1HXCOMM Use DEFHEADING() to define headings in both help text and texi 2HXCOMM Text between STEXI and ETEXI are copied to texi 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 texi and C 8 9DEFHEADING(Standard options:) 10STEXI 11@table @option 12ETEXI 13 14DEF("help", 0, QEMU_OPTION_h, 15 "-h or -help display this help and exit\n", QEMU_ARCH_ALL) 16STEXI 17@item -h 18@findex -h 19Display help and exit 20ETEXI 21 22DEF("version", 0, QEMU_OPTION_version, 23 "-version display version information and exit\n", QEMU_ARCH_ALL) 24STEXI 25@item -version 26@findex -version 27Display version information and exit 28ETEXI 29 30DEF("machine", HAS_ARG, QEMU_OPTION_machine, \ 31 "-machine [type=]name[,prop[=value][,...]]\n" 32 " selects emulated machine ('-machine help' for list)\n" 33 " property accel=accel1[:accel2[:...]] selects accelerator\n" 34 " supported accelerators are kvm, xen, hax, hvf, whpx or tcg (default: tcg)\n" 35 " kernel_irqchip=on|off|split controls accelerated irqchip support (default=off)\n" 36 " vmport=on|off|auto controls emulation of vmport (default: auto)\n" 37 " kvm_shadow_mem=size of KVM shadow MMU in bytes\n" 38 " dump-guest-core=on|off include guest memory in a core dump (default=on)\n" 39 " mem-merge=on|off controls memory merge support (default: on)\n" 40 " igd-passthru=on|off controls IGD GFX passthrough support (default=off)\n" 41 " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n" 42 " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n" 43 " suppress-vmdesc=on|off disables self-describing migration (default=off)\n" 44 " nvdimm=on|off controls NVDIMM support (default=off)\n" 45 " enforce-config-section=on|off enforce configuration section migration (default=off)\n" 46 " memory-encryption=@var{} memory encryption object to use (default=none)\n", 47 QEMU_ARCH_ALL) 48STEXI 49@item -machine [type=]@var{name}[,prop=@var{value}[,...]] 50@findex -machine 51Select the emulated machine by @var{name}. Use @code{-machine help} to list 52available machines. 53 54For architectures which aim to support live migration compatibility 55across releases, each release will introduce a new versioned machine 56type. For example, the 2.8.0 release introduced machine types 57``pc-i440fx-2.8'' and ``pc-q35-2.8'' for the x86_64/i686 architectures. 58 59To allow live migration of guests from QEMU version 2.8.0, to QEMU 60version 2.9.0, the 2.9.0 version must support the ``pc-i440fx-2.8'' 61and ``pc-q35-2.8'' machines too. To allow users live migrating VMs 62to skip multiple intermediate releases when upgrading, new releases 63of QEMU will support machine types from many previous versions. 64 65Supported machine properties are: 66@table @option 67@item accel=@var{accels1}[:@var{accels2}[:...]] 68This is used to enable an accelerator. Depending on the target architecture, 69kvm, xen, hax, hvf, whpx or tcg can be available. By default, tcg is used. If there is 70more than one accelerator specified, the next one is used if the previous one 71fails to initialize. 72@item kernel_irqchip=on|off 73Controls in-kernel irqchip support for the chosen accelerator when available. 74@item gfx_passthru=on|off 75Enables IGD GFX passthrough support for the chosen machine when available. 76@item vmport=on|off|auto 77Enables emulation of VMWare IO port, for vmmouse etc. auto says to select the 78value based on accel. For accel=xen the default is off otherwise the default 79is on. 80@item kvm_shadow_mem=size 81Defines the size of the KVM shadow MMU. 82@item dump-guest-core=on|off 83Include guest memory in a core dump. The default is on. 84@item mem-merge=on|off 85Enables or disables memory merge support. This feature, when supported by 86the host, de-duplicates identical memory pages among VMs instances 87(enabled by default). 88@item aes-key-wrap=on|off 89Enables or disables AES key wrapping support on s390-ccw hosts. This feature 90controls whether AES wrapping keys will be created to allow 91execution of AES cryptographic functions. The default is on. 92@item dea-key-wrap=on|off 93Enables or disables DEA key wrapping support on s390-ccw hosts. This feature 94controls whether DEA wrapping keys will be created to allow 95execution of DEA cryptographic functions. The default is on. 96@item nvdimm=on|off 97Enables or disables NVDIMM support. The default is off. 98@item enforce-config-section=on|off 99If @option{enforce-config-section} is set to @var{on}, force migration 100code to send configuration section even if the machine-type sets the 101@option{migration.send-configuration} property to @var{off}. 102NOTE: this parameter is deprecated. Please use @option{-global} 103@option{migration.send-configuration}=@var{on|off} instead. 104@item memory-encryption=@var{} 105Memory encryption object to use. The default is none. 106@end table 107ETEXI 108 109HXCOMM Deprecated by -machine 110DEF("M", HAS_ARG, QEMU_OPTION_M, "", QEMU_ARCH_ALL) 111 112DEF("cpu", HAS_ARG, QEMU_OPTION_cpu, 113 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL) 114STEXI 115@item -cpu @var{model} 116@findex -cpu 117Select CPU model (@code{-cpu help} for list and additional feature selection) 118ETEXI 119 120DEF("accel", HAS_ARG, QEMU_OPTION_accel, 121 "-accel [accel=]accelerator[,thread=single|multi]\n" 122 " select accelerator (kvm, xen, hax, hvf, whpx or tcg; use 'help' for a list)\n" 123 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL) 124STEXI 125@item -accel @var{name}[,prop=@var{value}[,...]] 126@findex -accel 127This is used to enable an accelerator. Depending on the target architecture, 128kvm, xen, hax, hvf, whpx or tcg can be available. By default, tcg is used. If there is 129more than one accelerator specified, the next one is used if the previous one 130fails to initialize. 131@table @option 132@item thread=single|multi 133Controls number of TCG threads. When the TCG is multi-threaded there will be one 134thread per vCPU therefor taking advantage of additional host cores. The default 135is to enable multi-threading where both the back-end and front-ends support it and 136no incompatible TCG features have been enabled (e.g. icount/replay). 137@end table 138ETEXI 139 140DEF("smp", HAS_ARG, QEMU_OPTION_smp, 141 "-smp [cpus=]n[,maxcpus=cpus][,cores=cores][,threads=threads][,dies=dies][,sockets=sockets]\n" 142 " set the number of CPUs to 'n' [default=1]\n" 143 " maxcpus= maximum number of total cpus, including\n" 144 " offline CPUs for hotplug, etc\n" 145 " cores= number of CPU cores on one socket (for PC, it's on one die)\n" 146 " threads= number of threads on one CPU core\n" 147 " dies= number of CPU dies on one socket (for PC only)\n" 148 " sockets= number of discrete sockets in the system\n", 149 QEMU_ARCH_ALL) 150STEXI 151@item -smp [cpus=]@var{n}[,cores=@var{cores}][,threads=@var{threads}][,dies=dies][,sockets=@var{sockets}][,maxcpus=@var{maxcpus}] 152@findex -smp 153Simulate an SMP system with @var{n} CPUs. On the PC target, up to 255 154CPUs are supported. On Sparc32 target, Linux limits the number of usable CPUs 155to 4. 156For the PC target, the number of @var{cores} per die, the number of @var{threads} 157per cores, the number of @var{dies} per packages and the total number of 158@var{sockets} can be specified. Missing values will be computed. 159If any on the three values is given, the total number of CPUs @var{n} can be omitted. 160@var{maxcpus} specifies the maximum number of hotpluggable CPUs. 161ETEXI 162 163DEF("numa", HAS_ARG, QEMU_OPTION_numa, 164 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node]\n" 165 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node]\n" 166 "-numa dist,src=source,dst=destination,val=distance\n" 167 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n", 168 QEMU_ARCH_ALL) 169STEXI 170@item -numa node[,mem=@var{size}][,cpus=@var{firstcpu}[-@var{lastcpu}]][,nodeid=@var{node}] 171@itemx -numa node[,memdev=@var{id}][,cpus=@var{firstcpu}[-@var{lastcpu}]][,nodeid=@var{node}] 172@itemx -numa dist,src=@var{source},dst=@var{destination},val=@var{distance} 173@itemx -numa cpu,node-id=@var{node}[,socket-id=@var{x}][,core-id=@var{y}][,thread-id=@var{z}] 174@findex -numa 175Define a NUMA node and assign RAM and VCPUs to it. 176Set the NUMA distance from a source node to a destination node. 177 178Legacy VCPU assignment uses @samp{cpus} option where 179@var{firstcpu} and @var{lastcpu} are CPU indexes. Each 180@samp{cpus} option represent a contiguous range of CPU indexes 181(or a single VCPU if @var{lastcpu} is omitted). A non-contiguous 182set of VCPUs can be represented by providing multiple @samp{cpus} 183options. If @samp{cpus} is omitted on all nodes, VCPUs are automatically 184split between them. 185 186For example, the following option assigns VCPUs 0, 1, 2 and 5 to 187a NUMA node: 188@example 189-numa node,cpus=0-2,cpus=5 190@end example 191 192@samp{cpu} option is a new alternative to @samp{cpus} option 193which uses @samp{socket-id|core-id|thread-id} properties to assign 194CPU objects to a @var{node} using topology layout properties of CPU. 195The set of properties is machine specific, and depends on used 196machine type/@samp{smp} options. It could be queried with 197@samp{hotpluggable-cpus} monitor command. 198@samp{node-id} property specifies @var{node} to which CPU object 199will be assigned, it's required for @var{node} to be declared 200with @samp{node} option before it's used with @samp{cpu} option. 201 202For example: 203@example 204-M pc \ 205-smp 1,sockets=2,maxcpus=2 \ 206-numa node,nodeid=0 -numa node,nodeid=1 \ 207-numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 208@end example 209 210@samp{mem} assigns a given RAM amount to a node. @samp{memdev} 211assigns RAM from a given memory backend device to a node. If 212@samp{mem} and @samp{memdev} are omitted in all nodes, RAM is 213split equally between them. 214 215@samp{mem} and @samp{memdev} are mutually exclusive. Furthermore, 216if one node uses @samp{memdev}, all of them have to use it. 217 218@var{source} and @var{destination} are NUMA node IDs. 219@var{distance} is the NUMA distance from @var{source} to @var{destination}. 220The distance from a node to itself is always 10. If any pair of nodes is 221given a distance, then all pairs must be given distances. Although, when 222distances are only given in one direction for each pair of nodes, then 223the distances in the opposite directions are assumed to be the same. If, 224however, an asymmetrical pair of distances is given for even one node 225pair, then all node pairs must be provided distance values for both 226directions, even when they are symmetrical. When a node is unreachable 227from another node, set the pair's distance to 255. 228 229Note that the -@option{numa} option doesn't allocate any of the 230specified resources, it just assigns existing resources to NUMA 231nodes. This means that one still has to use the @option{-m}, 232@option{-smp} options to allocate RAM and VCPUs respectively. 233 234ETEXI 235 236DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd, 237 "-add-fd fd=fd,set=set[,opaque=opaque]\n" 238 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL) 239STEXI 240@item -add-fd fd=@var{fd},set=@var{set}[,opaque=@var{opaque}] 241@findex -add-fd 242 243Add a file descriptor to an fd set. Valid options are: 244 245@table @option 246@item fd=@var{fd} 247This option defines the file descriptor of which a duplicate is added to fd set. 248The file descriptor cannot be stdin, stdout, or stderr. 249@item set=@var{set} 250This option defines the ID of the fd set to add the file descriptor to. 251@item opaque=@var{opaque} 252This option defines a free-form string that can be used to describe @var{fd}. 253@end table 254 255You can open an image using pre-opened file descriptors from an fd set: 256@example 257@value{qemu_system} \ 258 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \ 259 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \ 260 -drive file=/dev/fdset/2,index=0,media=disk 261@end example 262ETEXI 263 264DEF("set", HAS_ARG, QEMU_OPTION_set, 265 "-set group.id.arg=value\n" 266 " set <arg> parameter for item <id> of type <group>\n" 267 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL) 268STEXI 269@item -set @var{group}.@var{id}.@var{arg}=@var{value} 270@findex -set 271Set parameter @var{arg} for item @var{id} of type @var{group} 272ETEXI 273 274DEF("global", HAS_ARG, QEMU_OPTION_global, 275 "-global driver.property=value\n" 276 "-global driver=driver,property=property,value=value\n" 277 " set a global default for a driver property\n", 278 QEMU_ARCH_ALL) 279STEXI 280@item -global @var{driver}.@var{prop}=@var{value} 281@itemx -global driver=@var{driver},property=@var{property},value=@var{value} 282@findex -global 283Set default value of @var{driver}'s property @var{prop} to @var{value}, e.g.: 284 285@example 286@value{qemu_system_x86} -global ide-hd.physical_block_size=4096 disk-image.img 287@end example 288 289In particular, you can use this to set driver properties for devices which are 290created automatically by the machine model. To create a device which is not 291created automatically and set properties on it, use -@option{device}. 292 293-global @var{driver}.@var{prop}=@var{value} is shorthand for -global 294driver=@var{driver},property=@var{prop},value=@var{value}. The 295longhand syntax works even when @var{driver} contains a dot. 296ETEXI 297 298DEF("boot", HAS_ARG, QEMU_OPTION_boot, 299 "-boot [order=drives][,once=drives][,menu=on|off]\n" 300 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n" 301 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n" 302 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n" 303 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n" 304 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n", 305 QEMU_ARCH_ALL) 306STEXI 307@item -boot [order=@var{drives}][,once=@var{drives}][,menu=on|off][,splash=@var{sp_name}][,splash-time=@var{sp_time}][,reboot-timeout=@var{rb_timeout}][,strict=on|off] 308@findex -boot 309Specify boot order @var{drives} as a string of drive letters. Valid 310drive letters depend on the target architecture. The x86 PC uses: a, b 311(floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p (Etherboot 312from network adapter 1-4), hard disk boot is the default. To apply a 313particular boot order only on the first startup, specify it via 314@option{once}. Note that the @option{order} or @option{once} parameter 315should not be used together with the @option{bootindex} property of 316devices, since the firmware implementations normally do not support both 317at the same time. 318 319Interactive boot menus/prompts can be enabled via @option{menu=on} as far 320as firmware/BIOS supports them. The default is non-interactive boot. 321 322A splash picture could be passed to bios, enabling user to show it as logo, 323when option splash=@var{sp_name} is given and menu=on, If firmware/BIOS 324supports them. Currently Seabios for X86 system support it. 325limitation: The splash file could be a jpeg file or a BMP file in 24 BPP 326format(true color). The resolution should be supported by the SVGA mode, so 327the recommended is 320x240, 640x480, 800x640. 328 329A timeout could be passed to bios, guest will pause for @var{rb_timeout} ms 330when boot failed, then reboot. If @option{reboot-timeout} is not set, 331guest will not reboot by default. Currently Seabios for X86 332system support it. 333 334Do strict boot via @option{strict=on} as far as firmware/BIOS 335supports it. This only effects when boot priority is changed by 336bootindex options. The default is non-strict boot. 337 338@example 339# try to boot from network first, then from hard disk 340@value{qemu_system_x86} -boot order=nc 341# boot from CD-ROM first, switch back to default order after reboot 342@value{qemu_system_x86} -boot once=d 343# boot with a splash picture for 5 seconds. 344@value{qemu_system_x86} -boot menu=on,splash=/root/boot.bmp,splash-time=5000 345@end example 346 347Note: The legacy format '-boot @var{drives}' is still supported but its 348use is discouraged as it may be removed from future versions. 349ETEXI 350 351DEF("m", HAS_ARG, QEMU_OPTION_m, 352 "-m [size=]megs[,slots=n,maxmem=size]\n" 353 " configure guest RAM\n" 354 " size: initial amount of guest memory\n" 355 " slots: number of hotplug slots (default: none)\n" 356 " maxmem: maximum amount of guest memory (default: none)\n" 357 "NOTE: Some architectures might enforce a specific granularity\n", 358 QEMU_ARCH_ALL) 359STEXI 360@item -m [size=]@var{megs}[,slots=n,maxmem=size] 361@findex -m 362Sets guest startup RAM size to @var{megs} megabytes. Default is 128 MiB. 363Optionally, a suffix of ``M'' or ``G'' can be used to signify a value in 364megabytes or gigabytes respectively. Optional pair @var{slots}, @var{maxmem} 365could be used to set amount of hotpluggable memory slots and maximum amount of 366memory. Note that @var{maxmem} must be aligned to the page size. 367 368For example, the following command-line sets the guest startup RAM size to 3691GB, creates 3 slots to hotplug additional memory and sets the maximum 370memory the guest can reach to 4GB: 371 372@example 373@value{qemu_system} -m 1G,slots=3,maxmem=4G 374@end example 375 376If @var{slots} and @var{maxmem} are not specified, memory hotplug won't 377be enabled and the guest startup RAM will never increase. 378ETEXI 379 380DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath, 381 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL) 382STEXI 383@item -mem-path @var{path} 384@findex -mem-path 385Allocate guest RAM from a temporarily created file in @var{path}. 386ETEXI 387 388DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc, 389 "-mem-prealloc preallocate guest memory (use with -mem-path)\n", 390 QEMU_ARCH_ALL) 391STEXI 392@item -mem-prealloc 393@findex -mem-prealloc 394Preallocate memory when using -mem-path. 395ETEXI 396 397DEF("k", HAS_ARG, QEMU_OPTION_k, 398 "-k language use keyboard layout (for example 'fr' for French)\n", 399 QEMU_ARCH_ALL) 400STEXI 401@item -k @var{language} 402@findex -k 403Use keyboard layout @var{language} (for example @code{fr} for 404French). This option is only needed where it is not easy to get raw PC 405keycodes (e.g. on Macs, with some X11 servers or with a VNC or curses 406display). You don't normally need to use it on PC/Linux or PC/Windows 407hosts. 408 409The available layouts are: 410@example 411ar de-ch es fo fr-ca hu ja mk no pt-br sv 412da en-gb et fr fr-ch is lt nl pl ru th 413de en-us fi fr-be hr it lv nl-be pt sl tr 414@end example 415 416The default is @code{en-us}. 417ETEXI 418 419 420HXCOMM Deprecated by -audiodev 421DEF("audio-help", 0, QEMU_OPTION_audio_help, 422 "-audio-help show -audiodev equivalent of the currently specified audio settings\n", 423 QEMU_ARCH_ALL) 424STEXI 425@item -audio-help 426@findex -audio-help 427Will show the -audiodev equivalent of the currently specified 428(deprecated) environment variables. 429ETEXI 430 431DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, 432 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n" 433 " specifies the audio backend to use\n" 434 " id= identifier of the backend\n" 435 " timer-period= timer period in microseconds\n" 436 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n" 437 " in|out.fixed-settings= use fixed settings for host audio\n" 438 " in|out.frequency= frequency to use with fixed settings\n" 439 " in|out.channels= number of channels to use with fixed settings\n" 440 " in|out.format= sample format to use with fixed settings\n" 441 " valid values: s8, s16, s32, u8, u16, u32\n" 442 " in|out.voices= number of voices to use\n" 443 " in|out.buffer-length= length of buffer in microseconds\n" 444 "-audiodev none,id=id,[,prop[=value][,...]]\n" 445 " dummy driver that discards all output\n" 446#ifdef CONFIG_AUDIO_ALSA 447 "-audiodev alsa,id=id[,prop[=value][,...]]\n" 448 " in|out.dev= name of the audio device to use\n" 449 " in|out.period-length= length of period in microseconds\n" 450 " in|out.try-poll= attempt to use poll mode\n" 451 " threshold= threshold (in microseconds) when playback starts\n" 452#endif 453#ifdef CONFIG_AUDIO_COREAUDIO 454 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n" 455 " in|out.buffer-count= number of buffers\n" 456#endif 457#ifdef CONFIG_AUDIO_DSOUND 458 "-audiodev dsound,id=id[,prop[=value][,...]]\n" 459 " latency= add extra latency to playback in microseconds\n" 460#endif 461#ifdef CONFIG_AUDIO_OSS 462 "-audiodev oss,id=id[,prop[=value][,...]]\n" 463 " in|out.dev= path of the audio device to use\n" 464 " in|out.buffer-count= number of buffers\n" 465 " in|out.try-poll= attempt to use poll mode\n" 466 " try-mmap= try using memory mapped access\n" 467 " exclusive= open device in exclusive mode\n" 468 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n" 469#endif 470#ifdef CONFIG_AUDIO_PA 471 "-audiodev pa,id=id[,prop[=value][,...]]\n" 472 " server= PulseAudio server address\n" 473 " in|out.name= source/sink device name\n" 474 " in|out.latency= desired latency in microseconds\n" 475#endif 476#ifdef CONFIG_AUDIO_SDL 477 "-audiodev sdl,id=id[,prop[=value][,...]]\n" 478#endif 479#ifdef CONFIG_SPICE 480 "-audiodev spice,id=id[,prop[=value][,...]]\n" 481#endif 482 "-audiodev wav,id=id[,prop[=value][,...]]\n" 483 " path= path of wav file to record\n", 484 QEMU_ARCH_ALL) 485STEXI 486@item -audiodev [driver=]@var{driver},id=@var{id}[,@var{prop}[=@var{value}][,...]] 487@findex -audiodev 488Adds a new audio backend @var{driver} identified by @var{id}. There are 489global and driver specific properties. Some values can be set 490differently for input and output, they're marked with @code{in|out.}. 491You can set the input's property with @code{in.@var{prop}} and the 492output's property with @code{out.@var{prop}}. For example: 493@example 494-audiodev alsa,id=example,in.frequency=44110,out.frequency=8000 495-audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified 496@end example 497 498NOTE: parameter validation is known to be incomplete, in many cases 499specifying an invalid option causes QEMU to print an error message and 500continue emulation without sound. 501 502Valid global options are: 503 504@table @option 505@item id=@var{identifier} 506Identifies the audio backend. 507 508@item timer-period=@var{period} 509Sets the timer @var{period} used by the audio subsystem in microseconds. 510Default is 10000 (10 ms). 511 512@item in|out.mixing-engine=on|off 513Use QEMU's mixing engine to mix all streams inside QEMU and convert 514audio formats when not supported by the backend. When off, 515@var{fixed-settings} must be off too. Note that disabling this option 516means that the selected backend must support multiple streams and the 517audio formats used by the virtual cards, otherwise you'll get no sound. 518It's not recommended to disable this option unless you want to use 5.1 519or 7.1 audio, as mixing engine only supports mono and stereo audio. 520Default is on. 521 522@item in|out.fixed-settings=on|off 523Use fixed settings for host audio. When off, it will change based on 524how the guest opens the sound card. In this case you must not specify 525@var{frequency}, @var{channels} or @var{format}. Default is on. 526 527@item in|out.frequency=@var{frequency} 528Specify the @var{frequency} to use when using @var{fixed-settings}. 529Default is 44100Hz. 530 531@item in|out.channels=@var{channels} 532Specify the number of @var{channels} to use when using 533@var{fixed-settings}. Default is 2 (stereo). 534 535@item in|out.format=@var{format} 536Specify the sample @var{format} to use when using @var{fixed-settings}. 537Valid values are: @code{s8}, @code{s16}, @code{s32}, @code{u8}, 538@code{u16}, @code{u32}. Default is @code{s16}. 539 540@item in|out.voices=@var{voices} 541Specify the number of @var{voices} to use. Default is 1. 542 543@item in|out.buffer-length=@var{usecs} 544Sets the size of the buffer in microseconds. 545 546@end table 547 548@item -audiodev none,id=@var{id}[,@var{prop}[=@var{value}][,...]] 549Creates a dummy backend that discards all outputs. This backend has no 550backend specific properties. 551 552@item -audiodev alsa,id=@var{id}[,@var{prop}[=@var{value}][,...]] 553Creates backend using the ALSA. This backend is only available on 554Linux. 555 556ALSA specific options are: 557 558@table @option 559 560@item in|out.dev=@var{device} 561Specify the ALSA @var{device} to use for input and/or output. Default 562is @code{default}. 563 564@item in|out.period-length=@var{usecs} 565Sets the period length in microseconds. 566 567@item in|out.try-poll=on|off 568Attempt to use poll mode with the device. Default is on. 569 570@item threshold=@var{threshold} 571Threshold (in microseconds) when playback starts. Default is 0. 572 573@end table 574 575@item -audiodev coreaudio,id=@var{id}[,@var{prop}[=@var{value}][,...]] 576Creates a backend using Apple's Core Audio. This backend is only 577available on Mac OS and only supports playback. 578 579Core Audio specific options are: 580 581@table @option 582 583@item in|out.buffer-count=@var{count} 584Sets the @var{count} of the buffers. 585 586@end table 587 588@item -audiodev dsound,id=@var{id}[,@var{prop}[=@var{value}][,...]] 589Creates a backend using Microsoft's DirectSound. This backend is only 590available on Windows and only supports playback. 591 592DirectSound specific options are: 593 594@table @option 595 596@item latency=@var{usecs} 597Add extra @var{usecs} microseconds latency to playback. Default is 59810000 (10 ms). 599 600@end table 601 602@item -audiodev oss,id=@var{id}[,@var{prop}[=@var{value}][,...]] 603Creates a backend using OSS. This backend is available on most 604Unix-like systems. 605 606OSS specific options are: 607 608@table @option 609 610@item in|out.dev=@var{device} 611Specify the file name of the OSS @var{device} to use. Default is 612@code{/dev/dsp}. 613 614@item in|out.buffer-count=@var{count} 615Sets the @var{count} of the buffers. 616 617@item in|out.try-poll=on|of 618Attempt to use poll mode with the device. Default is on. 619 620@item try-mmap=on|off 621Try using memory mapped device access. Default is off. 622 623@item exclusive=on|off 624Open the device in exclusive mode (vmix won't work in this case). 625Default is off. 626 627@item dsp-policy=@var{policy} 628Sets the timing policy (between 0 and 10, where smaller number means 629smaller latency but higher CPU usage). Use -1 to use buffer sizes 630specified by @code{buffer} and @code{buffer-count}. This option is 631ignored if you do not have OSS 4. Default is 5. 632 633@end table 634 635@item -audiodev pa,id=@var{id}[,@var{prop}[=@var{value}][,...]] 636Creates a backend using PulseAudio. This backend is available on most 637systems. 638 639PulseAudio specific options are: 640 641@table @option 642 643@item server=@var{server} 644Sets the PulseAudio @var{server} to connect to. 645 646@item in|out.name=@var{sink} 647Use the specified source/sink for recording/playback. 648 649@item in|out.latency=@var{usecs} 650Desired latency in microseconds. The PulseAudio server will try to honor this 651value but actual latencies may be lower or higher. 652 653@end table 654 655@item -audiodev sdl,id=@var{id}[,@var{prop}[=@var{value}][,...]] 656Creates a backend using SDL. This backend is available on most systems, 657but you should use your platform's native backend if possible. This 658backend has no backend specific properties. 659 660@item -audiodev spice,id=@var{id}[,@var{prop}[=@var{value}][,...]] 661Creates a backend that sends audio through SPICE. This backend requires 662@code{-spice} and automatically selected in that case, so usually you 663can ignore this option. This backend has no backend specific 664properties. 665 666@item -audiodev wav,id=@var{id}[,@var{prop}[=@var{value}][,...]] 667Creates a backend that writes audio to a WAV file. 668 669Backend specific options are: 670 671@table @option 672 673@item path=@var{path} 674Write recorded audio into the specified file. Default is 675@code{qemu.wav}. 676 677@end table 678ETEXI 679 680DEF("soundhw", HAS_ARG, QEMU_OPTION_soundhw, 681 "-soundhw c1,... enable audio support\n" 682 " and only specified sound cards (comma separated list)\n" 683 " use '-soundhw help' to get the list of supported cards\n" 684 " use '-soundhw all' to enable all of them\n", QEMU_ARCH_ALL) 685STEXI 686@item -soundhw @var{card1}[,@var{card2},...] or -soundhw all 687@findex -soundhw 688Enable audio and selected sound hardware. Use 'help' to print all 689available sound hardware. For example: 690 691@example 692@value{qemu_system_x86} -soundhw sb16,adlib disk.img 693@value{qemu_system_x86} -soundhw es1370 disk.img 694@value{qemu_system_x86} -soundhw ac97 disk.img 695@value{qemu_system_x86} -soundhw hda disk.img 696@value{qemu_system_x86} -soundhw all disk.img 697@value{qemu_system_x86} -soundhw help 698@end example 699 700Note that Linux's i810_audio OSS kernel (for AC97) module might 701require manually specifying clocking. 702 703@example 704modprobe i810_audio clocking=48000 705@end example 706ETEXI 707 708DEF("device", HAS_ARG, QEMU_OPTION_device, 709 "-device driver[,prop[=value][,...]]\n" 710 " add device (based on driver)\n" 711 " prop=value,... sets driver properties\n" 712 " use '-device help' to print all possible drivers\n" 713 " use '-device driver,help' to print all possible properties\n", 714 QEMU_ARCH_ALL) 715STEXI 716@item -device @var{driver}[,@var{prop}[=@var{value}][,...]] 717@findex -device 718Add device @var{driver}. @var{prop}=@var{value} sets driver 719properties. Valid properties depend on the driver. To get help on 720possible drivers and properties, use @code{-device help} and 721@code{-device @var{driver},help}. 722 723Some drivers are: 724@item -device ipmi-bmc-sim,id=@var{id}[,slave_addr=@var{val}][,sdrfile=@var{file}][,furareasize=@var{val}][,furdatafile=@var{file}][,guid=@var{uuid}] 725 726Add an IPMI BMC. This is a simulation of a hardware management 727interface processor that normally sits on a system. It provides 728a watchdog and the ability to reset and power control the system. 729You need to connect this to an IPMI interface to make it useful 730 731The IPMI slave address to use for the BMC. The default is 0x20. 732This address is the BMC's address on the I2C network of management 733controllers. If you don't know what this means, it is safe to ignore 734it. 735 736@table @option 737@item id=@var{id} 738The BMC id for interfaces to use this device. 739@item slave_addr=@var{val} 740Define slave address to use for the BMC. The default is 0x20. 741@item sdrfile=@var{file} 742file containing raw Sensor Data Records (SDR) data. The default is none. 743@item fruareasize=@var{val} 744size of a Field Replaceable Unit (FRU) area. The default is 1024. 745@item frudatafile=@var{file} 746file containing raw Field Replaceable Unit (FRU) inventory data. The default is none. 747@item guid=@var{uuid} 748value for the GUID for the BMC, in standard UUID format. If this is set, 749get "Get GUID" command to the BMC will return it. Otherwise "Get GUID" 750will return an error. 751@end table 752 753@item -device ipmi-bmc-extern,id=@var{id},chardev=@var{id}[,slave_addr=@var{val}] 754 755Add a connection to an external IPMI BMC simulator. Instead of 756locally emulating the BMC like the above item, instead connect 757to an external entity that provides the IPMI services. 758 759A connection is made to an external BMC simulator. If you do this, it 760is strongly recommended that you use the "reconnect=" chardev option 761to reconnect to the simulator if the connection is lost. Note that if 762this is not used carefully, it can be a security issue, as the 763interface has the ability to send resets, NMIs, and power off the VM. 764It's best if QEMU makes a connection to an external simulator running 765on a secure port on localhost, so neither the simulator nor QEMU is 766exposed to any outside network. 767 768See the "lanserv/README.vm" file in the OpenIPMI library for more 769details on the external interface. 770 771@item -device isa-ipmi-kcs,bmc=@var{id}[,ioport=@var{val}][,irq=@var{val}] 772 773Add a KCS IPMI interafce on the ISA bus. This also adds a 774corresponding ACPI and SMBIOS entries, if appropriate. 775 776@table @option 777@item bmc=@var{id} 778The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above. 779@item ioport=@var{val} 780Define the I/O address of the interface. The default is 0xca0 for KCS. 781@item irq=@var{val} 782Define the interrupt to use. The default is 5. To disable interrupts, 783set this to 0. 784@end table 785 786@item -device isa-ipmi-bt,bmc=@var{id}[,ioport=@var{val}][,irq=@var{val}] 787 788Like the KCS interface, but defines a BT interface. The default port is 7890xe4 and the default interrupt is 5. 790 791ETEXI 792 793DEF("name", HAS_ARG, QEMU_OPTION_name, 794 "-name string1[,process=string2][,debug-threads=on|off]\n" 795 " set the name of the guest\n" 796 " string1 sets the window title and string2 the process name\n" 797 " When debug-threads is enabled, individual threads are given a separate name\n" 798 " NOTE: The thread names are for debugging and not a stable API.\n", 799 QEMU_ARCH_ALL) 800STEXI 801@item -name @var{name} 802@findex -name 803Sets the @var{name} of the guest. 804This name will be displayed in the SDL window caption. 805The @var{name} will also be used for the VNC server. 806Also optionally set the top visible process name in Linux. 807Naming of individual threads can also be enabled on Linux to aid debugging. 808ETEXI 809 810DEF("uuid", HAS_ARG, QEMU_OPTION_uuid, 811 "-uuid %08x-%04x-%04x-%04x-%012x\n" 812 " specify machine UUID\n", QEMU_ARCH_ALL) 813STEXI 814@item -uuid @var{uuid} 815@findex -uuid 816Set system UUID. 817ETEXI 818 819STEXI 820@end table 821ETEXI 822DEFHEADING() 823 824DEFHEADING(Block device options:) 825STEXI 826@table @option 827ETEXI 828 829DEF("fda", HAS_ARG, QEMU_OPTION_fda, 830 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL) 831DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL) 832STEXI 833@item -fda @var{file} 834@itemx -fdb @var{file} 835@findex -fda 836@findex -fdb 837Use @var{file} as floppy disk 0/1 image (@pxref{disk_images}). 838ETEXI 839 840DEF("hda", HAS_ARG, QEMU_OPTION_hda, 841 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL) 842DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL) 843DEF("hdc", HAS_ARG, QEMU_OPTION_hdc, 844 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL) 845DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL) 846STEXI 847@item -hda @var{file} 848@itemx -hdb @var{file} 849@itemx -hdc @var{file} 850@itemx -hdd @var{file} 851@findex -hda 852@findex -hdb 853@findex -hdc 854@findex -hdd 855Use @var{file} as hard disk 0, 1, 2 or 3 image (@pxref{disk_images}). 856ETEXI 857 858DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom, 859 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n", 860 QEMU_ARCH_ALL) 861STEXI 862@item -cdrom @var{file} 863@findex -cdrom 864Use @var{file} as CD-ROM image (you cannot use @option{-hdc} and 865@option{-cdrom} at the same time). You can use the host CD-ROM by 866using @file{/dev/cdrom} as filename (@pxref{host_drives}). 867ETEXI 868 869DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev, 870 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n" 871 " [,cache.direct=on|off][,cache.no-flush=on|off]\n" 872 " [,read-only=on|off][,auto-read-only=on|off]\n" 873 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n" 874 " [,driver specific parameters...]\n" 875 " configure a block backend\n", QEMU_ARCH_ALL) 876STEXI 877@item -blockdev @var{option}[,@var{option}[,@var{option}[,...]]] 878@findex -blockdev 879 880Define a new block driver node. Some of the options apply to all block drivers, 881other options are only accepted for a specific block driver. See below for a 882list of generic options and options for the most common block drivers. 883 884Options that expect a reference to another node (e.g. @code{file}) can be 885given in two ways. Either you specify the node name of an already existing node 886(file=@var{node-name}), or you define a new node inline, adding options 887for the referenced node after a dot (file.filename=@var{path},file.aio=native). 888 889A block driver node created with @option{-blockdev} can be used for a guest 890device by specifying its node name for the @code{drive} property in a 891@option{-device} argument that defines a block device. 892 893@table @option 894@item Valid options for any block driver node: 895 896@table @code 897@item driver 898Specifies the block driver to use for the given node. 899@item node-name 900This defines the name of the block driver node by which it will be referenced 901later. The name must be unique, i.e. it must not match the name of a different 902block driver node, or (if you use @option{-drive} as well) the ID of a drive. 903 904If no node name is specified, it is automatically generated. The generated node 905name is not intended to be predictable and changes between QEMU invocations. 906For the top level, an explicit node name must be specified. 907@item read-only 908Open the node read-only. Guest write attempts will fail. 909 910Note that some block drivers support only read-only access, either generally or 911in certain configurations. In this case, the default value 912@option{read-only=off} does not work and the option must be specified 913explicitly. 914@item auto-read-only 915If @option{auto-read-only=on} is set, QEMU may fall back to read-only usage 916even when @option{read-only=off} is requested, or even switch between modes as 917needed, e.g. depending on whether the image file is writable or whether a 918writing user is attached to the node. 919@item force-share 920Override the image locking system of QEMU by forcing the node to utilize 921weaker shared access for permissions where it would normally request exclusive 922access. When there is the potential for multiple instances to have the same 923file open (whether this invocation of QEMU is the first or the second 924instance), both instances must permit shared access for the second instance to 925succeed at opening the file. 926 927Enabling @option{force-share=on} requires @option{read-only=on}. 928@item cache.direct 929The host page cache can be avoided with @option{cache.direct=on}. This will 930attempt to do disk IO directly to the guest's memory. QEMU may still perform an 931internal copy of the data. 932@item cache.no-flush 933In case you don't care about data integrity over host failures, you can use 934@option{cache.no-flush=on}. This option tells QEMU that it never needs to write 935any data to the disk but can instead keep things in cache. If anything goes 936wrong, like your host losing power, the disk storage getting disconnected 937accidentally, etc. your image will most probably be rendered unusable. 938@item discard=@var{discard} 939@var{discard} is one of "ignore" (or "off") or "unmap" (or "on") and controls 940whether @code{discard} (also known as @code{trim} or @code{unmap}) requests are 941ignored or passed to the filesystem. Some machine types may not support 942discard requests. 943@item detect-zeroes=@var{detect-zeroes} 944@var{detect-zeroes} is "off", "on" or "unmap" and enables the automatic 945conversion of plain zero writes by the OS to driver specific optimized 946zero write commands. You may even choose "unmap" if @var{discard} is set 947to "unmap" to allow a zero write to be converted to an @code{unmap} operation. 948@end table 949 950@item Driver-specific options for @code{file} 951 952This is the protocol-level block driver for accessing regular files. 953 954@table @code 955@item filename 956The path to the image file in the local filesystem 957@item aio 958Specifies the AIO backend (threads/native, default: threads) 959@item locking 960Specifies whether the image file is protected with Linux OFD / POSIX locks. The 961default is to use the Linux Open File Descriptor API if available, otherwise no 962lock is applied. (auto/on/off, default: auto) 963@end table 964Example: 965@example 966-blockdev driver=file,node-name=disk,filename=disk.img 967@end example 968 969@item Driver-specific options for @code{raw} 970 971This is the image format block driver for raw images. It is usually 972stacked on top of a protocol level block driver such as @code{file}. 973 974@table @code 975@item file 976Reference to or definition of the data source block driver node 977(e.g. a @code{file} driver node) 978@end table 979Example 1: 980@example 981-blockdev driver=file,node-name=disk_file,filename=disk.img 982-blockdev driver=raw,node-name=disk,file=disk_file 983@end example 984Example 2: 985@example 986-blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img 987@end example 988 989@item Driver-specific options for @code{qcow2} 990 991This is the image format block driver for qcow2 images. It is usually 992stacked on top of a protocol level block driver such as @code{file}. 993 994@table @code 995@item file 996Reference to or definition of the data source block driver node 997(e.g. a @code{file} driver node) 998 999@item backing 1000Reference to or definition of the backing file block device (default is taken 1001from the image file). It is allowed to pass @code{null} here in order to disable 1002the default backing file. 1003 1004@item lazy-refcounts 1005Whether to enable the lazy refcounts feature (on/off; default is taken from the 1006image file) 1007 1008@item cache-size 1009The maximum total size of the L2 table and refcount block caches in bytes 1010(default: the sum of l2-cache-size and refcount-cache-size) 1011 1012@item l2-cache-size 1013The maximum size of the L2 table cache in bytes 1014(default: if cache-size is not specified - 32M on Linux platforms, and 8M on 1015non-Linux platforms; otherwise, as large as possible within the cache-size, 1016while permitting the requested or the minimal refcount cache size) 1017 1018@item refcount-cache-size 1019The maximum size of the refcount block cache in bytes 1020(default: 4 times the cluster size; or if cache-size is specified, the part of 1021it which is not used for the L2 cache) 1022 1023@item cache-clean-interval 1024Clean unused entries in the L2 and refcount caches. The interval is in seconds. 1025The default value is 600 on supporting platforms, and 0 on other platforms. 1026Setting it to 0 disables this feature. 1027 1028@item pass-discard-request 1029Whether discard requests to the qcow2 device should be forwarded to the data 1030source (on/off; default: on if discard=unmap is specified, off otherwise) 1031 1032@item pass-discard-snapshot 1033Whether discard requests for the data source should be issued when a snapshot 1034operation (e.g. deleting a snapshot) frees clusters in the qcow2 file (on/off; 1035default: on) 1036 1037@item pass-discard-other 1038Whether discard requests for the data source should be issued on other 1039occasions where a cluster gets freed (on/off; default: off) 1040 1041@item overlap-check 1042Which overlap checks to perform for writes to the image 1043(none/constant/cached/all; default: cached). For details or finer 1044granularity control refer to the QAPI documentation of @code{blockdev-add}. 1045@end table 1046 1047Example 1: 1048@example 1049-blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2 1050-blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216 1051@end example 1052Example 2: 1053@example 1054-blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2 1055@end example 1056 1057@item Driver-specific options for other drivers 1058Please refer to the QAPI documentation of the @code{blockdev-add} QMP command. 1059 1060@end table 1061 1062ETEXI 1063 1064DEF("drive", HAS_ARG, QEMU_OPTION_drive, 1065 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n" 1066 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n" 1067 " [,snapshot=on|off][,rerror=ignore|stop|report]\n" 1068 " [,werror=ignore|stop|report|enospc][,id=name][,aio=threads|native]\n" 1069 " [,readonly=on|off][,copy-on-read=on|off]\n" 1070 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n" 1071 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n" 1072 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n" 1073 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n" 1074 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n" 1075 " [[,iops_size=is]]\n" 1076 " [[,group=g]]\n" 1077 " use 'file' as a drive image\n", QEMU_ARCH_ALL) 1078STEXI 1079@item -drive @var{option}[,@var{option}[,@var{option}[,...]]] 1080@findex -drive 1081 1082Define a new drive. This includes creating a block driver node (the backend) as 1083well as a guest device, and is mostly a shortcut for defining the corresponding 1084@option{-blockdev} and @option{-device} options. 1085 1086@option{-drive} accepts all options that are accepted by @option{-blockdev}. In 1087addition, it knows the following options: 1088 1089@table @option 1090@item file=@var{file} 1091This option defines which disk image (@pxref{disk_images}) to use with 1092this drive. If the filename contains comma, you must double it 1093(for instance, "file=my,,file" to use file "my,file"). 1094 1095Special files such as iSCSI devices can be specified using protocol 1096specific URLs. See the section for "Device URL Syntax" for more information. 1097@item if=@var{interface} 1098This option defines on which type on interface the drive is connected. 1099Available types are: ide, scsi, sd, mtd, floppy, pflash, virtio, none. 1100@item bus=@var{bus},unit=@var{unit} 1101These options define where is connected the drive by defining the bus number and 1102the unit id. 1103@item index=@var{index} 1104This option defines where is connected the drive by using an index in the list 1105of available connectors of a given interface type. 1106@item media=@var{media} 1107This option defines the type of the media: disk or cdrom. 1108@item snapshot=@var{snapshot} 1109@var{snapshot} is "on" or "off" and controls snapshot mode for the given drive 1110(see @option{-snapshot}). 1111@item cache=@var{cache} 1112@var{cache} is "none", "writeback", "unsafe", "directsync" or "writethrough" 1113and controls how the host cache is used to access block data. This is a 1114shortcut that sets the @option{cache.direct} and @option{cache.no-flush} 1115options (as in @option{-blockdev}), and additionally @option{cache.writeback}, 1116which provides a default for the @option{write-cache} option of block guest 1117devices (as in @option{-device}). The modes correspond to the following 1118settings: 1119 1120@c Our texi2pod.pl script doesn't support @multitable, so fall back to using 1121@c plain ASCII art (well, UTF-8 art really). This looks okay both in the manpage 1122@c and the HTML output. 1123@example 1124@ │ cache.writeback cache.direct cache.no-flush 1125─────────────┼───────────────────────────────────────────────── 1126writeback │ on off off 1127none │ on on off 1128writethrough │ off off off 1129directsync │ off on off 1130unsafe │ on off on 1131@end example 1132 1133The default mode is @option{cache=writeback}. 1134 1135@item aio=@var{aio} 1136@var{aio} is "threads", or "native" and selects between pthread based disk I/O and native Linux AIO. 1137@item format=@var{format} 1138Specify which disk @var{format} will be used rather than detecting 1139the format. Can be used to specify format=raw to avoid interpreting 1140an untrusted format header. 1141@item werror=@var{action},rerror=@var{action} 1142Specify which @var{action} to take on write and read errors. Valid actions are: 1143"ignore" (ignore the error and try to continue), "stop" (pause QEMU), 1144"report" (report the error to the guest), "enospc" (pause QEMU only if the 1145host disk is full; report the error to the guest otherwise). 1146The default setting is @option{werror=enospc} and @option{rerror=report}. 1147@item copy-on-read=@var{copy-on-read} 1148@var{copy-on-read} is "on" or "off" and enables whether to copy read backing 1149file sectors into the image file. 1150@item bps=@var{b},bps_rd=@var{r},bps_wr=@var{w} 1151Specify bandwidth throttling limits in bytes per second, either for all request 1152types or for reads or writes only. Small values can lead to timeouts or hangs 1153inside the guest. A safe minimum for disks is 2 MB/s. 1154@item bps_max=@var{bm},bps_rd_max=@var{rm},bps_wr_max=@var{wm} 1155Specify bursts in bytes per second, either for all request types or for reads 1156or writes only. Bursts allow the guest I/O to spike above the limit 1157temporarily. 1158@item iops=@var{i},iops_rd=@var{r},iops_wr=@var{w} 1159Specify request rate limits in requests per second, either for all request 1160types or for reads or writes only. 1161@item iops_max=@var{bm},iops_rd_max=@var{rm},iops_wr_max=@var{wm} 1162Specify bursts in requests per second, either for all request types or for reads 1163or writes only. Bursts allow the guest I/O to spike above the limit 1164temporarily. 1165@item iops_size=@var{is} 1166Let every @var{is} bytes of a request count as a new request for iops 1167throttling purposes. Use this option to prevent guests from circumventing iops 1168limits by sending fewer but larger requests. 1169@item group=@var{g} 1170Join a throttling quota group with given name @var{g}. All drives that are 1171members of the same group are accounted for together. Use this option to 1172prevent guests from circumventing throttling limits by using many small disks 1173instead of a single larger disk. 1174@end table 1175 1176By default, the @option{cache.writeback=on} mode is used. It will report data 1177writes as completed as soon as the data is present in the host page cache. 1178This is safe as long as your guest OS makes sure to correctly flush disk caches 1179where needed. If your guest OS does not handle volatile disk write caches 1180correctly and your host crashes or loses power, then the guest may experience 1181data corruption. 1182 1183For such guests, you should consider using @option{cache.writeback=off}. This 1184means that the host page cache will be used to read and write data, but write 1185notification will be sent to the guest only after QEMU has made sure to flush 1186each write to the disk. Be aware that this has a major impact on performance. 1187 1188When using the @option{-snapshot} option, unsafe caching is always used. 1189 1190Copy-on-read avoids accessing the same backing file sectors repeatedly and is 1191useful when the backing file is over a slow network. By default copy-on-read 1192is off. 1193 1194Instead of @option{-cdrom} you can use: 1195@example 1196@value{qemu_system} -drive file=file,index=2,media=cdrom 1197@end example 1198 1199Instead of @option{-hda}, @option{-hdb}, @option{-hdc}, @option{-hdd}, you can 1200use: 1201@example 1202@value{qemu_system} -drive file=file,index=0,media=disk 1203@value{qemu_system} -drive file=file,index=1,media=disk 1204@value{qemu_system} -drive file=file,index=2,media=disk 1205@value{qemu_system} -drive file=file,index=3,media=disk 1206@end example 1207 1208You can open an image using pre-opened file descriptors from an fd set: 1209@example 1210@value{qemu_system} \ 1211 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \ 1212 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \ 1213 -drive file=/dev/fdset/2,index=0,media=disk 1214@end example 1215 1216You can connect a CDROM to the slave of ide0: 1217@example 1218@value{qemu_system_x86} -drive file=file,if=ide,index=1,media=cdrom 1219@end example 1220 1221If you don't specify the "file=" argument, you define an empty drive: 1222@example 1223@value{qemu_system_x86} -drive if=ide,index=1,media=cdrom 1224@end example 1225 1226Instead of @option{-fda}, @option{-fdb}, you can use: 1227@example 1228@value{qemu_system_x86} -drive file=file,index=0,if=floppy 1229@value{qemu_system_x86} -drive file=file,index=1,if=floppy 1230@end example 1231 1232By default, @var{interface} is "ide" and @var{index} is automatically 1233incremented: 1234@example 1235@value{qemu_system_x86} -drive file=a -drive file=b" 1236@end example 1237is interpreted like: 1238@example 1239@value{qemu_system_x86} -hda a -hdb b 1240@end example 1241ETEXI 1242 1243DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock, 1244 "-mtdblock file use 'file' as on-board Flash memory image\n", 1245 QEMU_ARCH_ALL) 1246STEXI 1247@item -mtdblock @var{file} 1248@findex -mtdblock 1249Use @var{file} as on-board Flash memory image. 1250ETEXI 1251 1252DEF("sd", HAS_ARG, QEMU_OPTION_sd, 1253 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL) 1254STEXI 1255@item -sd @var{file} 1256@findex -sd 1257Use @var{file} as SecureDigital card image. 1258ETEXI 1259 1260DEF("pflash", HAS_ARG, QEMU_OPTION_pflash, 1261 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL) 1262STEXI 1263@item -pflash @var{file} 1264@findex -pflash 1265Use @var{file} as a parallel flash image. 1266ETEXI 1267 1268DEF("snapshot", 0, QEMU_OPTION_snapshot, 1269 "-snapshot write to temporary files instead of disk image files\n", 1270 QEMU_ARCH_ALL) 1271STEXI 1272@item -snapshot 1273@findex -snapshot 1274Write to temporary files instead of disk image files. In this case, 1275the raw disk image you use is not written back. You can however force 1276the write back by pressing @key{C-a s} (@pxref{disk_images}). 1277ETEXI 1278 1279DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev, 1280 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1281 " [,writeout=immediate][,readonly][,fmode=fmode][,dmode=dmode]\n" 1282 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n" 1283 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n" 1284 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n" 1285 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n" 1286 " [[,throttling.iops-size=is]]\n" 1287 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly]\n" 1288 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly]\n" 1289 "-fsdev synth,id=id\n", 1290 QEMU_ARCH_ALL) 1291 1292STEXI 1293 1294@item -fsdev local,id=@var{id},path=@var{path},security_model=@var{security_model} [,writeout=@var{writeout}][,readonly][,fmode=@var{fmode}][,dmode=@var{dmode}] [,throttling.@var{option}=@var{value}[,throttling.@var{option}=@var{value}[,...]]] 1295@itemx -fsdev proxy,id=@var{id},socket=@var{socket}[,writeout=@var{writeout}][,readonly] 1296@itemx -fsdev proxy,id=@var{id},sock_fd=@var{sock_fd}[,writeout=@var{writeout}][,readonly] 1297@itemx -fsdev synth,id=@var{id}[,readonly] 1298@findex -fsdev 1299Define a new file system device. Valid options are: 1300@table @option 1301@item local 1302Accesses to the filesystem are done by QEMU. 1303@item proxy 1304Accesses to the filesystem are done by virtfs-proxy-helper(1). 1305@item synth 1306Synthetic filesystem, only used by QTests. 1307@item id=@var{id} 1308Specifies identifier for this device. 1309@item path=@var{path} 1310Specifies the export path for the file system device. Files under 1311this path will be available to the 9p client on the guest. 1312@item security_model=@var{security_model} 1313Specifies the security model to be used for this export path. 1314Supported security models are "passthrough", "mapped-xattr", "mapped-file" and "none". 1315In "passthrough" security model, files are stored using the same 1316credentials as they are created on the guest. This requires QEMU 1317to run as root. In "mapped-xattr" security model, some of the file 1318attributes like uid, gid, mode bits and link target are stored as 1319file attributes. For "mapped-file" these attributes are stored in the 1320hidden .virtfs_metadata directory. Directories exported by this security model cannot 1321interact with other unix tools. "none" security model is same as 1322passthrough except the sever won't report failures if it fails to 1323set file attributes like ownership. Security model is mandatory 1324only for local fsdriver. Other fsdrivers (like proxy) don't take 1325security model as a parameter. 1326@item writeout=@var{writeout} 1327This is an optional argument. The only supported value is "immediate". 1328This means that host page cache will be used to read and write data but 1329write notification will be sent to the guest only when the data has been 1330reported as written by the storage subsystem. 1331@item readonly 1332Enables exporting 9p share as a readonly mount for guests. By default 1333read-write access is given. 1334@item socket=@var{socket} 1335Enables proxy filesystem driver to use passed socket file for communicating 1336with virtfs-proxy-helper(1). 1337@item sock_fd=@var{sock_fd} 1338Enables proxy filesystem driver to use passed socket descriptor for 1339communicating with virtfs-proxy-helper(1). Usually a helper like libvirt 1340will create socketpair and pass one of the fds as sock_fd. 1341@item fmode=@var{fmode} 1342Specifies the default mode for newly created files on the host. Works only 1343with security models "mapped-xattr" and "mapped-file". 1344@item dmode=@var{dmode} 1345Specifies the default mode for newly created directories on the host. Works 1346only with security models "mapped-xattr" and "mapped-file". 1347@item throttling.bps-total=@var{b},throttling.bps-read=@var{r},throttling.bps-write=@var{w} 1348Specify bandwidth throttling limits in bytes per second, either for all request 1349types or for reads or writes only. 1350@item throttling.bps-total-max=@var{bm},bps-read-max=@var{rm},bps-write-max=@var{wm} 1351Specify bursts in bytes per second, either for all request types or for reads 1352or writes only. Bursts allow the guest I/O to spike above the limit 1353temporarily. 1354@item throttling.iops-total=@var{i},throttling.iops-read=@var{r}, throttling.iops-write=@var{w} 1355Specify request rate limits in requests per second, either for all request 1356types or for reads or writes only. 1357@item throttling.iops-total-max=@var{im},throttling.iops-read-max=@var{irm}, throttling.iops-write-max=@var{iwm} 1358Specify bursts in requests per second, either for all request types or for reads 1359or writes only. Bursts allow the guest I/O to spike above the limit temporarily. 1360@item throttling.iops-size=@var{is} 1361Let every @var{is} bytes of a request count as a new request for iops 1362throttling purposes. 1363@end table 1364 1365-fsdev option is used along with -device driver "virtio-9p-...". 1366@item -device virtio-9p-@var{type},fsdev=@var{id},mount_tag=@var{mount_tag} 1367Options for virtio-9p-... driver are: 1368@table @option 1369@item @var{type} 1370Specifies the variant to be used. Supported values are "pci", "ccw" or "device", 1371depending on the machine type. 1372@item fsdev=@var{id} 1373Specifies the id value specified along with -fsdev option. 1374@item mount_tag=@var{mount_tag} 1375Specifies the tag name to be used by the guest to mount this export point. 1376@end table 1377 1378ETEXI 1379 1380DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs, 1381 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n" 1382 " [,id=id][,writeout=immediate][,readonly][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n" 1383 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly]\n" 1384 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly]\n" 1385 "-virtfs synth,mount_tag=tag[,id=id][,readonly]\n", 1386 QEMU_ARCH_ALL) 1387 1388STEXI 1389 1390@item -virtfs local,path=@var{path},mount_tag=@var{mount_tag} ,security_model=@var{security_model}[,writeout=@var{writeout}][,readonly] [,fmode=@var{fmode}][,dmode=@var{dmode}][,multidevs=@var{multidevs}] 1391@itemx -virtfs proxy,socket=@var{socket},mount_tag=@var{mount_tag} [,writeout=@var{writeout}][,readonly] 1392@itemx -virtfs proxy,sock_fd=@var{sock_fd},mount_tag=@var{mount_tag} [,writeout=@var{writeout}][,readonly] 1393@itemx -virtfs synth,mount_tag=@var{mount_tag} 1394@findex -virtfs 1395 1396Define a new filesystem device and expose it to the guest using a virtio-9p-device. The general form of a Virtual File system pass-through options are: 1397@table @option 1398@item local 1399Accesses to the filesystem are done by QEMU. 1400@item proxy 1401Accesses to the filesystem are done by virtfs-proxy-helper(1). 1402@item synth 1403Synthetic filesystem, only used by QTests. 1404@item id=@var{id} 1405Specifies identifier for the filesystem device 1406@item path=@var{path} 1407Specifies the export path for the file system device. Files under 1408this path will be available to the 9p client on the guest. 1409@item security_model=@var{security_model} 1410Specifies the security model to be used for this export path. 1411Supported security models are "passthrough", "mapped-xattr", "mapped-file" and "none". 1412In "passthrough" security model, files are stored using the same 1413credentials as they are created on the guest. This requires QEMU 1414to run as root. In "mapped-xattr" security model, some of the file 1415attributes like uid, gid, mode bits and link target are stored as 1416file attributes. For "mapped-file" these attributes are stored in the 1417hidden .virtfs_metadata directory. Directories exported by this security model cannot 1418interact with other unix tools. "none" security model is same as 1419passthrough except the sever won't report failures if it fails to 1420set file attributes like ownership. Security model is mandatory only 1421for local fsdriver. Other fsdrivers (like proxy) don't take security 1422model as a parameter. 1423@item writeout=@var{writeout} 1424This is an optional argument. The only supported value is "immediate". 1425This means that host page cache will be used to read and write data but 1426write notification will be sent to the guest only when the data has been 1427reported as written by the storage subsystem. 1428@item readonly 1429Enables exporting 9p share as a readonly mount for guests. By default 1430read-write access is given. 1431@item socket=@var{socket} 1432Enables proxy filesystem driver to use passed socket file for 1433communicating with virtfs-proxy-helper(1). Usually a helper like libvirt 1434will create socketpair and pass one of the fds as sock_fd. 1435@item sock_fd 1436Enables proxy filesystem driver to use passed 'sock_fd' as the socket 1437descriptor for interfacing with virtfs-proxy-helper(1). 1438@item fmode=@var{fmode} 1439Specifies the default mode for newly created files on the host. Works only 1440with security models "mapped-xattr" and "mapped-file". 1441@item dmode=@var{dmode} 1442Specifies the default mode for newly created directories on the host. Works 1443only with security models "mapped-xattr" and "mapped-file". 1444@item mount_tag=@var{mount_tag} 1445Specifies the tag name to be used by the guest to mount this export point. 1446@item multidevs=@var{multidevs} 1447Specifies how to deal with multiple devices being shared with a 9p export. 1448Supported behaviours are either "remap", "forbid" or "warn". The latter is 1449the default behaviour on which virtfs 9p expects only one device to be 1450shared with the same export, and if more than one device is shared and 1451accessed via the same 9p export then only a warning message is logged 1452(once) by qemu on host side. In order to avoid file ID collisions on guest 1453you should either create a separate virtfs export for each device to be 1454shared with guests (recommended way) or you might use "remap" instead which 1455allows you to share multiple devices with only one export instead, which is 1456achieved by remapping the original inode numbers from host to guest in a 1457way that would prevent such collisions. Remapping inodes in such use cases 1458is required because the original device IDs from host are never passed and 1459exposed on guest. Instead all files of an export shared with virtfs always 1460share the same device id on guest. So two files with identical inode 1461numbers but from actually different devices on host would otherwise cause a 1462file ID collision and hence potential misbehaviours on guest. "forbid" on 1463the other hand assumes like "warn" that only one device is shared by the 1464same export, however it will not only log a warning message but also 1465deny access to additional devices on guest. Note though that "forbid" does 1466currently not block all possible file access operations (e.g. readdir() 1467would still return entries from other devices). 1468@end table 1469ETEXI 1470 1471DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi, 1472 "-iscsi [user=user][,password=password]\n" 1473 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE\n" 1474 " [,initiator-name=initiator-iqn][,id=target-iqn]\n" 1475 " [,timeout=timeout]\n" 1476 " iSCSI session parameters\n", QEMU_ARCH_ALL) 1477 1478STEXI 1479@item -iscsi 1480@findex -iscsi 1481Configure iSCSI session parameters. 1482ETEXI 1483 1484STEXI 1485@end table 1486ETEXI 1487DEFHEADING() 1488 1489DEFHEADING(USB options:) 1490STEXI 1491@table @option 1492ETEXI 1493 1494DEF("usb", 0, QEMU_OPTION_usb, 1495 "-usb enable on-board USB host controller (if not enabled by default)\n", 1496 QEMU_ARCH_ALL) 1497STEXI 1498@item -usb 1499@findex -usb 1500Enable USB emulation on machine types with an on-board USB host controller (if 1501not enabled by default). Note that on-board USB host controllers may not 1502support USB 3.0. In this case @option{-device qemu-xhci} can be used instead 1503on machines with PCI. 1504ETEXI 1505 1506DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice, 1507 "-usbdevice name add the host or guest USB device 'name'\n", 1508 QEMU_ARCH_ALL) 1509STEXI 1510 1511@item -usbdevice @var{devname} 1512@findex -usbdevice 1513Add the USB device @var{devname}. Note that this option is deprecated, 1514please use @code{-device usb-...} instead. @xref{usb_devices}. 1515 1516@table @option 1517 1518@item mouse 1519Virtual Mouse. This will override the PS/2 mouse emulation when activated. 1520 1521@item tablet 1522Pointer device that uses absolute coordinates (like a touchscreen). This 1523means QEMU is able to report the mouse position without having to grab the 1524mouse. Also overrides the PS/2 mouse emulation when activated. 1525 1526@item braille 1527Braille device. This will use BrlAPI to display the braille output on a real 1528or fake device. 1529 1530@end table 1531ETEXI 1532 1533STEXI 1534@end table 1535ETEXI 1536DEFHEADING() 1537 1538DEFHEADING(Display options:) 1539STEXI 1540@table @option 1541ETEXI 1542 1543DEF("display", HAS_ARG, QEMU_OPTION_display, 1544#if defined(CONFIG_SPICE) 1545 "-display spice-app[,gl=on|off]\n" 1546#endif 1547#if defined(CONFIG_SDL) 1548 "-display sdl[,alt_grab=on|off][,ctrl_grab=on|off]\n" 1549 " [,window_close=on|off][,gl=on|core|es|off]\n" 1550#endif 1551#if defined(CONFIG_GTK) 1552 "-display gtk[,grab_on_hover=on|off][,gl=on|off]|\n" 1553#endif 1554#if defined(CONFIG_VNC) 1555 "-display vnc=<display>[,<optargs>]\n" 1556#endif 1557#if defined(CONFIG_CURSES) 1558 "-display curses[,charset=<encoding>]\n" 1559#endif 1560#if defined(CONFIG_OPENGL) 1561 "-display egl-headless[,rendernode=<file>]\n" 1562#endif 1563 "-display none\n" 1564 " select display backend type\n" 1565 " The default display is equivalent to\n " 1566#if defined(CONFIG_GTK) 1567 "\"-display gtk\"\n" 1568#elif defined(CONFIG_SDL) 1569 "\"-display sdl\"\n" 1570#elif defined(CONFIG_COCOA) 1571 "\"-display cocoa\"\n" 1572#elif defined(CONFIG_VNC) 1573 "\"-vnc localhost:0,to=99,id=default\"\n" 1574#else 1575 "\"-display none\"\n" 1576#endif 1577 , QEMU_ARCH_ALL) 1578STEXI 1579@item -display @var{type} 1580@findex -display 1581Select type of display to use. This option is a replacement for the 1582old style -sdl/-curses/... options. Valid values for @var{type} are 1583@table @option 1584@item sdl 1585Display video output via SDL (usually in a separate graphics 1586window; see the SDL documentation for other possibilities). 1587@item curses 1588Display video output via curses. For graphics device models which 1589support a text mode, QEMU can display this output using a 1590curses/ncurses interface. Nothing is displayed when the graphics 1591device is in graphical mode or if the graphics device does not support 1592a text mode. Generally only the VGA device models support text mode. 1593The font charset used by the guest can be specified with the 1594@code{charset} option, for example @code{charset=CP850} for IBM CP850 1595encoding. The default is @code{CP437}. 1596@item none 1597Do not display video output. The guest will still see an emulated 1598graphics card, but its output will not be displayed to the QEMU 1599user. This option differs from the -nographic option in that it 1600only affects what is done with video output; -nographic also changes 1601the destination of the serial and parallel port data. 1602@item gtk 1603Display video output in a GTK window. This interface provides drop-down 1604menus and other UI elements to configure and control the VM during 1605runtime. 1606@item vnc 1607Start a VNC server on display <arg> 1608@item egl-headless 1609Offload all OpenGL operations to a local DRI device. For any graphical display, 1610this display needs to be paired with either VNC or SPICE displays. 1611@item spice-app 1612Start QEMU as a Spice server and launch the default Spice client 1613application. The Spice server will redirect the serial consoles and 1614QEMU monitors. (Since 4.0) 1615@end table 1616ETEXI 1617 1618DEF("nographic", 0, QEMU_OPTION_nographic, 1619 "-nographic disable graphical output and redirect serial I/Os to console\n", 1620 QEMU_ARCH_ALL) 1621STEXI 1622@item -nographic 1623@findex -nographic 1624Normally, if QEMU is compiled with graphical window support, it displays 1625output such as guest graphics, guest console, and the QEMU monitor in a 1626window. With this option, you can totally disable graphical output so 1627that QEMU is a simple command line application. The emulated serial port 1628is redirected on the console and muxed with the monitor (unless 1629redirected elsewhere explicitly). Therefore, you can still use QEMU to 1630debug a Linux kernel with a serial console. Use @key{C-a h} for help on 1631switching between the console and monitor. 1632ETEXI 1633 1634DEF("curses", 0, QEMU_OPTION_curses, 1635 "-curses shorthand for -display curses\n", 1636 QEMU_ARCH_ALL) 1637STEXI 1638@item -curses 1639@findex -curses 1640Normally, if QEMU is compiled with graphical window support, it displays 1641output such as guest graphics, guest console, and the QEMU monitor in a 1642window. With this option, QEMU can display the VGA output when in text 1643mode using a curses/ncurses interface. Nothing is displayed in graphical 1644mode. 1645ETEXI 1646 1647DEF("alt-grab", 0, QEMU_OPTION_alt_grab, 1648 "-alt-grab use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt)\n", 1649 QEMU_ARCH_ALL) 1650STEXI 1651@item -alt-grab 1652@findex -alt-grab 1653Use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt). Note that this also 1654affects the special keys (for fullscreen, monitor-mode switching, etc). 1655ETEXI 1656 1657DEF("ctrl-grab", 0, QEMU_OPTION_ctrl_grab, 1658 "-ctrl-grab use Right-Ctrl to grab mouse (instead of Ctrl-Alt)\n", 1659 QEMU_ARCH_ALL) 1660STEXI 1661@item -ctrl-grab 1662@findex -ctrl-grab 1663Use Right-Ctrl to grab mouse (instead of Ctrl-Alt). Note that this also 1664affects the special keys (for fullscreen, monitor-mode switching, etc). 1665ETEXI 1666 1667DEF("no-quit", 0, QEMU_OPTION_no_quit, 1668 "-no-quit disable SDL window close capability\n", QEMU_ARCH_ALL) 1669STEXI 1670@item -no-quit 1671@findex -no-quit 1672Disable SDL window close capability. 1673ETEXI 1674 1675DEF("sdl", 0, QEMU_OPTION_sdl, 1676 "-sdl shorthand for -display sdl\n", QEMU_ARCH_ALL) 1677STEXI 1678@item -sdl 1679@findex -sdl 1680Enable SDL. 1681ETEXI 1682 1683DEF("spice", HAS_ARG, QEMU_OPTION_spice, 1684 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n" 1685 " [,x509-key-file=<file>][,x509-key-password=<file>]\n" 1686 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n" 1687 " [,x509-dh-key-file=<file>][,addr=addr][,ipv4|ipv6|unix]\n" 1688 " [,tls-ciphers=<list>]\n" 1689 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n" 1690 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n" 1691 " [,sasl][,password=<secret>][,disable-ticketing]\n" 1692 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n" 1693 " [,jpeg-wan-compression=[auto|never|always]]\n" 1694 " [,zlib-glz-wan-compression=[auto|never|always]]\n" 1695 " [,streaming-video=[off|all|filter]][,disable-copy-paste]\n" 1696 " [,disable-agent-file-xfer][,agent-mouse=[on|off]]\n" 1697 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n" 1698 " [,gl=[on|off]][,rendernode=<file>]\n" 1699 " enable spice\n" 1700 " at least one of {port, tls-port} is mandatory\n", 1701 QEMU_ARCH_ALL) 1702STEXI 1703@item -spice @var{option}[,@var{option}[,...]] 1704@findex -spice 1705Enable the spice remote desktop protocol. Valid options are 1706 1707@table @option 1708 1709@item port=<nr> 1710Set the TCP port spice is listening on for plaintext channels. 1711 1712@item addr=<addr> 1713Set the IP address spice is listening on. Default is any address. 1714 1715@item ipv4 1716@itemx ipv6 1717@itemx unix 1718Force using the specified IP version. 1719 1720@item password=<secret> 1721Set the password you need to authenticate. 1722 1723@item sasl 1724Require that the client use SASL to authenticate with the spice. 1725The exact choice of authentication method used is controlled from the 1726system / user's SASL configuration file for the 'qemu' service. This 1727is typically found in /etc/sasl2/qemu.conf. If running QEMU as an 1728unprivileged user, an environment variable SASL_CONF_PATH can be used 1729to make it search alternate locations for the service config. 1730While some SASL auth methods can also provide data encryption (eg GSSAPI), 1731it is recommended that SASL always be combined with the 'tls' and 1732'x509' settings to enable use of SSL and server certificates. This 1733ensures a data encryption preventing compromise of authentication 1734credentials. 1735 1736@item disable-ticketing 1737Allow client connects without authentication. 1738 1739@item disable-copy-paste 1740Disable copy paste between the client and the guest. 1741 1742@item disable-agent-file-xfer 1743Disable spice-vdagent based file-xfer between the client and the guest. 1744 1745@item tls-port=<nr> 1746Set the TCP port spice is listening on for encrypted channels. 1747 1748@item x509-dir=<dir> 1749Set the x509 file directory. Expects same filenames as -vnc $display,x509=$dir 1750 1751@item x509-key-file=<file> 1752@itemx x509-key-password=<file> 1753@itemx x509-cert-file=<file> 1754@itemx x509-cacert-file=<file> 1755@itemx x509-dh-key-file=<file> 1756The x509 file names can also be configured individually. 1757 1758@item tls-ciphers=<list> 1759Specify which ciphers to use. 1760 1761@item tls-channel=[main|display|cursor|inputs|record|playback] 1762@itemx plaintext-channel=[main|display|cursor|inputs|record|playback] 1763Force specific channel to be used with or without TLS encryption. The 1764options can be specified multiple times to configure multiple 1765channels. The special name "default" can be used to set the default 1766mode. For channels which are not explicitly forced into one mode the 1767spice client is allowed to pick tls/plaintext as he pleases. 1768 1769@item image-compression=[auto_glz|auto_lz|quic|glz|lz|off] 1770Configure image compression (lossless). 1771Default is auto_glz. 1772 1773@item jpeg-wan-compression=[auto|never|always] 1774@itemx zlib-glz-wan-compression=[auto|never|always] 1775Configure wan image compression (lossy for slow links). 1776Default is auto. 1777 1778@item streaming-video=[off|all|filter] 1779Configure video stream detection. Default is off. 1780 1781@item agent-mouse=[on|off] 1782Enable/disable passing mouse events via vdagent. Default is on. 1783 1784@item playback-compression=[on|off] 1785Enable/disable audio stream compression (using celt 0.5.1). Default is on. 1786 1787@item seamless-migration=[on|off] 1788Enable/disable spice seamless migration. Default is off. 1789 1790@item gl=[on|off] 1791Enable/disable OpenGL context. Default is off. 1792 1793@item rendernode=<file> 1794DRM render node for OpenGL rendering. If not specified, it will pick 1795the first available. (Since 2.9) 1796 1797@end table 1798ETEXI 1799 1800DEF("portrait", 0, QEMU_OPTION_portrait, 1801 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n", 1802 QEMU_ARCH_ALL) 1803STEXI 1804@item -portrait 1805@findex -portrait 1806Rotate graphical output 90 deg left (only PXA LCD). 1807ETEXI 1808 1809DEF("rotate", HAS_ARG, QEMU_OPTION_rotate, 1810 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n", 1811 QEMU_ARCH_ALL) 1812STEXI 1813@item -rotate @var{deg} 1814@findex -rotate 1815Rotate graphical output some deg left (only PXA LCD). 1816ETEXI 1817 1818DEF("vga", HAS_ARG, QEMU_OPTION_vga, 1819 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n" 1820 " select video card type\n", QEMU_ARCH_ALL) 1821STEXI 1822@item -vga @var{type} 1823@findex -vga 1824Select type of VGA card to emulate. Valid values for @var{type} are 1825@table @option 1826@item cirrus 1827Cirrus Logic GD5446 Video card. All Windows versions starting from 1828Windows 95 should recognize and use this graphic card. For optimal 1829performances, use 16 bit color depth in the guest and the host OS. 1830(This card was the default before QEMU 2.2) 1831@item std 1832Standard VGA card with Bochs VBE extensions. If your guest OS 1833supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if you want 1834to use high resolution modes (>= 1280x1024x16) then you should use 1835this option. (This card is the default since QEMU 2.2) 1836@item vmware 1837VMWare SVGA-II compatible adapter. Use it if you have sufficiently 1838recent XFree86/XOrg server or Windows guest with a driver for this 1839card. 1840@item qxl 1841QXL paravirtual graphic card. It is VGA compatible (including VESA 18422.0 VBE support). Works best with qxl guest drivers installed though. 1843Recommended choice when using the spice protocol. 1844@item tcx 1845(sun4m only) Sun TCX framebuffer. This is the default framebuffer for 1846sun4m machines and offers both 8-bit and 24-bit colour depths at a 1847fixed resolution of 1024x768. 1848@item cg3 1849(sun4m only) Sun cgthree framebuffer. This is a simple 8-bit framebuffer 1850for sun4m machines available in both 1024x768 (OpenBIOS) and 1152x900 (OBP) 1851resolutions aimed at people wishing to run older Solaris versions. 1852@item virtio 1853Virtio VGA card. 1854@item none 1855Disable VGA card. 1856@end table 1857ETEXI 1858 1859DEF("full-screen", 0, QEMU_OPTION_full_screen, 1860 "-full-screen start in full screen\n", QEMU_ARCH_ALL) 1861STEXI 1862@item -full-screen 1863@findex -full-screen 1864Start in full screen. 1865ETEXI 1866 1867DEF("g", 1, QEMU_OPTION_g , 1868 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n", 1869 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K) 1870STEXI 1871@item -g @var{width}x@var{height}[x@var{depth}] 1872@findex -g 1873Set the initial graphical resolution and depth (PPC, SPARC only). 1874ETEXI 1875 1876DEF("vnc", HAS_ARG, QEMU_OPTION_vnc , 1877 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL) 1878STEXI 1879@item -vnc @var{display}[,@var{option}[,@var{option}[,...]]] 1880@findex -vnc 1881Normally, if QEMU is compiled with graphical window support, it displays 1882output such as guest graphics, guest console, and the QEMU monitor in a 1883window. With this option, you can have QEMU listen on VNC display 1884@var{display} and redirect the VGA display over the VNC session. It is 1885very useful to enable the usb tablet device when using this option 1886(option @option{-device usb-tablet}). When using the VNC display, you 1887must use the @option{-k} parameter to set the keyboard layout if you are 1888not using en-us. Valid syntax for the @var{display} is 1889 1890@table @option 1891 1892@item to=@var{L} 1893 1894With this option, QEMU will try next available VNC @var{display}s, until the 1895number @var{L}, if the origianlly defined "-vnc @var{display}" is not 1896available, e.g. port 5900+@var{display} is already used by another 1897application. By default, to=0. 1898 1899@item @var{host}:@var{d} 1900 1901TCP connections will only be allowed from @var{host} on display @var{d}. 1902By convention the TCP port is 5900+@var{d}. Optionally, @var{host} can 1903be omitted in which case the server will accept connections from any host. 1904 1905@item unix:@var{path} 1906 1907Connections will be allowed over UNIX domain sockets where @var{path} is the 1908location of a unix socket to listen for connections on. 1909 1910@item none 1911 1912VNC is initialized but not started. The monitor @code{change} command 1913can be used to later start the VNC server. 1914 1915@end table 1916 1917Following the @var{display} value there may be one or more @var{option} flags 1918separated by commas. Valid options are 1919 1920@table @option 1921 1922@item reverse 1923 1924Connect to a listening VNC client via a ``reverse'' connection. The 1925client is specified by the @var{display}. For reverse network 1926connections (@var{host}:@var{d},@code{reverse}), the @var{d} argument 1927is a TCP port number, not a display number. 1928 1929@item websocket 1930 1931Opens an additional TCP listening port dedicated to VNC Websocket connections. 1932If a bare @var{websocket} option is given, the Websocket port is 19335700+@var{display}. An alternative port can be specified with the 1934syntax @code{websocket}=@var{port}. 1935 1936If @var{host} is specified connections will only be allowed from this host. 1937It is possible to control the websocket listen address independently, using 1938the syntax @code{websocket}=@var{host}:@var{port}. 1939 1940If no TLS credentials are provided, the websocket connection runs in 1941unencrypted mode. If TLS credentials are provided, the websocket connection 1942requires encrypted client connections. 1943 1944@item password 1945 1946Require that password based authentication is used for client connections. 1947 1948The password must be set separately using the @code{set_password} command in 1949the @ref{pcsys_monitor}. The syntax to change your password is: 1950@code{set_password <protocol> <password>} where <protocol> could be either 1951"vnc" or "spice". 1952 1953If you would like to change <protocol> password expiration, you should use 1954@code{expire_password <protocol> <expiration-time>} where expiration time could 1955be one of the following options: now, never, +seconds or UNIX time of 1956expiration, e.g. +60 to make password expire in 60 seconds, or 1335196800 1957to make password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for this 1958date and time). 1959 1960You can also use keywords "now" or "never" for the expiration time to 1961allow <protocol> password to expire immediately or never expire. 1962 1963@item tls-creds=@var{ID} 1964 1965Provides the ID of a set of TLS credentials to use to secure the 1966VNC server. They will apply to both the normal VNC server socket 1967and the websocket socket (if enabled). Setting TLS credentials 1968will cause the VNC server socket to enable the VeNCrypt auth 1969mechanism. The credentials should have been previously created 1970using the @option{-object tls-creds} argument. 1971 1972@item tls-authz=@var{ID} 1973 1974Provides the ID of the QAuthZ authorization object against which 1975the client's x509 distinguished name will validated. This object is 1976only resolved at time of use, so can be deleted and recreated on the 1977fly while the VNC server is active. If missing, it will default 1978to denying access. 1979 1980@item sasl 1981 1982Require that the client use SASL to authenticate with the VNC server. 1983The exact choice of authentication method used is controlled from the 1984system / user's SASL configuration file for the 'qemu' service. This 1985is typically found in /etc/sasl2/qemu.conf. If running QEMU as an 1986unprivileged user, an environment variable SASL_CONF_PATH can be used 1987to make it search alternate locations for the service config. 1988While some SASL auth methods can also provide data encryption (eg GSSAPI), 1989it is recommended that SASL always be combined with the 'tls' and 1990'x509' settings to enable use of SSL and server certificates. This 1991ensures a data encryption preventing compromise of authentication 1992credentials. See the @ref{vnc_security} section for details on using 1993SASL authentication. 1994 1995@item sasl-authz=@var{ID} 1996 1997Provides the ID of the QAuthZ authorization object against which 1998the client's SASL username will validated. This object is 1999only resolved at time of use, so can be deleted and recreated on the 2000fly while the VNC server is active. If missing, it will default 2001to denying access. 2002 2003@item acl 2004 2005Legacy method for enabling authorization of clients against the 2006x509 distinguished name and SASL username. It results in the creation 2007of two @code{authz-list} objects with IDs of @code{vnc.username} and 2008@code{vnc.x509dname}. The rules for these objects must be configured 2009with the HMP ACL commands. 2010 2011This option is deprecated and should no longer be used. The new 2012@option{sasl-authz} and @option{tls-authz} options are a 2013replacement. 2014 2015@item lossy 2016 2017Enable lossy compression methods (gradient, JPEG, ...). If this 2018option is set, VNC client may receive lossy framebuffer updates 2019depending on its encoding settings. Enabling this option can save 2020a lot of bandwidth at the expense of quality. 2021 2022@item non-adaptive 2023 2024Disable adaptive encodings. Adaptive encodings are enabled by default. 2025An adaptive encoding will try to detect frequently updated screen regions, 2026and send updates in these regions using a lossy encoding (like JPEG). 2027This can be really helpful to save bandwidth when playing videos. Disabling 2028adaptive encodings restores the original static behavior of encodings 2029like Tight. 2030 2031@item share=[allow-exclusive|force-shared|ignore] 2032 2033Set display sharing policy. 'allow-exclusive' allows clients to ask 2034for exclusive access. As suggested by the rfb spec this is 2035implemented by dropping other connections. Connecting multiple 2036clients in parallel requires all clients asking for a shared session 2037(vncviewer: -shared switch). This is the default. 'force-shared' 2038disables exclusive client access. Useful for shared desktop sessions, 2039where you don't want someone forgetting specify -shared disconnect 2040everybody else. 'ignore' completely ignores the shared flag and 2041allows everybody connect unconditionally. Doesn't conform to the rfb 2042spec but is traditional QEMU behavior. 2043 2044@item key-delay-ms 2045 2046Set keyboard delay, for key down and key up events, in milliseconds. 2047Default is 10. Keyboards are low-bandwidth devices, so this slowdown 2048can help the device and guest to keep up and not lose events in case 2049events are arriving in bulk. Possible causes for the latter are flaky 2050network connections, or scripts for automated testing. 2051 2052@item audiodev=@var{audiodev} 2053 2054Use the specified @var{audiodev} when the VNC client requests audio 2055transmission. When not using an -audiodev argument, this option must 2056be omitted, otherwise is must be present and specify a valid audiodev. 2057 2058@end table 2059ETEXI 2060 2061STEXI 2062@end table 2063ETEXI 2064ARCHHEADING(, QEMU_ARCH_I386) 2065 2066ARCHHEADING(i386 target only:, QEMU_ARCH_I386) 2067STEXI 2068@table @option 2069ETEXI 2070 2071DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack, 2072 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n", 2073 QEMU_ARCH_I386) 2074STEXI 2075@item -win2k-hack 2076@findex -win2k-hack 2077Use it when installing Windows 2000 to avoid a disk full bug. After 2078Windows 2000 is installed, you no longer need this option (this option 2079slows down the IDE transfers). 2080ETEXI 2081 2082DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk, 2083 "-no-fd-bootchk disable boot signature checking for floppy disks\n", 2084 QEMU_ARCH_I386) 2085STEXI 2086@item -no-fd-bootchk 2087@findex -no-fd-bootchk 2088Disable boot signature checking for floppy disks in BIOS. May 2089be needed to boot from old floppy disks. 2090ETEXI 2091 2092DEF("no-acpi", 0, QEMU_OPTION_no_acpi, 2093 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2094STEXI 2095@item -no-acpi 2096@findex -no-acpi 2097Disable ACPI (Advanced Configuration and Power Interface) support. Use 2098it if your guest OS complains about ACPI problems (PC target machine 2099only). 2100ETEXI 2101 2102DEF("no-hpet", 0, QEMU_OPTION_no_hpet, 2103 "-no-hpet disable HPET\n", QEMU_ARCH_I386) 2104STEXI 2105@item -no-hpet 2106@findex -no-hpet 2107Disable HPET support. 2108ETEXI 2109 2110DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable, 2111 "-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" 2112 " ACPI table description\n", QEMU_ARCH_I386) 2113STEXI 2114@item -acpitable [sig=@var{str}][,rev=@var{n}][,oem_id=@var{str}][,oem_table_id=@var{str}][,oem_rev=@var{n}] [,asl_compiler_id=@var{str}][,asl_compiler_rev=@var{n}][,data=@var{file1}[:@var{file2}]...] 2115@findex -acpitable 2116Add ACPI table with specified header fields and context from specified files. 2117For file=, take whole ACPI table from the specified files, including all 2118ACPI headers (possible overridden by other options). 2119For data=, only data 2120portion of the table is used, all header information is specified in the 2121command line. 2122If a SLIC table is supplied to QEMU, then the SLIC's oem_id and oem_table_id 2123fields will override the same in the RSDT and the FADT (a.k.a. FACP), in order 2124to ensure the field matches required by the Microsoft SLIC spec and the ACPI 2125spec. 2126ETEXI 2127 2128DEF("smbios", HAS_ARG, QEMU_OPTION_smbios, 2129 "-smbios file=binary\n" 2130 " load SMBIOS entry from binary file\n" 2131 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n" 2132 " [,uefi=on|off]\n" 2133 " specify SMBIOS type 0 fields\n" 2134 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2135 " [,uuid=uuid][,sku=str][,family=str]\n" 2136 " specify SMBIOS type 1 fields\n" 2137 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n" 2138 " [,asset=str][,location=str]\n" 2139 " specify SMBIOS type 2 fields\n" 2140 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n" 2141 " [,sku=str]\n" 2142 " specify SMBIOS type 3 fields\n" 2143 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n" 2144 " [,asset=str][,part=str]\n" 2145 " specify SMBIOS type 4 fields\n" 2146 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n" 2147 " [,asset=str][,part=str][,speed=%d]\n" 2148 " specify SMBIOS type 17 fields\n", 2149 QEMU_ARCH_I386 | QEMU_ARCH_ARM) 2150STEXI 2151@item -smbios file=@var{binary} 2152@findex -smbios 2153Load SMBIOS entry from binary file. 2154 2155@item -smbios type=0[,vendor=@var{str}][,version=@var{str}][,date=@var{str}][,release=@var{%d.%d}][,uefi=on|off] 2156Specify SMBIOS type 0 fields 2157 2158@item -smbios type=1[,manufacturer=@var{str}][,product=@var{str}][,version=@var{str}][,serial=@var{str}][,uuid=@var{uuid}][,sku=@var{str}][,family=@var{str}] 2159Specify SMBIOS type 1 fields 2160 2161@item -smbios type=2[,manufacturer=@var{str}][,product=@var{str}][,version=@var{str}][,serial=@var{str}][,asset=@var{str}][,location=@var{str}] 2162Specify SMBIOS type 2 fields 2163 2164@item -smbios type=3[,manufacturer=@var{str}][,version=@var{str}][,serial=@var{str}][,asset=@var{str}][,sku=@var{str}] 2165Specify SMBIOS type 3 fields 2166 2167@item -smbios type=4[,sock_pfx=@var{str}][,manufacturer=@var{str}][,version=@var{str}][,serial=@var{str}][,asset=@var{str}][,part=@var{str}] 2168Specify SMBIOS type 4 fields 2169 2170@item -smbios type=17[,loc_pfx=@var{str}][,bank=@var{str}][,manufacturer=@var{str}][,serial=@var{str}][,asset=@var{str}][,part=@var{str}][,speed=@var{%d}] 2171Specify SMBIOS type 17 fields 2172ETEXI 2173 2174STEXI 2175@end table 2176ETEXI 2177DEFHEADING() 2178 2179DEFHEADING(Network options:) 2180STEXI 2181@table @option 2182ETEXI 2183 2184DEF("netdev", HAS_ARG, QEMU_OPTION_netdev, 2185#ifdef CONFIG_SLIRP 2186 "-netdev user,id=str[,ipv4[=on|off]][,net=addr[/mask]][,host=addr]\n" 2187 " [,ipv6[=on|off]][,ipv6-net=addr[/int]][,ipv6-host=addr]\n" 2188 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n" 2189 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n" 2190 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]" 2191#ifndef _WIN32 2192 "[,smb=dir[,smbserver=addr]]\n" 2193#endif 2194 " configure a user mode network backend with ID 'str',\n" 2195 " its DHCP server and optional services\n" 2196#endif 2197#ifdef _WIN32 2198 "-netdev tap,id=str,ifname=name\n" 2199 " configure a host TAP network backend with ID 'str'\n" 2200#else 2201 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n" 2202 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n" 2203 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n" 2204 " [,poll-us=n]\n" 2205 " configure a host TAP network backend with ID 'str'\n" 2206 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2207 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n" 2208 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n" 2209 " to deconfigure it\n" 2210 " use '[down]script=no' to disable script execution\n" 2211 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n" 2212 " configure it\n" 2213 " use 'fd=h' to connect to an already opened TAP interface\n" 2214 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n" 2215 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n" 2216 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n" 2217 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n" 2218 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n" 2219 " use vhost=on to enable experimental in kernel accelerator\n" 2220 " (only has effect for virtio guests which use MSIX)\n" 2221 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n" 2222 " use 'vhostfd=h' to connect to an already opened vhost net device\n" 2223 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n" 2224 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n" 2225 " use 'poll-us=n' to speciy the maximum number of microseconds that could be\n" 2226 " spent on busy polling for vhost net\n" 2227 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n" 2228 " configure a host TAP network backend with ID 'str' that is\n" 2229 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" 2230 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n" 2231#endif 2232#ifdef __linux__ 2233 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n" 2234 " [,rxsession=rxsession],txsession=txsession[,ipv6=on/off][,udp=on/off]\n" 2235 " [,cookie64=on/off][,counter][,pincounter][,txcookie=txcookie]\n" 2236 " [,rxcookie=rxcookie][,offset=offset]\n" 2237 " configure a network backend with ID 'str' connected to\n" 2238 " an Ethernet over L2TPv3 pseudowire.\n" 2239 " Linux kernel 3.3+ as well as most routers can talk\n" 2240 " L2TPv3. This transport allows connecting a VM to a VM,\n" 2241 " VM to a router and even VM to Host. It is a nearly-universal\n" 2242 " standard (RFC3391). Note - this implementation uses static\n" 2243 " pre-configured tunnels (same as the Linux kernel).\n" 2244 " use 'src=' to specify source address\n" 2245 " use 'dst=' to specify destination address\n" 2246 " use 'udp=on' to specify udp encapsulation\n" 2247 " use 'srcport=' to specify source udp port\n" 2248 " use 'dstport=' to specify destination udp port\n" 2249 " use 'ipv6=on' to force v6\n" 2250 " L2TPv3 uses cookies to prevent misconfiguration as\n" 2251 " well as a weak security measure\n" 2252 " use 'rxcookie=0x012345678' to specify a rxcookie\n" 2253 " use 'txcookie=0x012345678' to specify a txcookie\n" 2254 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n" 2255 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n" 2256 " use 'pincounter=on' to work around broken counter handling in peer\n" 2257 " use 'offset=X' to add an extra offset between header and data\n" 2258#endif 2259 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n" 2260 " configure a network backend to connect to another network\n" 2261 " using a socket connection\n" 2262 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n" 2263 " configure a network backend to connect to a multicast maddr and port\n" 2264 " use 'localaddr=addr' to specify the host address to send packets from\n" 2265 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n" 2266 " configure a network backend to connect to another network\n" 2267 " using an UDP tunnel\n" 2268#ifdef CONFIG_VDE 2269 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n" 2270 " configure a network backend to connect to port 'n' of a vde switch\n" 2271 " running on host and listening for incoming connections on 'socketpath'.\n" 2272 " Use group 'groupname' and mode 'octalmode' to change default\n" 2273 " ownership and permissions for communication port.\n" 2274#endif 2275#ifdef CONFIG_NETMAP 2276 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n" 2277 " attach to the existing netmap-enabled network interface 'name', or to a\n" 2278 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n" 2279 " netmap device, defaults to '/dev/netmap')\n" 2280#endif 2281#ifdef CONFIG_POSIX 2282 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n" 2283 " configure a vhost-user network, backed by a chardev 'dev'\n" 2284#endif 2285 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n" 2286 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL) 2287DEF("nic", HAS_ARG, QEMU_OPTION_nic, 2288 "-nic [tap|bridge|" 2289#ifdef CONFIG_SLIRP 2290 "user|" 2291#endif 2292#ifdef __linux__ 2293 "l2tpv3|" 2294#endif 2295#ifdef CONFIG_VDE 2296 "vde|" 2297#endif 2298#ifdef CONFIG_NETMAP 2299 "netmap|" 2300#endif 2301#ifdef CONFIG_POSIX 2302 "vhost-user|" 2303#endif 2304 "socket][,option][,...][mac=macaddr]\n" 2305 " initialize an on-board / default host NIC (using MAC address\n" 2306 " macaddr) and connect it to the given host network backend\n" 2307 "-nic none use it alone to have zero network devices (the default is to\n" 2308 " provided a 'user' network connection)\n", 2309 QEMU_ARCH_ALL) 2310DEF("net", HAS_ARG, QEMU_OPTION_net, 2311 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n" 2312 " configure or create an on-board (or machine default) NIC and\n" 2313 " connect it to hub 0 (please use -nic unless you need a hub)\n" 2314 "-net [" 2315#ifdef CONFIG_SLIRP 2316 "user|" 2317#endif 2318 "tap|" 2319 "bridge|" 2320#ifdef CONFIG_VDE 2321 "vde|" 2322#endif 2323#ifdef CONFIG_NETMAP 2324 "netmap|" 2325#endif 2326 "socket][,option][,option][,...]\n" 2327 " old way to initialize a host network interface\n" 2328 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL) 2329STEXI 2330@item -nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn] 2331@findex -nic 2332This option is a shortcut for configuring both the on-board (default) guest 2333NIC hardware and the host network backend in one go. The host backend options 2334are the same as with the corresponding @option{-netdev} options below. 2335The guest NIC model can be set with @option{model=@var{modelname}}. 2336Use @option{model=help} to list the available device types. 2337The hardware MAC address can be set with @option{mac=@var{macaddr}}. 2338 2339The following two example do exactly the same, to show how @option{-nic} can 2340be used to shorten the command line length (note that the e1000 is the default 2341on i386, so the @option{model=e1000} parameter could even be omitted here, too): 2342@example 2343@value{qemu_system} -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32 2344@value{qemu_system} -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32 2345@end example 2346 2347@item -nic none 2348Indicate that no network devices should be configured. It is used to override 2349the default configuration (default NIC with ``user'' host network backend) 2350which is activated if no other networking options are provided. 2351 2352@item -netdev user,id=@var{id}[,@var{option}][,@var{option}][,...] 2353@findex -netdev 2354Configure user mode host network backend which requires no administrator 2355privilege to run. Valid options are: 2356 2357@table @option 2358@item id=@var{id} 2359Assign symbolic name for use in monitor commands. 2360 2361@item ipv4=on|off and ipv6=on|off 2362Specify that either IPv4 or IPv6 must be enabled. If neither is specified 2363both protocols are enabled. 2364 2365@item net=@var{addr}[/@var{mask}] 2366Set IP network address the guest will see. Optionally specify the netmask, 2367either in the form a.b.c.d or as number of valid top-most bits. Default is 236810.0.2.0/24. 2369 2370@item host=@var{addr} 2371Specify the guest-visible address of the host. Default is the 2nd IP in the 2372guest network, i.e. x.x.x.2. 2373 2374@item ipv6-net=@var{addr}[/@var{int}] 2375Set IPv6 network address the guest will see (default is fec0::/64). The 2376network prefix is given in the usual hexadecimal IPv6 address 2377notation. The prefix size is optional, and is given as the number of 2378valid top-most bits (default is 64). 2379 2380@item ipv6-host=@var{addr} 2381Specify the guest-visible IPv6 address of the host. Default is the 2nd IPv6 in 2382the guest network, i.e. xxxx::2. 2383 2384@item restrict=on|off 2385If this option is enabled, the guest will be isolated, i.e. it will not be 2386able to contact the host and no guest IP packets will be routed over the host 2387to the outside. This option does not affect any explicitly set forwarding rules. 2388 2389@item hostname=@var{name} 2390Specifies the client hostname reported by the built-in DHCP server. 2391 2392@item dhcpstart=@var{addr} 2393Specify the first of the 16 IPs the built-in DHCP server can assign. Default 2394is the 15th to 31st IP in the guest network, i.e. x.x.x.15 to x.x.x.31. 2395 2396@item dns=@var{addr} 2397Specify the guest-visible address of the virtual nameserver. The address must 2398be different from the host address. Default is the 3rd IP in the guest network, 2399i.e. x.x.x.3. 2400 2401@item ipv6-dns=@var{addr} 2402Specify the guest-visible address of the IPv6 virtual nameserver. The address 2403must be different from the host address. Default is the 3rd IP in the guest 2404network, i.e. xxxx::3. 2405 2406@item dnssearch=@var{domain} 2407Provides an entry for the domain-search list sent by the built-in 2408DHCP server. More than one domain suffix can be transmitted by specifying 2409this option multiple times. If supported, this will cause the guest to 2410automatically try to append the given domain suffix(es) in case a domain name 2411can not be resolved. 2412 2413Example: 2414@example 2415@value{qemu_system} -nic user,dnssearch=mgmt.example.org,dnssearch=example.org 2416@end example 2417 2418@item domainname=@var{domain} 2419Specifies the client domain name reported by the built-in DHCP server. 2420 2421@item tftp=@var{dir} 2422When using the user mode network stack, activate a built-in TFTP 2423server. The files in @var{dir} will be exposed as the root of a TFTP server. 2424The TFTP client on the guest must be configured in binary mode (use the command 2425@code{bin} of the Unix TFTP client). 2426 2427@item tftp-server-name=@var{name} 2428In BOOTP reply, broadcast @var{name} as the "TFTP server name" (RFC2132 option 242966). This can be used to advise the guest to load boot files or configurations 2430from a different server than the host address. 2431 2432@item bootfile=@var{file} 2433When using the user mode network stack, broadcast @var{file} as the BOOTP 2434filename. In conjunction with @option{tftp}, this can be used to network boot 2435a guest from a local directory. 2436 2437Example (using pxelinux): 2438@example 2439@value{qemu_system} -hda linux.img -boot n -device e1000,netdev=n1 \ 2440 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0 2441@end example 2442 2443@item smb=@var{dir}[,smbserver=@var{addr}] 2444When using the user mode network stack, activate a built-in SMB 2445server so that Windows OSes can access to the host files in @file{@var{dir}} 2446transparently. The IP address of the SMB server can be set to @var{addr}. By 2447default the 4th IP in the guest network is used, i.e. x.x.x.4. 2448 2449In the guest Windows OS, the line: 2450@example 245110.0.2.4 smbserver 2452@end example 2453must be added in the file @file{C:\WINDOWS\LMHOSTS} (for windows 9x/Me) 2454or @file{C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS} (Windows NT/2000). 2455 2456Then @file{@var{dir}} can be accessed in @file{\\smbserver\qemu}. 2457 2458Note that a SAMBA server must be installed on the host OS. 2459 2460@item hostfwd=[tcp|udp]:[@var{hostaddr}]:@var{hostport}-[@var{guestaddr}]:@var{guestport} 2461Redirect incoming TCP or UDP connections to the host port @var{hostport} to 2462the guest IP address @var{guestaddr} on guest port @var{guestport}. If 2463@var{guestaddr} is not specified, its value is x.x.x.15 (default first address 2464given by the built-in DHCP server). By specifying @var{hostaddr}, the rule can 2465be bound to a specific host interface. If no connection type is set, TCP is 2466used. This option can be given multiple times. 2467 2468For example, to redirect host X11 connection from screen 1 to guest 2469screen 0, use the following: 2470 2471@example 2472# on the host 2473@value{qemu_system} -nic user,hostfwd=tcp:127.0.0.1:6001-:6000 2474# this host xterm should open in the guest X11 server 2475xterm -display :1 2476@end example 2477 2478To redirect telnet connections from host port 5555 to telnet port on 2479the guest, use the following: 2480 2481@example 2482# on the host 2483@value{qemu_system} -nic user,hostfwd=tcp::5555-:23 2484telnet localhost 5555 2485@end example 2486 2487Then when you use on the host @code{telnet localhost 5555}, you 2488connect to the guest telnet server. 2489 2490@item guestfwd=[tcp]:@var{server}:@var{port}-@var{dev} 2491@itemx guestfwd=[tcp]:@var{server}:@var{port}-@var{cmd:command} 2492Forward guest TCP connections to the IP address @var{server} on port @var{port} 2493to the character device @var{dev} or to a program executed by @var{cmd:command} 2494which gets spawned for each connection. This option can be given multiple times. 2495 2496You can either use a chardev directly and have that one used throughout QEMU's 2497lifetime, like in the following example: 2498 2499@example 2500# open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever 2501# the guest accesses it 2502@value{qemu_system} -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321 2503@end example 2504 2505Or you can execute a command on every TCP connection established by the guest, 2506so that QEMU behaves similar to an inetd process for that virtual server: 2507 2508@example 2509# call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234 2510# and connect the TCP stream to its stdin/stdout 2511@value{qemu_system} -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321' 2512@end example 2513 2514@end table 2515 2516@item -netdev tap,id=@var{id}[,fd=@var{h}][,ifname=@var{name}][,script=@var{file}][,downscript=@var{dfile}][,br=@var{bridge}][,helper=@var{helper}] 2517Configure a host TAP network backend with ID @var{id}. 2518 2519Use the network script @var{file} to configure it and the network script 2520@var{dfile} to deconfigure it. If @var{name} is not provided, the OS 2521automatically provides one. The default network configure script is 2522@file{/etc/qemu-ifup} and the default network deconfigure script is 2523@file{/etc/qemu-ifdown}. Use @option{script=no} or @option{downscript=no} 2524to disable script execution. 2525 2526If running QEMU as an unprivileged user, use the network helper 2527@var{helper} to configure the TAP interface and attach it to the bridge. 2528The default network helper executable is @file{/path/to/qemu-bridge-helper} 2529and the default bridge device is @file{br0}. 2530 2531@option{fd}=@var{h} can be used to specify the handle of an already 2532opened host TAP interface. 2533 2534Examples: 2535 2536@example 2537#launch a QEMU instance with the default network script 2538@value{qemu_system} linux.img -nic tap 2539@end example 2540 2541@example 2542#launch a QEMU instance with two NICs, each one connected 2543#to a TAP device 2544@value{qemu_system} linux.img \ 2545 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \ 2546 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1 2547@end example 2548 2549@example 2550#launch a QEMU instance with the default network helper to 2551#connect a TAP device to bridge br0 2552@value{qemu_system} linux.img -device virtio-net-pci,netdev=n1 \ 2553 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper" 2554@end example 2555 2556@item -netdev bridge,id=@var{id}[,br=@var{bridge}][,helper=@var{helper}] 2557Connect a host TAP network interface to a host bridge device. 2558 2559Use the network helper @var{helper} to configure the TAP interface and 2560attach it to the bridge. The default network helper executable is 2561@file{/path/to/qemu-bridge-helper} and the default bridge 2562device is @file{br0}. 2563 2564Examples: 2565 2566@example 2567#launch a QEMU instance with the default network helper to 2568#connect a TAP device to bridge br0 2569@value{qemu_system} linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1 2570@end example 2571 2572@example 2573#launch a QEMU instance with the default network helper to 2574#connect a TAP device to bridge qemubr0 2575@value{qemu_system} linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1 2576@end example 2577 2578@item -netdev socket,id=@var{id}[,fd=@var{h}][,listen=[@var{host}]:@var{port}][,connect=@var{host}:@var{port}] 2579 2580This host network backend can be used to connect the guest's network to 2581another QEMU virtual machine using a TCP socket connection. If @option{listen} 2582is specified, QEMU waits for incoming connections on @var{port} 2583(@var{host} is optional). @option{connect} is used to connect to 2584another QEMU instance using the @option{listen} option. @option{fd}=@var{h} 2585specifies an already opened TCP socket. 2586 2587Example: 2588@example 2589# launch a first QEMU instance 2590@value{qemu_system} linux.img \ 2591 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \ 2592 -netdev socket,id=n1,listen=:1234 2593# connect the network of this instance to the network of the first instance 2594@value{qemu_system} linux.img \ 2595 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \ 2596 -netdev socket,id=n2,connect=127.0.0.1:1234 2597@end example 2598 2599@item -netdev socket,id=@var{id}[,fd=@var{h}][,mcast=@var{maddr}:@var{port}[,localaddr=@var{addr}]] 2600 2601Configure a socket host network backend to share the guest's network traffic 2602with another QEMU virtual machines using a UDP multicast socket, effectively 2603making a bus for every QEMU with same multicast address @var{maddr} and @var{port}. 2604NOTES: 2605@enumerate 2606@item 2607Several QEMU can be running on different hosts and share same bus (assuming 2608correct multicast setup for these hosts). 2609@item 2610mcast support is compatible with User Mode Linux (argument @option{eth@var{N}=mcast}), see 2611@url{http://user-mode-linux.sf.net}. 2612@item 2613Use @option{fd=h} to specify an already opened UDP multicast socket. 2614@end enumerate 2615 2616Example: 2617@example 2618# launch one QEMU instance 2619@value{qemu_system} linux.img \ 2620 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \ 2621 -netdev socket,id=n1,mcast=230.0.0.1:1234 2622# launch another QEMU instance on same "bus" 2623@value{qemu_system} linux.img \ 2624 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \ 2625 -netdev socket,id=n2,mcast=230.0.0.1:1234 2626# launch yet another QEMU instance on same "bus" 2627@value{qemu_system} linux.img \ 2628 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \ 2629 -netdev socket,id=n3,mcast=230.0.0.1:1234 2630@end example 2631 2632Example (User Mode Linux compat.): 2633@example 2634# launch QEMU instance (note mcast address selected is UML's default) 2635@value{qemu_system} linux.img \ 2636 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \ 2637 -netdev socket,id=n1,mcast=239.192.168.1:1102 2638# launch UML 2639/path/to/linux ubd0=/path/to/root_fs eth0=mcast 2640@end example 2641 2642Example (send packets from host's 1.2.3.4): 2643@example 2644@value{qemu_system} linux.img \ 2645 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \ 2646 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4 2647@end example 2648 2649@item -netdev l2tpv3,id=@var{id},src=@var{srcaddr},dst=@var{dstaddr}[,srcport=@var{srcport}][,dstport=@var{dstport}],txsession=@var{txsession}[,rxsession=@var{rxsession}][,ipv6][,udp][,cookie64][,counter][,pincounter][,txcookie=@var{txcookie}][,rxcookie=@var{rxcookie}][,offset=@var{offset}] 2650Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3391) is a 2651popular protocol to transport Ethernet (and other Layer 2) data frames between 2652two systems. It is present in routers, firewalls and the Linux kernel 2653(from version 3.3 onwards). 2654 2655This transport allows a VM to communicate to another VM, router or firewall directly. 2656 2657@table @option 2658@item src=@var{srcaddr} 2659 source address (mandatory) 2660@item dst=@var{dstaddr} 2661 destination address (mandatory) 2662@item udp 2663 select udp encapsulation (default is ip). 2664@item srcport=@var{srcport} 2665 source udp port. 2666@item dstport=@var{dstport} 2667 destination udp port. 2668@item ipv6 2669 force v6, otherwise defaults to v4. 2670@item rxcookie=@var{rxcookie} 2671@itemx txcookie=@var{txcookie} 2672 Cookies are a weak form of security in the l2tpv3 specification. 2673Their function is mostly to prevent misconfiguration. By default they are 32 2674bit. 2675@item cookie64 2676 Set cookie size to 64 bit instead of the default 32 2677@item counter=off 2678 Force a 'cut-down' L2TPv3 with no counter as in 2679draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00 2680@item pincounter=on 2681 Work around broken counter handling in peer. This may also help on 2682networks which have packet reorder. 2683@item offset=@var{offset} 2684 Add an extra offset between header and data 2685@end table 2686 2687For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to the bridge br-lan 2688on the remote Linux host 1.2.3.4: 2689@example 2690# Setup tunnel on linux host using raw ip as encapsulation 2691# on 1.2.3.4 2692ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \ 2693 encap udp udp_sport 16384 udp_dport 16384 2694ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \ 2695 0xFFFFFFFF peer_session_id 0xFFFFFFFF 2696ifconfig vmtunnel0 mtu 1500 2697ifconfig vmtunnel0 up 2698brctl addif br-lan vmtunnel0 2699 2700 2701# on 4.3.2.1 2702# launch QEMU instance - if your network has reorder or is very lossy add ,pincounter 2703 2704@value{qemu_system} linux.img -device e1000,netdev=n1 \ 2705 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter 2706 2707@end example 2708 2709@item -netdev vde,id=@var{id}[,sock=@var{socketpath}][,port=@var{n}][,group=@var{groupname}][,mode=@var{octalmode}] 2710Configure VDE backend to connect to PORT @var{n} of a vde switch running on host and 2711listening for incoming connections on @var{socketpath}. Use GROUP @var{groupname} 2712and MODE @var{octalmode} to change default ownership and permissions for 2713communication port. This option is only available if QEMU has been compiled 2714with vde support enabled. 2715 2716Example: 2717@example 2718# launch vde switch 2719vde_switch -F -sock /tmp/myswitch 2720# launch QEMU instance 2721@value{qemu_system} linux.img -nic vde,sock=/tmp/myswitch 2722@end example 2723 2724@item -netdev vhost-user,chardev=@var{id}[,vhostforce=on|off][,queues=n] 2725 2726Establish a vhost-user netdev, backed by a chardev @var{id}. The chardev should 2727be a unix domain socket backed one. The vhost-user uses a specifically defined 2728protocol to pass vhost ioctl replacement messages to an application on the other 2729end of the socket. On non-MSIX guests, the feature can be forced with 2730@var{vhostforce}. Use 'queues=@var{n}' to specify the number of queues to 2731be created for multiqueue vhost-user. 2732 2733Example: 2734@example 2735qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \ 2736 -numa node,memdev=mem \ 2737 -chardev socket,id=chr0,path=/path/to/socket \ 2738 -netdev type=vhost-user,id=net0,chardev=chr0 \ 2739 -device virtio-net-pci,netdev=net0 2740@end example 2741 2742@item -netdev hubport,id=@var{id},hubid=@var{hubid}[,netdev=@var{nd}] 2743 2744Create a hub port on the emulated hub with ID @var{hubid}. 2745 2746The hubport netdev lets you connect a NIC to a QEMU emulated hub instead of a 2747single netdev. Alternatively, you can also connect the hubport to another 2748netdev with ID @var{nd} by using the @option{netdev=@var{nd}} option. 2749 2750@item -net nic[,netdev=@var{nd}][,macaddr=@var{mac}][,model=@var{type}] [,name=@var{name}][,addr=@var{addr}][,vectors=@var{v}] 2751@findex -net 2752Legacy option to configure or create an on-board (or machine default) Network 2753Interface Card(NIC) and connect it either to the emulated hub with ID 0 (i.e. 2754the default hub), or to the netdev @var{nd}. 2755The NIC is an e1000 by default on the PC target. Optionally, the MAC address 2756can be changed to @var{mac}, the device address set to @var{addr} (PCI cards 2757only), and a @var{name} can be assigned for use in monitor commands. 2758Optionally, for PCI cards, you can specify the number @var{v} of MSI-X vectors 2759that the card should have; this option currently only affects virtio cards; set 2760@var{v} = 0 to disable MSI-X. If no @option{-net} option is specified, a single 2761NIC is created. QEMU can emulate several different models of network card. 2762Use @code{-net nic,model=help} for a list of available devices for your target. 2763 2764@item -net user|tap|bridge|socket|l2tpv3|vde[,...][,name=@var{name}] 2765Configure a host network backend (with the options corresponding to the same 2766@option{-netdev} option) and connect it to the emulated hub 0 (the default 2767hub). Use @var{name} to specify the name of the hub port. 2768ETEXI 2769 2770STEXI 2771@end table 2772ETEXI 2773DEFHEADING() 2774 2775DEFHEADING(Character device options:) 2776 2777DEF("chardev", HAS_ARG, QEMU_OPTION_chardev, 2778 "-chardev help\n" 2779 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2780 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4][,ipv6][,nodelay][,reconnect=seconds]\n" 2781 " [,server][,nowait][,telnet][,websocket][,reconnect=seconds][,mux=on|off]\n" 2782 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n" 2783 "-chardev socket,id=id,path=path[,server][,nowait][,telnet][,websocket][,reconnect=seconds]\n" 2784 " [,mux=on|off][,logfile=PATH][,logappend=on|off] (unix)\n" 2785 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n" 2786 " [,localport=localport][,ipv4][,ipv6][,mux=on|off]\n" 2787 " [,logfile=PATH][,logappend=on|off]\n" 2788 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2789 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n" 2790 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2791 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n" 2792 "-chardev file,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2793 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2794#ifdef _WIN32 2795 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2796 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2797#else 2798 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2799 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n" 2800#endif 2801#ifdef CONFIG_BRLAPI 2802 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2803#endif 2804#if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \ 2805 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) 2806 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2807 "-chardev tty,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2808#endif 2809#if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__) 2810 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2811 "-chardev parport,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" 2812#endif 2813#if defined(CONFIG_SPICE) 2814 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 2815 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" 2816#endif 2817 , QEMU_ARCH_ALL 2818) 2819 2820STEXI 2821 2822The general form of a character device option is: 2823@table @option 2824@item -chardev @var{backend},id=@var{id}[,mux=on|off][,@var{options}] 2825@findex -chardev 2826Backend is one of: 2827@option{null}, 2828@option{socket}, 2829@option{udp}, 2830@option{msmouse}, 2831@option{vc}, 2832@option{ringbuf}, 2833@option{file}, 2834@option{pipe}, 2835@option{console}, 2836@option{serial}, 2837@option{pty}, 2838@option{stdio}, 2839@option{braille}, 2840@option{tty}, 2841@option{parallel}, 2842@option{parport}, 2843@option{spicevmc}, 2844@option{spiceport}. 2845The specific backend will determine the applicable options. 2846 2847Use @code{-chardev help} to print all available chardev backend types. 2848 2849All devices must have an id, which can be any string up to 127 characters long. 2850It is used to uniquely identify this device in other command line directives. 2851 2852A character device may be used in multiplexing mode by multiple front-ends. 2853Specify @option{mux=on} to enable this mode. 2854A multiplexer is a "1:N" device, and here the "1" end is your specified chardev 2855backend, and the "N" end is the various parts of QEMU that can talk to a chardev. 2856If you create a chardev with @option{id=myid} and @option{mux=on}, QEMU will 2857create a multiplexer with your specified ID, and you can then configure multiple 2858front ends to use that chardev ID for their input/output. Up to four different 2859front ends can be connected to a single multiplexed chardev. (Without 2860multiplexing enabled, a chardev can only be used by a single front end.) 2861For instance you could use this to allow a single stdio chardev to be used by 2862two serial ports and the QEMU monitor: 2863 2864@example 2865-chardev stdio,mux=on,id=char0 \ 2866-mon chardev=char0,mode=readline \ 2867-serial chardev:char0 \ 2868-serial chardev:char0 2869@end example 2870 2871You can have more than one multiplexer in a system configuration; for instance 2872you could have a TCP port multiplexed between UART 0 and UART 1, and stdio 2873multiplexed between the QEMU monitor and a parallel port: 2874 2875@example 2876-chardev stdio,mux=on,id=char0 \ 2877-mon chardev=char0,mode=readline \ 2878-parallel chardev:char0 \ 2879-chardev tcp,...,mux=on,id=char1 \ 2880-serial chardev:char1 \ 2881-serial chardev:char1 2882@end example 2883 2884When you're using a multiplexed character device, some escape sequences are 2885interpreted in the input. @xref{mux_keys, Keys in the character backend 2886multiplexer}. 2887 2888Note that some other command line options may implicitly create multiplexed 2889character backends; for instance @option{-serial mon:stdio} creates a 2890multiplexed stdio backend connected to the serial port and the QEMU monitor, 2891and @option{-nographic} also multiplexes the console and the monitor to 2892stdio. 2893 2894There is currently no support for multiplexing in the other direction 2895(where a single QEMU front end takes input and output from multiple chardevs). 2896 2897Every backend supports the @option{logfile} option, which supplies the path 2898to a file to record all data transmitted via the backend. The @option{logappend} 2899option controls whether the log file will be truncated or appended to when 2900opened. 2901 2902@end table 2903 2904The available backends are: 2905 2906@table @option 2907@item -chardev null,id=@var{id} 2908A void device. This device will not emit any data, and will drop any data it 2909receives. The null backend does not take any options. 2910 2911@item -chardev socket,id=@var{id}[,@var{TCP options} or @var{unix options}][,server][,nowait][,telnet][,websocket][,reconnect=@var{seconds}][,tls-creds=@var{id}][,tls-authz=@var{id}] 2912 2913Create a two-way stream socket, which can be either a TCP or a unix socket. A 2914unix socket will be created if @option{path} is specified. Behaviour is 2915undefined if TCP options are specified for a unix socket. 2916 2917@option{server} specifies that the socket shall be a listening socket. 2918 2919@option{nowait} specifies that QEMU should not block waiting for a client to 2920connect to a listening socket. 2921 2922@option{telnet} specifies that traffic on the socket should interpret telnet 2923escape sequences. 2924 2925@option{websocket} specifies that the socket uses WebSocket protocol for 2926communication. 2927 2928@option{reconnect} sets the timeout for reconnecting on non-server sockets when 2929the remote end goes away. qemu will delay this many seconds and then attempt 2930to reconnect. Zero disables reconnecting, and is the default. 2931 2932@option{tls-creds} requests enablement of the TLS protocol for encryption, 2933and specifies the id of the TLS credentials to use for the handshake. The 2934credentials must be previously created with the @option{-object tls-creds} 2935argument. 2936 2937@option{tls-auth} provides the ID of the QAuthZ authorization object against 2938which the client's x509 distinguished name will be validated. This object is 2939only resolved at time of use, so can be deleted and recreated on the fly 2940while the chardev server is active. If missing, it will default to denying 2941access. 2942 2943TCP and unix socket options are given below: 2944 2945@table @option 2946 2947@item TCP options: port=@var{port}[,host=@var{host}][,to=@var{to}][,ipv4][,ipv6][,nodelay] 2948 2949@option{host} for a listening socket specifies the local address to be bound. 2950For a connecting socket species the remote host to connect to. @option{host} is 2951optional for listening sockets. If not specified it defaults to @code{0.0.0.0}. 2952 2953@option{port} for a listening socket specifies the local port to be bound. For a 2954connecting socket specifies the port on the remote host to connect to. 2955@option{port} can be given as either a port number or a service name. 2956@option{port} is required. 2957 2958@option{to} is only relevant to listening sockets. If it is specified, and 2959@option{port} cannot be bound, QEMU will attempt to bind to subsequent ports up 2960to and including @option{to} until it succeeds. @option{to} must be specified 2961as a port number. 2962 2963@option{ipv4} and @option{ipv6} specify that either IPv4 or IPv6 must be used. 2964If neither is specified the socket may use either protocol. 2965 2966@option{nodelay} disables the Nagle algorithm. 2967 2968@item unix options: path=@var{path} 2969 2970@option{path} specifies the local path of the unix socket. @option{path} is 2971required. 2972 2973@end table 2974 2975@item -chardev udp,id=@var{id}[,host=@var{host}],port=@var{port}[,localaddr=@var{localaddr}][,localport=@var{localport}][,ipv4][,ipv6] 2976 2977Sends all traffic from the guest to a remote host over UDP. 2978 2979@option{host} specifies the remote host to connect to. If not specified it 2980defaults to @code{localhost}. 2981 2982@option{port} specifies the port on the remote host to connect to. @option{port} 2983is required. 2984 2985@option{localaddr} specifies the local address to bind to. If not specified it 2986defaults to @code{0.0.0.0}. 2987 2988@option{localport} specifies the local port to bind to. If not specified any 2989available local port will be used. 2990 2991@option{ipv4} and @option{ipv6} specify that either IPv4 or IPv6 must be used. 2992If neither is specified the device may use either protocol. 2993 2994@item -chardev msmouse,id=@var{id} 2995 2996Forward QEMU's emulated msmouse events to the guest. @option{msmouse} does not 2997take any options. 2998 2999@item -chardev vc,id=@var{id}[[,width=@var{width}][,height=@var{height}]][[,cols=@var{cols}][,rows=@var{rows}]] 3000 3001Connect to a QEMU text console. @option{vc} may optionally be given a specific 3002size. 3003 3004@option{width} and @option{height} specify the width and height respectively of 3005the console, in pixels. 3006 3007@option{cols} and @option{rows} specify that the console be sized to fit a text 3008console with the given dimensions. 3009 3010@item -chardev ringbuf,id=@var{id}[,size=@var{size}] 3011 3012Create a ring buffer with fixed size @option{size}. 3013@var{size} must be a power of two and defaults to @code{64K}. 3014 3015@item -chardev file,id=@var{id},path=@var{path} 3016 3017Log all traffic received from the guest to a file. 3018 3019@option{path} specifies the path of the file to be opened. This file will be 3020created if it does not already exist, and overwritten if it does. @option{path} 3021is required. 3022 3023@item -chardev pipe,id=@var{id},path=@var{path} 3024 3025Create a two-way connection to the guest. The behaviour differs slightly between 3026Windows hosts and other hosts: 3027 3028On Windows, a single duplex pipe will be created at 3029@file{\\.pipe\@option{path}}. 3030 3031On other hosts, 2 pipes will be created called @file{@option{path}.in} and 3032@file{@option{path}.out}. Data written to @file{@option{path}.in} will be 3033received by the guest. Data written by the guest can be read from 3034@file{@option{path}.out}. QEMU will not create these fifos, and requires them to 3035be present. 3036 3037@option{path} forms part of the pipe path as described above. @option{path} is 3038required. 3039 3040@item -chardev console,id=@var{id} 3041 3042Send traffic from the guest to QEMU's standard output. @option{console} does not 3043take any options. 3044 3045@option{console} is only available on Windows hosts. 3046 3047@item -chardev serial,id=@var{id},path=@option{path} 3048 3049Send traffic from the guest to a serial device on the host. 3050 3051On Unix hosts serial will actually accept any tty device, 3052not only serial lines. 3053 3054@option{path} specifies the name of the serial device to open. 3055 3056@item -chardev pty,id=@var{id} 3057 3058Create a new pseudo-terminal on the host and connect to it. @option{pty} does 3059not take any options. 3060 3061@option{pty} is not available on Windows hosts. 3062 3063@item -chardev stdio,id=@var{id}[,signal=on|off] 3064Connect to standard input and standard output of the QEMU process. 3065 3066@option{signal} controls if signals are enabled on the terminal, that includes 3067exiting QEMU with the key sequence @key{Control-c}. This option is enabled by 3068default, use @option{signal=off} to disable it. 3069 3070@item -chardev braille,id=@var{id} 3071 3072Connect to a local BrlAPI server. @option{braille} does not take any options. 3073 3074@item -chardev tty,id=@var{id},path=@var{path} 3075 3076@option{tty} is only available on Linux, Sun, FreeBSD, NetBSD, OpenBSD and 3077DragonFlyBSD hosts. It is an alias for @option{serial}. 3078 3079@option{path} specifies the path to the tty. @option{path} is required. 3080 3081@item -chardev parallel,id=@var{id},path=@var{path} 3082@itemx -chardev parport,id=@var{id},path=@var{path} 3083 3084@option{parallel} is only available on Linux, FreeBSD and DragonFlyBSD hosts. 3085 3086Connect to a local parallel port. 3087 3088@option{path} specifies the path to the parallel port device. @option{path} is 3089required. 3090 3091@item -chardev spicevmc,id=@var{id},debug=@var{debug},name=@var{name} 3092 3093@option{spicevmc} is only available when spice support is built in. 3094 3095@option{debug} debug level for spicevmc 3096 3097@option{name} name of spice channel to connect to 3098 3099Connect to a spice virtual machine channel, such as vdiport. 3100 3101@item -chardev spiceport,id=@var{id},debug=@var{debug},name=@var{name} 3102 3103@option{spiceport} is only available when spice support is built in. 3104 3105@option{debug} debug level for spicevmc 3106 3107@option{name} name of spice port to connect to 3108 3109Connect to a spice port, allowing a Spice client to handle the traffic 3110identified by a name (preferably a fqdn). 3111ETEXI 3112 3113STEXI 3114@end table 3115ETEXI 3116DEFHEADING() 3117 3118DEFHEADING(Bluetooth(R) options:) 3119STEXI 3120@table @option 3121ETEXI 3122 3123DEF("bt", HAS_ARG, QEMU_OPTION_bt, \ 3124 "-bt hci,null dumb bluetooth HCI - doesn't respond to commands\n" \ 3125 "-bt hci,host[:id]\n" \ 3126 " use host's HCI with the given name\n" \ 3127 "-bt hci[,vlan=n]\n" \ 3128 " emulate a standard HCI in virtual scatternet 'n'\n" \ 3129 "-bt vhci[,vlan=n]\n" \ 3130 " add host computer to virtual scatternet 'n' using VHCI\n" \ 3131 "-bt device:dev[,vlan=n]\n" \ 3132 " emulate a bluetooth device 'dev' in scatternet 'n'\n", 3133 QEMU_ARCH_ALL) 3134STEXI 3135@item -bt hci[...] 3136@findex -bt 3137Defines the function of the corresponding Bluetooth HCI. -bt options 3138are matched with the HCIs present in the chosen machine type. For 3139example when emulating a machine with only one HCI built into it, only 3140the first @code{-bt hci[...]} option is valid and defines the HCI's 3141logic. The Transport Layer is decided by the machine type. Currently 3142the machines @code{n800} and @code{n810} have one HCI and all other 3143machines have none. 3144 3145Note: This option and the whole bluetooth subsystem is considered as deprecated. 3146If you still use it, please send a mail to @email{qemu-devel@@nongnu.org} where 3147you describe your usecase. 3148 3149@anchor{bt-hcis} 3150The following three types are recognized: 3151 3152@table @option 3153@item -bt hci,null 3154(default) The corresponding Bluetooth HCI assumes no internal logic 3155and will not respond to any HCI commands or emit events. 3156 3157@item -bt hci,host[:@var{id}] 3158(@code{bluez} only) The corresponding HCI passes commands / events 3159to / from the physical HCI identified by the name @var{id} (default: 3160@code{hci0}) on the computer running QEMU. Only available on @code{bluez} 3161capable systems like Linux. 3162 3163@item -bt hci[,vlan=@var{n}] 3164Add a virtual, standard HCI that will participate in the Bluetooth 3165scatternet @var{n} (default @code{0}). Similarly to @option{-net} 3166VLANs, devices inside a bluetooth network @var{n} can only communicate 3167with other devices in the same network (scatternet). 3168@end table 3169 3170@item -bt vhci[,vlan=@var{n}] 3171(Linux-host only) Create a HCI in scatternet @var{n} (default 0) attached 3172to the host bluetooth stack instead of to the emulated target. This 3173allows the host and target machines to participate in a common scatternet 3174and communicate. Requires the Linux @code{vhci} driver installed. Can 3175be used as following: 3176 3177@example 3178@value{qemu_system} [...OPTIONS...] -bt hci,vlan=5 -bt vhci,vlan=5 3179@end example 3180 3181@item -bt device:@var{dev}[,vlan=@var{n}] 3182Emulate a bluetooth device @var{dev} and place it in network @var{n} 3183(default @code{0}). QEMU can only emulate one type of bluetooth devices 3184currently: 3185 3186@table @option 3187@item keyboard 3188Virtual wireless keyboard implementing the HIDP bluetooth profile. 3189@end table 3190ETEXI 3191 3192STEXI 3193@end table 3194ETEXI 3195DEFHEADING() 3196 3197#ifdef CONFIG_TPM 3198DEFHEADING(TPM device options:) 3199 3200DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \ 3201 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n" 3202 " use path to provide path to a character device; default is /dev/tpm0\n" 3203 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n" 3204 " not provided it will be searched for in /sys/class/misc/tpm?/device\n" 3205 "-tpmdev emulator,id=id,chardev=dev\n" 3206 " configure the TPM device using chardev backend\n", 3207 QEMU_ARCH_ALL) 3208STEXI 3209 3210The general form of a TPM device option is: 3211@table @option 3212 3213@item -tpmdev @var{backend},id=@var{id}[,@var{options}] 3214@findex -tpmdev 3215 3216The specific backend type will determine the applicable options. 3217The @code{-tpmdev} option creates the TPM backend and requires a 3218@code{-device} option that specifies the TPM frontend interface model. 3219 3220Use @code{-tpmdev help} to print all available TPM backend types. 3221 3222@end table 3223 3224The available backends are: 3225 3226@table @option 3227 3228@item -tpmdev passthrough,id=@var{id},path=@var{path},cancel-path=@var{cancel-path} 3229 3230(Linux-host only) Enable access to the host's TPM using the passthrough 3231driver. 3232 3233@option{path} specifies the path to the host's TPM device, i.e., on 3234a Linux host this would be @code{/dev/tpm0}. 3235@option{path} is optional and by default @code{/dev/tpm0} is used. 3236 3237@option{cancel-path} specifies the path to the host TPM device's sysfs 3238entry allowing for cancellation of an ongoing TPM command. 3239@option{cancel-path} is optional and by default QEMU will search for the 3240sysfs entry to use. 3241 3242Some notes about using the host's TPM with the passthrough driver: 3243 3244The TPM device accessed by the passthrough driver must not be 3245used by any other application on the host. 3246 3247Since the host's firmware (BIOS/UEFI) has already initialized the TPM, 3248the VM's firmware (BIOS/UEFI) will not be able to initialize the 3249TPM again and may therefore not show a TPM-specific menu that would 3250otherwise allow the user to configure the TPM, e.g., allow the user to 3251enable/disable or activate/deactivate the TPM. 3252Further, if TPM ownership is released from within a VM then the host's TPM 3253will get disabled and deactivated. To enable and activate the 3254TPM again afterwards, the host has to be rebooted and the user is 3255required to enter the firmware's menu to enable and activate the TPM. 3256If the TPM is left disabled and/or deactivated most TPM commands will fail. 3257 3258To create a passthrough TPM use the following two options: 3259@example 3260-tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0 3261@end example 3262Note that the @code{-tpmdev} id is @code{tpm0} and is referenced by 3263@code{tpmdev=tpm0} in the device option. 3264 3265@item -tpmdev emulator,id=@var{id},chardev=@var{dev} 3266 3267(Linux-host only) Enable access to a TPM emulator using Unix domain socket based 3268chardev backend. 3269 3270@option{chardev} specifies the unique ID of a character device backend that provides connection to the software TPM server. 3271 3272To create a TPM emulator backend device with chardev socket backend: 3273@example 3274 3275-chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0 3276 3277@end example 3278 3279ETEXI 3280 3281STEXI 3282@end table 3283ETEXI 3284DEFHEADING() 3285 3286#endif 3287 3288DEFHEADING(Linux/Multiboot boot specific:) 3289STEXI 3290 3291When using these options, you can use a given Linux or Multiboot 3292kernel without installing it in the disk image. It can be useful 3293for easier testing of various kernels. 3294 3295@table @option 3296ETEXI 3297 3298DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \ 3299 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL) 3300STEXI 3301@item -kernel @var{bzImage} 3302@findex -kernel 3303Use @var{bzImage} as kernel image. The kernel can be either a Linux kernel 3304or in multiboot format. 3305ETEXI 3306 3307DEF("append", HAS_ARG, QEMU_OPTION_append, \ 3308 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL) 3309STEXI 3310@item -append @var{cmdline} 3311@findex -append 3312Use @var{cmdline} as kernel command line 3313ETEXI 3314 3315DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \ 3316 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL) 3317STEXI 3318@item -initrd @var{file} 3319@findex -initrd 3320Use @var{file} as initial ram disk. 3321 3322@item -initrd "@var{file1} arg=foo,@var{file2}" 3323 3324This syntax is only available with multiboot. 3325 3326Use @var{file1} and @var{file2} as modules and pass arg=foo as parameter to the 3327first module. 3328ETEXI 3329 3330DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \ 3331 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL) 3332STEXI 3333@item -dtb @var{file} 3334@findex -dtb 3335Use @var{file} as a device tree binary (dtb) image and pass it to the kernel 3336on boot. 3337ETEXI 3338 3339STEXI 3340@end table 3341ETEXI 3342DEFHEADING() 3343 3344DEFHEADING(Debug/Expert options:) 3345STEXI 3346@table @option 3347ETEXI 3348 3349DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg, 3350 "-fw_cfg [name=]<name>,file=<file>\n" 3351 " add named fw_cfg entry with contents from file\n" 3352 "-fw_cfg [name=]<name>,string=<str>\n" 3353 " add named fw_cfg entry with contents from string\n", 3354 QEMU_ARCH_ALL) 3355STEXI 3356 3357@item -fw_cfg [name=]@var{name},file=@var{file} 3358@findex -fw_cfg 3359Add named fw_cfg entry with contents from file @var{file}. 3360 3361@item -fw_cfg [name=]@var{name},string=@var{str} 3362Add named fw_cfg entry with contents from string @var{str}. 3363 3364The terminating NUL character of the contents of @var{str} will not be 3365included as part of the fw_cfg item data. To insert contents with 3366embedded NUL characters, you have to use the @var{file} parameter. 3367 3368The fw_cfg entries are passed by QEMU through to the guest. 3369 3370Example: 3371@example 3372 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin 3373@end example 3374creates an fw_cfg entry named opt/com.mycompany/blob with contents 3375from ./my_blob.bin. 3376 3377ETEXI 3378 3379DEF("serial", HAS_ARG, QEMU_OPTION_serial, \ 3380 "-serial dev redirect the serial port to char device 'dev'\n", 3381 QEMU_ARCH_ALL) 3382STEXI 3383@item -serial @var{dev} 3384@findex -serial 3385Redirect the virtual serial port to host character device 3386@var{dev}. The default device is @code{vc} in graphical mode and 3387@code{stdio} in non graphical mode. 3388 3389This option can be used several times to simulate up to 4 serial 3390ports. 3391 3392Use @code{-serial none} to disable all serial ports. 3393 3394Available character devices are: 3395@table @option 3396@item vc[:@var{W}x@var{H}] 3397Virtual console. Optionally, a width and height can be given in pixel with 3398@example 3399vc:800x600 3400@end example 3401It is also possible to specify width or height in characters: 3402@example 3403vc:80Cx24C 3404@end example 3405@item pty 3406[Linux only] Pseudo TTY (a new PTY is automatically allocated) 3407@item none 3408No device is allocated. 3409@item null 3410void device 3411@item chardev:@var{id} 3412Use a named character device defined with the @code{-chardev} option. 3413@item /dev/XXX 3414[Linux only] Use host tty, e.g. @file{/dev/ttyS0}. The host serial port 3415parameters are set according to the emulated ones. 3416@item /dev/parport@var{N} 3417[Linux only, parallel port only] Use host parallel port 3418@var{N}. Currently SPP and EPP parallel port features can be used. 3419@item file:@var{filename} 3420Write output to @var{filename}. No character can be read. 3421@item stdio 3422[Unix only] standard input/output 3423@item pipe:@var{filename} 3424name pipe @var{filename} 3425@item COM@var{n} 3426[Windows only] Use host serial port @var{n} 3427@item udp:[@var{remote_host}]:@var{remote_port}[@@[@var{src_ip}]:@var{src_port}] 3428This implements UDP Net Console. 3429When @var{remote_host} or @var{src_ip} are not specified 3430they default to @code{0.0.0.0}. 3431When not using a specified @var{src_port} a random port is automatically chosen. 3432 3433If you just want a simple readonly console you can use @code{netcat} or 3434@code{nc}, by starting QEMU with: @code{-serial udp::4555} and nc as: 3435@code{nc -u -l -p 4555}. Any time QEMU writes something to that port it 3436will appear in the netconsole session. 3437 3438If you plan to send characters back via netconsole or you want to stop 3439and start QEMU a lot of times, you should have QEMU use the same 3440source port each time by using something like @code{-serial 3441udp::4555@@:4556} to QEMU. Another approach is to use a patched 3442version of netcat which can listen to a TCP port and send and receive 3443characters via udp. If you have a patched version of netcat which 3444activates telnet remote echo and single char transfer, then you can 3445use the following options to set up a netcat redirector to allow 3446telnet on port 5555 to access the QEMU port. 3447@table @code 3448@item QEMU Options: 3449-serial udp::4555@@:4556 3450@item netcat options: 3451-u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T 3452@item telnet options: 3453localhost 5555 3454@end table 3455 3456@item tcp:[@var{host}]:@var{port}[,@var{server}][,nowait][,nodelay][,reconnect=@var{seconds}] 3457The TCP Net Console has two modes of operation. It can send the serial 3458I/O to a location or wait for a connection from a location. By default 3459the TCP Net Console is sent to @var{host} at the @var{port}. If you use 3460the @var{server} option QEMU will wait for a client socket application 3461to connect to the port before continuing, unless the @code{nowait} 3462option was specified. The @code{nodelay} option disables the Nagle buffering 3463algorithm. The @code{reconnect} option only applies if @var{noserver} is 3464set, if the connection goes down it will attempt to reconnect at the 3465given interval. If @var{host} is omitted, 0.0.0.0 is assumed. Only 3466one TCP connection at a time is accepted. You can use @code{telnet} to 3467connect to the corresponding character device. 3468@table @code 3469@item Example to send tcp console to 192.168.0.2 port 4444 3470-serial tcp:192.168.0.2:4444 3471@item Example to listen and wait on port 4444 for connection 3472-serial tcp::4444,server 3473@item Example to not wait and listen on ip 192.168.0.100 port 4444 3474-serial tcp:192.168.0.100:4444,server,nowait 3475@end table 3476 3477@item telnet:@var{host}:@var{port}[,server][,nowait][,nodelay] 3478The telnet protocol is used instead of raw tcp sockets. The options 3479work the same as if you had specified @code{-serial tcp}. The 3480difference is that the port acts like a telnet server or client using 3481telnet option negotiation. This will also allow you to send the 3482MAGIC_SYSRQ sequence if you use a telnet that supports sending the break 3483sequence. Typically in unix telnet you do it with Control-] and then 3484type "send break" followed by pressing the enter key. 3485 3486@item websocket:@var{host}:@var{port},server[,nowait][,nodelay] 3487The WebSocket protocol is used instead of raw tcp socket. The port acts as 3488a WebSocket server. Client mode is not supported. 3489 3490@item unix:@var{path}[,server][,nowait][,reconnect=@var{seconds}] 3491A unix domain socket is used instead of a tcp socket. The option works the 3492same as if you had specified @code{-serial tcp} except the unix domain socket 3493@var{path} is used for connections. 3494 3495@item mon:@var{dev_string} 3496This is a special option to allow the monitor to be multiplexed onto 3497another serial port. The monitor is accessed with key sequence of 3498@key{Control-a} and then pressing @key{c}. 3499@var{dev_string} should be any one of the serial devices specified 3500above. An example to multiplex the monitor onto a telnet server 3501listening on port 4444 would be: 3502@table @code 3503@item -serial mon:telnet::4444,server,nowait 3504@end table 3505When the monitor is multiplexed to stdio in this way, Ctrl+C will not terminate 3506QEMU any more but will be passed to the guest instead. 3507 3508@item braille 3509Braille device. This will use BrlAPI to display the braille output on a real 3510or fake device. 3511 3512@item msmouse 3513Three button serial mouse. Configure the guest to use Microsoft protocol. 3514@end table 3515ETEXI 3516 3517DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \ 3518 "-parallel dev redirect the parallel port to char device 'dev'\n", 3519 QEMU_ARCH_ALL) 3520STEXI 3521@item -parallel @var{dev} 3522@findex -parallel 3523Redirect the virtual parallel port to host device @var{dev} (same 3524devices as the serial port). On Linux hosts, @file{/dev/parportN} can 3525be used to use hardware devices connected on the corresponding host 3526parallel port. 3527 3528This option can be used several times to simulate up to 3 parallel 3529ports. 3530 3531Use @code{-parallel none} to disable all parallel ports. 3532ETEXI 3533 3534DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \ 3535 "-monitor dev redirect the monitor to char device 'dev'\n", 3536 QEMU_ARCH_ALL) 3537STEXI 3538@item -monitor @var{dev} 3539@findex -monitor 3540Redirect the monitor to host device @var{dev} (same devices as the 3541serial port). 3542The default device is @code{vc} in graphical mode and @code{stdio} in 3543non graphical mode. 3544Use @code{-monitor none} to disable the default monitor. 3545ETEXI 3546DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \ 3547 "-qmp dev like -monitor but opens in 'control' mode\n", 3548 QEMU_ARCH_ALL) 3549STEXI 3550@item -qmp @var{dev} 3551@findex -qmp 3552Like -monitor but opens in 'control' mode. 3553ETEXI 3554DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \ 3555 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n", 3556 QEMU_ARCH_ALL) 3557STEXI 3558@item -qmp-pretty @var{dev} 3559@findex -qmp-pretty 3560Like -qmp but uses pretty JSON formatting. 3561ETEXI 3562 3563DEF("mon", HAS_ARG, QEMU_OPTION_mon, \ 3564 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL) 3565STEXI 3566@item -mon [chardev=]name[,mode=readline|control][,pretty[=on|off]] 3567@findex -mon 3568Setup monitor on chardev @var{name}. @code{pretty} turns on JSON pretty printing 3569easing human reading and debugging. 3570ETEXI 3571 3572DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \ 3573 "-debugcon dev redirect the debug console to char device 'dev'\n", 3574 QEMU_ARCH_ALL) 3575STEXI 3576@item -debugcon @var{dev} 3577@findex -debugcon 3578Redirect the debug console to host device @var{dev} (same devices as the 3579serial port). The debug console is an I/O port which is typically port 35800xe9; writing to that I/O port sends output to this device. 3581The default device is @code{vc} in graphical mode and @code{stdio} in 3582non graphical mode. 3583ETEXI 3584 3585DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \ 3586 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL) 3587STEXI 3588@item -pidfile @var{file} 3589@findex -pidfile 3590Store the QEMU process PID in @var{file}. It is useful if you launch QEMU 3591from a script. 3592ETEXI 3593 3594DEF("singlestep", 0, QEMU_OPTION_singlestep, \ 3595 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL) 3596STEXI 3597@item -singlestep 3598@findex -singlestep 3599Run the emulation in single step mode. 3600ETEXI 3601 3602DEF("preconfig", 0, QEMU_OPTION_preconfig, \ 3603 "--preconfig pause QEMU before machine is initialized (experimental)\n", 3604 QEMU_ARCH_ALL) 3605STEXI 3606@item --preconfig 3607@findex --preconfig 3608Pause QEMU for interactive configuration before the machine is created, 3609which allows querying and configuring properties that will affect 3610machine initialization. Use QMP command 'x-exit-preconfig' to exit 3611the preconfig state and move to the next state (i.e. run guest if -S 3612isn't used or pause the second time if -S is used). This option is 3613experimental. 3614ETEXI 3615 3616DEF("S", 0, QEMU_OPTION_S, \ 3617 "-S freeze CPU at startup (use 'c' to start execution)\n", 3618 QEMU_ARCH_ALL) 3619STEXI 3620@item -S 3621@findex -S 3622Do not start CPU at startup (you must type 'c' in the monitor). 3623ETEXI 3624 3625DEF("realtime", HAS_ARG, QEMU_OPTION_realtime, 3626 "-realtime [mlock=on|off]\n" 3627 " run qemu with realtime features\n" 3628 " mlock=on|off controls mlock support (default: on)\n", 3629 QEMU_ARCH_ALL) 3630STEXI 3631@item -realtime mlock=on|off 3632@findex -realtime 3633Run qemu with realtime features. 3634mlocking qemu and guest memory can be enabled via @option{mlock=on} 3635(enabled by default). 3636ETEXI 3637 3638DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit, 3639 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n" 3640 " run qemu with overcommit hints\n" 3641 " mem-lock=on|off controls memory lock support (default: off)\n" 3642 " cpu-pm=on|off controls cpu power management (default: off)\n", 3643 QEMU_ARCH_ALL) 3644STEXI 3645@item -overcommit mem-lock=on|off 3646@item -overcommit cpu-pm=on|off 3647@findex -overcommit 3648Run qemu with hints about host resource overcommit. The default is 3649to assume that host overcommits all resources. 3650 3651Locking qemu and guest memory can be enabled via @option{mem-lock=on} (disabled 3652by default). This works when host memory is not overcommitted and reduces the 3653worst-case latency for guest. This is equivalent to @option{realtime}. 3654 3655Guest ability to manage power state of host cpus (increasing latency for other 3656processes on the same host cpu, but decreasing latency for guest) can be 3657enabled via @option{cpu-pm=on} (disabled by default). This works best when 3658host CPU is not overcommitted. When used, host estimates of CPU cycle and power 3659utilization will be incorrect, not taking into account guest idle time. 3660ETEXI 3661 3662DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \ 3663 "-gdb dev wait for gdb connection on 'dev'\n", QEMU_ARCH_ALL) 3664STEXI 3665@item -gdb @var{dev} 3666@findex -gdb 3667Wait for gdb connection on device @var{dev} (@pxref{gdb_usage}). Typical 3668connections will likely be TCP-based, but also UDP, pseudo TTY, or even 3669stdio are reasonable use case. The latter is allowing to start QEMU from 3670within gdb and establish the connection via a pipe: 3671@example 3672(gdb) target remote | exec @value{qemu_system} -gdb stdio ... 3673@end example 3674ETEXI 3675 3676DEF("s", 0, QEMU_OPTION_s, \ 3677 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n", 3678 QEMU_ARCH_ALL) 3679STEXI 3680@item -s 3681@findex -s 3682Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234 3683(@pxref{gdb_usage}). 3684ETEXI 3685 3686DEF("d", HAS_ARG, QEMU_OPTION_d, \ 3687 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n", 3688 QEMU_ARCH_ALL) 3689STEXI 3690@item -d @var{item1}[,...] 3691@findex -d 3692Enable logging of specified items. Use '-d help' for a list of log items. 3693ETEXI 3694 3695DEF("D", HAS_ARG, QEMU_OPTION_D, \ 3696 "-D logfile output log to logfile (default stderr)\n", 3697 QEMU_ARCH_ALL) 3698STEXI 3699@item -D @var{logfile} 3700@findex -D 3701Output log in @var{logfile} instead of to stderr 3702ETEXI 3703 3704DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \ 3705 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n", 3706 QEMU_ARCH_ALL) 3707STEXI 3708@item -dfilter @var{range1}[,...] 3709@findex -dfilter 3710Filter debug output to that relevant to a range of target addresses. The filter 3711spec can be either @var{start}+@var{size}, @var{start}-@var{size} or 3712@var{start}..@var{end} where @var{start} @var{end} and @var{size} are the 3713addresses and sizes required. For example: 3714@example 3715 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000 3716@end example 3717Will dump output for any code in the 0x1000 sized block starting at 0x8000 and 3718the 0x200 sized block starting at 0xffffffc000080000 and another 0x1000 sized 3719block starting at 0xffffffc00005f000. 3720ETEXI 3721 3722DEF("seed", HAS_ARG, QEMU_OPTION_seed, \ 3723 "-seed number seed the pseudo-random number generator\n", 3724 QEMU_ARCH_ALL) 3725STEXI 3726@item -seed @var{number} 3727@findex -seed 3728Force the guest to use a deterministic pseudo-random number generator, seeded 3729with @var{number}. This does not affect crypto routines within the host. 3730ETEXI 3731 3732DEF("L", HAS_ARG, QEMU_OPTION_L, \ 3733 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n", 3734 QEMU_ARCH_ALL) 3735STEXI 3736@item -L @var{path} 3737@findex -L 3738Set the directory for the BIOS, VGA BIOS and keymaps. 3739 3740To list all the data directories, use @code{-L help}. 3741ETEXI 3742 3743DEF("bios", HAS_ARG, QEMU_OPTION_bios, \ 3744 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL) 3745STEXI 3746@item -bios @var{file} 3747@findex -bios 3748Set the filename for the BIOS. 3749ETEXI 3750 3751DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \ 3752 "-enable-kvm enable KVM full virtualization support\n", QEMU_ARCH_ALL) 3753STEXI 3754@item -enable-kvm 3755@findex -enable-kvm 3756Enable KVM full virtualization support. This option is only available 3757if KVM support is enabled when compiling. 3758ETEXI 3759 3760DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, 3761 "-xen-domid id specify xen guest domain id\n", QEMU_ARCH_ALL) 3762DEF("xen-attach", 0, QEMU_OPTION_xen_attach, 3763 "-xen-attach attach to existing xen domain\n" 3764 " libxl will use this when starting QEMU\n", 3765 QEMU_ARCH_ALL) 3766DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, 3767 "-xen-domid-restrict restrict set of available xen operations\n" 3768 " to specified domain id. (Does not affect\n" 3769 " xenpv machine type).\n", 3770 QEMU_ARCH_ALL) 3771STEXI 3772@item -xen-domid @var{id} 3773@findex -xen-domid 3774Specify xen guest domain @var{id} (XEN only). 3775@item -xen-attach 3776@findex -xen-attach 3777Attach to existing xen domain. 3778libxl will use this when starting QEMU (XEN only). 3779@findex -xen-domid-restrict 3780Restrict set of available xen operations to specified domain id (XEN only). 3781ETEXI 3782 3783DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \ 3784 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL) 3785STEXI 3786@item -no-reboot 3787@findex -no-reboot 3788Exit instead of rebooting. 3789ETEXI 3790 3791DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \ 3792 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL) 3793STEXI 3794@item -no-shutdown 3795@findex -no-shutdown 3796Don't exit QEMU on guest shutdown, but instead only stop the emulation. 3797This allows for instance switching to monitor to commit changes to the 3798disk image. 3799ETEXI 3800 3801DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \ 3802 "-loadvm [tag|id]\n" \ 3803 " start right away with a saved state (loadvm in monitor)\n", 3804 QEMU_ARCH_ALL) 3805STEXI 3806@item -loadvm @var{file} 3807@findex -loadvm 3808Start right away with a saved state (@code{loadvm} in monitor) 3809ETEXI 3810 3811#ifndef _WIN32 3812DEF("daemonize", 0, QEMU_OPTION_daemonize, \ 3813 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL) 3814#endif 3815STEXI 3816@item -daemonize 3817@findex -daemonize 3818Daemonize the QEMU process after initialization. QEMU will not detach from 3819standard IO until it is ready to receive connections on any of its devices. 3820This option is a useful way for external programs to launch QEMU without having 3821to cope with initialization race conditions. 3822ETEXI 3823 3824DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \ 3825 "-option-rom rom load a file, rom, into the option ROM space\n", 3826 QEMU_ARCH_ALL) 3827STEXI 3828@item -option-rom @var{file} 3829@findex -option-rom 3830Load the contents of @var{file} as an option ROM. 3831This option is useful to load things like EtherBoot. 3832ETEXI 3833 3834DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \ 3835 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \ 3836 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n", 3837 QEMU_ARCH_ALL) 3838 3839STEXI 3840 3841@item -rtc [base=utc|localtime|@var{datetime}][,clock=host|rt|vm][,driftfix=none|slew] 3842@findex -rtc 3843Specify @option{base} as @code{utc} or @code{localtime} to let the RTC start at the current 3844UTC or local time, respectively. @code{localtime} is required for correct date in 3845MS-DOS or Windows. To start at a specific point in time, provide @var{datetime} in the 3846format @code{2006-06-17T16:01:21} or @code{2006-06-17}. The default base is UTC. 3847 3848By default the RTC is driven by the host system time. This allows using of the 3849RTC as accurate reference clock inside the guest, specifically if the host 3850time is smoothly following an accurate external reference clock, e.g. via NTP. 3851If you want to isolate the guest time from the host, you can set @option{clock} 3852to @code{rt} instead, which provides a host monotonic clock if host support it. 3853To even prevent the RTC from progressing during suspension, you can set @option{clock} 3854to @code{vm} (virtual clock). @samp{clock=vm} is recommended especially in 3855icount mode in order to preserve determinism; however, note that in icount mode 3856the speed of the virtual clock is variable and can in general differ from the 3857host clock. 3858 3859Enable @option{driftfix} (i386 targets only) if you experience time drift problems, 3860specifically with Windows' ACPI HAL. This option will try to figure out how 3861many timer interrupts were not processed by the Windows guest and will 3862re-inject them. 3863ETEXI 3864 3865DEF("icount", HAS_ARG, QEMU_OPTION_icount, \ 3866 "-icount [shift=N|auto][,align=on|off][,sleep=on|off,rr=record|replay,rrfile=<filename>,rrsnapshot=<snapshot>]\n" \ 3867 " enable virtual instruction counter with 2^N clock ticks per\n" \ 3868 " instruction, enable aligning the host and virtual clocks\n" \ 3869 " or disable real time cpu sleeping\n", QEMU_ARCH_ALL) 3870STEXI 3871@item -icount [shift=@var{N}|auto][,rr=record|replay,rrfile=@var{filename},rrsnapshot=@var{snapshot}] 3872@findex -icount 3873Enable virtual instruction counter. The virtual cpu will execute one 3874instruction every 2^@var{N} ns of virtual time. If @code{auto} is specified 3875then the virtual cpu speed will be automatically adjusted to keep virtual 3876time within a few seconds of real time. 3877 3878When the virtual cpu is sleeping, the virtual time will advance at default 3879speed unless @option{sleep=on|off} is specified. 3880With @option{sleep=on|off}, the virtual time will jump to the next timer deadline 3881instantly whenever the virtual cpu goes to sleep mode and will not advance 3882if no timer is enabled. This behavior give deterministic execution times from 3883the guest point of view. 3884 3885Note that while this option can give deterministic behavior, it does not 3886provide cycle accurate emulation. Modern CPUs contain superscalar out of 3887order cores with complex cache hierarchies. The number of instructions 3888executed often has little or no correlation with actual performance. 3889 3890@option{align=on} will activate the delay algorithm which will try 3891to synchronise the host clock and the virtual clock. The goal is to 3892have a guest running at the real frequency imposed by the shift option. 3893Whenever the guest clock is behind the host clock and if 3894@option{align=on} is specified then we print a message to the user 3895to inform about the delay. 3896Currently this option does not work when @option{shift} is @code{auto}. 3897Note: The sync algorithm will work for those shift values for which 3898the guest clock runs ahead of the host clock. Typically this happens 3899when the shift value is high (how high depends on the host machine). 3900 3901When @option{rr} option is specified deterministic record/replay is enabled. 3902Replay log is written into @var{filename} file in record mode and 3903read from this file in replay mode. 3904 3905Option rrsnapshot is used to create new vm snapshot named @var{snapshot} 3906at the start of execution recording. In replay mode this option is used 3907to load the initial VM state. 3908ETEXI 3909 3910DEF("watchdog", HAS_ARG, QEMU_OPTION_watchdog, \ 3911 "-watchdog model\n" \ 3912 " enable virtual hardware watchdog [default=none]\n", 3913 QEMU_ARCH_ALL) 3914STEXI 3915@item -watchdog @var{model} 3916@findex -watchdog 3917Create a virtual hardware watchdog device. Once enabled (by a guest 3918action), the watchdog must be periodically polled by an agent inside 3919the guest or else the guest will be restarted. Choose a model for 3920which your guest has drivers. 3921 3922The @var{model} is the model of hardware watchdog to emulate. Use 3923@code{-watchdog help} to list available hardware models. Only one 3924watchdog can be enabled for a guest. 3925 3926The following models may be available: 3927@table @option 3928@item ib700 3929iBASE 700 is a very simple ISA watchdog with a single timer. 3930@item i6300esb 3931Intel 6300ESB I/O controller hub is a much more featureful PCI-based 3932dual-timer watchdog. 3933@item diag288 3934A virtual watchdog for s390x backed by the diagnose 288 hypercall 3935(currently KVM only). 3936@end table 3937ETEXI 3938 3939DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \ 3940 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \ 3941 " action when watchdog fires [default=reset]\n", 3942 QEMU_ARCH_ALL) 3943STEXI 3944@item -watchdog-action @var{action} 3945@findex -watchdog-action 3946 3947The @var{action} controls what QEMU will do when the watchdog timer 3948expires. 3949The default is 3950@code{reset} (forcefully reset the guest). 3951Other possible actions are: 3952@code{shutdown} (attempt to gracefully shutdown the guest), 3953@code{poweroff} (forcefully poweroff the guest), 3954@code{inject-nmi} (inject a NMI into the guest), 3955@code{pause} (pause the guest), 3956@code{debug} (print a debug message and continue), or 3957@code{none} (do nothing). 3958 3959Note that the @code{shutdown} action requires that the guest responds 3960to ACPI signals, which it may not be able to do in the sort of 3961situations where the watchdog would have expired, and thus 3962@code{-watchdog-action shutdown} is not recommended for production use. 3963 3964Examples: 3965 3966@table @code 3967@item -watchdog i6300esb -watchdog-action pause 3968@itemx -watchdog ib700 3969@end table 3970ETEXI 3971 3972DEF("echr", HAS_ARG, QEMU_OPTION_echr, \ 3973 "-echr chr set terminal escape character instead of ctrl-a\n", 3974 QEMU_ARCH_ALL) 3975STEXI 3976 3977@item -echr @var{numeric_ascii_value} 3978@findex -echr 3979Change the escape character used for switching to the monitor when using 3980monitor and serial sharing. The default is @code{0x01} when using the 3981@code{-nographic} option. @code{0x01} is equal to pressing 3982@code{Control-a}. You can select a different character from the ascii 3983control keys where 1 through 26 map to Control-a through Control-z. For 3984instance you could use the either of the following to change the escape 3985character to Control-t. 3986@table @code 3987@item -echr 0x14 3988@itemx -echr 20 3989@end table 3990ETEXI 3991 3992DEF("show-cursor", 0, QEMU_OPTION_show_cursor, \ 3993 "-show-cursor show cursor\n", QEMU_ARCH_ALL) 3994STEXI 3995@item -show-cursor 3996@findex -show-cursor 3997Show cursor. 3998ETEXI 3999 4000DEF("tb-size", HAS_ARG, QEMU_OPTION_tb_size, \ 4001 "-tb-size n set TB size\n", QEMU_ARCH_ALL) 4002STEXI 4003@item -tb-size @var{n} 4004@findex -tb-size 4005Set TB size. 4006ETEXI 4007 4008DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \ 4009 "-incoming tcp:[host]:port[,to=maxport][,ipv4][,ipv6]\n" \ 4010 "-incoming rdma:host:port[,ipv4][,ipv6]\n" \ 4011 "-incoming unix:socketpath\n" \ 4012 " prepare for incoming migration, listen on\n" \ 4013 " specified protocol and socket address\n" \ 4014 "-incoming fd:fd\n" \ 4015 "-incoming exec:cmdline\n" \ 4016 " accept incoming migration on given file descriptor\n" \ 4017 " or from given external command\n" \ 4018 "-incoming defer\n" \ 4019 " wait for the URI to be specified via migrate_incoming\n", 4020 QEMU_ARCH_ALL) 4021STEXI 4022@item -incoming tcp:[@var{host}]:@var{port}[,to=@var{maxport}][,ipv4][,ipv6] 4023@itemx -incoming rdma:@var{host}:@var{port}[,ipv4][,ipv6] 4024@findex -incoming 4025Prepare for incoming migration, listen on a given tcp port. 4026 4027@item -incoming unix:@var{socketpath} 4028Prepare for incoming migration, listen on a given unix socket. 4029 4030@item -incoming fd:@var{fd} 4031Accept incoming migration from a given filedescriptor. 4032 4033@item -incoming exec:@var{cmdline} 4034Accept incoming migration as an output from specified external command. 4035 4036@item -incoming defer 4037Wait for the URI to be specified via migrate_incoming. The monitor can 4038be used to change settings (such as migration parameters) prior to issuing 4039the migrate_incoming to allow the migration to begin. 4040ETEXI 4041 4042DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \ 4043 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL) 4044STEXI 4045@item -only-migratable 4046@findex -only-migratable 4047Only allow migratable devices. Devices will not be allowed to enter an 4048unmigratable state. 4049ETEXI 4050 4051DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \ 4052 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL) 4053STEXI 4054@item -nodefaults 4055@findex -nodefaults 4056Don't create default devices. Normally, QEMU sets the default devices like serial 4057port, parallel port, virtual console, monitor device, VGA adapter, floppy and 4058CD-ROM drive and others. The @code{-nodefaults} option will disable all those 4059default devices. 4060ETEXI 4061 4062#ifndef _WIN32 4063DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \ 4064 "-chroot dir chroot to dir just before starting the VM\n", 4065 QEMU_ARCH_ALL) 4066#endif 4067STEXI 4068@item -chroot @var{dir} 4069@findex -chroot 4070Immediately before starting guest execution, chroot to the specified 4071directory. Especially useful in combination with -runas. 4072ETEXI 4073 4074#ifndef _WIN32 4075DEF("runas", HAS_ARG, QEMU_OPTION_runas, \ 4076 "-runas user change to user id user just before starting the VM\n" \ 4077 " user can be numeric uid:gid instead\n", 4078 QEMU_ARCH_ALL) 4079#endif 4080STEXI 4081@item -runas @var{user} 4082@findex -runas 4083Immediately before starting guest execution, drop root privileges, switching 4084to the specified user. 4085ETEXI 4086 4087DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env, 4088 "-prom-env variable=value\n" 4089 " set OpenBIOS nvram variables\n", 4090 QEMU_ARCH_PPC | QEMU_ARCH_SPARC) 4091STEXI 4092@item -prom-env @var{variable}=@var{value} 4093@findex -prom-env 4094Set OpenBIOS nvram @var{variable} to given @var{value} (PPC, SPARC only). 4095ETEXI 4096DEF("semihosting", 0, QEMU_OPTION_semihosting, 4097 "-semihosting semihosting mode\n", 4098 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | QEMU_ARCH_LM32 | 4099 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2) 4100STEXI 4101@item -semihosting 4102@findex -semihosting 4103Enable semihosting mode (ARM, M68K, Xtensa, MIPS, Nios II only). 4104ETEXI 4105DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config, 4106 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]\n" \ 4107 " semihosting configuration\n", 4108QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | QEMU_ARCH_LM32 | 4109QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2) 4110STEXI 4111@item -semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]] 4112@findex -semihosting-config 4113Enable and configure semihosting (ARM, M68K, Xtensa, MIPS, Nios II only). 4114@table @option 4115@item target=@code{native|gdb|auto} 4116Defines where the semihosting calls will be addressed, to QEMU (@code{native}) 4117or to GDB (@code{gdb}). The default is @code{auto}, which means @code{gdb} 4118during debug sessions and @code{native} otherwise. 4119@item chardev=@var{str1} 4120Send the output to a chardev backend output for native or auto output when not in gdb 4121@item arg=@var{str1},arg=@var{str2},... 4122Allows the user to pass input arguments, and can be used multiple times to build 4123up a list. The old-style @code{-kernel}/@code{-append} method of passing a 4124command line is still supported for backward compatibility. If both the 4125@code{--semihosting-config arg} and the @code{-kernel}/@code{-append} are 4126specified, the former is passed to semihosting as it always takes precedence. 4127@end table 4128ETEXI 4129DEF("old-param", 0, QEMU_OPTION_old_param, 4130 "-old-param old param mode\n", QEMU_ARCH_ARM) 4131STEXI 4132@item -old-param 4133@findex -old-param (ARM) 4134Old param mode (ARM only). 4135ETEXI 4136 4137DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \ 4138 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \ 4139 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \ 4140 " Enable seccomp mode 2 system call filter (default 'off').\n" \ 4141 " use 'obsolete' to allow obsolete system calls that are provided\n" \ 4142 " by the kernel, but typically no longer used by modern\n" \ 4143 " C library implementations.\n" \ 4144 " use 'elevateprivileges' to allow or deny QEMU process to elevate\n" \ 4145 " its privileges by blacklisting all set*uid|gid system calls.\n" \ 4146 " The value 'children' will deny set*uid|gid system calls for\n" \ 4147 " main QEMU process but will allow forks and execves to run unprivileged\n" \ 4148 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \ 4149 " blacklisting *fork and execve\n" \ 4150 " use 'resourcecontrol' to disable process affinity and schedular priority\n", 4151 QEMU_ARCH_ALL) 4152STEXI 4153@item -sandbox @var{arg}[,obsolete=@var{string}][,elevateprivileges=@var{string}][,spawn=@var{string}][,resourcecontrol=@var{string}] 4154@findex -sandbox 4155Enable Seccomp mode 2 system call filter. 'on' will enable syscall filtering and 'off' will 4156disable it. The default is 'off'. 4157@table @option 4158@item obsolete=@var{string} 4159Enable Obsolete system calls 4160@item elevateprivileges=@var{string} 4161Disable set*uid|gid system calls 4162@item spawn=@var{string} 4163Disable *fork and execve 4164@item resourcecontrol=@var{string} 4165Disable process affinity and schedular priority 4166@end table 4167ETEXI 4168 4169DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig, 4170 "-readconfig <file>\n", QEMU_ARCH_ALL) 4171STEXI 4172@item -readconfig @var{file} 4173@findex -readconfig 4174Read device configuration from @var{file}. This approach is useful when you want to spawn 4175QEMU process with many command line options but you don't want to exceed the command line 4176character limit. 4177ETEXI 4178DEF("writeconfig", HAS_ARG, QEMU_OPTION_writeconfig, 4179 "-writeconfig <file>\n" 4180 " read/write config file\n", QEMU_ARCH_ALL) 4181STEXI 4182@item -writeconfig @var{file} 4183@findex -writeconfig 4184Write device configuration to @var{file}. The @var{file} can be either filename to save 4185command line and device configuration into file or dash @code{-}) character to print the 4186output to stdout. This can be later used as input file for @code{-readconfig} option. 4187ETEXI 4188 4189DEF("no-user-config", 0, QEMU_OPTION_nouserconfig, 4190 "-no-user-config\n" 4191 " do not load default user-provided config files at startup\n", 4192 QEMU_ARCH_ALL) 4193STEXI 4194@item -no-user-config 4195@findex -no-user-config 4196The @code{-no-user-config} option makes QEMU not load any of the user-provided 4197config files on @var{sysconfdir}. 4198ETEXI 4199 4200DEF("trace", HAS_ARG, QEMU_OPTION_trace, 4201 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n" 4202 " specify tracing options\n", 4203 QEMU_ARCH_ALL) 4204STEXI 4205HXCOMM This line is not accurate, as some sub-options are backend-specific but 4206HXCOMM HX does not support conditional compilation of text. 4207@item -trace [[enable=]@var{pattern}][,events=@var{file}][,file=@var{file}] 4208@findex -trace 4209@include qemu-option-trace.texi 4210ETEXI 4211DEF("plugin", HAS_ARG, QEMU_OPTION_plugin, 4212 "-plugin [file=]<file>[,arg=<string>]\n" 4213 " load a plugin\n", 4214 QEMU_ARCH_ALL) 4215STEXI 4216@item -plugin file=@var{file}[,arg=@var{string}] 4217@findex -plugin 4218 4219Load a plugin. 4220 4221@table @option 4222@item file=@var{file} 4223Load the given plugin from a shared library file. 4224@item arg=@var{string} 4225Argument string passed to the plugin. (Can be given multiple times.) 4226@end table 4227ETEXI 4228 4229HXCOMM Internal use 4230DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL) 4231DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL) 4232 4233#ifdef __linux__ 4234DEF("enable-fips", 0, QEMU_OPTION_enablefips, 4235 "-enable-fips enable FIPS 140-2 compliance\n", 4236 QEMU_ARCH_ALL) 4237#endif 4238STEXI 4239@item -enable-fips 4240@findex -enable-fips 4241Enable FIPS 140-2 compliance mode. 4242ETEXI 4243 4244HXCOMM Deprecated by -accel tcg 4245DEF("no-kvm", 0, QEMU_OPTION_no_kvm, "", QEMU_ARCH_I386) 4246 4247DEF("msg", HAS_ARG, QEMU_OPTION_msg, 4248 "-msg timestamp[=on|off]\n" 4249 " change the format of messages\n" 4250 " on|off controls leading timestamps (default:on)\n", 4251 QEMU_ARCH_ALL) 4252STEXI 4253@item -msg timestamp[=on|off] 4254@findex -msg 4255prepend a timestamp to each log message.(default:on) 4256ETEXI 4257 4258DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate, 4259 "-dump-vmstate <file>\n" 4260 " Output vmstate information in JSON format to file.\n" 4261 " Use the scripts/vmstate-static-checker.py file to\n" 4262 " check for possible regressions in migration code\n" 4263 " by comparing two such vmstate dumps.\n", 4264 QEMU_ARCH_ALL) 4265STEXI 4266@item -dump-vmstate @var{file} 4267@findex -dump-vmstate 4268Dump json-encoded vmstate information for current machine type to file 4269in @var{file} 4270ETEXI 4271 4272DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile, 4273 "-enable-sync-profile\n" 4274 " enable synchronization profiling\n", 4275 QEMU_ARCH_ALL) 4276STEXI 4277@item -enable-sync-profile 4278@findex -enable-sync-profile 4279Enable synchronization profiling. 4280ETEXI 4281 4282STEXI 4283@end table 4284ETEXI 4285DEFHEADING() 4286 4287DEFHEADING(Generic object creation:) 4288STEXI 4289@table @option 4290ETEXI 4291 4292DEF("object", HAS_ARG, QEMU_OPTION_object, 4293 "-object TYPENAME[,PROP1=VALUE1,...]\n" 4294 " create a new object of type TYPENAME setting properties\n" 4295 " in the order they are specified. Note that the 'id'\n" 4296 " property must be set. These objects are placed in the\n" 4297 " '/objects' path.\n", 4298 QEMU_ARCH_ALL) 4299STEXI 4300@item -object @var{typename}[,@var{prop1}=@var{value1},...] 4301@findex -object 4302Create a new object of type @var{typename} setting properties 4303in the order they are specified. Note that the 'id' 4304property must be set. These objects are placed in the 4305'/objects' path. 4306 4307@table @option 4308 4309@item -object memory-backend-file,id=@var{id},size=@var{size},mem-path=@var{dir},share=@var{on|off},discard-data=@var{on|off},merge=@var{on|off},dump=@var{on|off},prealloc=@var{on|off},host-nodes=@var{host-nodes},policy=@var{default|preferred|bind|interleave},align=@var{align} 4310 4311Creates a memory file backend object, which can be used to back 4312the guest RAM with huge pages. 4313 4314The @option{id} parameter is a unique ID that will be used to reference this 4315memory region when configuring the @option{-numa} argument. 4316 4317The @option{size} option provides the size of the memory region, and accepts 4318common suffixes, eg @option{500M}. 4319 4320The @option{mem-path} provides the path to either a shared memory or huge page 4321filesystem mount. 4322 4323The @option{share} boolean option determines whether the memory 4324region is marked as private to QEMU, or shared. The latter allows 4325a co-operating external process to access the QEMU memory region. 4326 4327The @option{share} is also required for pvrdma devices due to 4328limitations in the RDMA API provided by Linux. 4329 4330Setting share=on might affect the ability to configure NUMA 4331bindings for the memory backend under some circumstances, see 4332Documentation/vm/numa_memory_policy.txt on the Linux kernel 4333source tree for additional details. 4334 4335Setting the @option{discard-data} boolean option to @var{on} 4336indicates that file contents can be destroyed when QEMU exits, 4337to avoid unnecessarily flushing data to the backing file. Note 4338that @option{discard-data} is only an optimization, and QEMU 4339might not discard file contents if it aborts unexpectedly or is 4340terminated using SIGKILL. 4341 4342The @option{merge} boolean option enables memory merge, also known as 4343MADV_MERGEABLE, so that Kernel Samepage Merging will consider the pages for 4344memory deduplication. 4345 4346Setting the @option{dump} boolean option to @var{off} excludes the memory from 4347core dumps. This feature is also known as MADV_DONTDUMP. 4348 4349The @option{prealloc} boolean option enables memory preallocation. 4350 4351The @option{host-nodes} option binds the memory range to a list of NUMA host 4352nodes. 4353 4354The @option{policy} option sets the NUMA policy to one of the following values: 4355 4356@table @option 4357@item @var{default} 4358default host policy 4359 4360@item @var{preferred} 4361prefer the given host node list for allocation 4362 4363@item @var{bind} 4364restrict memory allocation to the given host node list 4365 4366@item @var{interleave} 4367interleave memory allocations across the given host node list 4368@end table 4369 4370The @option{align} option specifies the base address alignment when 4371QEMU mmap(2) @option{mem-path}, and accepts common suffixes, eg 4372@option{2M}. Some backend store specified by @option{mem-path} 4373requires an alignment different than the default one used by QEMU, eg 4374the device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In 4375such cases, users can specify the required alignment via this option. 4376 4377The @option{pmem} option specifies whether the backing file specified 4378by @option{mem-path} is in host persistent memory that can be accessed 4379using the SNIA NVM programming model (e.g. Intel NVDIMM). 4380If @option{pmem} is set to 'on', QEMU will take necessary operations to 4381guarantee the persistence of its own writes to @option{mem-path} 4382(e.g. in vNVDIMM label emulation and live migration). 4383Also, we will map the backend-file with MAP_SYNC flag, which ensures the 4384file metadata is in sync for @option{mem-path} in case of host crash 4385or a power failure. MAP_SYNC requires support from both the host kernel 4386(since Linux kernel 4.15) and the filesystem of @option{mem-path} mounted 4387with DAX option. 4388 4389@item -object memory-backend-ram,id=@var{id},merge=@var{on|off},dump=@var{on|off},share=@var{on|off},prealloc=@var{on|off},size=@var{size},host-nodes=@var{host-nodes},policy=@var{default|preferred|bind|interleave} 4390 4391Creates a memory backend object, which can be used to back the guest RAM. 4392Memory backend objects offer more control than the @option{-m} option that is 4393traditionally used to define guest RAM. Please refer to 4394@option{memory-backend-file} for a description of the options. 4395 4396@item -object memory-backend-memfd,id=@var{id},merge=@var{on|off},dump=@var{on|off},share=@var{on|off},prealloc=@var{on|off},size=@var{size},host-nodes=@var{host-nodes},policy=@var{default|preferred|bind|interleave},seal=@var{on|off},hugetlb=@var{on|off},hugetlbsize=@var{size} 4397 4398Creates an anonymous memory file backend object, which allows QEMU to 4399share the memory with an external process (e.g. when using 4400vhost-user). The memory is allocated with memfd and optional 4401sealing. (Linux only) 4402 4403The @option{seal} option creates a sealed-file, that will block 4404further resizing the memory ('on' by default). 4405 4406The @option{hugetlb} option specify the file to be created resides in 4407the hugetlbfs filesystem (since Linux 4.14). Used in conjunction with 4408the @option{hugetlb} option, the @option{hugetlbsize} option specify 4409the hugetlb page size on systems that support multiple hugetlb page 4410sizes (it must be a power of 2 value supported by the system). 4411 4412In some versions of Linux, the @option{hugetlb} option is incompatible 4413with the @option{seal} option (requires at least Linux 4.16). 4414 4415Please refer to @option{memory-backend-file} for a description of the 4416other options. 4417 4418The @option{share} boolean option is @var{on} by default with memfd. 4419 4420@item -object rng-builtin,id=@var{id} 4421 4422Creates a random number generator backend which obtains entropy from 4423QEMU builtin functions. The @option{id} parameter is a unique ID that 4424will be used to reference this entropy backend from the @option{virtio-rng} 4425device. By default, the @option{virtio-rng} device uses this RNG backend. 4426 4427@item -object rng-random,id=@var{id},filename=@var{/dev/random} 4428 4429Creates a random number generator backend which obtains entropy from 4430a device on the host. The @option{id} parameter is a unique ID that 4431will be used to reference this entropy backend from the @option{virtio-rng} 4432device. The @option{filename} parameter specifies which file to obtain 4433entropy from and if omitted defaults to @option{/dev/urandom}. 4434 4435@item -object rng-egd,id=@var{id},chardev=@var{chardevid} 4436 4437Creates a random number generator backend which obtains entropy from 4438an external daemon running on the host. The @option{id} parameter is 4439a unique ID that will be used to reference this entropy backend from 4440the @option{virtio-rng} device. The @option{chardev} parameter is 4441the unique ID of a character device backend that provides the connection 4442to the RNG daemon. 4443 4444@item -object tls-creds-anon,id=@var{id},endpoint=@var{endpoint},dir=@var{/path/to/cred/dir},verify-peer=@var{on|off} 4445 4446Creates a TLS anonymous credentials object, which can be used to provide 4447TLS support on network backends. The @option{id} parameter is a unique 4448ID which network backends will use to access the credentials. The 4449@option{endpoint} is either @option{server} or @option{client} depending 4450on whether the QEMU network backend that uses the credentials will be 4451acting as a client or as a server. If @option{verify-peer} is enabled 4452(the default) then once the handshake is completed, the peer credentials 4453will be verified, though this is a no-op for anonymous credentials. 4454 4455The @var{dir} parameter tells QEMU where to find the credential 4456files. For server endpoints, this directory may contain a file 4457@var{dh-params.pem} providing diffie-hellman parameters to use 4458for the TLS server. If the file is missing, QEMU will generate 4459a set of DH parameters at startup. This is a computationally 4460expensive operation that consumes random pool entropy, so it is 4461recommended that a persistent set of parameters be generated 4462upfront and saved. 4463 4464@item -object tls-creds-psk,id=@var{id},endpoint=@var{endpoint},dir=@var{/path/to/keys/dir}[,username=@var{username}] 4465 4466Creates a TLS Pre-Shared Keys (PSK) credentials object, which can be used to provide 4467TLS support on network backends. The @option{id} parameter is a unique 4468ID which network backends will use to access the credentials. The 4469@option{endpoint} is either @option{server} or @option{client} depending 4470on whether the QEMU network backend that uses the credentials will be 4471acting as a client or as a server. For clients only, @option{username} 4472is the username which will be sent to the server. If omitted 4473it defaults to ``qemu''. 4474 4475The @var{dir} parameter tells QEMU where to find the keys file. 4476It is called ``@var{dir}/keys.psk'' and contains ``username:key'' 4477pairs. This file can most easily be created using the GnuTLS 4478@code{psktool} program. 4479 4480For server endpoints, @var{dir} may also contain a file 4481@var{dh-params.pem} providing diffie-hellman parameters to use 4482for the TLS server. If the file is missing, QEMU will generate 4483a set of DH parameters at startup. This is a computationally 4484expensive operation that consumes random pool entropy, so it is 4485recommended that a persistent set of parameters be generated 4486up front and saved. 4487 4488@item -object tls-creds-x509,id=@var{id},endpoint=@var{endpoint},dir=@var{/path/to/cred/dir},priority=@var{priority},verify-peer=@var{on|off},passwordid=@var{id} 4489 4490Creates a TLS anonymous credentials object, which can be used to provide 4491TLS support on network backends. The @option{id} parameter is a unique 4492ID which network backends will use to access the credentials. The 4493@option{endpoint} is either @option{server} or @option{client} depending 4494on whether the QEMU network backend that uses the credentials will be 4495acting as a client or as a server. If @option{verify-peer} is enabled 4496(the default) then once the handshake is completed, the peer credentials 4497will be verified. With x509 certificates, this implies that the clients 4498must be provided with valid client certificates too. 4499 4500The @var{dir} parameter tells QEMU where to find the credential 4501files. For server endpoints, this directory may contain a file 4502@var{dh-params.pem} providing diffie-hellman parameters to use 4503for the TLS server. If the file is missing, QEMU will generate 4504a set of DH parameters at startup. This is a computationally 4505expensive operation that consumes random pool entropy, so it is 4506recommended that a persistent set of parameters be generated 4507upfront and saved. 4508 4509For x509 certificate credentials the directory will contain further files 4510providing the x509 certificates. The certificates must be stored 4511in PEM format, in filenames @var{ca-cert.pem}, @var{ca-crl.pem} (optional), 4512@var{server-cert.pem} (only servers), @var{server-key.pem} (only servers), 4513@var{client-cert.pem} (only clients), and @var{client-key.pem} (only clients). 4514 4515For the @var{server-key.pem} and @var{client-key.pem} files which 4516contain sensitive private keys, it is possible to use an encrypted 4517version by providing the @var{passwordid} parameter. This provides 4518the ID of a previously created @code{secret} object containing the 4519password for decryption. 4520 4521The @var{priority} parameter allows to override the global default 4522priority used by gnutls. This can be useful if the system administrator 4523needs to use a weaker set of crypto priorities for QEMU without 4524potentially forcing the weakness onto all applications. Or conversely 4525if one wants wants a stronger default for QEMU than for all other 4526applications, they can do this through this parameter. Its format is 4527a gnutls priority string as described at 4528@url{https://gnutls.org/manual/html_node/Priority-Strings.html}. 4529 4530@item -object filter-buffer,id=@var{id},netdev=@var{netdevid},interval=@var{t}[,queue=@var{all|rx|tx}][,status=@var{on|off}] 4531 4532Interval @var{t} can't be 0, this filter batches the packet delivery: all 4533packets arriving in a given interval on netdev @var{netdevid} are delayed 4534until the end of the interval. Interval is in microseconds. 4535@option{status} is optional that indicate whether the netfilter is 4536on (enabled) or off (disabled), the default status for netfilter will be 'on'. 4537 4538queue @var{all|rx|tx} is an option that can be applied to any netfilter. 4539 4540@option{all}: the filter is attached both to the receive and the transmit 4541 queue of the netdev (default). 4542 4543@option{rx}: the filter is attached to the receive queue of the netdev, 4544 where it will receive packets sent to the netdev. 4545 4546@option{tx}: the filter is attached to the transmit queue of the netdev, 4547 where it will receive packets sent by the netdev. 4548 4549@item -object filter-mirror,id=@var{id},netdev=@var{netdevid},outdev=@var{chardevid},queue=@var{all|rx|tx}[,vnet_hdr_support] 4550 4551filter-mirror on netdev @var{netdevid},mirror net packet to chardev@var{chardevid}, if it has the vnet_hdr_support flag, filter-mirror will mirror packet with vnet_hdr_len. 4552 4553@item -object filter-redirector,id=@var{id},netdev=@var{netdevid},indev=@var{chardevid},outdev=@var{chardevid},queue=@var{all|rx|tx}[,vnet_hdr_support] 4554 4555filter-redirector on netdev @var{netdevid},redirect filter's net packet to chardev 4556@var{chardevid},and redirect indev's packet to filter.if it has the vnet_hdr_support flag, 4557filter-redirector will redirect packet with vnet_hdr_len. 4558Create a filter-redirector we need to differ outdev id from indev id, id can not 4559be the same. we can just use indev or outdev, but at least one of indev or outdev 4560need to be specified. 4561 4562@item -object filter-rewriter,id=@var{id},netdev=@var{netdevid},queue=@var{all|rx|tx},[vnet_hdr_support] 4563 4564Filter-rewriter is a part of COLO project.It will rewrite tcp packet to 4565secondary from primary to keep secondary tcp connection,and rewrite 4566tcp packet to primary from secondary make tcp packet can be handled by 4567client.if it has the vnet_hdr_support flag, we can parse packet with vnet header. 4568 4569usage: 4570colo secondary: 4571-object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 4572-object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 4573-object filter-rewriter,id=rew0,netdev=hn0,queue=all 4574 4575@item -object filter-dump,id=@var{id},netdev=@var{dev}[,file=@var{filename}][,maxlen=@var{len}] 4576 4577Dump the network traffic on netdev @var{dev} to the file specified by 4578@var{filename}. At most @var{len} bytes (64k by default) per packet are stored. 4579The file format is libpcap, so it can be analyzed with tools such as tcpdump 4580or Wireshark. 4581 4582@item -object colo-compare,id=@var{id},primary_in=@var{chardevid},secondary_in=@var{chardevid},outdev=@var{chardevid},iothread=@var{id}[,vnet_hdr_support][,notify_dev=@var{id}] 4583 4584Colo-compare gets packet from primary_in@var{chardevid} and secondary_in@var{chardevid}, than compare primary packet with 4585secondary packet. If the packets are same, we will output primary 4586packet to outdev@var{chardevid}, else we will notify colo-frame 4587do checkpoint and send primary packet to outdev@var{chardevid}. 4588In order to improve efficiency, we need to put the task of comparison 4589in another thread. If it has the vnet_hdr_support flag, colo compare 4590will send/recv packet with vnet_hdr_len. 4591If you want to use Xen COLO, will need the notify_dev to notify Xen 4592colo-frame to do checkpoint. 4593 4594we must use it with the help of filter-mirror and filter-redirector. 4595 4596@example 4597 4598KVM COLO 4599 4600primary: 4601-netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 4602-device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 4603-chardev socket,id=mirror0,host=3.3.3.3,port=9003,server,nowait 4604-chardev socket,id=compare1,host=3.3.3.3,port=9004,server,nowait 4605-chardev socket,id=compare0,host=3.3.3.3,port=9001,server,nowait 4606-chardev socket,id=compare0-0,host=3.3.3.3,port=9001 4607-chardev socket,id=compare_out,host=3.3.3.3,port=9005,server,nowait 4608-chardev socket,id=compare_out0,host=3.3.3.3,port=9005 4609-object iothread,id=iothread1 4610-object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 4611-object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 4612-object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 4613-object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1 4614 4615secondary: 4616-netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 4617-device e1000,netdev=hn0,mac=52:a4:00:12:78:66 4618-chardev socket,id=red0,host=3.3.3.3,port=9003 4619-chardev socket,id=red1,host=3.3.3.3,port=9004 4620-object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 4621-object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 4622 4623 4624Xen COLO 4625 4626primary: 4627-netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown 4628-device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 4629-chardev socket,id=mirror0,host=3.3.3.3,port=9003,server,nowait 4630-chardev socket,id=compare1,host=3.3.3.3,port=9004,server,nowait 4631-chardev socket,id=compare0,host=3.3.3.3,port=9001,server,nowait 4632-chardev socket,id=compare0-0,host=3.3.3.3,port=9001 4633-chardev socket,id=compare_out,host=3.3.3.3,port=9005,server,nowait 4634-chardev socket,id=compare_out0,host=3.3.3.3,port=9005 4635-chardev socket,id=notify_way,host=3.3.3.3,port=9009,server,nowait 4636-object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 4637-object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out 4638-object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 4639-object iothread,id=iothread1 4640-object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1 4641 4642secondary: 4643-netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown 4644-device e1000,netdev=hn0,mac=52:a4:00:12:78:66 4645-chardev socket,id=red0,host=3.3.3.3,port=9003 4646-chardev socket,id=red1,host=3.3.3.3,port=9004 4647-object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 4648-object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 4649 4650@end example 4651 4652If you want to know the detail of above command line, you can read 4653the colo-compare git log. 4654 4655@item -object cryptodev-backend-builtin,id=@var{id}[,queues=@var{queues}] 4656 4657Creates a cryptodev backend which executes crypto opreation from 4658the QEMU cipher APIS. The @var{id} parameter is 4659a unique ID that will be used to reference this cryptodev backend from 4660the @option{virtio-crypto} device. The @var{queues} parameter is optional, 4661which specify the queue number of cryptodev backend, the default of 4662@var{queues} is 1. 4663 4664@example 4665 4666 # @value{qemu_system} \ 4667 [...] \ 4668 -object cryptodev-backend-builtin,id=cryptodev0 \ 4669 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \ 4670 [...] 4671@end example 4672 4673@item -object cryptodev-vhost-user,id=@var{id},chardev=@var{chardevid}[,queues=@var{queues}] 4674 4675Creates a vhost-user cryptodev backend, backed by a chardev @var{chardevid}. 4676The @var{id} parameter is a unique ID that will be used to reference this 4677cryptodev backend from the @option{virtio-crypto} device. 4678The chardev should be a unix domain socket backed one. The vhost-user uses 4679a specifically defined protocol to pass vhost ioctl replacement messages 4680to an application on the other end of the socket. 4681The @var{queues} parameter is optional, which specify the queue number 4682of cryptodev backend for multiqueue vhost-user, the default of @var{queues} is 1. 4683 4684@example 4685 4686 # @value{qemu_system} \ 4687 [...] \ 4688 -chardev socket,id=chardev0,path=/path/to/socket \ 4689 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \ 4690 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \ 4691 [...] 4692@end example 4693 4694@item -object secret,id=@var{id},data=@var{string},format=@var{raw|base64}[,keyid=@var{secretid},iv=@var{string}] 4695@item -object secret,id=@var{id},file=@var{filename},format=@var{raw|base64}[,keyid=@var{secretid},iv=@var{string}] 4696 4697Defines a secret to store a password, encryption key, or some other sensitive 4698data. The sensitive data can either be passed directly via the @var{data} 4699parameter, or indirectly via the @var{file} parameter. Using the @var{data} 4700parameter is insecure unless the sensitive data is encrypted. 4701 4702The sensitive data can be provided in raw format (the default), or base64. 4703When encoded as JSON, the raw format only supports valid UTF-8 characters, 4704so base64 is recommended for sending binary data. QEMU will convert from 4705which ever format is provided to the format it needs internally. eg, an 4706RBD password can be provided in raw format, even though it will be base64 4707encoded when passed onto the RBD sever. 4708 4709For added protection, it is possible to encrypt the data associated with 4710a secret using the AES-256-CBC cipher. Use of encryption is indicated 4711by providing the @var{keyid} and @var{iv} parameters. The @var{keyid} 4712parameter provides the ID of a previously defined secret that contains 4713the AES-256 decryption key. This key should be 32-bytes long and be 4714base64 encoded. The @var{iv} parameter provides the random initialization 4715vector used for encryption of this particular secret and should be a 4716base64 encrypted string of the 16-byte IV. 4717 4718The simplest (insecure) usage is to provide the secret inline 4719 4720@example 4721 4722 # @value{qemu_system} -object secret,id=sec0,data=letmein,format=raw 4723 4724@end example 4725 4726The simplest secure usage is to provide the secret via a file 4727 4728 # printf "letmein" > mypasswd.txt 4729 # @value{qemu_system} -object secret,id=sec0,file=mypasswd.txt,format=raw 4730 4731For greater security, AES-256-CBC should be used. To illustrate usage, 4732consider the openssl command line tool which can encrypt the data. Note 4733that when encrypting, the plaintext must be padded to the cipher block 4734size (32 bytes) using the standard PKCS#5/6 compatible padding algorithm. 4735 4736First a master key needs to be created in base64 encoding: 4737 4738@example 4739 # openssl rand -base64 32 > key.b64 4740 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"') 4741@end example 4742 4743Each secret to be encrypted needs to have a random initialization vector 4744generated. These do not need to be kept secret 4745 4746@example 4747 # openssl rand -base64 16 > iv.b64 4748 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"') 4749@end example 4750 4751The secret to be defined can now be encrypted, in this case we're 4752telling openssl to base64 encode the result, but it could be left 4753as raw bytes if desired. 4754 4755@example 4756 # SECRET=$(printf "letmein" | 4757 openssl enc -aes-256-cbc -a -K $KEY -iv $IV) 4758@end example 4759 4760When launching QEMU, create a master secret pointing to @code{key.b64} 4761and specify that to be used to decrypt the user password. Pass the 4762contents of @code{iv.b64} to the second secret 4763 4764@example 4765 # @value{qemu_system} \ 4766 -object secret,id=secmaster0,format=base64,file=key.b64 \ 4767 -object secret,id=sec0,keyid=secmaster0,format=base64,\ 4768 data=$SECRET,iv=$(<iv.b64) 4769@end example 4770 4771@item -object sev-guest,id=@var{id},cbitpos=@var{cbitpos},reduced-phys-bits=@var{val},[sev-device=@var{string},policy=@var{policy},handle=@var{handle},dh-cert-file=@var{file},session-file=@var{file}] 4772 4773Create a Secure Encrypted Virtualization (SEV) guest object, which can be used 4774to provide the guest memory encryption support on AMD processors. 4775 4776When memory encryption is enabled, one of the physical address bit (aka the 4777C-bit) is utilized to mark if a memory page is protected. The @option{cbitpos} 4778is used to provide the C-bit position. The C-bit position is Host family dependent 4779hence user must provide this value. On EPYC, the value should be 47. 4780 4781When memory encryption is enabled, we loose certain bits in physical address space. 4782The @option{reduced-phys-bits} is used to provide the number of bits we loose in 4783physical address space. Similar to C-bit, the value is Host family dependent. 4784On EPYC, the value should be 5. 4785 4786The @option{sev-device} provides the device file to use for communicating with 4787the SEV firmware running inside AMD Secure Processor. The default device is 4788'/dev/sev'. If hardware supports memory encryption then /dev/sev devices are 4789created by CCP driver. 4790 4791The @option{policy} provides the guest policy to be enforced by the SEV firmware 4792and restrict what configuration and operational commands can be performed on this 4793guest by the hypervisor. The policy should be provided by the guest owner and is 4794bound to the guest and cannot be changed throughout the lifetime of the guest. 4795The default is 0. 4796 4797If guest @option{policy} allows sharing the key with another SEV guest then 4798@option{handle} can be use to provide handle of the guest from which to share 4799the key. 4800 4801The @option{dh-cert-file} and @option{session-file} provides the guest owner's 4802Public Diffie-Hillman key defined in SEV spec. The PDH and session parameters 4803are used for establishing a cryptographic session with the guest owner to 4804negotiate keys used for attestation. The file must be encoded in base64. 4805 4806e.g to launch a SEV guest 4807@example 4808 # @value{qemu_system_x86} \ 4809 ...... 4810 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \ 4811 -machine ...,memory-encryption=sev0 4812 ..... 4813 4814@end example 4815 4816 4817@item -object authz-simple,id=@var{id},identity=@var{string} 4818 4819Create an authorization object that will control access to network services. 4820 4821The @option{identity} parameter is identifies the user and its format 4822depends on the network service that authorization object is associated 4823with. For authorizing based on TLS x509 certificates, the identity must 4824be the x509 distinguished name. Note that care must be taken to escape 4825any commas in the distinguished name. 4826 4827An example authorization object to validate a x509 distinguished name 4828would look like: 4829@example 4830 # @value{qemu_system} \ 4831 ... 4832 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \ 4833 ... 4834@end example 4835 4836Note the use of quotes due to the x509 distinguished name containing 4837whitespace, and escaping of ','. 4838 4839@item -object authz-listfile,id=@var{id},filename=@var{path},refresh=@var{yes|no} 4840 4841Create an authorization object that will control access to network services. 4842 4843The @option{filename} parameter is the fully qualified path to a file 4844containing the access control list rules in JSON format. 4845 4846An example set of rules that match against SASL usernames might look 4847like: 4848 4849@example 4850 @{ 4851 "rules": [ 4852 @{ "match": "fred", "policy": "allow", "format": "exact" @}, 4853 @{ "match": "bob", "policy": "allow", "format": "exact" @}, 4854 @{ "match": "danb", "policy": "deny", "format": "glob" @}, 4855 @{ "match": "dan*", "policy": "allow", "format": "exact" @}, 4856 ], 4857 "policy": "deny" 4858 @} 4859@end example 4860 4861When checking access the object will iterate over all the rules and 4862the first rule to match will have its @option{policy} value returned 4863as the result. If no rules match, then the default @option{policy} 4864value is returned. 4865 4866The rules can either be an exact string match, or they can use the 4867simple UNIX glob pattern matching to allow wildcards to be used. 4868 4869If @option{refresh} is set to true the file will be monitored 4870and automatically reloaded whenever its content changes. 4871 4872As with the @code{authz-simple} object, the format of the identity 4873strings being matched depends on the network service, but is usually 4874a TLS x509 distinguished name, or a SASL username. 4875 4876An example authorization object to validate a SASL username 4877would look like: 4878@example 4879 # @value{qemu_system} \ 4880 ... 4881 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=yes 4882 ... 4883@end example 4884 4885@item -object authz-pam,id=@var{id},service=@var{string} 4886 4887Create an authorization object that will control access to network services. 4888 4889The @option{service} parameter provides the name of a PAM service to use 4890for authorization. It requires that a file @code{/etc/pam.d/@var{service}} 4891exist to provide the configuration for the @code{account} subsystem. 4892 4893An example authorization object to validate a TLS x509 distinguished 4894name would look like: 4895 4896@example 4897 # @value{qemu_system} \ 4898 ... 4899 -object authz-pam,id=auth0,service=qemu-vnc 4900 ... 4901@end example 4902 4903There would then be a corresponding config file for PAM at 4904@code{/etc/pam.d/qemu-vnc} that contains: 4905 4906@example 4907account requisite pam_listfile.so item=user sense=allow \ 4908 file=/etc/qemu/vnc.allow 4909@end example 4910 4911Finally the @code{/etc/qemu/vnc.allow} file would contain 4912the list of x509 distingished names that are permitted 4913access 4914 4915@example 4916CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB 4917@end example 4918 4919@item -object iothread,id=@var{id},poll-max-ns=@var{poll-max-ns},poll-grow=@var{poll-grow},poll-shrink=@var{poll-shrink} 4920 4921Creates a dedicated event loop thread that devices can be assigned to. This is 4922known as an IOThread. By default device emulation happens in vCPU threads or 4923the main event loop thread. This can become a scalability bottleneck. 4924IOThreads allow device emulation and I/O to run on other host CPUs. 4925 4926The @option{id} parameter is a unique ID that will be used to reference this 4927IOThread from @option{-device ...,iothread=@var{id}}. Multiple devices can be 4928assigned to an IOThread. Note that not all devices support an 4929@option{iothread} parameter. 4930 4931The @code{query-iothreads} QMP command lists IOThreads and reports their thread 4932IDs so that the user can configure host CPU pinning/affinity. 4933 4934IOThreads use an adaptive polling algorithm to reduce event loop latency. 4935Instead of entering a blocking system call to monitor file descriptors and then 4936pay the cost of being woken up when an event occurs, the polling algorithm 4937spins waiting for events for a short time. The algorithm's default parameters 4938are suitable for many cases but can be adjusted based on knowledge of the 4939workload and/or host device latency. 4940 4941The @option{poll-max-ns} parameter is the maximum number of nanoseconds to busy 4942wait for events. Polling can be disabled by setting this value to 0. 4943 4944The @option{poll-grow} parameter is the multiplier used to increase the polling 4945time when the algorithm detects it is missing events due to not polling long 4946enough. 4947 4948The @option{poll-shrink} parameter is the divisor used to decrease the polling 4949time when the algorithm detects it is spending too long polling without 4950encountering events. 4951 4952The polling parameters can be modified at run-time using the @code{qom-set} command (where @code{iothread1} is the IOThread's @code{id}): 4953 4954@example 4955(qemu) qom-set /objects/iothread1 poll-max-ns 100000 4956@end example 4957 4958@end table 4959 4960ETEXI 4961 4962 4963HXCOMM This is the last statement. Insert new options before this line! 4964STEXI 4965@end table 4966ETEXI 4967
