specifies the port on the remote host to connect to. 3182 ``port`` is required. 3183 3184 ``localaddr`` specifies the local address to bind to. If not 3185 specified it defaults to ``0.0.0.0``. 3186 3187 ``localport`` specifies the local port to bind to. If not specified 3188 any available local port will be used. 3189 3190 ``ipv4`` and ``ipv6`` specify that either IPv4 or IPv6 must be used. 3191 If neither is specified the device may use either protocol. 3192 3193``-chardev msmouse,id=id`` 3194 Forward QEMU's emulated msmouse events to the guest. ``msmouse`` 3195 does not take any options. 3196 3197``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]`` 3198 Connect to a QEMU text console. ``vc`` may optionally be given a 3199 specific size. 3200 3201 ``width`` and ``height`` specify the width and height respectively 3202 of the console, in pixels. 3203 3204 ``cols`` and ``rows`` specify that the console be sized to fit a 3205 text console with the given dimensions. 3206 3207``-chardev ringbuf,id=id[,size=size]`` 3208 Create a ring buffer with fixed size ``size``. size must be a power 3209 of two and defaults to ``64K``. 3210 3211``-chardev file,id=id,path=path`` 3212 Log all traffic received from the guest to a file. 3213 3214 ``path`` specifies the path of the file to be opened. This file will 3215 be created if it does not already exist, and overwritten if it does. 3216 ``path`` is required. 3217 3218``-chardev pipe,id=id,path=path`` 3219 Create a two-way connection to the guest. The behaviour differs 3220 slightly between Windows hosts and other hosts: 3221 3222 On Windows, a single duplex pipe will be created at 3223 ``\\.pipe\path``. 3224 3225 On other hosts, 2 pipes will be created called ``path.in`` and 3226 ``path.out``. Data written to ``path.in`` will be received by the 3227 guest. Data written by the guest can be read from ``path.out``. QEMU 3228 will not create these fifos, and requires them to be present. 3229 3230 ``path`` forms part of the pipe path as described above. ``path`` is 3231 required. 3232 3233``-chardev console,id=id`` 3234 Send traffic from the guest to QEMU's standard output. ``console`` 3235 does not take any options. 3236 3237 ``console`` is only available on Windows hosts. 3238 3239``-chardev serial,id=id,path=path`` 3240 Send traffic from the guest to a serial device on the host. 3241 3242 On Unix hosts serial will actually accept any tty device, not only 3243 serial lines. 3244 3245 ``path`` specifies the name of the serial device to open. 3246 3247``-chardev pty,id=id`` 3248 Create a new pseudo-terminal on the host and connect to it. ``pty`` 3249 does not take any options. 3250 3251 ``pty`` is not available on Windows hosts. 3252 3253``-chardev stdio,id=id[,signal=on|off]`` 3254 Connect to standard input and standard output of the QEMU process. 3255 3256 ``signal`` controls if signals are enabled on the terminal, that 3257 includes exiting QEMU with the key sequence Control-c. This option 3258 is enabled by default, use ``signal=off`` to disable it. 3259 3260``-chardev braille,id=id`` 3261 Connect to a local BrlAPI server. ``braille`` does not take any 3262 options. 3263 3264``-chardev tty,id=id,path=path`` 3265 ``tty`` is only available on Linux, Sun, FreeBSD, NetBSD, OpenBSD 3266 and DragonFlyBSD hosts. It is an alias for ``serial``. 3267 3268 ``path`` specifies the path to the tty. ``path`` is required. 3269 3270``-chardev parallel,id=id,path=path`` 3271 \ 3272``-chardev parport,id=id,path=path`` 3273 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD 3274 hosts. 3275 3276 Connect to a local parallel port. 3277 3278 ``path`` specifies the path to the parallel port device. ``path`` is 3279 required. 3280 3281``-chardev spicevmc,id=id,debug=debug,name=name`` 3282 ``spicevmc`` is only available when spice support is built in. 3283 3284 ``debug`` debug level for spicevmc 3285 3286 ``name`` name of spice channel to connect to 3287 3288 Connect to a spice virtual machine channel, such as vdiport. 3289 3290``-chardev spiceport,id=id,debug=debug,name=name`` 3291 ``spiceport`` is only available when spice support is built in. 3292 3293 ``debug`` debug level for spicevmc 3294 3295 ``name`` name of spice port to connect to 3296 3297 Connect to a spice port, allowing a Spice client to handle the 3298 traffic identified by a name (preferably a fqdn). 3299ERST 3300 3301DEFHEADING() 3302 3303#ifdef CONFIG_TPM 3304DEFHEADING(TPM device options:) 3305 3306DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \ 3307 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n" 3308 " use path to provide path to a character device; default is /dev/tpm0\n" 3309 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n" 3310 " not provided it will be searched for in /sys/class/misc/tpm?/device\n" 3311 "-tpmdev emulator,id=id,chardev=dev\n" 3312 " configure the TPM device using chardev backend\n", 3313 QEMU_ARCH_ALL) 3314SRST 3315The general form of a TPM device option is: 3316 3317``-tpmdev backend,id=id[,options]`` 3318 The specific backend type will determine the applicable options. The 3319 ``-tpmdev`` option creates the TPM backend and requires a 3320 ``-device`` option that specifies the TPM frontend interface model. 3321 3322 Use ``-tpmdev help`` to print all available TPM backend types. 3323 3324The available backends are: 3325 3326``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path`` 3327 (Linux-host only) Enable access to the host's TPM using the 3328 passthrough driver. 3329 3330 ``path`` specifies the path to the host's TPM device, i.e., on a 3331 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by 3332 default ``/dev/tpm0`` is used. 3333 3334 ``cancel-path`` specifies the path to the host TPM device's sysfs 3335 entry allowing for cancellation of an ongoing TPM command. 3336 ``cancel-path`` is optional and by default QEMU will search for the 3337 sysfs entry to use. 3338 3339 Some notes about using the host's TPM with the passthrough driver: 3340 3341 The TPM device accessed by the passthrough driver must not be used 3342 by any other application on the host. 3343 3344 Since the host's firmware (BIOS/UEFI) has already initialized the 3345 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize 3346 the TPM again and may therefore not show a TPM-specific menu that 3347 would otherwise allow the user to configure the TPM, e.g., allow the 3348 user to enable/disable or activate/deactivate the TPM. Further, if 3349 TPM ownership is released from within a VM then the host's TPM will 3350 get disabled and deactivated. To enable and activate the TPM again 3351 afterwards, the host has to be rebooted and the user is required to 3352 enter the firmware's menu to enable and activate the TPM. If the TPM 3353 is left disabled and/or deactivated most TPM commands will fail. 3354 3355 To create a passthrough TPM use the following two options: 3356 3357 :: 3358 3359 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0 3360 3361 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by 3362 ``tpmdev=tpm0`` in the device option. 3363 3364``-tpmdev emulator,id=id,chardev=dev`` 3365 (Linux-host only) Enable access to a TPM emulator using Unix domain 3366 socket based chardev backend. 3367 3368 ``chardev`` specifies the unique ID of a character device backend 3369 that provides connection to the software TPM server. 3370 3371 To create a TPM emulator backend device with chardev socket backend: 3372 3373 :: 3374 3375 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0 3376ERST 3377 3378DEFHEADING() 3379 3380#endif 3381 3382DEFHEADING(Linux/Multiboot boot specific:) 3383SRST 3384When using these options, you can use a given Linux or Multiboot kernel 3385without installing it in the disk image. It can be useful for easier 3386testing of various kernels. 3387 3388 3389ERST 3390 3391DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \ 3392 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL) 3393SRST 3394``-kernel bzImage`` 3395 Use bzImage as kernel image. The kernel can be either a Linux kernel 3396 or in multiboot format. 3397ERST 3398 3399DEF("append", HAS_ARG, QEMU_OPTION_append, \ 3400 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL) 3401SRST 3402``-append cmdline`` 3403 Use cmdline as kernel command line 3404ERST 3405 3406DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \ 3407 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL) 3408SRST 3409``-initrd file`` 3410 Use file as initial ram disk. 3411 3412``-initrd "file1 arg=foo,file2"`` 3413 This syntax is only available with multiboot. 3414 3415 Use file1 and file2 as modules and pass arg=foo as parameter to the 3416 first module. 3417ERST 3418 3419DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \ 3420 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL) 3421SRST 3422``-dtb file`` 3423 Use file as a device tree binary (dtb) image and pass it to the 3424 kernel on boot. 3425ERST 3426 3427DEFHEADING() 3428 3429DEFHEADING(Debug/Expert options:) 3430 3431DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg, 3432 "-fw_cfg [name=]<name>,file=<file>\n" 3433 " add named fw_cfg entry with contents from file\n" 3434 "-fw_cfg [name=]<name>,string=<str>\n" 3435 " add named fw_cfg entry with contents from string\n", 3436 QEMU_ARCH_ALL) 3437SRST 3438``-fw_cfg [name=]name,file=file`` 3439 Add named fw\_cfg entry with contents from file file. 3440 3441``-fw_cfg [name=]name,string=str`` 3442 Add named fw\_cfg entry with contents from string str. 3443 3444 The terminating NUL character of the contents of str will not be 3445 included as part of the fw\_cfg item data. To insert contents with 3446 embedded NUL characters, you have to use the file parameter. 3447 3448 The fw\_cfg entries are passed by QEMU through to the guest. 3449 3450 Example: 3451 3452 :: 3453 3454 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin 3455 3456 creates an fw\_cfg entry named opt/com.mycompany/blob with contents 3457 from ./my\_blob.bin. 3458ERST 3459 3460DEF("serial", HAS_ARG, QEMU_OPTION_serial, \ 3461 "-serial dev redirect the serial port to char device 'dev'\n", 3462 QEMU_ARCH_ALL) 3463SRST 3464``-serial dev`` 3465 Redirect the virtual serial port to host character device dev. The 3466 default device is ``vc`` in graphical mode and ``stdio`` in non 3467 graphical mode. 3468 3469 This option can be used several times to simulate up to 4 serial 3470 ports. 3471 3472 Use ``-serial none`` to disable all serial ports. 3473 3474 Available character devices are: 3475 3476 ``vc[:WxH]`` 3477 Virtual console. Optionally, a width and height can be given in 3478 pixel with 3479 3480 :: 3481 3482 vc:800x600 3483 3484 It is also possible to specify width or height in characters: 3485 3486 :: 3487 3488 vc:80Cx24C 3489 3490 ``pty`` 3491 [Linux only] Pseudo TTY (a new PTY is automatically allocated) 3492 3493 ``none`` 3494 No device is allocated. 3495 3496 ``null`` 3497 void device 3498 3499 ``chardev:id`` 3500 Use a named character device defined with the ``-chardev`` 3501 option. 3502 3503 ``/dev/XXX`` 3504 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial 3505 port parameters are set according to the emulated ones. 3506 3507 ``/dev/parportN`` 3508 [Linux only, parallel port only] Use host parallel port N. 3509 Currently SPP and EPP parallel port features can be used. 3510 3511 ``file:filename`` 3512 Write output to filename. No character can be read. 3513 3514 ``stdio`` 3515 [Unix only] standard input/output 3516 3517 ``pipe:filename`` 3518 name pipe filename 3519 3520 ``COMn`` 3521 [Windows only] Use host serial port n 3522 3523 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]`` 3524 This implements UDP Net Console. When remote\_host or src\_ip 3525 are not specified they default to ``0.0.0.0``. When not using a 3526 specified src\_port a random port is automatically chosen. 3527 3528 If you just want a simple readonly console you can use 3529 ``netcat`` or ``nc``, by starting QEMU with: 3530 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time 3531 QEMU writes something to that port it will appear in the 3532 netconsole session. 3533 3534 If you plan to send characters back via netconsole or you want 3535 to stop and start QEMU a lot of times, you should have QEMU use 3536 the same source port each time by using something like ``-serial 3537 udp::4555@:4556`` to QEMU. Another approach is to use a patched 3538 version of netcat which can listen to a TCP port and send and 3539 receive characters via udp. If you have a patched version of 3540 netcat which activates telnet remote echo and single char 3541 transfer, then you can use the following options to set up a 3542 netcat redirector to allow telnet on port 5555 to access the 3543 QEMU port. 3544 3545 ``QEMU Options:`` 3546 -serial udp::4555@:4556 3547 3548 ``netcat options:`` 3549 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T 3550 3551 ``telnet options:`` 3552 localhost 5555 3553 3554 ``tcp:[host]:port[,server][,nowait][,nodelay][,reconnect=seconds]`` 3555 The TCP Net Console has two modes of operation. It can send the 3556 serial I/O to a location or wait for a connection from a 3557 location. By default the TCP Net Console is sent to host at the 3558 port. If you use the server option QEMU will wait for a client 3559 socket application to connect to the port before continuing, 3560 unless the ``nowait`` option was specified. The ``nodelay`` 3561 option disables the Nagle buffering algorithm. The ``reconnect`` 3562 option only applies if noserver is set, if the connection goes 3563 down it will attempt to reconnect at the given interval. If host 3564 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a 3565 time is accepted. You can use ``telnet`` to connect to the 3566 corresponding character device. 3567 3568 ``Example to send tcp console to 192.168.0.2 port 4444`` 3569 -serial tcp:192.168.0.2:4444 3570 3571 ``Example to listen and wait on port 4444 for connection`` 3572 -serial tcp::4444,server 3573 3574 ``Example to not wait and listen on ip 192.168.0.100 port 4444`` 3575 -serial tcp:192.168.0.100:4444,server,nowait 3576 3577 ``telnet:host:port[,server][,nowait][,nodelay]`` 3578 The telnet protocol is used instead of raw tcp sockets. The 3579 options work the same as if you had specified ``-serial tcp``. 3580 The difference is that the port acts like a telnet server or 3581 client using telnet option negotiation. This will also allow you 3582 to send the MAGIC\_SYSRQ sequence if you use a telnet that 3583 supports sending the break sequence. Typically in unix telnet 3584 you do it with Control-] and then type "send break" followed by 3585 pressing the enter key. 3586 3587 ``websocket:host:port,server[,nowait][,nodelay]`` 3588 The WebSocket protocol is used instead of raw tcp socket. The 3589 port acts as a WebSocket server. Client mode is not supported. 3590 3591 ``unix:path[,server][,nowait][,reconnect=seconds]`` 3592 A unix domain socket is used instead of a tcp socket. The option 3593 works the same as if you had specified ``-serial tcp`` except 3594 the unix domain socket path is used for connections. 3595 3596 ``mon:dev_string`` 3597 This is a special option to allow the monitor to be multiplexed 3598 onto another serial port. The monitor is accessed with key 3599 sequence of Control-a and then pressing c. dev\_string should be 3600 any one of the serial devices specified above. An example to 3601 multiplex the monitor onto a telnet server listening on port 3602 4444 would be: 3603 3604 ``-serial mon:telnet::4444,server,nowait`` 3605 3606 When the monitor is multiplexed to stdio in this way, Ctrl+C 3607 will not terminate QEMU any more but will be passed to the guest 3608 instead. 3609 3610 ``braille`` 3611 Braille device. This will use BrlAPI to display the braille 3612 output on a real or fake device. 3613 3614 ``msmouse`` 3615 Three button serial mouse. Configure the guest to use Microsoft 3616 protocol. 3617ERST 3618 3619DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \ 3620 "-parallel dev redirect the parallel port to char device 'dev'\n", 3621 QEMU_ARCH_ALL) 3622SRST 3623``-parallel dev`` 3624 Redirect the virtual parallel port to host device dev (same devices 3625 as the serial port). On Linux hosts, ``/dev/parportN`` can be used 3626 to use hardware devices connected on the corresponding host parallel 3627 port. 3628 3629 This option can be used several times to simulate up to 3 parallel 3630 ports. 3631 3632 Use ``-parallel none`` to disable all parallel ports. 3633ERST 3634 3635DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \ 3636 "-monitor dev redirect the monitor to char device 'dev'\n", 3637 QEMU_ARCH_ALL) 3638SRST 3639``-monitor dev`` 3640 Redirect the monitor to host device dev (same devices as the serial 3641 port). The default device is ``vc`` in graphical mode and ``stdio`` 3642 in non graphical mode. Use ``-monitor none`` to disable the default 3643 monitor. 3644ERST 3645DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \ 3646 "-qmp dev like -monitor but opens in 'control' mode\n", 3647 QEMU_ARCH_ALL) 3648SRST 3649``-qmp dev`` 3650 Like -monitor but opens in 'control' mode. 3651ERST 3652DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \ 3653 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n", 3654 QEMU_ARCH_ALL) 3655SRST 3656``-qmp-pretty dev`` 3657 Like -qmp but uses pretty JSON formatting. 3658ERST 3659 3660DEF("mon", HAS_ARG, QEMU_OPTION_mon, \ 3661 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL) 3662SRST 3663``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]`` 3664 Setup monitor on chardev name. ``pretty`` turns on JSON pretty 3665 printing easing human reading and debugging. 3666ERST 3667 3668DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \ 3669 "-debugcon dev redirect the debug console to char device 'dev'\n", 3670 QEMU_ARCH_ALL) 3671SRST 3672``-debugcon dev`` 3673 Redirect the debug console to host device dev (same devices as the 3674 serial port). The debug console is an I/O port which is typically 3675 port 0xe9; writing to that I/O port sends output to this device. The 3676 default device is ``vc`` in graphical mode and ``stdio`` in non 3677 graphical mode. 3678ERST 3679 3680DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \ 3681 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL) 3682SRST 3683``-pidfile file`` 3684 Store the QEMU process PID in file. It is useful if you launch QEMU 3685 from a script. 3686ERST 3687 3688DEF("singlestep", 0, QEMU_OPTION_singlestep, \ 3689 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL) 3690SRST 3691``-singlestep`` 3692 Run the emulation in single step mode. 3693ERST 3694 3695DEF("preconfig", 0, QEMU_OPTION_preconfig, \ 3696 "--preconfig pause QEMU before machine is initialized (experimental)\n", 3697 QEMU_ARCH_ALL) 3698SRST 3699``--preconfig`` 3700 Pause QEMU for interactive configuration before the machine is 3701 created, which allows querying and configuring properties that will 3702 affect machine initialization. Use QMP command 'x-exit-preconfig' to 3703 exit the preconfig state and move to the next state (i.e. run guest 3704 if -S isn't used or pause the second time if -S is used). This 3705 option is experimental. 3706ERST 3707 3708DEF("S", 0, QEMU_OPTION_S, \ 3709 "-S freeze CPU at startup (use 'c' to start execution)\n", 3710 QEMU_ARCH_ALL) 3711SRST 3712``-S`` 3713 Do not start CPU at startup (you must type 'c' in the monitor). 3714ERST 3715 3716DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit, 3717 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n" 3718 " run qemu with overcommit hints\n" 3719 " mem-lock=on|off controls memory lock support (default: off)\n" 3720 " cpu-pm=on|off controls cpu power management (default: off)\n", 3721 QEMU_ARCH_ALL) 3722SRST 3723``-overcommit mem-lock=on|off`` 3724 \ 3725``-overcommit cpu-pm=on|off`` 3726 Run qemu with hints about host resource overcommit. The default is 3727 to assume that host overcommits all resources. 3728 3729 Locking qemu and guest memory can be enabled via ``mem-lock=on`` 3730 (disabled by default). This works when host memory is not 3731 overcommitted and reduces the worst-case latency for guest. 3732 3733 Guest ability to manage power state of host cpus (increasing latency 3734 for other processes on the same host cpu, but decreasing latency for 3735 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This 3736 works best when host CPU is not overcommitted. When used, host 3737 estimates of CPU cycle and power utilization will be incorrect, not 3738 taking into account guest idle time. 3739ERST 3740 3741DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \ 3742 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n" 3743 " the guest without waiting for gdb to connect; use -S too\n" 3744 " if you want it to not start execution.)\n", 3745 QEMU_ARCH_ALL) 3746SRST 3747``-gdb dev`` 3748 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter 3749 in the System Emulation Users Guide). Note that this option does not pause QEMU 3750 execution -- if you want QEMU to not start the guest until you 3751 connect with gdb and issue a ``continue`` command, you will need to 3752 also pass the ``-S`` option to QEMU. 3753 3754 The most usual configuration is to listen on a local TCP socket:: 3755 3756 -gdb tcp::3117 3757 3758 but you can specify other backends; UDP, pseudo TTY, or even stdio 3759 are all reasonable use cases. For example, a stdio connection 3760 allows you to start QEMU from within gdb and establish the 3761 connection via a pipe: 3762 3763 .. parsed-literal:: 3764 3765 (gdb) target remote | exec |qemu_system| -gdb stdio ... 3766ERST 3767 3768DEF("s", 0, QEMU_OPTION_s, \ 3769 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n", 3770 QEMU_ARCH_ALL) 3771SRST 3772``-s`` 3773 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234 3774 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide). 3775ERST 3776 3777DEF("d", HAS_ARG, QEMU_OPTION_d, \ 3778 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n", 3779 QEMU_ARCH_ALL) 3780SRST 3781``-d item1[,...]`` 3782 Enable logging of specified items. Use '-d help' for a list of log 3783 items. 3784ERST 3785 3786DEF("D", HAS_ARG, QEMU_OPTION_D, \ 3787 "-D logfile output log to logfile (default stderr)\n", 3788 QEMU_ARCH_ALL) 3789SRST 3790``-D logfile`` 3791 Output log in logfile instead of to stderr 3792ERST 3793 3794DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \ 3795 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n", 3796 QEMU_ARCH_ALL) 3797SRST 3798``-dfilter range1[,...]`` 3799 Filter debug output to that relevant to a range of target addresses. 3800 The filter spec can be either start+size, start-size or start..end 3801 where start end and size are the addresses and sizes required. For 3802 example: 3803 3804 :: 3805 3806 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000 3807 3808 Will dump output for any code in the 0x1000 sized block starting at 3809 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and 3810 another 0x1000 sized block starting at 0xffffffc00005f000. 3811ERST 3812 3813DEF("seed", HAS_ARG, QEMU_OPTION_seed, \ 3814 "-seed number seed the pseudo-random number generator\n", 3815 QEMU_ARCH_ALL) 3816SRST 3817``-seed number`` 3818 Force the guest to use a deterministic pseudo-random number 3819 generator, seeded with number. This does not affect crypto routines 3820 within the host. 3821ERST 3822 3823DEF("L", HAS_ARG, QEMU_OPTION_L, \ 3824 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n", 3825 QEMU_ARCH_ALL) 3826SRST 3827``-L path`` 3828 Set the directory for the BIOS, VGA BIOS and keymaps. 3829 3830 To list all the data directories, use ``-L help``. 3831ERST 3832 3833DEF("bios", HAS_ARG, QEMU_OPTION_bios, \ 3834 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL) 3835SRST 3836``-bios file`` 3837 Set the filename for the BIOS. 3838ERST 3839 3840DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \ 3841 "-enable-kvm enable KVM full virtualization support\n", QEMU_ARCH_ALL) 3842SRST 3843``-enable-kvm`` 3844 Enable KVM full virtualization support. This option is only 3845 available if KVM support is enabled when compiling. 3846ERST 3847 3848DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, 3849 "-xen-domid id specify xen guest domain id\n", QEMU_ARCH_ALL) 3850DEF("xen-attach", 0, QEMU_OPTION_xen_attach, 3851 "-xen-attach attach to existing xen domain\n" 3852 " libxl will use this when starting QEMU\n", 3853 QEMU_ARCH_ALL) 3854DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, 3855 "-xen-domid-restrict restrict set of available xen operations\n" 3856 " to specified domain id. (Does not affect\n" 3857 " xenpv machine type).\n", 3858 QEMU_ARCH_ALL) 3859SRST 3860``-xen-domid id`` 3861 Specify xen guest domain id (XEN only). 3862 3863``-xen-attach`` 3864 Attach to existing xen domain. libxl will use this when starting 3865 QEMU (XEN only). Restrict set of available xen operations to 3866 specified domain id (XEN only). 3867ERST 3868 3869DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \ 3870 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL) 3871SRST 3872``-no-reboot`` 3873 Exit instead of rebooting. 3874ERST 3875 3876DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \ 3877 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL) 3878SRST 3879``-no-shutdown`` 3880 Don't exit QEMU on guest shutdown, but instead only stop the 3881 emulation. This allows for instance switching to monitor to commit 3882 changes to the disk image. 3883ERST 3884 3885DEF("action", HAS_ARG, QEMU_OPTION_action, 3886 "-action reboot=none|shutdown\n" 3887 " action when guest reboots [default=none]\n" 3888 "-action shutdown=poweroff|pause\n" 3889 " action when guest shuts down [default=poweroff]\n" 3890 "-action panic=poweroff|pause|none\n" 3891 " action when guest panics [default=poweroff]\n" 3892 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" 3893 " action when watchdog fires [default=reset]\n", 3894 QEMU_ARCH_ALL) 3895SRST 3896``-action event=action`` 3897 The action parameter serves to modify QEMU's default behavior when 3898 certain guest events occur. It provides a generic method for specifying the 3899 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown`` 3900 parameters. 3901 3902 Examples: 3903 3904 ``-action panic=none`` 3905 ``-action reboot=shutdown,shutdown=pause`` 3906 ``-watchdog i6300esb -action watchdog=pause`` 3907 3908ERST 3909 3910DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \ 3911 "-loadvm [tag|id]\n" \ 3912 " start right away with a saved state (loadvm in monitor)\n", 3913 QEMU_ARCH_ALL) 3914SRST 3915``-loadvm file`` 3916 Start right away with a saved state (``loadvm`` in monitor) 3917ERST 3918 3919#ifndef _WIN32 3920DEF("daemonize", 0, QEMU_OPTION_daemonize, \ 3921 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL) 3922#endif 3923SRST 3924``-daemonize`` 3925 Daemonize the QEMU process after initialization. QEMU will not 3926 detach from standard IO until it is ready to receive connections on 3927 any of its devices. This option is a useful way for external 3928 programs to launch QEMU without having to cope with initialization 3929 race conditions. 3930ERST 3931 3932DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \ 3933 "-option-rom rom load a file, rom, into the option ROM space\n", 3934 QEMU_ARCH_ALL) 3935SRST 3936``-option-rom file`` 3937 Load the contents of file as an option ROM. This option is useful to 3938 load things like EtherBoot. 3939ERST 3940 3941DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \ 3942 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \ 3943 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n", 3944 QEMU_ARCH_ALL) 3945 3946SRST 3947``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]`` 3948 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at 3949 the current UTC or local time, respectively. ``localtime`` is 3950 required for correct date in MS-DOS or Windows. To start at a 3951 specific point in time, provide datetime in the format 3952 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC. 3953 3954 By default the RTC is driven by the host system time. This allows 3955 using of the RTC as accurate reference clock inside the guest, 3956 specifically if the host time is smoothly following an accurate 3957 external reference clock, e.g. via NTP. If you want to isolate the 3958 guest time from the host, you can set ``clock`` to ``rt`` instead, 3959 which provides a host monotonic clock if host support it. To even 3960 prevent the RTC from progressing during suspension, you can set 3961 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is 3962 recommended especially in icount mode in order to preserve 3963 determinism; however, note that in icount mode the speed of the 3964 virtual clock is variable and can in general differ from the host 3965 clock. 3966 3967 Enable ``driftfix`` (i386 targets only) if you experience time drift 3968 problems, specifically with Windows' ACPI HAL. This option will try 3969 to figure out how many timer interrupts were not processed by the 3970 Windows guest and will re-inject them. 3971ERST 3972 3973DEF("icount", HAS_ARG, QEMU_OPTION_icount, \ 3974 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \ 3975 " enable virtual instruction counter with 2^N clock ticks per\n" \ 3976 " instruction, enable aligning the host and virtual clocks\n" \ 3977 " or disable real time cpu sleeping, and optionally enable\n" \ 3978 " record-and-replay mode\n", QEMU_ARCH_ALL) 3979SRST 3980``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]`` 3981 Enable virtual instruction counter. The virtual cpu will execute one 3982 instruction every 2^N ns of virtual time. If ``auto`` is specified 3983 then the virtual cpu speed will be automatically adjusted to keep 3984 virtual time within a few seconds of real time. 3985 3986 Note that while this option can give deterministic behavior, it does 3987 not provide cycle accurate emulation. Modern CPUs contain 3988 superscalar out of order cores with complex cache hierarchies. The 3989 number of instructions executed often has little or no correlation 3990 with actual performance. 3991 3992 When the virtual cpu is sleeping, the virtual time will advance at 3993 default speed unless ``sleep=on`` is specified. With 3994 ``sleep=on``, the virtual time will jump to the next timer 3995 deadline instantly whenever the virtual cpu goes to sleep mode and 3996 will not advance if no timer is enabled. This behavior gives 3997 deterministic execution times from the guest point of view. 3998 The default if icount is enabled is ``sleep=off``. 3999 ``sleep=on`` cannot be used together with either ``shift=auto`` 4000 or ``align=on``. 4001 4002 ``align=on`` will activate the delay algorithm which will try to 4003 synchronise the host clock and the virtual clock. The goal is to 4004 have a guest running at the real frequency imposed by the shift 4005 option. Whenever the guest clock is behind the host clock and if 4006 ``align=on`` is specified then we print a message to the user to 4007 inform about the delay. Currently this option does not work when 4008 ``shift`` is ``auto``. Note: The sync algorithm will work for those 4009 shift values for which the guest clock runs ahead of the host clock. 4010 Typically this happens when the shift value is high (how high 4011 depends on the host machine). The default if icount is enabled 4012 is ``align=off``. 4013 4014 When the ``rr`` option is specified deterministic record/replay is 4015 enabled. The ``rrfile=`` option must also be provided to 4016 specify the path to the replay log. In record mode data is written 4017 to this file, and in replay mode it is read back. 4018 If the ``rrsnapshot`` option is given then it specifies a VM snapshot 4019 name. In record mode, a new VM snapshot with the given name is created 4020 at the start of execution recording. In replay mode this option 4021 specifies the snapshot name used to load the initial VM state. 4022ERST 4023 4024DEF("watchdog", HAS_ARG, QEMU_OPTION_watchdog, \ 4025 "-watchdog model\n" \ 4026 " enable virtual hardware watchdog [default=none]\n", 4027 QEMU_ARCH_ALL) 4028SRST 4029``-watchdog model`` 4030 Create a virtual hardware watchdog device. Once enabled (by a guest 4031 action), the watchdog must be periodically polled by an agent inside 4032 the guest or else the guest will be restarted. Choose a model for 4033 which your guest has drivers. 4034 4035 The model is the model of hardware watchdog to emulate. Use 4036 ``-watchdog help`` to list available hardware models. Only one 4037 watchdog can be enabled for a guest. 4038 4039 The following models may be available: 4040 4041 ``ib700`` 4042 iBASE 700 is a very simple ISA watchdog with a single timer. 4043 4044 ``i6300esb`` 4045 Intel 6300ESB I/O controller hub is a much more featureful 4046 PCI-based dual-timer watchdog. 4047 4048 ``diag288`` 4049 A virtual watchdog for s390x backed by the diagnose 288 4050 hypercall (currently KVM only). 4051ERST 4052 4053DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \ 4054 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \ 4055 " action when watchdog fires [default=reset]\n", 4056 QEMU_ARCH_ALL) 4057SRST 4058``-watchdog-action action`` 4059 The action controls what QEMU will do when the watchdog timer 4060 expires. The default is ``reset`` (forcefully reset the guest). 4061 Other possible actions are: ``shutdown`` (attempt to gracefully 4062 shutdown the guest), ``poweroff`` (forcefully poweroff the guest), 4063 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the 4064 guest), ``debug`` (print a debug message and continue), or ``none`` 4065 (do nothing). 4066 4067 Note that the ``shutdown`` action requires that the guest responds 4068 to ACPI signals, which it may not be able to do in the sort of 4069 situations where the watchdog would have expired, and thus 4070 ``-watchdog-action shutdown`` is not recommended for production use. 4071 4072 Examples: 4073 4074 ``-watchdog i6300esb -watchdog-action pause``; \ ``-watchdog ib700`` 4075 4076ERST 4077 4078DEF("echr", HAS_ARG, QEMU_OPTION_echr, \ 4079 "-echr chr set terminal escape character instead of ctrl-a\n", 4080 QEMU_ARCH_ALL) 4081SRST 4082``-echr numeric_ascii_value`` 4083 Change the escape character used for switching to the monitor when 4084 using monitor and serial sharing. The default is ``0x01`` when using 4085 the ``-nographic`` option. ``0x01`` is equal to pressing 4086 ``Control-a``. You can select a different character from the ascii 4087 control keys where 1 through 26 map to Control-a through Control-z. 4088 For instance you could use the either of the following to change the 4089 escape character to Control-t. 4090 4091 ``-echr 0x14``; \ ``-echr 20`` 4092 4093ERST 4094 4095DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \ 4096 "-incoming tcp:[host]:port[,to=maxport][,ipv4][,ipv6]\n" \ 4097 "-incoming rdma:host:port[,ipv4][,ipv6]\n" \ 4098 "-incoming unix:socketpath\n" \ 4099 " prepare for incoming migration, listen on\n" \ 4100 " specified protocol and socket address\n" \ 4101 "-incoming fd:fd\n" \ 4102 "-incoming exec:cmdline\n" \ 4103 " accept incoming migration on given file descriptor\n" \ 4104 " or from given external command\n" \ 4105 "-incoming defer\n" \ 4106 " wait for the URI to be specified via migrate_incoming\n", 4107 QEMU_ARCH_ALL) 4108SRST 4109``-incoming tcp:[host]:port[,to=maxport][,ipv4][,ipv6]`` 4110 \ 4111``-incoming rdma:host:port[,ipv4][,ipv6]`` 4112 Prepare for incoming migration, listen on a given tcp port. 4113 4114``-incoming unix:socketpath`` 4115 Prepare for incoming migration, listen on a given unix socket. 4116 4117``-incoming fd:fd`` 4118 Accept incoming migration from a given filedescriptor. 4119 4120``-incoming exec:cmdline`` 4121 Accept incoming migration as an output from specified external 4122 command. 4123 4124``-incoming defer`` 4125 Wait for the URI to be specified via migrate\_incoming. The monitor 4126 can be used to change settings (such as migration parameters) prior 4127 to issuing the migrate\_incoming to allow the migration to begin. 4128ERST 4129 4130DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \ 4131 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL) 4132SRST 4133``-only-migratable`` 4134 Only allow migratable devices. Devices will not be allowed to enter 4135 an unmigratable state. 4136ERST 4137 4138DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \ 4139 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL) 4140SRST 4141``-nodefaults`` 4142 Don't create default devices. Normally, QEMU sets the default 4143 devices like serial port, parallel port, virtual console, monitor 4144 device, VGA adapter, floppy and CD-ROM drive and others. The 4145 ``-nodefaults`` option will disable all those default devices. 4146ERST 4147 4148#ifndef _WIN32 4149DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \ 4150 "-chroot dir chroot to dir just before starting the VM\n", 4151 QEMU_ARCH_ALL) 4152#endif 4153SRST 4154``-chroot dir`` 4155 Immediately before starting guest execution, chroot to the specified 4156 directory. Especially useful in combination with -runas. 4157ERST 4158 4159#ifndef _WIN32 4160DEF("runas", HAS_ARG, QEMU_OPTION_runas, \ 4161 "-runas user change to user id user just before starting the VM\n" \ 4162 " user can be numeric uid:gid instead\n", 4163 QEMU_ARCH_ALL) 4164#endif 4165SRST 4166``-runas user`` 4167 Immediately before starting guest execution, drop root privileges, 4168 switching to the specified user. 4169ERST 4170 4171DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env, 4172 "-prom-env variable=value\n" 4173 " set OpenBIOS nvram variables\n", 4174 QEMU_ARCH_PPC | QEMU_ARCH_SPARC) 4175SRST 4176``-prom-env variable=value`` 4177 Set OpenBIOS nvram variable to given value (PPC, SPARC only). 4178 4179 :: 4180 4181 qemu-system-sparc -prom-env 'auto-boot?=false' \ 4182 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single' 4183 4184 :: 4185 4186 qemu-system-ppc -prom-env 'auto-boot?=false' \ 4187 -prom-env 'boot-device=hd:2,\yaboot' \ 4188 -prom-env 'boot-args=conf=hd:2,\yaboot.conf' 4189ERST 4190DEF("semihosting", 0, QEMU_OPTION_semihosting, 4191 "-semihosting semihosting mode\n", 4192 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | QEMU_ARCH_LM32 | 4193 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2) 4194SRST 4195``-semihosting`` 4196 Enable semihosting mode (ARM, M68K, Xtensa, MIPS, Nios II only). 4197 4198 Note that this allows guest direct access to the host filesystem, so 4199 should only be used with a trusted guest OS. 4200 4201 See the -semihosting-config option documentation for further 4202 information about the facilities this enables. 4203ERST 4204DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config, 4205 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]\n" \ 4206 " semihosting configuration\n", 4207QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | QEMU_ARCH_LM32 | 4208QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2) 4209SRST 4210``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]`` 4211 Enable and configure semihosting (ARM, M68K, Xtensa, MIPS, Nios II 4212 only). 4213 4214 Note that this allows guest direct access to the host filesystem, so 4215 should only be used with a trusted guest OS. 4216 4217 On Arm this implements the standard semihosting API, version 2.0. 4218 4219 On M68K this implements the "ColdFire GDB" interface used by 4220 libgloss. 4221 4222 Xtensa semihosting provides basic file IO calls, such as 4223 open/read/write/seek/select. Tensilica baremetal libc for ISS and 4224 linux platform "sim" use this interface. 4225 4226 ``target=native|gdb|auto`` 4227 Defines where the semihosting calls will be addressed, to QEMU 4228 (``native``) or to GDB (``gdb``). The default is ``auto``, which 4229 means ``gdb`` during debug sessions and ``native`` otherwise. 4230 4231 ``chardev=str1`` 4232 Send the output to a chardev backend output for native or auto 4233 output when not in gdb 4234 4235 ``arg=str1,arg=str2,...`` 4236 Allows the user to pass input arguments, and can be used 4237 multiple times to build up a list. The old-style 4238 ``-kernel``/``-append`` method of passing a command line is 4239 still supported for backward compatibility. If both the 4240 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are 4241 specified, the former is passed to semihosting as it always 4242 takes precedence. 4243ERST 4244DEF("old-param", 0, QEMU_OPTION_old_param, 4245 "-old-param old param mode\n", QEMU_ARCH_ARM) 4246SRST 4247``-old-param`` 4248 Old param mode (ARM only). 4249ERST 4250 4251DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \ 4252 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \ 4253 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \ 4254 " Enable seccomp mode 2 system call filter (default 'off').\n" \ 4255 " use 'obsolete' to allow obsolete system calls that are provided\n" \ 4256 " by the kernel, but typically no longer used by modern\n" \ 4257 " C library implementations.\n" \ 4258 " use 'elevateprivileges' to allow or deny QEMU process to elevate\n" \ 4259 " its privileges by blacklisting all set*uid|gid system calls.\n" \ 4260 " The value 'children' will deny set*uid|gid system calls for\n" \ 4261 " main QEMU process but will allow forks and execves to run unprivileged\n" \ 4262 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \ 4263 " blacklisting *fork and execve\n" \ 4264 " use 'resourcecontrol' to disable process affinity and schedular priority\n", 4265 QEMU_ARCH_ALL) 4266SRST 4267``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]`` 4268 Enable Seccomp mode 2 system call filter. 'on' will enable syscall 4269 filtering and 'off' will disable it. The default is 'off'. 4270 4271 ``obsolete=string`` 4272 Enable Obsolete system calls 4273 4274 ``elevateprivileges=string`` 4275 Disable set\*uid\|gid system calls 4276 4277 ``spawn=string`` 4278 Disable \*fork and execve 4279 4280 ``resourcecontrol=string`` 4281 Disable process affinity and schedular priority 4282ERST 4283 4284DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig, 4285 "-readconfig <file>\n", QEMU_ARCH_ALL) 4286SRST 4287``-readconfig file`` 4288 Read device configuration from file. This approach is useful when 4289 you want to spawn QEMU process with many command line options but 4290 you don't want to exceed the command line character limit. 4291ERST 4292DEF("writeconfig", HAS_ARG, QEMU_OPTION_writeconfig, 4293 "-writeconfig <file>\n" 4294 " read/write config file\n", QEMU_ARCH_ALL) 4295SRST 4296``-writeconfig file`` 4297 Write device configuration to file. The file can be either filename 4298 to save command line and device configuration into file or dash 4299 ``-``) character to print the output to stdout. This can be later 4300 used as input file for ``-readconfig`` option. 4301ERST 4302 4303DEF("no-user-config", 0, QEMU_OPTION_nouserconfig, 4304 "-no-user-config\n" 4305 " do not load default user-provided config files at startup\n", 4306 QEMU_ARCH_ALL) 4307SRST 4308``-no-user-config`` 4309 The ``-no-user-config`` option makes QEMU not load any of the 4310 user-provided config files on sysconfdir. 4311ERST 4312 4313DEF("trace", HAS_ARG, QEMU_OPTION_trace, 4314 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n" 4315 " specify tracing options\n", 4316 QEMU_ARCH_ALL) 4317SRST 4318``-trace [[enable=]pattern][,events=file][,file=file]`` 4319 .. include:: ../qemu-option-trace.rst.inc 4320 4321ERST 4322DEF("plugin", HAS_ARG, QEMU_OPTION_plugin, 4323 "-plugin [file=]<file>[,arg=<string>]\n" 4324 " load a plugin\n", 4325 QEMU_ARCH_ALL) 4326SRST 4327``-plugin file=file[,arg=string]`` 4328 Load a plugin. 4329 4330 ``file=file`` 4331 Load the given plugin from a shared library file. 4332 4333 ``arg=string`` 4334 Argument string passed to the plugin. (Can be given multiple 4335 times.) 4336ERST 4337 4338HXCOMM Internal use 4339DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL) 4340DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL) 4341 4342#ifdef __linux__ 4343DEF("enable-fips", 0, QEMU_OPTION_enablefips, 4344 "-enable-fips enable FIPS 140-2 compliance\n", 4345 QEMU_ARCH_ALL) 4346#endif 4347SRST 4348``-enable-fips`` 4349 Enable FIPS 140-2 compliance mode. 4350ERST 4351 4352DEF("msg", HAS_ARG, QEMU_OPTION_msg, 4353 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n" 4354 " control error message format\n" 4355 " timestamp=on enables timestamps (default: off)\n" 4356 " guest-name=on enables guest name prefix but only if\n" 4357 " -name guest option is set (default: off)\n", 4358 QEMU_ARCH_ALL) 4359SRST 4360``-msg [timestamp[=on|off]][,guest-name[=on|off]]`` 4361 Control error message format. 4362 4363 ``timestamp=on|off`` 4364 Prefix messages with a timestamp. Default is off. 4365 4366 ``guest-name=on|off`` 4367 Prefix messages with guest name but only if -name guest option is set 4368 otherwise the option is ignored. Default is off. 4369ERST 4370 4371DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate, 4372 "-dump-vmstate <file>\n" 4373 " Output vmstate information in JSON format to file.\n" 4374 " Use the scripts/vmstate-static-checker.py file to\n" 4375 " check for possible regressions in migration code\n" 4376 " by comparing two such vmstate dumps.\n", 4377 QEMU_ARCH_ALL) 4378SRST 4379``-dump-vmstate file`` 4380 Dump json-encoded vmstate information for current machine type to 4381 file in file 4382ERST 4383 4384DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile, 4385 "-enable-sync-profile\n" 4386 " enable synchronization profiling\n", 4387 QEMU_ARCH_ALL) 4388SRST 4389``-enable-sync-profile`` 4390 Enable synchronization profiling. 4391ERST 4392 4393DEFHEADING() 4394 4395DEFHEADING(Generic object creation:) 4396 4397DEF("object", HAS_ARG, QEMU_OPTION_object, 4398 "-object TYPENAME[,PROP1=VALUE1,...]\n" 4399 " create a new object of type TYPENAME setting properties\n" 4400 " in the order they are specified. Note that the 'id'\n" 4401 " property must be set. These objects are placed in the\n" 4402 " '/objects' path.\n", 4403 QEMU_ARCH_ALL) 4404SRST 4405``-object typename[,prop1=value1,...]`` 4406 Create a new object of type typename setting properties in the order 4407 they are specified. Note that the 'id' property must be set. These 4408 objects are placed in the '/objects' path. 4409 4410 ``-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`` 4411 Creates a memory file backend object, which can be used to back 4412 the guest RAM with huge pages. 4413 4414 The ``id`` parameter is a unique ID that will be used to 4415 reference this memory region when configuring the ``-numa`` 4416 argument. 4417 4418 The ``size`` option provides the size of the memory region, and 4419 accepts common suffixes, eg ``500M``. 4420 4421 The ``mem-path`` provides the path to either a shared memory or 4422 huge page filesystem mount. 4423 4424 The ``share`` boolean option determines whether the memory 4425 region is marked as private to QEMU, or shared. The latter 4426 allows a co-operating external process to access the QEMU memory 4427 region. 4428 4429 The ``share`` is also required for pvrdma devices due to 4430 limitations in the RDMA API provided by Linux. 4431 4432 Setting share=on might affect the ability to configure NUMA 4433 bindings for the memory backend under some circumstances, see 4434 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel 4435 source tree for additional details. 4436 4437 Setting the ``discard-data`` boolean option to on indicates that 4438 file contents can be destroyed when QEMU exits, to avoid 4439 unnecessarily flushing data to the backing file. Note that 4440 ``discard-data`` is only an optimization, and QEMU might not 4441 discard file contents if it aborts unexpectedly or is terminated 4442 using SIGKILL. 4443 4444 The ``merge`` boolean option enables memory merge, also known as 4445 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider 4446 the pages for memory deduplication. 4447 4448 Setting the ``dump`` boolean option to off excludes the memory 4449 from core dumps. This feature is also known as MADV\_DONTDUMP. 4450 4451 The ``prealloc`` boolean option enables memory preallocation. 4452 4453 The ``host-nodes`` option binds the memory range to a list of 4454 NUMA host nodes. 4455 4456 The ``policy`` option sets the NUMA policy to one of the 4457 following values: 4458 4459 ``default`` 4460 default host policy 4461 4462 ``preferred`` 4463 prefer the given host node list for allocation 4464 4465 ``bind`` 4466 restrict memory allocation to the given host node list 4467 4468 ``interleave`` 4469 interleave memory allocations across the given host node 4470 list 4471 4472 The ``align`` option specifies the base address alignment when 4473 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg 4474 ``2M``. Some backend store specified by ``mem-path`` requires an 4475 alignment different than the default one used by QEMU, eg the 4476 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In 4477 such cases, users can specify the required alignment via this 4478 option. 4479 4480 The ``pmem`` option specifies whether the backing file specified 4481 by ``mem-path`` is in host persistent memory that can be 4482 accessed using the SNIA NVM programming model (e.g. Intel 4483 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary 4484 operations to guarantee the persistence of its own writes to 4485 ``mem-path`` (e.g. in vNVDIMM label emulation and live 4486 migration). Also, we will map the backend-file with MAP\_SYNC 4487 flag, which ensures the file metadata is in sync for 4488 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC 4489 requires support from both the host kernel (since Linux kernel 4490 4.15) and the filesystem of ``mem-path`` mounted with DAX 4491 option. 4492 4493 ``-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`` 4494 Creates a memory backend object, which can be used to back the 4495 guest RAM. Memory backend objects offer more control than the 4496 ``-m`` option that is traditionally used to define guest RAM. 4497 Please refer to ``memory-backend-file`` for a description of the 4498 options. 4499 4500 ``-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`` 4501 Creates an anonymous memory file backend object, which allows 4502 QEMU to share the memory with an external process (e.g. when 4503 using vhost-user). The memory is allocated with memfd and 4504 optional sealing. (Linux only) 4505 4506 The ``seal`` option creates a sealed-file, that will block 4507 further resizing the memory ('on' by default). 4508 4509 The ``hugetlb`` option specify the file to be created resides in 4510 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction 4511 with the ``hugetlb`` option, the ``hugetlbsize`` option specify 4512 the hugetlb page size on systems that support multiple hugetlb 4513 page sizes (it must be a power of 2 value supported by the 4514 system). 4515 4516 In some versions of Linux, the ``hugetlb`` option is 4517 incompatible with the ``seal`` option (requires at least Linux 4518 4.16). 4519 4520 Please refer to ``memory-backend-file`` for a description of the 4521 other options. 4522 4523 The ``share`` boolean option is on by default with memfd. 4524 4525 ``-object rng-builtin,id=id`` 4526 Creates a random number generator backend which obtains entropy 4527 from QEMU builtin functions. The ``id`` parameter is a unique ID 4528 that will be used to reference this entropy backend from the 4529 ``virtio-rng`` device. By default, the ``virtio-rng`` device 4530 uses this RNG backend. 4531 4532 ``-object rng-random,id=id,filename=/dev/random`` 4533 Creates a random number generator backend which obtains entropy 4534 from a device on the host. The ``id`` parameter is a unique ID 4535 that will be used to reference this entropy backend from the 4536 ``virtio-rng`` device. The ``filename`` parameter specifies 4537 which file to obtain entropy from and if omitted defaults to 4538 ``/dev/urandom``. 4539 4540 ``-object rng-egd,id=id,chardev=chardevid`` 4541 Creates a random number generator backend which obtains entropy 4542 from an external daemon running on the host. The ``id`` 4543 parameter is a unique ID that will be used to reference this 4544 entropy backend from the ``virtio-rng`` device. The ``chardev`` 4545 parameter is the unique ID of a character device backend that 4546 provides the connection to the RNG daemon. 4547 4548 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off`` 4549 Creates a TLS anonymous credentials object, which can be used to 4550 provide TLS support on network backends. The ``id`` parameter is 4551 a unique ID which network backends will use to access the 4552 credentials. The ``endpoint`` is either ``server`` or ``client`` 4553 depending on whether the QEMU network backend that uses the 4554 credentials will be acting as a client or as a server. If 4555 ``verify-peer`` is enabled (the default) then once the handshake 4556 is completed, the peer credentials will be verified, though this 4557 is a no-op for anonymous credentials. 4558 4559 The dir parameter tells QEMU where to find the credential files. 4560 For server endpoints, this directory may contain a file 4561 dh-params.pem providing diffie-hellman parameters to use for the 4562 TLS server. If the file is missing, QEMU will generate a set of 4563 DH parameters at startup. This is a computationally expensive 4564 operation that consumes random pool entropy, so it is 4565 recommended that a persistent set of parameters be generated 4566 upfront and saved. 4567 4568 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]`` 4569 Creates a TLS Pre-Shared Keys (PSK) credentials object, which 4570 can be used to provide TLS support on network backends. The 4571 ``id`` parameter is a unique ID which network backends will use 4572 to access the credentials. The ``endpoint`` is either ``server`` 4573 or ``client`` depending on whether the QEMU network backend that 4574 uses the credentials will be acting as a client or as a server. 4575 For clients only, ``username`` is the username which will be 4576 sent to the server. If omitted it defaults to "qemu". 4577 4578 The dir parameter tells QEMU where to find the keys file. It is 4579 called "dir/keys.psk" and contains "username:key" pairs. This 4580 file can most easily be created using the GnuTLS ``psktool`` 4581 program. 4582 4583 For server endpoints, dir may also contain a file dh-params.pem 4584 providing diffie-hellman parameters to use for the TLS server. 4585 If the file is missing, QEMU will generate a set of DH 4586 parameters at startup. This is a computationally expensive 4587 operation that consumes random pool entropy, so it is 4588 recommended that a persistent set of parameters be generated up 4589 front and saved. 4590 4591 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id`` 4592 Creates a TLS anonymous credentials object, which can be used to 4593 provide TLS support on network backends. The ``id`` parameter is 4594 a unique ID which network backends will use to access the 4595 credentials. The ``endpoint`` is either ``server`` or ``client`` 4596 depending on whether the QEMU network backend that uses the 4597 credentials will be acting as a client or as a server. If 4598 ``verify-peer`` is enabled (the default) then once the handshake 4599 is completed, the peer credentials will be verified. With x509 4600 certificates, this implies that the clients must be provided 4601 with valid client certificates too. 4602 4603 The dir parameter tells QEMU where to find the credential files. 4604 For server endpoints, this directory may contain a file 4605 dh-params.pem providing diffie-hellman parameters to use for the 4606 TLS server. If the file is missing, QEMU will generate a set of 4607 DH parameters at startup. This is a computationally expensive 4608 operation that consumes random pool entropy, so it is 4609 recommended that a persistent set of parameters be generated 4610 upfront and saved. 4611 4612 For x509 certificate credentials the directory will contain 4613 further files providing the x509 certificates. The certificates 4614 must be stored in PEM format, in filenames ca-cert.pem, 4615 ca-crl.pem (optional), server-cert.pem (only servers), 4616 server-key.pem (only servers), client-cert.pem (only clients), 4617 and client-key.pem (only clients). 4618 4619 For the server-key.pem and client-key.pem files which contain 4620 sensitive private keys, it is possible to use an encrypted 4621 version by providing the passwordid parameter. This provides the 4622 ID of a previously created ``secret`` object containing the 4623 password for decryption. 4624 4625 The priority parameter allows to override the global default 4626 priority used by gnutls. This can be useful if the system 4627 administrator needs to use a weaker set of crypto priorities for 4628 QEMU without potentially forcing the weakness onto all 4629 applications. Or conversely if one wants wants a stronger 4630 default for QEMU than for all other applications, they can do 4631 this through this parameter. Its format is a gnutls priority 4632 string as described at 4633 https://gnutls.org/manual/html_node/Priority-Strings.html. 4634 4635 ``-object tls-cipher-suites,id=id,priority=priority`` 4636 Creates a TLS cipher suites object, which can be used to control 4637 the TLS cipher/protocol algorithms that applications are permitted 4638 to use. 4639 4640 The ``id`` parameter is a unique ID which frontends will use to 4641 access the ordered list of permitted TLS cipher suites from the 4642 host. 4643 4644 The ``priority`` parameter allows to override the global default 4645 priority used by gnutls. This can be useful if the system 4646 administrator needs to use a weaker set of crypto priorities for 4647 QEMU without potentially forcing the weakness onto all 4648 applications. Or conversely if one wants wants a stronger 4649 default for QEMU than for all other applications, they can do 4650 this through this parameter. Its format is a gnutls priority 4651 string as described at 4652 https://gnutls.org/manual/html_node/Priority-Strings.html. 4653 4654 An example of use of this object is to control UEFI HTTPS Boot. 4655 The tls-cipher-suites object exposes the ordered list of permitted 4656 TLS cipher suites from the host side to the guest firmware, via 4657 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER 4658 objects. The firmware uses the IANA_TLS_CIPHER array for configuring 4659 guest-side TLS. 4660 4661 In the following example, the priority at which the host-side policy 4662 is retrieved is given by the ``priority`` property. 4663 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to 4664 refer to /etc/crypto-policies/back-ends/gnutls.config. 4665 4666 .. parsed-literal:: 4667 4668 # |qemu_system| \\ 4669 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\ 4670 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0 4671 4672 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]`` 4673 Interval t can't be 0, this filter batches the packet delivery: 4674 all packets arriving in a given interval on netdev netdevid are 4675 delayed until the end of the interval. Interval is in 4676 microseconds. ``status`` is optional that indicate whether the 4677 netfilter is on (enabled) or off (disabled), the default status 4678 for netfilter will be 'on'. 4679 4680 queue all\|rx\|tx is an option that can be applied to any 4681 netfilter. 4682 4683 ``all``: the filter is attached both to the receive and the 4684 transmit queue of the netdev (default). 4685 4686 ``rx``: the filter is attached to the receive queue of the 4687 netdev, where it will receive packets sent to the netdev. 4688 4689 ``tx``: the filter is attached to the transmit queue of the 4690 netdev, where it will receive packets sent by the netdev. 4691 4692 position head\|tail\|id=<id> is an option to specify where the 4693 filter should be inserted in the filter list. It can be applied 4694 to any netfilter. 4695 4696 ``head``: the filter is inserted at the head of the filter list, 4697 before any existing filters. 4698 4699 ``tail``: the filter is inserted at the tail of the filter list, 4700 behind any existing filters (default). 4701 4702 ``id=<id>``: the filter is inserted before or behind the filter 4703 specified by <id>, see the insert option below. 4704 4705 insert behind\|before is an option to specify where to insert 4706 the new filter relative to the one specified with 4707 position=id=<id>. It can be applied to any netfilter. 4708 4709 ``before``: insert before the specified filter. 4710 4711 ``behind``: insert behind the specified filter (default). 4712 4713 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 4714 filter-mirror on netdev netdevid,mirror net packet to 4715 chardevchardevid, if it has the vnet\_hdr\_support flag, 4716 filter-mirror will mirror packet with vnet\_hdr\_len. 4717 4718 ``-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]`` 4719 filter-redirector on netdev netdevid,redirect filter's net 4720 packet to chardev chardevid,and redirect indev's packet to 4721 filter.if it has the vnet\_hdr\_support flag, filter-redirector 4722 will redirect packet with vnet\_hdr\_len. Create a 4723 filter-redirector we need to differ outdev id from indev id, id 4724 can not be the same. we can just use indev or outdev, but at 4725 least one of indev or outdev need to be specified. 4726 4727 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` 4728 Filter-rewriter is a part of COLO project.It will rewrite tcp 4729 packet to secondary from primary to keep secondary tcp 4730 connection,and rewrite tcp packet to primary from secondary make 4731 tcp packet can be handled by client.if it has the 4732 vnet\_hdr\_support flag, we can parse packet with vnet header. 4733 4734 usage: colo secondary: -object 4735 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object 4736 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object 4737 filter-rewriter,id=rew0,netdev=hn0,queue=all 4738 4739 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]`` 4740 Dump the network traffic on netdev dev to the file specified by 4741 filename. At most len bytes (64k by default) per packet are 4742 stored. The file format is libpcap, so it can be analyzed with 4743 tools such as tcpdump or Wireshark. 4744 4745 ``-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}]`` 4746 Colo-compare gets packet from primary\_in chardevid and 4747 secondary\_in, then compare whether the payload of primary packet 4748 and secondary packet are the same. If same, it will output 4749 primary packet to out\_dev, else it will notify COLO-framework to do 4750 checkpoint and send primary packet to out\_dev. In order to 4751 improve efficiency, we need to put the task of comparison in 4752 another iothread. If it has the vnet\_hdr\_support flag, 4753 colo compare will send/recv packet with vnet\_hdr\_len. 4754 The compare\_timeout=@var{ms} determines the maximum time of the 4755 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms} 4756 is to set the period of scanning expired primary node network packets. 4757 The max\_queue\_size=@var{size} is to set the max compare queue 4758 size depend on user environment. 4759 If user want to use Xen COLO, need to add the notify\_dev to 4760 notify Xen colo-frame to do checkpoint. 4761 4762 COLO-compare must be used with the help of filter-mirror, 4763 filter-redirector and filter-rewriter. 4764 4765 :: 4766 4767 KVM COLO 4768 4769 primary: 4770 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 4771 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 4772 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server,nowait 4773 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server,nowait 4774 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server,nowait 4775 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 4776 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server,nowait 4777 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 4778 -object iothread,id=iothread1 4779 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 4780 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 4781 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 4782 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1 4783 4784 secondary: 4785 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 4786 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 4787 -chardev socket,id=red0,host=3.3.3.3,port=9003 4788 -chardev socket,id=red1,host=3.3.3.3,port=9004 4789 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 4790 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 4791 4792 4793 Xen COLO 4794 4795 primary: 4796 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 4797 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 4798 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server,nowait 4799 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server,nowait 4800 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server,nowait 4801 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 4802 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server,nowait 4803 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 4804 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server,nowait 4805 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 4806 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 4807 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 4808 -object iothread,id=iothread1 4809 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1 4810 4811 secondary: 4812 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 4813 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 4814 -chardev socket,id=red0,host=3.3.3.3,port=9003 4815 -chardev socket,id=red1,host=3.3.3.3,port=9004 4816 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 4817 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 4818 4819 If you want to know the detail of above command line, you can 4820 read the colo-compare git log. 4821 4822 ``-object cryptodev-backend-builtin,id=id[,queues=queues]`` 4823 Creates a cryptodev backend which executes crypto opreation from 4824 the QEMU cipher APIS. The id parameter is a unique ID that will 4825 be used to reference this cryptodev backend from the 4826 ``virtio-crypto`` device. The queues parameter is optional, 4827 which specify the queue number of cryptodev backend, the default 4828 of queues is 1. 4829 4830 .. parsed-literal:: 4831 4832 # |qemu_system| \\ 4833 [...] \\ 4834 -object cryptodev-backend-builtin,id=cryptodev0 \\ 4835 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 4836 [...] 4837 4838 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]`` 4839 Creates a vhost-user cryptodev backend, backed by a chardev 4840 chardevid. The id parameter is a unique ID that will be used to 4841 reference this cryptodev backend from the ``virtio-crypto`` 4842 device. The chardev should be a unix domain socket backed one. 4843 The vhost-user uses a specifically defined protocol to pass 4844 vhost ioctl replacement messages to an application on the other 4845 end of the socket. The queues parameter is optional, which 4846 specify the queue number of cryptodev backend for multiqueue 4847 vhost-user, the default of queues is 1. 4848 4849 .. parsed-literal:: 4850 4851 # |qemu_system| \\ 4852 [...] \\ 4853 -chardev socket,id=chardev0,path=/path/to/socket \\ 4854 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\ 4855 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ 4856 [...] 4857 4858 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]`` 4859 \ 4860 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]`` 4861 Defines a secret to store a password, encryption key, or some 4862 other sensitive data. The sensitive data can either be passed 4863 directly via the data parameter, or indirectly via the file 4864 parameter. Using the data parameter is insecure unless the 4865 sensitive data is encrypted. 4866 4867 The sensitive data can be provided in raw format (the default), 4868 or base64. When encoded as JSON, the raw format only supports 4869 valid UTF-8 characters, so base64 is recommended for sending 4870 binary data. QEMU will convert from which ever format is 4871 provided to the format it needs internally. eg, an RBD password 4872 can be provided in raw format, even though it will be base64 4873 encoded when passed onto the RBD sever. 4874 4875 For added protection, it is possible to encrypt the data 4876 associated with a secret using the AES-256-CBC cipher. Use of 4877 encryption is indicated by providing the keyid and iv 4878 parameters. The keyid parameter provides the ID of a previously 4879 defined secret that contains the AES-256 decryption key. This 4880 key should be 32-bytes long and be base64 encoded. The iv 4881 parameter provides the random initialization vector used for 4882 encryption of this particular secret and should be a base64 4883 encrypted string of the 16-byte IV. 4884 4885 The simplest (insecure) usage is to provide the secret inline 4886 4887 .. parsed-literal:: 4888 4889 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw 4890 4891 The simplest secure usage is to provide the secret via a file 4892 4893 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object 4894 secret,id=sec0,file=mypasswd.txt,format=raw 4895 4896 For greater security, AES-256-CBC should be used. To illustrate 4897 usage, consider the openssl command line tool which can encrypt 4898 the data. Note that when encrypting, the plaintext must be 4899 padded to the cipher block size (32 bytes) using the standard 4900 PKCS#5/6 compatible padding algorithm. 4901 4902 First a master key needs to be created in base64 encoding: 4903 4904 :: 4905 4906 # openssl rand -base64 32 > key.b64 4907 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"') 4908 4909 Each secret to be encrypted needs to have a random 4910 initialization vector generated. These do not need to be kept 4911 secret 4912 4913 :: 4914 4915 # openssl rand -base64 16 > iv.b64 4916 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"') 4917 4918 The secret to be defined can now be encrypted, in this case 4919 we're telling openssl to base64 encode the result, but it could 4920 be left as raw bytes if desired. 4921 4922 :: 4923 4924 # SECRET=$(printf "letmein" | 4925 openssl enc -aes-256-cbc -a -K $KEY -iv $IV) 4926 4927 When launching QEMU, create a master secret pointing to 4928 ``key.b64`` and specify that to be used to decrypt the user 4929 password. Pass the contents of ``iv.b64`` to the second secret 4930 4931 .. parsed-literal:: 4932 4933 # |qemu_system| \\ 4934 -object secret,id=secmaster0,format=base64,file=key.b64 \\ 4935 -object secret,id=sec0,keyid=secmaster0,format=base64,\\ 4936 data=$SECRET,iv=$(<iv.b64) 4937 4938 ``-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]`` 4939 Create a Secure Encrypted Virtualization (SEV) guest object, 4940 which can be used to provide the guest memory encryption support 4941 on AMD processors. 4942 4943 When memory encryption is enabled, one of the physical address 4944 bit (aka the C-bit) is utilized to mark if a memory page is 4945 protected. The ``cbitpos`` is used to provide the C-bit 4946 position. The C-bit position is Host family dependent hence user 4947 must provide this value. On EPYC, the value should be 47. 4948 4949 When memory encryption is enabled, we loose certain bits in 4950 physical address space. The ``reduced-phys-bits`` is used to 4951 provide the number of bits we loose in physical address space. 4952 Similar to C-bit, the value is Host family dependent. On EPYC, 4953 the value should be 5. 4954 4955 The ``sev-device`` provides the device file to use for 4956 communicating with the SEV firmware running inside AMD Secure 4957 Processor. The default device is '/dev/sev'. If hardware 4958 supports memory encryption then /dev/sev devices are created by 4959 CCP driver. 4960 4961 The ``policy`` provides the guest policy to be enforced by the 4962 SEV firmware and restrict what configuration and operational 4963 commands can be performed on this guest by the hypervisor. The 4964 policy should be provided by the guest owner and is bound to the 4965 guest and cannot be changed throughout the lifetime of the 4966 guest. The default is 0. 4967 4968 If guest ``policy`` allows sharing the key with another SEV 4969 guest then ``handle`` can be use to provide handle of the guest 4970 from which to share the key. 4971 4972 The ``dh-cert-file`` and ``session-file`` provides the guest 4973 owner's Public Diffie-Hillman key defined in SEV spec. The PDH 4974 and session parameters are used for establishing a cryptographic 4975 session with the guest owner to negotiate keys used for 4976 attestation. The file must be encoded in base64. 4977 4978 e.g to launch a SEV guest 4979 4980 .. parsed-literal:: 4981 4982 # |qemu_system_x86| \\ 4983 ...... \\ 4984 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \\ 4985 -machine ...,memory-encryption=sev0 \\ 4986 ..... 4987 4988 ``-object authz-simple,id=id,identity=string`` 4989 Create an authorization object that will control access to 4990 network services. 4991 4992 The ``identity`` parameter is identifies the user and its format 4993 depends on the network service that authorization object is 4994 associated with. For authorizing based on TLS x509 certificates, 4995 the identity must be the x509 distinguished name. Note that care 4996 must be taken to escape any commas in the distinguished name. 4997 4998 An example authorization object to validate a x509 distinguished 4999 name would look like: 5000 5001 .. parsed-literal:: 5002 5003 # |qemu_system| \\ 5004 ... \\ 5005 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\ 5006 ... 5007 5008 Note the use of quotes due to the x509 distinguished name 5009 containing whitespace, and escaping of ','. 5010 5011 ``-object authz-listfile,id=id,filename=path,refresh=yes|no`` 5012 Create an authorization object that will control access to 5013 network services. 5014 5015 The ``filename`` parameter is the fully qualified path to a file 5016 containing the access control list rules in JSON format. 5017 5018 An example set of rules that match against SASL usernames might 5019 look like: 5020 5021 :: 5022 5023 { 5024 "rules": [ 5025 { "match": "fred", "policy": "allow", "format": "exact" }, 5026 { "match": "bob", "policy": "allow", "format": "exact" }, 5027 { "match": "danb", "policy": "deny", "format": "glob" }, 5028 { "match": "dan*", "policy": "allow", "format": "exact" }, 5029 ], 5030 "policy": "deny" 5031 } 5032 5033 When checking access the object will iterate over all the rules 5034 and the first rule to match will have its ``policy`` value 5035 returned as the result. If no rules match, then the default 5036 ``policy`` value is returned. 5037 5038 The rules can either be an exact string match, or they can use 5039 the simple UNIX glob pattern matching to allow wildcards to be 5040 used. 5041 5042 If ``refresh`` is set to true the file will be monitored and 5043 automatically reloaded whenever its content changes. 5044 5045 As with the ``authz-simple`` object, the format of the identity 5046 strings being matched depends on the network service, but is 5047 usually a TLS x509 distinguished name, or a SASL username. 5048 5049 An example authorization object to validate a SASL username 5050 would look like: 5051 5052 .. parsed-literal:: 5053 5054 # |qemu_system| \\ 5055 ... \\ 5056 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=yes \\ 5057 ... 5058 5059 ``-object authz-pam,id=id,service=string`` 5060 Create an authorization object that will control access to 5061 network services. 5062 5063 The ``service`` parameter provides the name of a PAM service to 5064 use for authorization. It requires that a file 5065 ``/etc/pam.d/service`` exist to provide the configuration for 5066 the ``account`` subsystem. 5067 5068 An example authorization object to validate a TLS x509 5069 distinguished name would look like: 5070 5071 .. parsed-literal:: 5072 5073 # |qemu_system| \\ 5074 ... \\ 5075 -object authz-pam,id=auth0,service=qemu-vnc \\ 5076 ... 5077 5078 There would then be a corresponding config file for PAM at 5079 ``/etc/pam.d/qemu-vnc`` that contains: 5080 5081 :: 5082 5083 account requisite pam_listfile.so item=user sense=allow \ 5084 file=/etc/qemu/vnc.allow 5085 5086 Finally the ``/etc/qemu/vnc.allow`` file would contain the list 5087 of x509 distingished names that are permitted access 5088 5089 :: 5090 5091 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB 5092 5093 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink`` 5094 Creates a dedicated event loop thread that devices can be 5095 assigned to. This is known as an IOThread. By default device 5096 emulation happens in vCPU threads or the main event loop thread. 5097 This can become a scalability bottleneck. IOThreads allow device 5098 emulation and I/O to run on other host CPUs. 5099 5100 The ``id`` parameter is a unique ID that will be used to 5101 reference this IOThread from ``-device ...,iothread=id``. 5102 Multiple devices can be assigned to an IOThread. Note that not 5103 all devices support an ``iothread`` parameter. 5104 5105 The ``query-iothreads`` QMP command lists IOThreads and reports 5106 their thread IDs so that the user can configure host CPU 5107 pinning/affinity. 5108 5109 IOThreads use an adaptive polling algorithm to reduce event loop 5110 latency. Instead of entering a blocking system call to monitor 5111 file descriptors and then pay the cost of being woken up when an 5112 event occurs, the polling algorithm spins waiting for events for 5113 a short time. The algorithm's default parameters are suitable 5114 for many cases but can be adjusted based on knowledge of the 5115 workload and/or host device latency. 5116 5117 The ``poll-max-ns`` parameter is the maximum number of 5118 nanoseconds to busy wait for events. Polling can be disabled by 5119 setting this value to 0. 5120 5121 The ``poll-grow`` parameter is the multiplier used to increase 5122 the polling time when the algorithm detects it is missing events 5123 due to not polling long enough. 5124 5125 The ``poll-shrink`` parameter is the divisor used to decrease 5126 the polling time when the algorithm detects it is spending too 5127 long polling without encountering events. 5128 5129 The polling parameters can be modified at run-time using the 5130 ``qom-set`` command (where ``iothread1`` is the IOThread's 5131 ``id``): 5132 5133 :: 5134 5135 (qemu) qom-set /objects/iothread1 poll-max-ns 100000 5136ERST 5137 5138 5139HXCOMM This is the last statement. Insert new options before this line! 5140
