1 HXCOMM Use DEFHEADING() to define headings in both help text and rST.
2 HXCOMM Text between SRST and ERST is copied to the rST version and
3 HXCOMM discarded from C version.
4 HXCOMM DEF(option, HAS_ARG/0, opt_enum, opt_help, arch_mask) is used to
5 HXCOMM construct option structures, enums and help message for specified
7 HXCOMM HXCOMM can be used for comments, discarded from both rST and C.
9 DEFHEADING(Standard options:)
11 DEF("help", 0, QEMU_OPTION_h,
12 "-h or -help display this help and exit\n", QEMU_ARCH_ALL)
18 DEF("version", 0, QEMU_OPTION_version,
19 "-version display version information and exit\n", QEMU_ARCH_ALL)
22 Display version information and exit
25 DEF("machine", HAS_ARG, QEMU_OPTION_machine, \
26 "-machine [type=]name[,prop[=value][,...]]\n"
27 " selects emulated machine ('-machine help' for list)\n"
28 " property accel=accel1[:accel2[:...]] selects accelerator\n"
29 " supported accelerators are kvm, xen, hax, hvf, whpx or tcg (default: tcg)\n"
30 " vmport=on|off|auto controls emulation of vmport (default: auto)\n"
31 " dump-guest-core=on|off include guest memory in a core dump (default=on)\n"
32 " mem-merge=on|off controls memory merge support (default: on)\n"
33 " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n"
34 " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n"
35 " suppress-vmdesc=on|off disables self-describing migration (default=off)\n"
36 " nvdimm=on|off controls NVDIMM support (default=off)\n"
37 " enforce-config-section=on|off enforce configuration section migration (default=off)\n"
38 " memory-encryption=@var{} memory encryption object to use (default=none)\n"
39 " hmat=on|off controls ACPI HMAT support (default=off)\n",
42 ``-machine [type=]name[,prop=value[,...]]``
43 Select the emulated machine by name. Use ``-machine help`` to list
46 For architectures which aim to support live migration compatibility
47 across releases, each release will introduce a new versioned machine
48 type. For example, the 2.8.0 release introduced machine types
49 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures.
51 To allow live migration of guests from QEMU version 2.8.0, to QEMU
52 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8"
53 and "pc-q35-2.8" machines too. To allow users live migrating VMs to
54 skip multiple intermediate releases when upgrading, new releases of
55 QEMU will support machine types from many previous versions.
57 Supported machine properties are:
59 ``accel=accels1[:accels2[:...]]``
60 This is used to enable an accelerator. Depending on the target
61 architecture, kvm, xen, hax, hvf, whpx or tcg can be available.
62 By default, tcg is used. If there is more than one accelerator
63 specified, the next one is used if the previous one fails to
66 ``vmport=on|off|auto``
67 Enables emulation of VMWare IO port, for vmmouse etc. auto says
68 to select the value based on accel. For accel=xen the default is
69 off otherwise the default is on.
71 ``dump-guest-core=on|off``
72 Include guest memory in a core dump. The default is on.
75 Enables or disables memory merge support. This feature, when
76 supported by the host, de-duplicates identical memory pages
77 among VMs instances (enabled by default).
79 ``aes-key-wrap=on|off``
80 Enables or disables AES key wrapping support on s390-ccw hosts.
81 This feature controls whether AES wrapping keys will be created
82 to allow execution of AES cryptographic functions. The default
85 ``dea-key-wrap=on|off``
86 Enables or disables DEA key wrapping support on s390-ccw hosts.
87 This feature controls whether DEA wrapping keys will be created
88 to allow execution of DEA cryptographic functions. The default
92 Enables or disables NVDIMM support. The default is off.
94 ``enforce-config-section=on|off``
95 If ``enforce-config-section`` is set to on, force migration code
96 to send configuration section even if the machine-type sets the
97 ``migration.send-configuration`` property to off. NOTE: this
98 parameter is deprecated. Please use ``-global``
99 ``migration.send-configuration``\ =on\|off instead.
101 ``memory-encryption=``
102 Memory encryption object to use. The default is none.
105 Enables or disables ACPI Heterogeneous Memory Attribute Table
106 (HMAT) support. The default is off.
109 HXCOMM Deprecated by -machine
110 DEF("M", HAS_ARG, QEMU_OPTION_M, "", QEMU_ARCH_ALL)
112 DEF("cpu", HAS_ARG, QEMU_OPTION_cpu,
113 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL)
116 Select CPU model (``-cpu help`` for list and additional feature
120 DEF("accel", HAS_ARG, QEMU_OPTION_accel,
121 "-accel [accel=]accelerator[,prop[=value][,...]]\n"
122 " select accelerator (kvm, xen, hax, hvf, whpx or tcg; use 'help' for a list)\n"
123 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n"
124 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n"
125 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n"
126 " tb-size=n (TCG translation block cache size)\n"
127 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL)
129 ``-accel name[,prop=value[,...]]``
130 This is used to enable an accelerator. Depending on the target
131 architecture, kvm, xen, hax, hvf, whpx or tcg can be available. By
132 default, tcg is used. If there is more than one accelerator
133 specified, the next one is used if the previous one fails to
136 ``igd-passthru=on|off``
137 When Xen is in use, this option controls whether Intel
138 integrated graphics devices can be passed through to the guest
141 ``kernel-irqchip=on|off|split``
142 Controls KVM in-kernel irqchip support. The default is full
143 acceleration of the interrupt controllers. On x86, split irqchip
144 reduces the kernel attack surface, at a performance cost for
145 non-MSI interrupts. Disabling the in-kernel irqchip completely
146 is not recommended except for debugging purposes.
148 ``kvm-shadow-mem=size``
149 Defines the size of the KVM shadow MMU.
152 Controls the size (in MiB) of the TCG translation block cache.
154 ``thread=single|multi``
155 Controls number of TCG threads. When the TCG is multi-threaded
156 there will be one thread per vCPU therefor taking advantage of
157 additional host cores. The default is to enable multi-threading
158 where both the back-end and front-ends support it and no
159 incompatible TCG features have been enabled (e.g.
163 DEF("smp", HAS_ARG, QEMU_OPTION_smp,
164 "-smp [cpus=]n[,maxcpus=cpus][,cores=cores][,threads=threads][,dies=dies][,sockets=sockets]\n"
165 " set the number of CPUs to 'n' [default=1]\n"
166 " maxcpus= maximum number of total cpus, including\n"
167 " offline CPUs for hotplug, etc\n"
168 " cores= number of CPU cores on one socket (for PC, it's on one die)\n"
169 " threads= number of threads on one CPU core\n"
170 " dies= number of CPU dies on one socket (for PC only)\n"
171 " sockets= number of discrete sockets in the system\n",
174 ``-smp [cpus=]n[,cores=cores][,threads=threads][,dies=dies][,sockets=sockets][,maxcpus=maxcpus]``
175 Simulate an SMP system with n CPUs. On the PC target, up to 255 CPUs
176 are supported. On Sparc32 target, Linux limits the number of usable
177 CPUs to 4. For the PC target, the number of cores per die, the
178 number of threads per cores, the number of dies per packages and the
179 total number of sockets can be specified. Missing values will be
180 computed. If any on the three values is given, the total number of
181 CPUs n can be omitted. maxcpus specifies the maximum number of
185 DEF("numa", HAS_ARG, QEMU_OPTION_numa,
186 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
187 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
188 "-numa dist,src=source,dst=destination,val=distance\n"
189 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n"
190 "-numa hmat-lb,initiator=node,target=node,hierarchy=memory|first-level|second-level|third-level,data-type=access-latency|read-latency|write-latency[,latency=lat][,bandwidth=bw]\n"
191 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n",
194 ``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
196 ``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
198 ``-numa dist,src=source,dst=destination,val=distance``
200 ``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]``
202 ``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=tpye[,latency=lat][,bandwidth=bw]``
204 ``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]``
205 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA
206 distance from a source node to a destination node. Set the ACPI
207 Heterogeneous Memory Attributes for the given nodes.
209 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and
210 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a
211 contiguous range of CPU indexes (or a single VCPU if lastcpu is
212 omitted). A non-contiguous set of VCPUs can be represented by
213 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is
214 omitted on all nodes, VCPUs are automatically split between them.
216 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a
221 -numa node,cpus=0-2,cpus=5
223 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option
224 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to
225 assign CPU objects to a node using topology layout properties of
226 CPU. The set of properties is machine specific, and depends on used
227 machine type/'\ ``smp``\ ' options. It could be queried with
228 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ '
229 property specifies node to which CPU object will be assigned, it's
230 required for node to be declared with '\ ``node``\ ' option before
231 it's used with '\ ``cpu``\ ' option.
238 -smp 1,sockets=2,maxcpus=2 \
239 -numa node,nodeid=0 -numa node,nodeid=1 \
240 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1
242 Legacy '\ ``mem``\ ' assigns a given RAM amount to a node (not supported
243 for 5.1 and newer machine types). '\ ``memdev``\ ' assigns RAM from
244 a given memory backend device to a node. If '\ ``mem``\ ' and
245 '\ ``memdev``\ ' are omitted in all nodes, RAM is split equally between them.
248 '\ ``mem``\ ' and '\ ``memdev``\ ' are mutually exclusive.
249 Furthermore, if one node uses '\ ``memdev``\ ', all of them have to
252 '\ ``initiator``\ ' is an additional option that points to an
253 initiator NUMA node that has best performance (the lowest latency or
254 largest bandwidth) to this NUMA node. Note that this option can be
255 set only when the machine property 'hmat' is set to 'on'.
257 Following example creates a machine with 2 NUMA nodes, node 0 has
258 CPU. node 1 has only memory, and its initiator is node 0. Note that
259 because node 0 has CPU, by default the initiator of node 0 is itself
265 -m 2G,slots=2,maxmem=4G \
266 -object memory-backend-ram,size=1G,id=m0 \
267 -object memory-backend-ram,size=1G,id=m1 \
268 -numa node,nodeid=0,memdev=m0 \
269 -numa node,nodeid=1,memdev=m1,initiator=0 \
270 -smp 2,sockets=2,maxcpus=2 \
271 -numa cpu,node-id=0,socket-id=0 \
272 -numa cpu,node-id=0,socket-id=1
274 source and destination are NUMA node IDs. distance is the NUMA
275 distance from source to destination. The distance from a node to
276 itself is always 10. If any pair of nodes is given a distance, then
277 all pairs must be given distances. Although, when distances are only
278 given in one direction for each pair of nodes, then the distances in
279 the opposite directions are assumed to be the same. If, however, an
280 asymmetrical pair of distances is given for even one node pair, then
281 all node pairs must be provided distance values for both directions,
282 even when they are symmetrical. When a node is unreachable from
283 another node, set the pair's distance to 255.
285 Note that the -``numa`` option doesn't allocate any of the specified
286 resources, it just assigns existing resources to NUMA nodes. This
287 means that one still has to use the ``-m``, ``-smp`` options to
288 allocate RAM and VCPUs respectively.
290 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth
291 Information between initiator and target NUMA nodes in ACPI
292 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can
293 create memory requests, usually it has one or more processors.
294 Target NUMA node contains addressable memory.
296 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is
297 the memory hierarchy of the target NUMA node: if hierarchy is
298 'memory', the structure represents the memory performance; if
299 hierarchy is 'first-level\|second-level\|third-level', this
300 structure represents aggregated performance of memory side caches
301 for each domain. type of 'data-type' is type of data represented by
302 this structure instance: if 'hierarchy' is 'memory', 'data-type' is
303 'access\|read\|write' latency or 'access\|read\|write' bandwidth of
304 the target memory; if 'hierarchy' is
305 'first-level\|second-level\|third-level', 'data-type' is
306 'access\|read\|write' hit latency or 'access\|read\|write' hit
307 bandwidth of the target memory side cache.
309 lat is latency value in nanoseconds. bw is bandwidth value, the
310 possible value and units are NUM[M\|G\|T], mean that the bandwidth
311 value are NUM byte per second (or MB/s, GB/s or TB/s depending on
312 used suffix). Note that if latency or bandwidth value is 0, means
313 the corresponding latency or bandwidth information is not provided.
315 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory
316 belongs. size is the size of memory side cache in bytes. level is
317 the cache level described in this structure, note that the cache
318 level 0 should not be used with '\ ``hmat-cache``\ ' option.
319 associativity is the cache associativity, the possible value is
320 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy
321 is the write policy. line is the cache Line size in bytes.
323 For example, the following options describe 2 NUMA nodes. Node 0 has
324 2 cpus and a ram, node 1 has only a ram. The processors in node 0
325 access memory in node 0 with access-latency 5 nanoseconds,
326 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access
327 memory in NUMA node 1 with access-latency 10 nanoseconds,
328 access-bandwidth is 100 MB/s. And for memory side cache information,
329 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB,
330 policy is write-back, the cache Line size is 8 bytes:
336 -object memory-backend-ram,size=1G,id=m0 \
337 -object memory-backend-ram,size=1G,id=m1 \
339 -numa node,nodeid=0,memdev=m0 \
340 -numa node,nodeid=1,memdev=m1,initiator=0 \
341 -numa cpu,node-id=0,socket-id=0 \
342 -numa cpu,node-id=0,socket-id=1 \
343 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \
344 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \
345 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \
346 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \
347 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \
348 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8
351 DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd,
352 "-add-fd fd=fd,set=set[,opaque=opaque]\n"
353 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL)
355 ``-add-fd fd=fd,set=set[,opaque=opaque]``
356 Add a file descriptor to an fd set. Valid options are:
359 This option defines the file descriptor of which a duplicate is
360 added to fd set. The file descriptor cannot be stdin, stdout, or
364 This option defines the ID of the fd set to add the file
368 This option defines a free-form string that can be used to
371 You can open an image using pre-opened file descriptors from an fd
377 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \
378 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \
379 -drive file=/dev/fdset/2,index=0,media=disk
382 DEF("set", HAS_ARG, QEMU_OPTION_set,
383 "-set group.id.arg=value\n"
384 " set <arg> parameter for item <id> of type <group>\n"
385 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL)
387 ``-set group.id.arg=value``
388 Set parameter arg for item id of type group
391 DEF("global", HAS_ARG, QEMU_OPTION_global,
392 "-global driver.property=value\n"
393 "-global driver=driver,property=property,value=value\n"
394 " set a global default for a driver property\n",
397 ``-global driver.prop=value``
399 ``-global driver=driver,property=property,value=value``
400 Set default value of driver's property prop to value, e.g.:
404 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img
406 In particular, you can use this to set driver properties for devices
407 which are created automatically by the machine model. To create a
408 device which is not created automatically and set properties on it,
411 -global driver.prop=value is shorthand for -global
412 driver=driver,property=prop,value=value. The longhand syntax works
413 even when driver contains a dot.
416 DEF("boot", HAS_ARG, QEMU_OPTION_boot,
417 "-boot [order=drives][,once=drives][,menu=on|off]\n"
418 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n"
419 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n"
420 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n"
421 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n"
422 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n",
425 ``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]``
426 Specify boot order drives as a string of drive letters. Valid drive
427 letters depend on the target architecture. The x86 PC uses: a, b
428 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p
429 (Etherboot from network adapter 1-4), hard disk boot is the default.
430 To apply a particular boot order only on the first startup, specify
431 it via ``once``. Note that the ``order`` or ``once`` parameter
432 should not be used together with the ``bootindex`` property of
433 devices, since the firmware implementations normally do not support
434 both at the same time.
436 Interactive boot menus/prompts can be enabled via ``menu=on`` as far
437 as firmware/BIOS supports them. The default is non-interactive boot.
439 A splash picture could be passed to bios, enabling user to show it
440 as logo, when option splash=sp\_name is given and menu=on, If
441 firmware/BIOS supports them. Currently Seabios for X86 system
442 support it. limitation: The splash file could be a jpeg file or a
443 BMP file in 24 BPP format(true color). The resolution should be
444 supported by the SVGA mode, so the recommended is 320x240, 640x480,
447 A timeout could be passed to bios, guest will pause for rb\_timeout
448 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will
449 not reboot, qemu passes '-1' to bios by default. Currently Seabios
450 for X86 system support it.
452 Do strict boot via ``strict=on`` as far as firmware/BIOS supports
453 it. This only effects when boot priority is changed by bootindex
454 options. The default is non-strict boot.
458 # try to boot from network first, then from hard disk
459 |qemu_system_x86| -boot order=nc
460 # boot from CD-ROM first, switch back to default order after reboot
461 |qemu_system_x86| -boot once=d
462 # boot with a splash picture for 5 seconds.
463 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000
465 Note: The legacy format '-boot drives' is still supported but its
466 use is discouraged as it may be removed from future versions.
469 DEF("m", HAS_ARG, QEMU_OPTION_m,
470 "-m [size=]megs[,slots=n,maxmem=size]\n"
471 " configure guest RAM\n"
472 " size: initial amount of guest memory\n"
473 " slots: number of hotplug slots (default: none)\n"
474 " maxmem: maximum amount of guest memory (default: none)\n"
475 "NOTE: Some architectures might enforce a specific granularity\n",
478 ``-m [size=]megs[,slots=n,maxmem=size]``
479 Sets guest startup RAM size to megs megabytes. Default is 128 MiB.
480 Optionally, a suffix of "M" or "G" can be used to signify a value in
481 megabytes or gigabytes respectively. Optional pair slots, maxmem
482 could be used to set amount of hotpluggable memory slots and maximum
483 amount of memory. Note that maxmem must be aligned to the page size.
485 For example, the following command-line sets the guest startup RAM
486 size to 1GB, creates 3 slots to hotplug additional memory and sets
487 the maximum memory the guest can reach to 4GB:
491 |qemu_system| -m 1G,slots=3,maxmem=4G
493 If slots and maxmem are not specified, memory hotplug won't be
494 enabled and the guest startup RAM will never increase.
497 DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath,
498 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL)
501 Allocate guest RAM from a temporarily created file in path.
504 DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc,
505 "-mem-prealloc preallocate guest memory (use with -mem-path)\n",
509 Preallocate memory when using -mem-path.
512 DEF("k", HAS_ARG, QEMU_OPTION_k,
513 "-k language use keyboard layout (for example 'fr' for French)\n",
517 Use keyboard layout language (for example ``fr`` for French). This
518 option is only needed where it is not easy to get raw PC keycodes
519 (e.g. on Macs, with some X11 servers or with a VNC or curses
520 display). You don't normally need to use it on PC/Linux or
523 The available layouts are:
527 ar de-ch es fo fr-ca hu ja mk no pt-br sv
528 da en-gb et fr fr-ch is lt nl pl ru th
529 de en-us fi fr-be hr it lv nl-be pt sl tr
531 The default is ``en-us``.
535 HXCOMM Deprecated by -audiodev
536 DEF("audio-help", 0, QEMU_OPTION_audio_help,
537 "-audio-help show -audiodev equivalent of the currently specified audio settings\n",
541 Will show the -audiodev equivalent of the currently specified
542 (deprecated) environment variables.
545 DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev,
546 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n"
547 " specifies the audio backend to use\n"
548 " id= identifier of the backend\n"
549 " timer-period= timer period in microseconds\n"
550 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n"
551 " in|out.fixed-settings= use fixed settings for host audio\n"
552 " in|out.frequency= frequency to use with fixed settings\n"
553 " in|out.channels= number of channels to use with fixed settings\n"
554 " in|out.format= sample format to use with fixed settings\n"
555 " valid values: s8, s16, s32, u8, u16, u32, f32\n"
556 " in|out.voices= number of voices to use\n"
557 " in|out.buffer-length= length of buffer in microseconds\n"
558 "-audiodev none,id=id,[,prop[=value][,...]]\n"
559 " dummy driver that discards all output\n"
560 #ifdef CONFIG_AUDIO_ALSA
561 "-audiodev alsa,id=id[,prop[=value][,...]]\n"
562 " in|out.dev= name of the audio device to use\n"
563 " in|out.period-length= length of period in microseconds\n"
564 " in|out.try-poll= attempt to use poll mode\n"
565 " threshold= threshold (in microseconds) when playback starts\n"
567 #ifdef CONFIG_AUDIO_COREAUDIO
568 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n"
569 " in|out.buffer-count= number of buffers\n"
571 #ifdef CONFIG_AUDIO_DSOUND
572 "-audiodev dsound,id=id[,prop[=value][,...]]\n"
573 " latency= add extra latency to playback in microseconds\n"
575 #ifdef CONFIG_AUDIO_OSS
576 "-audiodev oss,id=id[,prop[=value][,...]]\n"
577 " in|out.dev= path of the audio device to use\n"
578 " in|out.buffer-count= number of buffers\n"
579 " in|out.try-poll= attempt to use poll mode\n"
580 " try-mmap= try using memory mapped access\n"
581 " exclusive= open device in exclusive mode\n"
582 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n"
584 #ifdef CONFIG_AUDIO_PA
585 "-audiodev pa,id=id[,prop[=value][,...]]\n"
586 " server= PulseAudio server address\n"
587 " in|out.name= source/sink device name\n"
588 " in|out.latency= desired latency in microseconds\n"
590 #ifdef CONFIG_AUDIO_SDL
591 "-audiodev sdl,id=id[,prop[=value][,...]]\n"
594 "-audiodev spice,id=id[,prop[=value][,...]]\n"
596 "-audiodev wav,id=id[,prop[=value][,...]]\n"
597 " path= path of wav file to record\n",
600 ``-audiodev [driver=]driver,id=id[,prop[=value][,...]]``
601 Adds a new audio backend driver identified by id. There are global
602 and driver specific properties. Some values can be set differently
603 for input and output, they're marked with ``in|out.``. You can set
604 the input's property with ``in.prop`` and the output's property with
605 ``out.prop``. For example:
609 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000
610 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified
612 NOTE: parameter validation is known to be incomplete, in many cases
613 specifying an invalid option causes QEMU to print an error message
614 and continue emulation without sound.
616 Valid global options are:
619 Identifies the audio backend.
621 ``timer-period=period``
622 Sets the timer period used by the audio subsystem in
623 microseconds. Default is 10000 (10 ms).
625 ``in|out.mixing-engine=on|off``
626 Use QEMU's mixing engine to mix all streams inside QEMU and
627 convert audio formats when not supported by the backend. When
628 off, fixed-settings must be off too. Note that disabling this
629 option means that the selected backend must support multiple
630 streams and the audio formats used by the virtual cards,
631 otherwise you'll get no sound. It's not recommended to disable
632 this option unless you want to use 5.1 or 7.1 audio, as mixing
633 engine only supports mono and stereo audio. Default is on.
635 ``in|out.fixed-settings=on|off``
636 Use fixed settings for host audio. When off, it will change
637 based on how the guest opens the sound card. In this case you
638 must not specify frequency, channels or format. Default is on.
640 ``in|out.frequency=frequency``
641 Specify the frequency to use when using fixed-settings. Default
644 ``in|out.channels=channels``
645 Specify the number of channels to use when using fixed-settings.
646 Default is 2 (stereo).
648 ``in|out.format=format``
649 Specify the sample format to use when using fixed-settings.
650 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``,
651 ``u32``, ``f32``. Default is ``s16``.
653 ``in|out.voices=voices``
654 Specify the number of voices to use. Default is 1.
656 ``in|out.buffer-length=usecs``
657 Sets the size of the buffer in microseconds.
659 ``-audiodev none,id=id[,prop[=value][,...]]``
660 Creates a dummy backend that discards all outputs. This backend has
661 no backend specific properties.
663 ``-audiodev alsa,id=id[,prop[=value][,...]]``
664 Creates backend using the ALSA. This backend is only available on
667 ALSA specific options are:
669 ``in|out.dev=device``
670 Specify the ALSA device to use for input and/or output. Default
673 ``in|out.period-length=usecs``
674 Sets the period length in microseconds.
676 ``in|out.try-poll=on|off``
677 Attempt to use poll mode with the device. Default is on.
679 ``threshold=threshold``
680 Threshold (in microseconds) when playback starts. Default is 0.
682 ``-audiodev coreaudio,id=id[,prop[=value][,...]]``
683 Creates a backend using Apple's Core Audio. This backend is only
684 available on Mac OS and only supports playback.
686 Core Audio specific options are:
688 ``in|out.buffer-count=count``
689 Sets the count of the buffers.
691 ``-audiodev dsound,id=id[,prop[=value][,...]]``
692 Creates a backend using Microsoft's DirectSound. This backend is
693 only available on Windows and only supports playback.
695 DirectSound specific options are:
698 Add extra usecs microseconds latency to playback. Default is
701 ``-audiodev oss,id=id[,prop[=value][,...]]``
702 Creates a backend using OSS. This backend is available on most
705 OSS specific options are:
707 ``in|out.dev=device``
708 Specify the file name of the OSS device to use. Default is
711 ``in|out.buffer-count=count``
712 Sets the count of the buffers.
714 ``in|out.try-poll=on|of``
715 Attempt to use poll mode with the device. Default is on.
718 Try using memory mapped device access. Default is off.
721 Open the device in exclusive mode (vmix won't work in this
722 case). Default is off.
724 ``dsp-policy=policy``
725 Sets the timing policy (between 0 and 10, where smaller number
726 means smaller latency but higher CPU usage). Use -1 to use
727 buffer sizes specified by ``buffer`` and ``buffer-count``. This
728 option is ignored if you do not have OSS 4. Default is 5.
730 ``-audiodev pa,id=id[,prop[=value][,...]]``
731 Creates a backend using PulseAudio. This backend is available on
734 PulseAudio specific options are:
737 Sets the PulseAudio server to connect to.
740 Use the specified source/sink for recording/playback.
742 ``in|out.latency=usecs``
743 Desired latency in microseconds. The PulseAudio server will try
744 to honor this value but actual latencies may be lower or higher.
746 ``-audiodev sdl,id=id[,prop[=value][,...]]``
747 Creates a backend using SDL. This backend is available on most
748 systems, but you should use your platform's native backend if
749 possible. This backend has no backend specific properties.
751 ``-audiodev spice,id=id[,prop[=value][,...]]``
752 Creates a backend that sends audio through SPICE. This backend
753 requires ``-spice`` and automatically selected in that case, so
754 usually you can ignore this option. This backend has no backend
757 ``-audiodev wav,id=id[,prop[=value][,...]]``
758 Creates a backend that writes audio to a WAV file.
760 Backend specific options are:
763 Write recorded audio into the specified file. Default is
767 DEF("soundhw", HAS_ARG, QEMU_OPTION_soundhw,
768 "-soundhw c1,... enable audio support\n"
769 " and only specified sound cards (comma separated list)\n"
770 " use '-soundhw help' to get the list of supported cards\n"
771 " use '-soundhw all' to enable all of them\n", QEMU_ARCH_ALL)
773 ``-soundhw card1[,card2,...] or -soundhw all``
774 Enable audio and selected sound hardware. Use 'help' to print all
775 available sound hardware. For example:
779 |qemu_system_x86| -soundhw sb16,adlib disk.img
780 |qemu_system_x86| -soundhw es1370 disk.img
781 |qemu_system_x86| -soundhw ac97 disk.img
782 |qemu_system_x86| -soundhw hda disk.img
783 |qemu_system_x86| -soundhw all disk.img
784 |qemu_system_x86| -soundhw help
786 Note that Linux's i810\_audio OSS kernel (for AC97) module might
787 require manually specifying clocking.
791 modprobe i810_audio clocking=48000
794 DEF("device", HAS_ARG, QEMU_OPTION_device,
795 "-device driver[,prop[=value][,...]]\n"
796 " add device (based on driver)\n"
797 " prop=value,... sets driver properties\n"
798 " use '-device help' to print all possible drivers\n"
799 " use '-device driver,help' to print all possible properties\n",
802 ``-device driver[,prop[=value][,...]]``
803 Add device driver. prop=value sets driver properties. Valid
804 properties depend on the driver. To get help on possible drivers and
805 properties, use ``-device help`` and ``-device driver,help``.
809 ``-device ipmi-bmc-sim,id=id[,slave_addr=val][,sdrfile=file][,furareasize=val][,furdatafile=file][,guid=uuid]``
810 Add an IPMI BMC. This is a simulation of a hardware management
811 interface processor that normally sits on a system. It provides a
812 watchdog and the ability to reset and power control the system. You
813 need to connect this to an IPMI interface to make it useful
815 The IPMI slave address to use for the BMC. The default is 0x20. This
816 address is the BMC's address on the I2C network of management
817 controllers. If you don't know what this means, it is safe to ignore
821 The BMC id for interfaces to use this device.
824 Define slave address to use for the BMC. The default is 0x20.
827 file containing raw Sensor Data Records (SDR) data. The default
831 size of a Field Replaceable Unit (FRU) area. The default is
835 file containing raw Field Replaceable Unit (FRU) inventory data.
839 value for the GUID for the BMC, in standard UUID format. If this
840 is set, get "Get GUID" command to the BMC will return it.
841 Otherwise "Get GUID" will return an error.
843 ``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]``
844 Add a connection to an external IPMI BMC simulator. Instead of
845 locally emulating the BMC like the above item, instead connect to an
846 external entity that provides the IPMI services.
848 A connection is made to an external BMC simulator. If you do this,
849 it is strongly recommended that you use the "reconnect=" chardev
850 option to reconnect to the simulator if the connection is lost. Note
851 that if this is not used carefully, it can be a security issue, as
852 the interface has the ability to send resets, NMIs, and power off
853 the VM. It's best if QEMU makes a connection to an external
854 simulator running on a secure port on localhost, so neither the
855 simulator nor QEMU is exposed to any outside network.
857 See the "lanserv/README.vm" file in the OpenIPMI library for more
858 details on the external interface.
860 ``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]``
861 Add a KCS IPMI interafce on the ISA bus. This also adds a
862 corresponding ACPI and SMBIOS entries, if appropriate.
865 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern
869 Define the I/O address of the interface. The default is 0xca0
873 Define the interrupt to use. The default is 5. To disable
874 interrupts, set this to 0.
876 ``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]``
877 Like the KCS interface, but defines a BT interface. The default port
878 is 0xe4 and the default interrupt is 5.
881 DEF("name", HAS_ARG, QEMU_OPTION_name,
882 "-name string1[,process=string2][,debug-threads=on|off]\n"
883 " set the name of the guest\n"
884 " string1 sets the window title and string2 the process name\n"
885 " When debug-threads is enabled, individual threads are given a separate name\n"
886 " NOTE: The thread names are for debugging and not a stable API.\n",
890 Sets the name of the guest. This name will be displayed in the SDL
891 window caption. The name will also be used for the VNC server. Also
892 optionally set the top visible process name in Linux. Naming of
893 individual threads can also be enabled on Linux to aid debugging.
896 DEF("uuid", HAS_ARG, QEMU_OPTION_uuid,
897 "-uuid %08x-%04x-%04x-%04x-%012x\n"
898 " specify machine UUID\n", QEMU_ARCH_ALL)
906 DEFHEADING(Block device options:)
908 DEF("fda", HAS_ARG, QEMU_OPTION_fda,
909 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
910 DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
915 Use file as floppy disk 0/1 image (see
916 :ref:`disk_005fimages`).
919 DEF("hda", HAS_ARG, QEMU_OPTION_hda,
920 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL)
921 DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL)
922 DEF("hdc", HAS_ARG, QEMU_OPTION_hdc,
923 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL)
924 DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL)
933 Use file as hard disk 0, 1, 2 or 3 image (see
934 :ref:`disk_005fimages`).
937 DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom,
938 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n",
942 Use file as CD-ROM image (you cannot use ``-hdc`` and ``-cdrom`` at
943 the same time). You can use the host CD-ROM by using ``/dev/cdrom``
947 DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev,
948 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n"
949 " [,cache.direct=on|off][,cache.no-flush=on|off]\n"
950 " [,read-only=on|off][,auto-read-only=on|off]\n"
951 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n"
952 " [,driver specific parameters...]\n"
953 " configure a block backend\n", QEMU_ARCH_ALL)
955 ``-blockdev option[,option[,option[,...]]]``
956 Define a new block driver node. Some of the options apply to all
957 block drivers, other options are only accepted for a specific block
958 driver. See below for a list of generic options and options for the
959 most common block drivers.
961 Options that expect a reference to another node (e.g. ``file``) can
962 be given in two ways. Either you specify the node name of an already
963 existing node (file=node-name), or you define a new node inline,
964 adding options for the referenced node after a dot
965 (file.filename=path,file.aio=native).
967 A block driver node created with ``-blockdev`` can be used for a
968 guest device by specifying its node name for the ``drive`` property
969 in a ``-device`` argument that defines a block device.
971 ``Valid options for any block driver node:``
973 Specifies the block driver to use for the given node.
976 This defines the name of the block driver node by which it
977 will be referenced later. The name must be unique, i.e. it
978 must not match the name of a different block driver node, or
979 (if you use ``-drive`` as well) the ID of a drive.
981 If no node name is specified, it is automatically generated.
982 The generated node name is not intended to be predictable
983 and changes between QEMU invocations. For the top level, an
984 explicit node name must be specified.
987 Open the node read-only. Guest write attempts will fail.
989 Note that some block drivers support only read-only access,
990 either generally or in certain configurations. In this case,
991 the default value ``read-only=off`` does not work and the
992 option must be specified explicitly.
995 If ``auto-read-only=on`` is set, QEMU may fall back to
996 read-only usage even when ``read-only=off`` is requested, or
997 even switch between modes as needed, e.g. depending on
998 whether the image file is writable or whether a writing user
999 is attached to the node.
1002 Override the image locking system of QEMU by forcing the
1003 node to utilize weaker shared access for permissions where
1004 it would normally request exclusive access. When there is
1005 the potential for multiple instances to have the same file
1006 open (whether this invocation of QEMU is the first or the
1007 second instance), both instances must permit shared access
1008 for the second instance to succeed at opening the file.
1010 Enabling ``force-share=on`` requires ``read-only=on``.
1013 The host page cache can be avoided with ``cache.direct=on``.
1014 This will attempt to do disk IO directly to the guest's
1015 memory. QEMU may still perform an internal copy of the data.
1018 In case you don't care about data integrity over host
1019 failures, you can use ``cache.no-flush=on``. This option
1020 tells QEMU that it never needs to write any data to the disk
1021 but can instead keep things in cache. If anything goes
1022 wrong, like your host losing power, the disk storage getting
1023 disconnected accidentally, etc. your image will most
1024 probably be rendered unusable.
1027 discard is one of "ignore" (or "off") or "unmap" (or "on")
1028 and controls whether ``discard`` (also known as ``trim`` or
1029 ``unmap``) requests are ignored or passed to the filesystem.
1030 Some machine types may not support discard requests.
1032 ``detect-zeroes=detect-zeroes``
1033 detect-zeroes is "off", "on" or "unmap" and enables the
1034 automatic conversion of plain zero writes by the OS to
1035 driver specific optimized zero write commands. You may even
1036 choose "unmap" if discard is set to "unmap" to allow a zero
1037 write to be converted to an ``unmap`` operation.
1039 ``Driver-specific options for file``
1040 This is the protocol-level block driver for accessing regular
1044 The path to the image file in the local filesystem
1047 Specifies the AIO backend (threads/native, default: threads)
1050 Specifies whether the image file is protected with Linux OFD
1051 / POSIX locks. The default is to use the Linux Open File
1052 Descriptor API if available, otherwise no lock is applied.
1053 (auto/on/off, default: auto)
1059 -blockdev driver=file,node-name=disk,filename=disk.img
1061 ``Driver-specific options for raw``
1062 This is the image format block driver for raw images. It is
1063 usually stacked on top of a protocol level block driver such as
1067 Reference to or definition of the data source block driver
1068 node (e.g. a ``file`` driver node)
1074 -blockdev driver=file,node-name=disk_file,filename=disk.img
1075 -blockdev driver=raw,node-name=disk,file=disk_file
1081 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img
1083 ``Driver-specific options for qcow2``
1084 This is the image format block driver for qcow2 images. It is
1085 usually stacked on top of a protocol level block driver such as
1089 Reference to or definition of the data source block driver
1090 node (e.g. a ``file`` driver node)
1093 Reference to or definition of the backing file block device
1094 (default is taken from the image file). It is allowed to
1095 pass ``null`` here in order to disable the default backing
1099 Whether to enable the lazy refcounts feature (on/off;
1100 default is taken from the image file)
1103 The maximum total size of the L2 table and refcount block
1104 caches in bytes (default: the sum of l2-cache-size and
1105 refcount-cache-size)
1108 The maximum size of the L2 table cache in bytes (default: if
1109 cache-size is not specified - 32M on Linux platforms, and 8M
1110 on non-Linux platforms; otherwise, as large as possible
1111 within the cache-size, while permitting the requested or the
1112 minimal refcount cache size)
1114 ``refcount-cache-size``
1115 The maximum size of the refcount block cache in bytes
1116 (default: 4 times the cluster size; or if cache-size is
1117 specified, the part of it which is not used for the L2
1120 ``cache-clean-interval``
1121 Clean unused entries in the L2 and refcount caches. The
1122 interval is in seconds. The default value is 600 on
1123 supporting platforms, and 0 on other platforms. Setting it
1124 to 0 disables this feature.
1126 ``pass-discard-request``
1127 Whether discard requests to the qcow2 device should be
1128 forwarded to the data source (on/off; default: on if
1129 discard=unmap is specified, off otherwise)
1131 ``pass-discard-snapshot``
1132 Whether discard requests for the data source should be
1133 issued when a snapshot operation (e.g. deleting a snapshot)
1134 frees clusters in the qcow2 file (on/off; default: on)
1136 ``pass-discard-other``
1137 Whether discard requests for the data source should be
1138 issued on other occasions where a cluster gets freed
1139 (on/off; default: off)
1142 Which overlap checks to perform for writes to the image
1143 (none/constant/cached/all; default: cached). For details or
1144 finer granularity control refer to the QAPI documentation of
1151 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2
1152 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216
1158 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2
1160 ``Driver-specific options for other drivers``
1161 Please refer to the QAPI documentation of the ``blockdev-add``
1165 DEF("drive", HAS_ARG, QEMU_OPTION_drive,
1166 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n"
1167 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n"
1168 " [,snapshot=on|off][,rerror=ignore|stop|report]\n"
1169 " [,werror=ignore|stop|report|enospc][,id=name][,aio=threads|native]\n"
1170 " [,readonly=on|off][,copy-on-read=on|off]\n"
1171 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n"
1172 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n"
1173 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n"
1174 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n"
1175 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n"
1176 " [[,iops_size=is]]\n"
1178 " use 'file' as a drive image\n", QEMU_ARCH_ALL)
1180 ``-drive option[,option[,option[,...]]]``
1181 Define a new drive. This includes creating a block driver node (the
1182 backend) as well as a guest device, and is mostly a shortcut for
1183 defining the corresponding ``-blockdev`` and ``-device`` options.
1185 ``-drive`` accepts all options that are accepted by ``-blockdev``.
1186 In addition, it knows the following options:
1189 This option defines which disk image (see
1190 :ref:`disk_005fimages`) to use with this drive. If
1191 the filename contains comma, you must double it (for instance,
1192 "file=my,,file" to use file "my,file").
1194 Special files such as iSCSI devices can be specified using
1195 protocol specific URLs. See the section for "Device URL Syntax"
1196 for more information.
1199 This option defines on which type on interface the drive is
1200 connected. Available types are: ide, scsi, sd, mtd, floppy,
1201 pflash, virtio, none.
1203 ``bus=bus,unit=unit``
1204 These options define where is connected the drive by defining
1205 the bus number and the unit id.
1208 This option defines where is connected the drive by using an
1209 index in the list of available connectors of a given interface
1213 This option defines the type of the media: disk or cdrom.
1215 ``snapshot=snapshot``
1216 snapshot is "on" or "off" and controls snapshot mode for the
1217 given drive (see ``-snapshot``).
1220 cache is "none", "writeback", "unsafe", "directsync" or
1221 "writethrough" and controls how the host cache is used to access
1222 block data. This is a shortcut that sets the ``cache.direct``
1223 and ``cache.no-flush`` options (as in ``-blockdev``), and
1224 additionally ``cache.writeback``, which provides a default for
1225 the ``write-cache`` option of block guest devices (as in
1226 ``-device``). The modes correspond to the following settings:
1228 ============= =============== ============ ==============
1229 \ cache.writeback cache.direct cache.no-flush
1230 ============= =============== ============ ==============
1231 writeback on off off
1233 writethrough off off off
1234 directsync off on off
1236 ============= =============== ============ ==============
1238 The default mode is ``cache=writeback``.
1241 aio is "threads", or "native" and selects between pthread based
1242 disk I/O and native Linux AIO.
1245 Specify which disk format will be used rather than detecting the
1246 format. Can be used to specify format=raw to avoid interpreting
1247 an untrusted format header.
1249 ``werror=action,rerror=action``
1250 Specify which action to take on write and read errors. Valid
1251 actions are: "ignore" (ignore the error and try to continue),
1252 "stop" (pause QEMU), "report" (report the error to the guest),
1253 "enospc" (pause QEMU only if the host disk is full; report the
1254 error to the guest otherwise). The default setting is
1255 ``werror=enospc`` and ``rerror=report``.
1257 ``copy-on-read=copy-on-read``
1258 copy-on-read is "on" or "off" and enables whether to copy read
1259 backing file sectors into the image file.
1261 ``bps=b,bps_rd=r,bps_wr=w``
1262 Specify bandwidth throttling limits in bytes per second, either
1263 for all request types or for reads or writes only. Small values
1264 can lead to timeouts or hangs inside the guest. A safe minimum
1265 for disks is 2 MB/s.
1267 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm``
1268 Specify bursts in bytes per second, either for all request types
1269 or for reads or writes only. Bursts allow the guest I/O to spike
1270 above the limit temporarily.
1272 ``iops=i,iops_rd=r,iops_wr=w``
1273 Specify request rate limits in requests per second, either for
1274 all request types or for reads or writes only.
1276 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm``
1277 Specify bursts in requests per second, either for all request
1278 types or for reads or writes only. Bursts allow the guest I/O to
1279 spike above the limit temporarily.
1282 Let every is bytes of a request count as a new request for iops
1283 throttling purposes. Use this option to prevent guests from
1284 circumventing iops limits by sending fewer but larger requests.
1287 Join a throttling quota group with given name g. All drives that
1288 are members of the same group are accounted for together. Use
1289 this option to prevent guests from circumventing throttling
1290 limits by using many small disks instead of a single larger
1293 By default, the ``cache.writeback=on`` mode is used. It will report
1294 data writes as completed as soon as the data is present in the host
1295 page cache. This is safe as long as your guest OS makes sure to
1296 correctly flush disk caches where needed. If your guest OS does not
1297 handle volatile disk write caches correctly and your host crashes or
1298 loses power, then the guest may experience data corruption.
1300 For such guests, you should consider using ``cache.writeback=off``.
1301 This means that the host page cache will be used to read and write
1302 data, but write notification will be sent to the guest only after
1303 QEMU has made sure to flush each write to the disk. Be aware that
1304 this has a major impact on performance.
1306 When using the ``-snapshot`` option, unsafe caching is always used.
1308 Copy-on-read avoids accessing the same backing file sectors
1309 repeatedly and is useful when the backing file is over a slow
1310 network. By default copy-on-read is off.
1312 Instead of ``-cdrom`` you can use:
1316 |qemu_system| -drive file=file,index=2,media=cdrom
1318 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use:
1322 |qemu_system| -drive file=file,index=0,media=disk
1323 |qemu_system| -drive file=file,index=1,media=disk
1324 |qemu_system| -drive file=file,index=2,media=disk
1325 |qemu_system| -drive file=file,index=3,media=disk
1327 You can open an image using pre-opened file descriptors from an fd
1333 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \
1334 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \
1335 -drive file=/dev/fdset/2,index=0,media=disk
1337 You can connect a CDROM to the slave of ide0:
1341 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom
1343 If you don't specify the "file=" argument, you define an empty
1348 |qemu_system_x86| -drive if=ide,index=1,media=cdrom
1350 Instead of ``-fda``, ``-fdb``, you can use:
1354 |qemu_system_x86| -drive file=file,index=0,if=floppy
1355 |qemu_system_x86| -drive file=file,index=1,if=floppy
1357 By default, interface is "ide" and index is automatically
1362 |qemu_system_x86| -drive file=a -drive file=b"
1364 is interpreted like:
1368 |qemu_system_x86| -hda a -hdb b
1371 DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock,
1372 "-mtdblock file use 'file' as on-board Flash memory image\n",
1376 Use file as on-board Flash memory image.
1379 DEF("sd", HAS_ARG, QEMU_OPTION_sd,
1380 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL)
1383 Use file as SecureDigital card image.
1386 DEF("pflash", HAS_ARG, QEMU_OPTION_pflash,
1387 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL)
1390 Use file as a parallel flash image.
1393 DEF("snapshot", 0, QEMU_OPTION_snapshot,
1394 "-snapshot write to temporary files instead of disk image files\n",
1398 Write to temporary files instead of disk image files. In this case,
1399 the raw disk image you use is not written back. You can however
1400 force the write back by pressing C-a s (see
1401 :ref:`disk_005fimages`).
1404 DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev,
1405 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1406 " [,writeout=immediate][,readonly][,fmode=fmode][,dmode=dmode]\n"
1407 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n"
1408 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n"
1409 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n"
1410 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n"
1411 " [[,throttling.iops-size=is]]\n"
1412 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly]\n"
1413 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly]\n"
1414 "-fsdev synth,id=id\n",
1418 ``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]``
1420 ``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly]``
1422 ``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly]``
1424 ``-fsdev synth,id=id[,readonly]``
1425 Define a new file system device. Valid options are:
1428 Accesses to the filesystem are done by QEMU.
1431 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1434 Synthetic filesystem, only used by QTests.
1437 Specifies identifier for this device.
1440 Specifies the export path for the file system device. Files
1441 under this path will be available to the 9p client on the guest.
1443 ``security_model=security_model``
1444 Specifies the security model to be used for this export path.
1445 Supported security models are "passthrough", "mapped-xattr",
1446 "mapped-file" and "none". In "passthrough" security model, files
1447 are stored using the same credentials as they are created on the
1448 guest. This requires QEMU to run as root. In "mapped-xattr"
1449 security model, some of the file attributes like uid, gid, mode
1450 bits and link target are stored as file attributes. For
1451 "mapped-file" these attributes are stored in the hidden
1452 .virtfs\_metadata directory. Directories exported by this
1453 security model cannot interact with other unix tools. "none"
1454 security model is same as passthrough except the sever won't
1455 report failures if it fails to set file attributes like
1456 ownership. Security model is mandatory only for local fsdriver.
1457 Other fsdrivers (like proxy) don't take security model as a
1460 ``writeout=writeout``
1461 This is an optional argument. The only supported value is
1462 "immediate". This means that host page cache will be used to
1463 read and write data but write notification will be sent to the
1464 guest only when the data has been reported as written by the
1468 Enables exporting 9p share as a readonly mount for guests. By
1469 default read-write access is given.
1472 Enables proxy filesystem driver to use passed socket file for
1473 communicating with virtfs-proxy-helper(1).
1476 Enables proxy filesystem driver to use passed socket descriptor
1477 for communicating with virtfs-proxy-helper(1). Usually a helper
1478 like libvirt will create socketpair and pass one of the fds as
1482 Specifies the default mode for newly created files on the host.
1483 Works only with security models "mapped-xattr" and
1487 Specifies the default mode for newly created directories on the
1488 host. Works only with security models "mapped-xattr" and
1491 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w``
1492 Specify bandwidth throttling limits in bytes per second, either
1493 for all request types or for reads or writes only.
1495 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm``
1496 Specify bursts in bytes per second, either for all request types
1497 or for reads or writes only. Bursts allow the guest I/O to spike
1498 above the limit temporarily.
1500 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w``
1501 Specify request rate limits in requests per second, either for
1502 all request types or for reads or writes only.
1504 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm``
1505 Specify bursts in requests per second, either for all request
1506 types or for reads or writes only. Bursts allow the guest I/O to
1507 spike above the limit temporarily.
1509 ``throttling.iops-size=is``
1510 Let every is bytes of a request count as a new request for iops
1511 throttling purposes.
1513 -fsdev option is used along with -device driver "virtio-9p-...".
1515 ``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag``
1516 Options for virtio-9p-... driver are:
1519 Specifies the variant to be used. Supported values are "pci",
1520 "ccw" or "device", depending on the machine type.
1523 Specifies the id value specified along with -fsdev option.
1525 ``mount_tag=mount_tag``
1526 Specifies the tag name to be used by the guest to mount this
1530 DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs,
1531 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1532 " [,id=id][,writeout=immediate][,readonly][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n"
1533 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly]\n"
1534 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly]\n"
1535 "-virtfs synth,mount_tag=tag[,id=id][,readonly]\n",
1539 ``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]``
1541 ``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly]``
1543 ``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly]``
1545 ``-virtfs synth,mount_tag=mount_tag``
1546 Define a new virtual filesystem device and expose it to the guest using
1547 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain
1548 directory on host is made directly accessible by guest as a pass-through
1549 file system by using the 9P network protocol for communication between
1550 host and guests, if desired even accessible, shared by several guests
1553 Note that ``-virtfs`` is actually just a convenience shortcut for its
1554 generalized form ``-fsdev -device virtio-9p-pci``.
1556 The general form of pass-through file system options are:
1559 Accesses to the filesystem are done by QEMU.
1562 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1565 Synthetic filesystem, only used by QTests.
1568 Specifies identifier for the filesystem device
1571 Specifies the export path for the file system device. Files
1572 under this path will be available to the 9p client on the guest.
1574 ``security_model=security_model``
1575 Specifies the security model to be used for this export path.
1576 Supported security models are "passthrough", "mapped-xattr",
1577 "mapped-file" and "none". In "passthrough" security model, files
1578 are stored using the same credentials as they are created on the
1579 guest. This requires QEMU to run as root. In "mapped-xattr"
1580 security model, some of the file attributes like uid, gid, mode
1581 bits and link target are stored as file attributes. For
1582 "mapped-file" these attributes are stored in the hidden
1583 .virtfs\_metadata directory. Directories exported by this
1584 security model cannot interact with other unix tools. "none"
1585 security model is same as passthrough except the sever won't
1586 report failures if it fails to set file attributes like
1587 ownership. Security model is mandatory only for local fsdriver.
1588 Other fsdrivers (like proxy) don't take security model as a
1591 ``writeout=writeout``
1592 This is an optional argument. The only supported value is
1593 "immediate". This means that host page cache will be used to
1594 read and write data but write notification will be sent to the
1595 guest only when the data has been reported as written by the
1599 Enables exporting 9p share as a readonly mount for guests. By
1600 default read-write access is given.
1603 Enables proxy filesystem driver to use passed socket file for
1604 communicating with virtfs-proxy-helper(1). Usually a helper like
1605 libvirt will create socketpair and pass one of the fds as
1609 Enables proxy filesystem driver to use passed 'sock\_fd' as the
1610 socket descriptor for interfacing with virtfs-proxy-helper(1).
1613 Specifies the default mode for newly created files on the host.
1614 Works only with security models "mapped-xattr" and
1618 Specifies the default mode for newly created directories on the
1619 host. Works only with security models "mapped-xattr" and
1622 ``mount_tag=mount_tag``
1623 Specifies the tag name to be used by the guest to mount this
1626 ``multidevs=multidevs``
1627 Specifies how to deal with multiple devices being shared with a
1628 9p export. Supported behaviours are either "remap", "forbid" or
1629 "warn". The latter is the default behaviour on which virtfs 9p
1630 expects only one device to be shared with the same export, and
1631 if more than one device is shared and accessed via the same 9p
1632 export then only a warning message is logged (once) by qemu on
1633 host side. In order to avoid file ID collisions on guest you
1634 should either create a separate virtfs export for each device to
1635 be shared with guests (recommended way) or you might use "remap"
1636 instead which allows you to share multiple devices with only one
1637 export instead, which is achieved by remapping the original
1638 inode numbers from host to guest in a way that would prevent
1639 such collisions. Remapping inodes in such use cases is required
1640 because the original device IDs from host are never passed and
1641 exposed on guest. Instead all files of an export shared with
1642 virtfs always share the same device id on guest. So two files
1643 with identical inode numbers but from actually different devices
1644 on host would otherwise cause a file ID collision and hence
1645 potential misbehaviours on guest. "forbid" on the other hand
1646 assumes like "warn" that only one device is shared by the same
1647 export, however it will not only log a warning message but also
1648 deny access to additional devices on guest. Note though that
1649 "forbid" does currently not block all possible file access
1650 operations (e.g. readdir() would still return entries from other
1654 DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi,
1655 "-iscsi [user=user][,password=password]\n"
1656 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE\n"
1657 " [,initiator-name=initiator-iqn][,id=target-iqn]\n"
1658 " [,timeout=timeout]\n"
1659 " iSCSI session parameters\n", QEMU_ARCH_ALL)
1663 Configure iSCSI session parameters.
1668 DEFHEADING(USB options:)
1670 DEF("usb", 0, QEMU_OPTION_usb,
1671 "-usb enable on-board USB host controller (if not enabled by default)\n",
1675 Enable USB emulation on machine types with an on-board USB host
1676 controller (if not enabled by default). Note that on-board USB host
1677 controllers may not support USB 3.0. In this case
1678 ``-device qemu-xhci`` can be used instead on machines with PCI.
1681 DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice,
1682 "-usbdevice name add the host or guest USB device 'name'\n",
1685 ``-usbdevice devname``
1686 Add the USB device devname. Note that this option is deprecated,
1687 please use ``-device usb-...`` instead. See
1688 :ref:`usb_005fdevices`.
1691 Virtual Mouse. This will override the PS/2 mouse emulation when
1695 Pointer device that uses absolute coordinates (like a
1696 touchscreen). This means QEMU is able to report the mouse
1697 position without having to grab the mouse. Also overrides the
1698 PS/2 mouse emulation when activated.
1701 Braille device. This will use BrlAPI to display the braille
1702 output on a real or fake device.
1707 DEFHEADING(Display options:)
1709 DEF("display", HAS_ARG, QEMU_OPTION_display,
1710 #if defined(CONFIG_SPICE)
1711 "-display spice-app[,gl=on|off]\n"
1713 #if defined(CONFIG_SDL)
1714 "-display sdl[,alt_grab=on|off][,ctrl_grab=on|off]\n"
1715 " [,window_close=on|off][,gl=on|core|es|off]\n"
1717 #if defined(CONFIG_GTK)
1718 "-display gtk[,grab_on_hover=on|off][,gl=on|off]|\n"
1720 #if defined(CONFIG_VNC)
1721 "-display vnc=<display>[,<optargs>]\n"
1723 #if defined(CONFIG_CURSES)
1724 "-display curses[,charset=<encoding>]\n"
1726 #if defined(CONFIG_OPENGL)
1727 "-display egl-headless[,rendernode=<file>]\n"
1730 " select display backend type\n"
1731 " The default display is equivalent to\n "
1732 #if defined(CONFIG_GTK)
1733 "\"-display gtk\"\n"
1734 #elif defined(CONFIG_SDL)
1735 "\"-display sdl\"\n"
1736 #elif defined(CONFIG_COCOA)
1737 "\"-display cocoa\"\n"
1738 #elif defined(CONFIG_VNC)
1739 "\"-vnc localhost:0,to=99,id=default\"\n"
1741 "\"-display none\"\n"
1746 Select type of display to use. This option is a replacement for the
1747 old style -sdl/-curses/... options. Use ``-display help`` to list
1748 the available display types. Valid values for type are
1751 Display video output via SDL (usually in a separate graphics
1752 window; see the SDL documentation for other possibilities).
1755 Display video output via curses. For graphics device models
1756 which support a text mode, QEMU can display this output using a
1757 curses/ncurses interface. Nothing is displayed when the graphics
1758 device is in graphical mode or if the graphics device does not
1759 support a text mode. Generally only the VGA device models
1760 support text mode. The font charset used by the guest can be
1761 specified with the ``charset`` option, for example
1762 ``charset=CP850`` for IBM CP850 encoding. The default is
1766 Do not display video output. The guest will still see an
1767 emulated graphics card, but its output will not be displayed to
1768 the QEMU user. This option differs from the -nographic option in
1769 that it only affects what is done with video output; -nographic
1770 also changes the destination of the serial and parallel port
1774 Display video output in a GTK window. This interface provides
1775 drop-down menus and other UI elements to configure and control
1776 the VM during runtime.
1779 Start a VNC server on display <arg>
1782 Offload all OpenGL operations to a local DRI device. For any
1783 graphical display, this display needs to be paired with either
1784 VNC or SPICE displays.
1787 Start QEMU as a Spice server and launch the default Spice client
1788 application. The Spice server will redirect the serial consoles
1789 and QEMU monitors. (Since 4.0)
1792 DEF("nographic", 0, QEMU_OPTION_nographic,
1793 "-nographic disable graphical output and redirect serial I/Os to console\n",
1797 Normally, if QEMU is compiled with graphical window support, it
1798 displays output such as guest graphics, guest console, and the QEMU
1799 monitor in a window. With this option, you can totally disable
1800 graphical output so that QEMU is a simple command line application.
1801 The emulated serial port is redirected on the console and muxed with
1802 the monitor (unless redirected elsewhere explicitly). Therefore, you
1803 can still use QEMU to debug a Linux kernel with a serial console.
1804 Use C-a h for help on switching between the console and monitor.
1807 DEF("curses", 0, QEMU_OPTION_curses,
1808 "-curses shorthand for -display curses\n",
1812 Normally, if QEMU is compiled with graphical window support, it
1813 displays output such as guest graphics, guest console, and the QEMU
1814 monitor in a window. With this option, QEMU can display the VGA
1815 output when in text mode using a curses/ncurses interface. Nothing
1816 is displayed in graphical mode.
1819 DEF("alt-grab", 0, QEMU_OPTION_alt_grab,
1820 "-alt-grab use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt)\n",
1824 Use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt). Note that
1825 this also affects the special keys (for fullscreen, monitor-mode
1829 DEF("ctrl-grab", 0, QEMU_OPTION_ctrl_grab,
1830 "-ctrl-grab use Right-Ctrl to grab mouse (instead of Ctrl-Alt)\n",
1834 Use Right-Ctrl to grab mouse (instead of Ctrl-Alt). Note that this
1835 also affects the special keys (for fullscreen, monitor-mode
1839 DEF("no-quit", 0, QEMU_OPTION_no_quit,
1840 "-no-quit disable SDL window close capability\n", QEMU_ARCH_ALL)
1843 Disable SDL window close capability.
1846 DEF("sdl", 0, QEMU_OPTION_sdl,
1847 "-sdl shorthand for -display sdl\n", QEMU_ARCH_ALL)
1853 DEF("spice", HAS_ARG, QEMU_OPTION_spice,
1854 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n"
1855 " [,x509-key-file=<file>][,x509-key-password=<file>]\n"
1856 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n"
1857 " [,x509-dh-key-file=<file>][,addr=addr][,ipv4|ipv6|unix]\n"
1858 " [,tls-ciphers=<list>]\n"
1859 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n"
1860 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n"
1861 " [,sasl][,password=<secret>][,disable-ticketing]\n"
1862 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n"
1863 " [,jpeg-wan-compression=[auto|never|always]]\n"
1864 " [,zlib-glz-wan-compression=[auto|never|always]]\n"
1865 " [,streaming-video=[off|all|filter]][,disable-copy-paste]\n"
1866 " [,disable-agent-file-xfer][,agent-mouse=[on|off]]\n"
1867 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n"
1868 " [,gl=[on|off]][,rendernode=<file>]\n"
1870 " at least one of {port, tls-port} is mandatory\n",
1873 ``-spice option[,option[,...]]``
1874 Enable the spice remote desktop protocol. Valid options are
1877 Set the TCP port spice is listening on for plaintext channels.
1880 Set the IP address spice is listening on. Default is any
1883 ``ipv4``; \ ``ipv6``; \ ``unix``
1884 Force using the specified IP version.
1886 ``password=<secret>``
1887 Set the password you need to authenticate.
1890 Require that the client use SASL to authenticate with the spice.
1891 The exact choice of authentication method used is controlled
1892 from the system / user's SASL configuration file for the 'qemu'
1893 service. This is typically found in /etc/sasl2/qemu.conf. If
1894 running QEMU as an unprivileged user, an environment variable
1895 SASL\_CONF\_PATH can be used to make it search alternate
1896 locations for the service config. While some SASL auth methods
1897 can also provide data encryption (eg GSSAPI), it is recommended
1898 that SASL always be combined with the 'tls' and 'x509' settings
1899 to enable use of SSL and server certificates. This ensures a
1900 data encryption preventing compromise of authentication
1903 ``disable-ticketing``
1904 Allow client connects without authentication.
1906 ``disable-copy-paste``
1907 Disable copy paste between the client and the guest.
1909 ``disable-agent-file-xfer``
1910 Disable spice-vdagent based file-xfer between the client and the
1914 Set the TCP port spice is listening on for encrypted channels.
1917 Set the x509 file directory. Expects same filenames as -vnc
1920 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>``
1921 The x509 file names can also be configured individually.
1923 ``tls-ciphers=<list>``
1924 Specify which ciphers to use.
1926 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]``
1927 Force specific channel to be used with or without TLS
1928 encryption. The options can be specified multiple times to
1929 configure multiple channels. The special name "default" can be
1930 used to set the default mode. For channels which are not
1931 explicitly forced into one mode the spice client is allowed to
1932 pick tls/plaintext as he pleases.
1934 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]``
1935 Configure image compression (lossless). Default is auto\_glz.
1937 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]``
1938 Configure wan image compression (lossy for slow links). Default
1941 ``streaming-video=[off|all|filter]``
1942 Configure video stream detection. Default is off.
1944 ``agent-mouse=[on|off]``
1945 Enable/disable passing mouse events via vdagent. Default is on.
1947 ``playback-compression=[on|off]``
1948 Enable/disable audio stream compression (using celt 0.5.1).
1951 ``seamless-migration=[on|off]``
1952 Enable/disable spice seamless migration. Default is off.
1955 Enable/disable OpenGL context. Default is off.
1957 ``rendernode=<file>``
1958 DRM render node for OpenGL rendering. If not specified, it will
1959 pick the first available. (Since 2.9)
1962 DEF("portrait", 0, QEMU_OPTION_portrait,
1963 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n",
1967 Rotate graphical output 90 deg left (only PXA LCD).
1970 DEF("rotate", HAS_ARG, QEMU_OPTION_rotate,
1971 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n",
1975 Rotate graphical output some deg left (only PXA LCD).
1978 DEF("vga", HAS_ARG, QEMU_OPTION_vga,
1979 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n"
1980 " select video card type\n", QEMU_ARCH_ALL)
1983 Select type of VGA card to emulate. Valid values for type are
1986 Cirrus Logic GD5446 Video card. All Windows versions starting
1987 from Windows 95 should recognize and use this graphic card. For
1988 optimal performances, use 16 bit color depth in the guest and
1989 the host OS. (This card was the default before QEMU 2.2)
1992 Standard VGA card with Bochs VBE extensions. If your guest OS
1993 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if
1994 you want to use high resolution modes (>= 1280x1024x16) then you
1995 should use this option. (This card is the default since QEMU
1999 VMWare SVGA-II compatible adapter. Use it if you have
2000 sufficiently recent XFree86/XOrg server or Windows guest with a
2001 driver for this card.
2004 QXL paravirtual graphic card. It is VGA compatible (including
2005 VESA 2.0 VBE support). Works best with qxl guest drivers
2006 installed though. Recommended choice when using the spice
2010 (sun4m only) Sun TCX framebuffer. This is the default
2011 framebuffer for sun4m machines and offers both 8-bit and 24-bit
2012 colour depths at a fixed resolution of 1024x768.
2015 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit
2016 framebuffer for sun4m machines available in both 1024x768
2017 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people
2018 wishing to run older Solaris versions.
2027 DEF("full-screen", 0, QEMU_OPTION_full_screen,
2028 "-full-screen start in full screen\n", QEMU_ARCH_ALL)
2031 Start in full screen.
2034 DEF("g", HAS_ARG, QEMU_OPTION_g ,
2035 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n",
2036 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K)
2038 ``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]``
2039 Set the initial graphical resolution and depth (PPC, SPARC only).
2041 For PPC the default is 800x600x32.
2043 For SPARC with the TCX graphics device, the default is 1024x768x8
2044 with the option of 1024x768x24. For cgthree, the default is
2045 1024x768x8 with the option of 1152x900x8 for people who wish to use
2049 DEF("vnc", HAS_ARG, QEMU_OPTION_vnc ,
2050 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL)
2052 ``-vnc display[,option[,option[,...]]]``
2053 Normally, if QEMU is compiled with graphical window support, it
2054 displays output such as guest graphics, guest console, and the QEMU
2055 monitor in a window. With this option, you can have QEMU listen on
2056 VNC display display and redirect the VGA display over the VNC
2057 session. It is very useful to enable the usb tablet device when
2058 using this option (option ``-device usb-tablet``). When using the
2059 VNC display, you must use the ``-k`` parameter to set the keyboard
2060 layout if you are not using en-us. Valid syntax for the display is
2063 With this option, QEMU will try next available VNC displays,
2064 until the number L, if the origianlly defined "-vnc display" is
2065 not available, e.g. port 5900+display is already used by another
2066 application. By default, to=0.
2069 TCP connections will only be allowed from host on display d. By
2070 convention the TCP port is 5900+d. Optionally, host can be
2071 omitted in which case the server will accept connections from
2075 Connections will be allowed over UNIX domain sockets where path
2076 is the location of a unix socket to listen for connections on.
2079 VNC is initialized but not started. The monitor ``change``
2080 command can be used to later start the VNC server.
2082 Following the display value there may be one or more option flags
2083 separated by commas. Valid options are
2086 Connect to a listening VNC client via a "reverse" connection.
2087 The client is specified by the display. For reverse network
2088 connections (host:d,``reverse``), the d argument is a TCP port
2089 number, not a display number.
2092 Opens an additional TCP listening port dedicated to VNC
2093 Websocket connections. If a bare websocket option is given, the
2094 Websocket port is 5700+display. An alternative port can be
2095 specified with the syntax ``websocket``\ =port.
2097 If host is specified connections will only be allowed from this
2098 host. It is possible to control the websocket listen address
2099 independently, using the syntax ``websocket``\ =host:port.
2101 If no TLS credentials are provided, the websocket connection
2102 runs in unencrypted mode. If TLS credentials are provided, the
2103 websocket connection requires encrypted client connections.
2106 Require that password based authentication is used for client
2109 The password must be set separately using the ``set_password``
2110 command in the :ref:`pcsys_005fmonitor`. The
2111 syntax to change your password is:
2112 ``set_password <protocol> <password>`` where <protocol> could be
2113 either "vnc" or "spice".
2115 If you would like to change <protocol> password expiration, you
2116 should use ``expire_password <protocol> <expiration-time>``
2117 where expiration time could be one of the following options:
2118 now, never, +seconds or UNIX time of expiration, e.g. +60 to
2119 make password expire in 60 seconds, or 1335196800 to make
2120 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for
2121 this date and time).
2123 You can also use keywords "now" or "never" for the expiration
2124 time to allow <protocol> password to expire immediately or never
2128 Provides the ID of a set of TLS credentials to use to secure the
2129 VNC server. They will apply to both the normal VNC server socket
2130 and the websocket socket (if enabled). Setting TLS credentials
2131 will cause the VNC server socket to enable the VeNCrypt auth
2132 mechanism. The credentials should have been previously created
2133 using the ``-object tls-creds`` argument.
2136 Provides the ID of the QAuthZ authorization object against which
2137 the client's x509 distinguished name will validated. This object
2138 is only resolved at time of use, so can be deleted and recreated
2139 on the fly while the VNC server is active. If missing, it will
2140 default to denying access.
2143 Require that the client use SASL to authenticate with the VNC
2144 server. The exact choice of authentication method used is
2145 controlled from the system / user's SASL configuration file for
2146 the 'qemu' service. This is typically found in
2147 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user,
2148 an environment variable SASL\_CONF\_PATH can be used to make it
2149 search alternate locations for the service config. While some
2150 SASL auth methods can also provide data encryption (eg GSSAPI),
2151 it is recommended that SASL always be combined with the 'tls'
2152 and 'x509' settings to enable use of SSL and server
2153 certificates. This ensures a data encryption preventing
2154 compromise of authentication credentials. See the
2155 :ref:`vnc_005fsecurity` section for details on
2156 using SASL authentication.
2159 Provides the ID of the QAuthZ authorization object against which
2160 the client's SASL username will validated. This object is only
2161 resolved at time of use, so can be deleted and recreated on the
2162 fly while the VNC server is active. If missing, it will default
2166 Legacy method for enabling authorization of clients against the
2167 x509 distinguished name and SASL username. It results in the
2168 creation of two ``authz-list`` objects with IDs of
2169 ``vnc.username`` and ``vnc.x509dname``. The rules for these
2170 objects must be configured with the HMP ACL commands.
2172 This option is deprecated and should no longer be used. The new
2173 ``sasl-authz`` and ``tls-authz`` options are a replacement.
2176 Enable lossy compression methods (gradient, JPEG, ...). If this
2177 option is set, VNC client may receive lossy framebuffer updates
2178 depending on its encoding settings. Enabling this option can
2179 save a lot of bandwidth at the expense of quality.
2182 Disable adaptive encodings. Adaptive encodings are enabled by
2183 default. An adaptive encoding will try to detect frequently
2184 updated screen regions, and send updates in these regions using
2185 a lossy encoding (like JPEG). This can be really helpful to save
2186 bandwidth when playing videos. Disabling adaptive encodings
2187 restores the original static behavior of encodings like Tight.
2189 ``share=[allow-exclusive|force-shared|ignore]``
2190 Set display sharing policy. 'allow-exclusive' allows clients to
2191 ask for exclusive access. As suggested by the rfb spec this is
2192 implemented by dropping other connections. Connecting multiple
2193 clients in parallel requires all clients asking for a shared
2194 session (vncviewer: -shared switch). This is the default.
2195 'force-shared' disables exclusive client access. Useful for
2196 shared desktop sessions, where you don't want someone forgetting
2197 specify -shared disconnect everybody else. 'ignore' completely
2198 ignores the shared flag and allows everybody connect
2199 unconditionally. Doesn't conform to the rfb spec but is
2200 traditional QEMU behavior.
2203 Set keyboard delay, for key down and key up events, in
2204 milliseconds. Default is 10. Keyboards are low-bandwidth
2205 devices, so this slowdown can help the device and guest to keep
2206 up and not lose events in case events are arriving in bulk.
2207 Possible causes for the latter are flaky network connections, or
2208 scripts for automated testing.
2210 ``audiodev=audiodev``
2211 Use the specified audiodev when the VNC client requests audio
2212 transmission. When not using an -audiodev argument, this option
2213 must be omitted, otherwise is must be present and specify a
2217 ARCHHEADING(, QEMU_ARCH_I386)
2219 ARCHHEADING(i386 target only:, QEMU_ARCH_I386)
2221 DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack,
2222 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n",
2226 Use it when installing Windows 2000 to avoid a disk full bug. After
2227 Windows 2000 is installed, you no longer need this option (this
2228 option slows down the IDE transfers).
2231 DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk,
2232 "-no-fd-bootchk disable boot signature checking for floppy disks\n",
2236 Disable boot signature checking for floppy disks in BIOS. May be
2237 needed to boot from old floppy disks.
2240 DEF("no-acpi", 0, QEMU_OPTION_no_acpi,
2241 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM)
2244 Disable ACPI (Advanced Configuration and Power Interface) support.
2245 Use it if your guest OS complains about ACPI problems (PC target
2249 DEF("no-hpet", 0, QEMU_OPTION_no_hpet,
2250 "-no-hpet disable HPET\n", QEMU_ARCH_I386)
2253 Disable HPET support.
2256 DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable,
2257 "-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"
2258 " ACPI table description\n", QEMU_ARCH_I386)
2260 ``-acpitable [sig=str][,rev=n][,oem_id=str][,oem_table_id=str][,oem_rev=n] [,asl_compiler_id=str][,asl_compiler_rev=n][,data=file1[:file2]...]``
2261 Add ACPI table with specified header fields and context from
2262 specified files. For file=, take whole ACPI table from the specified
2263 files, including all ACPI headers (possible overridden by other
2264 options). For data=, only data portion of the table is used, all
2265 header information is specified in the command line. If a SLIC table
2266 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id
2267 fields will override the same in the RSDT and the FADT (a.k.a.
2268 FACP), in order to ensure the field matches required by the
2269 Microsoft SLIC spec and the ACPI spec.
2272 DEF("smbios", HAS_ARG, QEMU_OPTION_smbios,
2273 "-smbios file=binary\n"
2274 " load SMBIOS entry from binary file\n"
2275 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n"
2277 " specify SMBIOS type 0 fields\n"
2278 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2279 " [,uuid=uuid][,sku=str][,family=str]\n"
2280 " specify SMBIOS type 1 fields\n"
2281 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2282 " [,asset=str][,location=str]\n"
2283 " specify SMBIOS type 2 fields\n"
2284 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n"
2286 " specify SMBIOS type 3 fields\n"
2287 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n"
2288 " [,asset=str][,part=str]\n"
2289 " specify SMBIOS type 4 fields\n"
2290 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n"
2291 " [,asset=str][,part=str][,speed=%d]\n"
2292 " specify SMBIOS type 17 fields\n",
2293 QEMU_ARCH_I386 | QEMU_ARCH_ARM)
2295 ``-smbios file=binary``
2296 Load SMBIOS entry from binary file.
2298 ``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]``
2299 Specify SMBIOS type 0 fields
2301 ``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]``
2302 Specify SMBIOS type 1 fields
2304 ``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]``
2305 Specify SMBIOS type 2 fields
2307 ``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]``
2308 Specify SMBIOS type 3 fields
2310 ``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str]``
2311 Specify SMBIOS type 4 fields
2313 ``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]``
2314 Specify SMBIOS type 17 fields
2319 DEFHEADING(Network options:)
2321 DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
2323 "-netdev user,id=str[,ipv4[=on|off]][,net=addr[/mask]][,host=addr]\n"
2324 " [,ipv6[=on|off]][,ipv6-net=addr[/int]][,ipv6-host=addr]\n"
2325 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n"
2326 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n"
2327 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]"
2329 "[,smb=dir[,smbserver=addr]]\n"
2331 " configure a user mode network backend with ID 'str',\n"
2332 " its DHCP server and optional services\n"
2335 "-netdev tap,id=str,ifname=name\n"
2336 " configure a host TAP network backend with ID 'str'\n"
2338 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n"
2339 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n"
2340 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n"
2342 " configure a host TAP network backend with ID 'str'\n"
2343 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2344 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n"
2345 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n"
2346 " to deconfigure it\n"
2347 " use '[down]script=no' to disable script execution\n"
2348 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n"
2350 " use 'fd=h' to connect to an already opened TAP interface\n"
2351 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n"
2352 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n"
2353 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n"
2354 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n"
2355 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n"
2356 " use vhost=on to enable experimental in kernel accelerator\n"
2357 " (only has effect for virtio guests which use MSIX)\n"
2358 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n"
2359 " use 'vhostfd=h' to connect to an already opened vhost net device\n"
2360 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n"
2361 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n"
2362 " use 'poll-us=n' to speciy the maximum number of microseconds that could be\n"
2363 " spent on busy polling for vhost net\n"
2364 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n"
2365 " configure a host TAP network backend with ID 'str' that is\n"
2366 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2367 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n"
2370 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n"
2371 " [,rxsession=rxsession],txsession=txsession[,ipv6=on/off][,udp=on/off]\n"
2372 " [,cookie64=on/off][,counter][,pincounter][,txcookie=txcookie]\n"
2373 " [,rxcookie=rxcookie][,offset=offset]\n"
2374 " configure a network backend with ID 'str' connected to\n"
2375 " an Ethernet over L2TPv3 pseudowire.\n"
2376 " Linux kernel 3.3+ as well as most routers can talk\n"
2377 " L2TPv3. This transport allows connecting a VM to a VM,\n"
2378 " VM to a router and even VM to Host. It is a nearly-universal\n"
2379 " standard (RFC3931). Note - this implementation uses static\n"
2380 " pre-configured tunnels (same as the Linux kernel).\n"
2381 " use 'src=' to specify source address\n"
2382 " use 'dst=' to specify destination address\n"
2383 " use 'udp=on' to specify udp encapsulation\n"
2384 " use 'srcport=' to specify source udp port\n"
2385 " use 'dstport=' to specify destination udp port\n"
2386 " use 'ipv6=on' to force v6\n"
2387 " L2TPv3 uses cookies to prevent misconfiguration as\n"
2388 " well as a weak security measure\n"
2389 " use 'rxcookie=0x012345678' to specify a rxcookie\n"
2390 " use 'txcookie=0x012345678' to specify a txcookie\n"
2391 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n"
2392 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n"
2393 " use 'pincounter=on' to work around broken counter handling in peer\n"
2394 " use 'offset=X' to add an extra offset between header and data\n"
2396 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n"
2397 " configure a network backend to connect to another network\n"
2398 " using a socket connection\n"
2399 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n"
2400 " configure a network backend to connect to a multicast maddr and port\n"
2401 " use 'localaddr=addr' to specify the host address to send packets from\n"
2402 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n"
2403 " configure a network backend to connect to another network\n"
2404 " using an UDP tunnel\n"
2406 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n"
2407 " configure a network backend to connect to port 'n' of a vde switch\n"
2408 " running on host and listening for incoming connections on 'socketpath'.\n"
2409 " Use group 'groupname' and mode 'octalmode' to change default\n"
2410 " ownership and permissions for communication port.\n"
2412 #ifdef CONFIG_NETMAP
2413 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n"
2414 " attach to the existing netmap-enabled network interface 'name', or to a\n"
2415 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n"
2416 " netmap device, defaults to '/dev/netmap')\n"
2419 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n"
2420 " configure a vhost-user network, backed by a chardev 'dev'\n"
2422 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n"
2423 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL)
2424 DEF("nic", HAS_ARG, QEMU_OPTION_nic,
2435 #ifdef CONFIG_NETMAP
2441 "socket][,option][,...][mac=macaddr]\n"
2442 " initialize an on-board / default host NIC (using MAC address\n"
2443 " macaddr) and connect it to the given host network backend\n"
2444 "-nic none use it alone to have zero network devices (the default is to\n"
2445 " provided a 'user' network connection)\n",
2447 DEF("net", HAS_ARG, QEMU_OPTION_net,
2448 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n"
2449 " configure or create an on-board (or machine default) NIC and\n"
2450 " connect it to hub 0 (please use -nic unless you need a hub)\n"
2460 #ifdef CONFIG_NETMAP
2463 "socket][,option][,option][,...]\n"
2464 " old way to initialize a host network interface\n"
2465 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL)
2467 ``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]``
2468 This option is a shortcut for configuring both the on-board
2469 (default) guest NIC hardware and the host network backend in one go.
2470 The host backend options are the same as with the corresponding
2471 ``-netdev`` options below. The guest NIC model can be set with
2472 ``model=modelname``. Use ``model=help`` to list the available device
2473 types. The hardware MAC address can be set with ``mac=macaddr``.
2475 The following two example do exactly the same, to show how ``-nic``
2476 can be used to shorten the command line length:
2480 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32
2481 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32
2484 Indicate that no network devices should be configured. It is used to
2485 override the default configuration (default NIC with "user" host
2486 network backend) which is activated if no other networking options
2489 ``-netdev user,id=id[,option][,option][,...]``
2490 Configure user mode host network backend which requires no
2491 administrator privilege to run. Valid options are:
2494 Assign symbolic name for use in monitor commands.
2496 ``ipv4=on|off and ipv6=on|off``
2497 Specify that either IPv4 or IPv6 must be enabled. If neither is
2498 specified both protocols are enabled.
2501 Set IP network address the guest will see. Optionally specify
2502 the netmask, either in the form a.b.c.d or as number of valid
2503 top-most bits. Default is 10.0.2.0/24.
2506 Specify the guest-visible address of the host. Default is the
2507 2nd IP in the guest network, i.e. x.x.x.2.
2509 ``ipv6-net=addr[/int]``
2510 Set IPv6 network address the guest will see (default is
2511 fec0::/64). The network prefix is given in the usual hexadecimal
2512 IPv6 address notation. The prefix size is optional, and is given
2513 as the number of valid top-most bits (default is 64).
2516 Specify the guest-visible IPv6 address of the host. Default is
2517 the 2nd IPv6 in the guest network, i.e. xxxx::2.
2520 If this option is enabled, the guest will be isolated, i.e. it
2521 will not be able to contact the host and no guest IP packets
2522 will be routed over the host to the outside. This option does
2523 not affect any explicitly set forwarding rules.
2526 Specifies the client hostname reported by the built-in DHCP
2530 Specify the first of the 16 IPs the built-in DHCP server can
2531 assign. Default is the 15th to 31st IP in the guest network,
2532 i.e. x.x.x.15 to x.x.x.31.
2535 Specify the guest-visible address of the virtual nameserver. The
2536 address must be different from the host address. Default is the
2537 3rd IP in the guest network, i.e. x.x.x.3.
2540 Specify the guest-visible address of the IPv6 virtual
2541 nameserver. The address must be different from the host address.
2542 Default is the 3rd IP in the guest network, i.e. xxxx::3.
2544 ``dnssearch=domain``
2545 Provides an entry for the domain-search list sent by the
2546 built-in DHCP server. More than one domain suffix can be
2547 transmitted by specifying this option multiple times. If
2548 supported, this will cause the guest to automatically try to
2549 append the given domain suffix(es) in case a domain name can not
2556 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org
2558 ``domainname=domain``
2559 Specifies the client domain name reported by the built-in DHCP
2563 When using the user mode network stack, activate a built-in TFTP
2564 server. The files in dir will be exposed as the root of a TFTP
2565 server. The TFTP client on the guest must be configured in
2566 binary mode (use the command ``bin`` of the Unix TFTP client).
2568 ``tftp-server-name=name``
2569 In BOOTP reply, broadcast name as the "TFTP server name"
2570 (RFC2132 option 66). This can be used to advise the guest to
2571 load boot files or configurations from a different server than
2575 When using the user mode network stack, broadcast file as the
2576 BOOTP filename. In conjunction with ``tftp``, this can be used
2577 to network boot a guest from a local directory.
2579 Example (using pxelinux):
2583 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \
2584 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0
2586 ``smb=dir[,smbserver=addr]``
2587 When using the user mode network stack, activate a built-in SMB
2588 server so that Windows OSes can access to the host files in
2589 ``dir`` transparently. The IP address of the SMB server can be
2590 set to addr. By default the 4th IP in the guest network is used,
2593 In the guest Windows OS, the line:
2599 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows
2600 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows
2603 Then ``dir`` can be accessed in ``\\smbserver\qemu``.
2605 Note that a SAMBA server must be installed on the host OS.
2607 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport``
2608 Redirect incoming TCP or UDP connections to the host port
2609 hostport to the guest IP address guestaddr on guest port
2610 guestport. If guestaddr is not specified, its value is x.x.x.15
2611 (default first address given by the built-in DHCP server). By
2612 specifying hostaddr, the rule can be bound to a specific host
2613 interface. If no connection type is set, TCP is used. This
2614 option can be given multiple times.
2616 For example, to redirect host X11 connection from screen 1 to
2617 guest screen 0, use the following:
2622 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000
2623 # this host xterm should open in the guest X11 server
2626 To redirect telnet connections from host port 5555 to telnet
2627 port on the guest, use the following:
2632 |qemu_system| -nic user,hostfwd=tcp::5555-:23
2633 telnet localhost 5555
2635 Then when you use on the host ``telnet localhost 5555``, you
2636 connect to the guest telnet server.
2638 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command``
2639 Forward guest TCP connections to the IP address server on port
2640 port to the character device dev or to a program executed by
2641 cmd:command which gets spawned for each connection. This option
2642 can be given multiple times.
2644 You can either use a chardev directly and have that one used
2645 throughout QEMU's lifetime, like in the following example:
2649 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever
2650 # the guest accesses it
2651 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321
2653 Or you can execute a command on every TCP connection established
2654 by the guest, so that QEMU behaves similar to an inetd process
2655 for that virtual server:
2659 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234
2660 # and connect the TCP stream to its stdin/stdout
2661 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321'
2663 ``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]``
2664 Configure a host TAP network backend with ID id.
2666 Use the network script file to configure it and the network script
2667 dfile to deconfigure it. If name is not provided, the OS
2668 automatically provides one. The default network configure script is
2669 ``/etc/qemu-ifup`` and the default network deconfigure script is
2670 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to
2671 disable script execution.
2673 If running QEMU as an unprivileged user, use the network helper
2674 helper to configure the TAP interface and attach it to the bridge.
2675 The default network helper executable is
2676 ``/path/to/qemu-bridge-helper`` and the default bridge device is
2679 ``fd``\ =h can be used to specify the handle of an already opened
2686 #launch a QEMU instance with the default network script
2687 |qemu_system| linux.img -nic tap
2691 #launch a QEMU instance with two NICs, each one connected
2693 |qemu_system| linux.img \
2694 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \
2695 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1
2699 #launch a QEMU instance with the default network helper to
2700 #connect a TAP device to bridge br0
2701 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \
2702 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper"
2704 ``-netdev bridge,id=id[,br=bridge][,helper=helper]``
2705 Connect a host TAP network interface to a host bridge device.
2707 Use the network helper helper to configure the TAP interface and
2708 attach it to the bridge. The default network helper executable is
2709 ``/path/to/qemu-bridge-helper`` and the default bridge device is
2716 #launch a QEMU instance with the default network helper to
2717 #connect a TAP device to bridge br0
2718 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1
2722 #launch a QEMU instance with the default network helper to
2723 #connect a TAP device to bridge qemubr0
2724 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1
2726 ``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]``
2727 This host network backend can be used to connect the guest's network
2728 to another QEMU virtual machine using a TCP socket connection. If
2729 ``listen`` is specified, QEMU waits for incoming connections on port
2730 (host is optional). ``connect`` is used to connect to another QEMU
2731 instance using the ``listen`` option. ``fd``\ =h specifies an
2732 already opened TCP socket.
2738 # launch a first QEMU instance
2739 |qemu_system| linux.img \
2740 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \
2741 -netdev socket,id=n1,listen=:1234
2742 # connect the network of this instance to the network of the first instance
2743 |qemu_system| linux.img \
2744 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \
2745 -netdev socket,id=n2,connect=127.0.0.1:1234
2747 ``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]``
2748 Configure a socket host network backend to share the guest's network
2749 traffic with another QEMU virtual machines using a UDP multicast
2750 socket, effectively making a bus for every QEMU with same multicast
2751 address maddr and port. NOTES:
2753 1. Several QEMU can be running on different hosts and share same bus
2754 (assuming correct multicast setup for these hosts).
2756 2. mcast support is compatible with User Mode Linux (argument
2757 ``ethN=mcast``), see http://user-mode-linux.sf.net.
2759 3. Use ``fd=h`` to specify an already opened UDP multicast socket.
2765 # launch one QEMU instance
2766 |qemu_system| linux.img \
2767 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \
2768 -netdev socket,id=n1,mcast=230.0.0.1:1234
2769 # launch another QEMU instance on same "bus"
2770 |qemu_system| linux.img \
2771 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \
2772 -netdev socket,id=n2,mcast=230.0.0.1:1234
2773 # launch yet another QEMU instance on same "bus"
2774 |qemu_system| linux.img \
2775 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \
2776 -netdev socket,id=n3,mcast=230.0.0.1:1234
2778 Example (User Mode Linux compat.):
2782 # launch QEMU instance (note mcast address selected is UML's default)
2783 |qemu_system| linux.img \
2784 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \
2785 -netdev socket,id=n1,mcast=239.192.168.1:1102
2787 /path/to/linux ubd0=/path/to/root_fs eth0=mcast
2789 Example (send packets from host's 1.2.3.4):
2793 |qemu_system| linux.img \
2794 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \
2795 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4
2797 ``-netdev l2tpv3,id=id,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport],txsession=txsession[,rxsession=rxsession][,ipv6][,udp][,cookie64][,counter][,pincounter][,txcookie=txcookie][,rxcookie=rxcookie][,offset=offset]``
2798 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931)
2799 is a popular protocol to transport Ethernet (and other Layer 2) data
2800 frames between two systems. It is present in routers, firewalls and
2801 the Linux kernel (from version 3.3 onwards).
2803 This transport allows a VM to communicate to another VM, router or
2807 source address (mandatory)
2810 destination address (mandatory)
2813 select udp encapsulation (default is ip).
2819 destination udp port.
2822 force v6, otherwise defaults to v4.
2824 ``rxcookie=rxcookie``; \ ``txcookie=txcookie``
2825 Cookies are a weak form of security in the l2tpv3 specification.
2826 Their function is mostly to prevent misconfiguration. By default
2830 Set cookie size to 64 bit instead of the default 32
2833 Force a 'cut-down' L2TPv3 with no counter as in
2834 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00
2837 Work around broken counter handling in peer. This may also help
2838 on networks which have packet reorder.
2841 Add an extra offset between header and data
2843 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to
2844 the bridge br-lan on the remote Linux host 1.2.3.4:
2848 # Setup tunnel on linux host using raw ip as encapsulation
2850 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \
2851 encap udp udp_sport 16384 udp_dport 16384
2852 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \
2853 0xFFFFFFFF peer_session_id 0xFFFFFFFF
2854 ifconfig vmtunnel0 mtu 1500
2855 ifconfig vmtunnel0 up
2856 brctl addif br-lan vmtunnel0
2860 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter
2862 |qemu_system| linux.img -device e1000,netdev=n1 \
2863 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter
2865 ``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]``
2866 Configure VDE backend to connect to PORT n of a vde switch running
2867 on host and listening for incoming connections on socketpath. Use
2868 GROUP groupname and MODE octalmode to change default ownership and
2869 permissions for communication port. This option is only available if
2870 QEMU has been compiled with vde support enabled.
2877 vde_switch -F -sock /tmp/myswitch
2878 # launch QEMU instance
2879 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch
2881 ``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]``
2882 Establish a vhost-user netdev, backed by a chardev id. The chardev
2883 should be a unix domain socket backed one. The vhost-user uses a
2884 specifically defined protocol to pass vhost ioctl replacement
2885 messages to an application on the other end of the socket. On
2886 non-MSIX guests, the feature can be forced with vhostforce. Use
2887 'queues=n' to specify the number of queues to be created for
2888 multiqueue vhost-user.
2894 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
2895 -numa node,memdev=mem \
2896 -chardev socket,id=chr0,path=/path/to/socket \
2897 -netdev type=vhost-user,id=net0,chardev=chr0 \
2898 -device virtio-net-pci,netdev=net0
2900 ``-netdev hubport,id=id,hubid=hubid[,netdev=nd]``
2901 Create a hub port on the emulated hub with ID hubid.
2903 The hubport netdev lets you connect a NIC to a QEMU emulated hub
2904 instead of a single netdev. Alternatively, you can also connect the
2905 hubport to another netdev with ID nd by using the ``netdev=nd``
2908 ``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]``
2909 Legacy option to configure or create an on-board (or machine
2910 default) Network Interface Card(NIC) and connect it either to the
2911 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd.
2912 If model is omitted, then the default NIC model associated with the
2913 machine type is used. Note that the default NIC model may change in
2914 future QEMU releases, so it is highly recommended to always specify
2915 a model. Optionally, the MAC address can be changed to mac, the
2916 device address set to addr (PCI cards only), and a name can be
2917 assigned for use in monitor commands. Optionally, for PCI cards, you
2918 can specify the number v of MSI-X vectors that the card should have;
2919 this option currently only affects virtio cards; set v = 0 to
2920 disable MSI-X. If no ``-net`` option is specified, a single NIC is
2921 created. QEMU can emulate several different models of network card.
2922 Use ``-net nic,model=help`` for a list of available devices for your
2925 ``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]``
2926 Configure a host network backend (with the options corresponding to
2927 the same ``-netdev`` option) and connect it to the emulated hub 0
2928 (the default hub). Use name to specify the name of the hub port.
2933 DEFHEADING(Character device options:)
2935 DEF("chardev", HAS_ARG, QEMU_OPTION_chardev,
2937 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2938 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4][,ipv6][,nodelay][,reconnect=seconds]\n"
2939 " [,server][,nowait][,telnet][,websocket][,reconnect=seconds][,mux=on|off]\n"
2940 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n"
2941 "-chardev socket,id=id,path=path[,server][,nowait][,telnet][,websocket][,reconnect=seconds]\n"
2942 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n"
2943 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n"
2944 " [,localport=localport][,ipv4][,ipv6][,mux=on|off]\n"
2945 " [,logfile=PATH][,logappend=on|off]\n"
2946 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2947 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n"
2948 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2949 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n"
2950 "-chardev file,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2951 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2953 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2954 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2956 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2957 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n"
2959 #ifdef CONFIG_BRLAPI
2960 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2962 #if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \
2963 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__)
2964 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2965 "-chardev tty,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2967 #if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__)
2968 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2969 "-chardev parport,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
2971 #if defined(CONFIG_SPICE)
2972 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
2973 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
2979 The general form of a character device option is:
2981 ``-chardev backend,id=id[,mux=on|off][,options]``
2982 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``,
2983 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``,
2984 ``pty``, ``stdio``, ``braille``, ``tty``, ``parallel``, ``parport``,
2985 ``spicevmc``, ``spiceport``. The specific backend will determine the
2988 Use ``-chardev help`` to print all available chardev backend types.
2990 All devices must have an id, which can be any string up to 127
2991 characters long. It is used to uniquely identify this device in
2992 other command line directives.
2994 A character device may be used in multiplexing mode by multiple
2995 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is
2996 a "1:N" device, and here the "1" end is your specified chardev
2997 backend, and the "N" end is the various parts of QEMU that can talk
2998 to a chardev. If you create a chardev with ``id=myid`` and
2999 ``mux=on``, QEMU will create a multiplexer with your specified ID,
3000 and you can then configure multiple front ends to use that chardev
3001 ID for their input/output. Up to four different front ends can be
3002 connected to a single multiplexed chardev. (Without multiplexing
3003 enabled, a chardev can only be used by a single front end.) For
3004 instance you could use this to allow a single stdio chardev to be
3005 used by two serial ports and the QEMU monitor:
3009 -chardev stdio,mux=on,id=char0 \
3010 -mon chardev=char0,mode=readline \
3011 -serial chardev:char0 \
3012 -serial chardev:char0
3014 You can have more than one multiplexer in a system configuration;
3015 for instance you could have a TCP port multiplexed between UART 0
3016 and UART 1, and stdio multiplexed between the QEMU monitor and a
3021 -chardev stdio,mux=on,id=char0 \
3022 -mon chardev=char0,mode=readline \
3023 -parallel chardev:char0 \
3024 -chardev tcp,...,mux=on,id=char1 \
3025 -serial chardev:char1 \
3026 -serial chardev:char1
3028 When you're using a multiplexed character device, some escape
3029 sequences are interpreted in the input. See :ref:`mux_005fkeys`.
3031 Note that some other command line options may implicitly create
3032 multiplexed character backends; for instance ``-serial mon:stdio``
3033 creates a multiplexed stdio backend connected to the serial port and
3034 the QEMU monitor, and ``-nographic`` also multiplexes the console
3035 and the monitor to stdio.
3037 There is currently no support for multiplexing in the other
3038 direction (where a single QEMU front end takes input and output from
3041 Every backend supports the ``logfile`` option, which supplies the
3042 path to a file to record all data transmitted via the backend. The
3043 ``logappend`` option controls whether the log file will be truncated
3044 or appended to when opened.
3046 The available backends are:
3048 ``-chardev null,id=id``
3049 A void device. This device will not emit any data, and will drop any
3050 data it receives. The null backend does not take any options.
3052 ``-chardev socket,id=id[,TCP options or unix options][,server][,nowait][,telnet][,websocket][,reconnect=seconds][,tls-creds=id][,tls-authz=id]``
3053 Create a two-way stream socket, which can be either a TCP or a unix
3054 socket. A unix socket will be created if ``path`` is specified.
3055 Behaviour is undefined if TCP options are specified for a unix
3058 ``server`` specifies that the socket shall be a listening socket.
3060 ``nowait`` specifies that QEMU should not block waiting for a client
3061 to connect to a listening socket.
3063 ``telnet`` specifies that traffic on the socket should interpret
3064 telnet escape sequences.
3066 ``websocket`` specifies that the socket uses WebSocket protocol for
3069 ``reconnect`` sets the timeout for reconnecting on non-server
3070 sockets when the remote end goes away. qemu will delay this many
3071 seconds and then attempt to reconnect. Zero disables reconnecting,
3074 ``tls-creds`` requests enablement of the TLS protocol for
3075 encryption, and specifies the id of the TLS credentials to use for
3076 the handshake. The credentials must be previously created with the
3077 ``-object tls-creds`` argument.
3079 ``tls-auth`` provides the ID of the QAuthZ authorization object
3080 against which the client's x509 distinguished name will be
3081 validated. This object is only resolved at time of use, so can be
3082 deleted and recreated on the fly while the chardev server is active.
3083 If missing, it will default to denying access.
3085 TCP and unix socket options are given below:
3087 ``TCP options: port=port[,host=host][,to=to][,ipv4][,ipv6][,nodelay]``
3088 ``host`` for a listening socket specifies the local address to
3089 be bound. For a connecting socket species the remote host to
3090 connect to. ``host`` is optional for listening sockets. If not
3091 specified it defaults to ``0.0.0.0``.
3093 ``port`` for a listening socket specifies the local port to be
3094 bound. For a connecting socket specifies the port on the remote
3095 host to connect to. ``port`` can be given as either a port
3096 number or a service name. ``port`` is required.
3098 ``to`` is only relevant to listening sockets. If it is
3099 specified, and ``port`` cannot be bound, QEMU will attempt to
3100 bind to subsequent ports up to and including ``to`` until it
3101 succeeds. ``to`` must be specified as a port number.
3103 ``ipv4`` and ``ipv6`` specify that either IPv4 or IPv6 must be
3104 used. If neither is specified the socket may use either
3107 ``nodelay`` disables the Nagle algorithm.
3109 ``unix options: path=path[,abstract=on|off][,tight=on|off]``
3110 ``path`` specifies the local path of the unix socket. ``path``
3112 ``abstract`` specifies the use of the abstract socket namespace,
3113 rather than the filesystem. Optional, defaults to false.
3114 ``tight`` sets the socket length of abstract sockets to their minimum,
3115 rather than the full sun_path length. Optional, defaults to true.
3117 ``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4][,ipv6]``
3118 Sends all traffic from the guest to a remote host over UDP.
3120 ``host`` specifies the remote host to connect to. If not specified
3121 it defaults to ``localhost``.
3123 ``port`` specifies the port on the remote host to connect to.
3124 ``port`` is required.
3126 ``localaddr`` specifies the local address to bind to. If not
3127 specified it defaults to ``0.0.0.0``.
3129 ``localport`` specifies the local port to bind to. If not specified
3130 any available local port will be used.
3132 ``ipv4`` and ``ipv6`` specify that either IPv4 or IPv6 must be used.
3133 If neither is specified the device may use either protocol.
3135 ``-chardev msmouse,id=id``
3136 Forward QEMU's emulated msmouse events to the guest. ``msmouse``
3137 does not take any options.
3139 ``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]``
3140 Connect to a QEMU text console. ``vc`` may optionally be given a
3143 ``width`` and ``height`` specify the width and height respectively
3144 of the console, in pixels.
3146 ``cols`` and ``rows`` specify that the console be sized to fit a
3147 text console with the given dimensions.
3149 ``-chardev ringbuf,id=id[,size=size]``
3150 Create a ring buffer with fixed size ``size``. size must be a power
3151 of two and defaults to ``64K``.
3153 ``-chardev file,id=id,path=path``
3154 Log all traffic received from the guest to a file.
3156 ``path`` specifies the path of the file to be opened. This file will
3157 be created if it does not already exist, and overwritten if it does.
3158 ``path`` is required.
3160 ``-chardev pipe,id=id,path=path``
3161 Create a two-way connection to the guest. The behaviour differs
3162 slightly between Windows hosts and other hosts:
3164 On Windows, a single duplex pipe will be created at
3167 On other hosts, 2 pipes will be created called ``path.in`` and
3168 ``path.out``. Data written to ``path.in`` will be received by the
3169 guest. Data written by the guest can be read from ``path.out``. QEMU
3170 will not create these fifos, and requires them to be present.
3172 ``path`` forms part of the pipe path as described above. ``path`` is
3175 ``-chardev console,id=id``
3176 Send traffic from the guest to QEMU's standard output. ``console``
3177 does not take any options.
3179 ``console`` is only available on Windows hosts.
3181 ``-chardev serial,id=id,path=path``
3182 Send traffic from the guest to a serial device on the host.
3184 On Unix hosts serial will actually accept any tty device, not only
3187 ``path`` specifies the name of the serial device to open.
3189 ``-chardev pty,id=id``
3190 Create a new pseudo-terminal on the host and connect to it. ``pty``
3191 does not take any options.
3193 ``pty`` is not available on Windows hosts.
3195 ``-chardev stdio,id=id[,signal=on|off]``
3196 Connect to standard input and standard output of the QEMU process.
3198 ``signal`` controls if signals are enabled on the terminal, that
3199 includes exiting QEMU with the key sequence Control-c. This option
3200 is enabled by default, use ``signal=off`` to disable it.
3202 ``-chardev braille,id=id``
3203 Connect to a local BrlAPI server. ``braille`` does not take any
3206 ``-chardev tty,id=id,path=path``
3207 ``tty`` is only available on Linux, Sun, FreeBSD, NetBSD, OpenBSD
3208 and DragonFlyBSD hosts. It is an alias for ``serial``.
3210 ``path`` specifies the path to the tty. ``path`` is required.
3212 ``-chardev parallel,id=id,path=path``
3214 ``-chardev parport,id=id,path=path``
3215 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD
3218 Connect to a local parallel port.
3220 ``path`` specifies the path to the parallel port device. ``path`` is
3223 ``-chardev spicevmc,id=id,debug=debug,name=name``
3224 ``spicevmc`` is only available when spice support is built in.
3226 ``debug`` debug level for spicevmc
3228 ``name`` name of spice channel to connect to
3230 Connect to a spice virtual machine channel, such as vdiport.
3232 ``-chardev spiceport,id=id,debug=debug,name=name``
3233 ``spiceport`` is only available when spice support is built in.
3235 ``debug`` debug level for spicevmc
3237 ``name`` name of spice port to connect to
3239 Connect to a spice port, allowing a Spice client to handle the
3240 traffic identified by a name (preferably a fqdn).
3246 DEFHEADING(TPM device options:)
3248 DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \
3249 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n"
3250 " use path to provide path to a character device; default is /dev/tpm0\n"
3251 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n"
3252 " not provided it will be searched for in /sys/class/misc/tpm?/device\n"
3253 "-tpmdev emulator,id=id,chardev=dev\n"
3254 " configure the TPM device using chardev backend\n",
3257 The general form of a TPM device option is:
3259 ``-tpmdev backend,id=id[,options]``
3260 The specific backend type will determine the applicable options. The
3261 ``-tpmdev`` option creates the TPM backend and requires a
3262 ``-device`` option that specifies the TPM frontend interface model.
3264 Use ``-tpmdev help`` to print all available TPM backend types.
3266 The available backends are:
3268 ``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path``
3269 (Linux-host only) Enable access to the host's TPM using the
3272 ``path`` specifies the path to the host's TPM device, i.e., on a
3273 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by
3274 default ``/dev/tpm0`` is used.
3276 ``cancel-path`` specifies the path to the host TPM device's sysfs
3277 entry allowing for cancellation of an ongoing TPM command.
3278 ``cancel-path`` is optional and by default QEMU will search for the
3281 Some notes about using the host's TPM with the passthrough driver:
3283 The TPM device accessed by the passthrough driver must not be used
3284 by any other application on the host.
3286 Since the host's firmware (BIOS/UEFI) has already initialized the
3287 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize
3288 the TPM again and may therefore not show a TPM-specific menu that
3289 would otherwise allow the user to configure the TPM, e.g., allow the
3290 user to enable/disable or activate/deactivate the TPM. Further, if
3291 TPM ownership is released from within a VM then the host's TPM will
3292 get disabled and deactivated. To enable and activate the TPM again
3293 afterwards, the host has to be rebooted and the user is required to
3294 enter the firmware's menu to enable and activate the TPM. If the TPM
3295 is left disabled and/or deactivated most TPM commands will fail.
3297 To create a passthrough TPM use the following two options:
3301 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0
3303 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by
3304 ``tpmdev=tpm0`` in the device option.
3306 ``-tpmdev emulator,id=id,chardev=dev``
3307 (Linux-host only) Enable access to a TPM emulator using Unix domain
3308 socket based chardev backend.
3310 ``chardev`` specifies the unique ID of a character device backend
3311 that provides connection to the software TPM server.
3313 To create a TPM emulator backend device with chardev socket backend:
3317 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0
3324 DEFHEADING(Linux/Multiboot boot specific:)
3326 When using these options, you can use a given Linux or Multiboot kernel
3327 without installing it in the disk image. It can be useful for easier
3328 testing of various kernels.
3333 DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \
3334 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL)
3337 Use bzImage as kernel image. The kernel can be either a Linux kernel
3338 or in multiboot format.
3341 DEF("append", HAS_ARG, QEMU_OPTION_append, \
3342 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL)
3345 Use cmdline as kernel command line
3348 DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \
3349 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL)
3352 Use file as initial ram disk.
3354 ``-initrd "file1 arg=foo,file2"``
3355 This syntax is only available with multiboot.
3357 Use file1 and file2 as modules and pass arg=foo as parameter to the
3361 DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \
3362 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL)
3365 Use file as a device tree binary (dtb) image and pass it to the
3371 DEFHEADING(Debug/Expert options:)
3373 DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg,
3374 "-fw_cfg [name=]<name>,file=<file>\n"
3375 " add named fw_cfg entry with contents from file\n"
3376 "-fw_cfg [name=]<name>,string=<str>\n"
3377 " add named fw_cfg entry with contents from string\n",
3380 ``-fw_cfg [name=]name,file=file``
3381 Add named fw\_cfg entry with contents from file file.
3383 ``-fw_cfg [name=]name,string=str``
3384 Add named fw\_cfg entry with contents from string str.
3386 The terminating NUL character of the contents of str will not be
3387 included as part of the fw\_cfg item data. To insert contents with
3388 embedded NUL characters, you have to use the file parameter.
3390 The fw\_cfg entries are passed by QEMU through to the guest.
3396 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin
3398 creates an fw\_cfg entry named opt/com.mycompany/blob with contents
3399 from ./my\_blob.bin.
3402 DEF("serial", HAS_ARG, QEMU_OPTION_serial, \
3403 "-serial dev redirect the serial port to char device 'dev'\n",
3407 Redirect the virtual serial port to host character device dev. The
3408 default device is ``vc`` in graphical mode and ``stdio`` in non
3411 This option can be used several times to simulate up to 4 serial
3414 Use ``-serial none`` to disable all serial ports.
3416 Available character devices are:
3419 Virtual console. Optionally, a width and height can be given in
3426 It is also possible to specify width or height in characters:
3433 [Linux only] Pseudo TTY (a new PTY is automatically allocated)
3436 No device is allocated.
3442 Use a named character device defined with the ``-chardev``
3446 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial
3447 port parameters are set according to the emulated ones.
3450 [Linux only, parallel port only] Use host parallel port N.
3451 Currently SPP and EPP parallel port features can be used.
3454 Write output to filename. No character can be read.
3457 [Unix only] standard input/output
3463 [Windows only] Use host serial port n
3465 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]``
3466 This implements UDP Net Console. When remote\_host or src\_ip
3467 are not specified they default to ``0.0.0.0``. When not using a
3468 specified src\_port a random port is automatically chosen.
3470 If you just want a simple readonly console you can use
3471 ``netcat`` or ``nc``, by starting QEMU with:
3472 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time
3473 QEMU writes something to that port it will appear in the
3476 If you plan to send characters back via netconsole or you want
3477 to stop and start QEMU a lot of times, you should have QEMU use
3478 the same source port each time by using something like ``-serial
3479 udp::4555@:4556`` to QEMU. Another approach is to use a patched
3480 version of netcat which can listen to a TCP port and send and
3481 receive characters via udp. If you have a patched version of
3482 netcat which activates telnet remote echo and single char
3483 transfer, then you can use the following options to set up a
3484 netcat redirector to allow telnet on port 5555 to access the
3488 -serial udp::4555@:4556
3491 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T
3496 ``tcp:[host]:port[,server][,nowait][,nodelay][,reconnect=seconds]``
3497 The TCP Net Console has two modes of operation. It can send the
3498 serial I/O to a location or wait for a connection from a
3499 location. By default the TCP Net Console is sent to host at the
3500 port. If you use the server option QEMU will wait for a client
3501 socket application to connect to the port before continuing,
3502 unless the ``nowait`` option was specified. The ``nodelay``
3503 option disables the Nagle buffering algorithm. The ``reconnect``
3504 option only applies if noserver is set, if the connection goes
3505 down it will attempt to reconnect at the given interval. If host
3506 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a
3507 time is accepted. You can use ``telnet`` to connect to the
3508 corresponding character device.
3510 ``Example to send tcp console to 192.168.0.2 port 4444``
3511 -serial tcp:192.168.0.2:4444
3513 ``Example to listen and wait on port 4444 for connection``
3514 -serial tcp::4444,server
3516 ``Example to not wait and listen on ip 192.168.0.100 port 4444``
3517 -serial tcp:192.168.0.100:4444,server,nowait
3519 ``telnet:host:port[,server][,nowait][,nodelay]``
3520 The telnet protocol is used instead of raw tcp sockets. The
3521 options work the same as if you had specified ``-serial tcp``.
3522 The difference is that the port acts like a telnet server or
3523 client using telnet option negotiation. This will also allow you
3524 to send the MAGIC\_SYSRQ sequence if you use a telnet that
3525 supports sending the break sequence. Typically in unix telnet
3526 you do it with Control-] and then type "send break" followed by
3527 pressing the enter key.
3529 ``websocket:host:port,server[,nowait][,nodelay]``
3530 The WebSocket protocol is used instead of raw tcp socket. The
3531 port acts as a WebSocket server. Client mode is not supported.
3533 ``unix:path[,server][,nowait][,reconnect=seconds]``
3534 A unix domain socket is used instead of a tcp socket. The option
3535 works the same as if you had specified ``-serial tcp`` except
3536 the unix domain socket path is used for connections.
3539 This is a special option to allow the monitor to be multiplexed
3540 onto another serial port. The monitor is accessed with key
3541 sequence of Control-a and then pressing c. dev\_string should be
3542 any one of the serial devices specified above. An example to
3543 multiplex the monitor onto a telnet server listening on port
3546 ``-serial mon:telnet::4444,server,nowait``
3548 When the monitor is multiplexed to stdio in this way, Ctrl+C
3549 will not terminate QEMU any more but will be passed to the guest
3553 Braille device. This will use BrlAPI to display the braille
3554 output on a real or fake device.
3557 Three button serial mouse. Configure the guest to use Microsoft
3561 DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \
3562 "-parallel dev redirect the parallel port to char device 'dev'\n",
3566 Redirect the virtual parallel port to host device dev (same devices
3567 as the serial port). On Linux hosts, ``/dev/parportN`` can be used
3568 to use hardware devices connected on the corresponding host parallel
3571 This option can be used several times to simulate up to 3 parallel
3574 Use ``-parallel none`` to disable all parallel ports.
3577 DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \
3578 "-monitor dev redirect the monitor to char device 'dev'\n",
3582 Redirect the monitor to host device dev (same devices as the serial
3583 port). The default device is ``vc`` in graphical mode and ``stdio``
3584 in non graphical mode. Use ``-monitor none`` to disable the default
3587 DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \
3588 "-qmp dev like -monitor but opens in 'control' mode\n",
3592 Like -monitor but opens in 'control' mode.
3594 DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \
3595 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n",
3599 Like -qmp but uses pretty JSON formatting.
3602 DEF("mon", HAS_ARG, QEMU_OPTION_mon, \
3603 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL)
3605 ``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]``
3606 Setup monitor on chardev name. ``pretty`` turns on JSON pretty
3607 printing easing human reading and debugging.
3610 DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \
3611 "-debugcon dev redirect the debug console to char device 'dev'\n",
3615 Redirect the debug console to host device dev (same devices as the
3616 serial port). The debug console is an I/O port which is typically
3617 port 0xe9; writing to that I/O port sends output to this device. The
3618 default device is ``vc`` in graphical mode and ``stdio`` in non
3622 DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \
3623 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL)
3626 Store the QEMU process PID in file. It is useful if you launch QEMU
3630 DEF("singlestep", 0, QEMU_OPTION_singlestep, \
3631 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL)
3634 Run the emulation in single step mode.
3637 DEF("preconfig", 0, QEMU_OPTION_preconfig, \
3638 "--preconfig pause QEMU before machine is initialized (experimental)\n",
3642 Pause QEMU for interactive configuration before the machine is
3643 created, which allows querying and configuring properties that will
3644 affect machine initialization. Use QMP command 'x-exit-preconfig' to
3645 exit the preconfig state and move to the next state (i.e. run guest
3646 if -S isn't used or pause the second time if -S is used). This
3647 option is experimental.
3650 DEF("S", 0, QEMU_OPTION_S, \
3651 "-S freeze CPU at startup (use 'c' to start execution)\n",
3655 Do not start CPU at startup (you must type 'c' in the monitor).
3658 DEF("realtime", HAS_ARG, QEMU_OPTION_realtime,
3659 "-realtime [mlock=on|off]\n"
3660 " run qemu with realtime features\n"
3661 " mlock=on|off controls mlock support (default: on)\n",
3664 ``-realtime mlock=on|off``
3665 Run qemu with realtime features. mlocking qemu and guest memory can
3666 be enabled via ``mlock=on`` (enabled by default).
3669 DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit,
3670 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n"
3671 " run qemu with overcommit hints\n"
3672 " mem-lock=on|off controls memory lock support (default: off)\n"
3673 " cpu-pm=on|off controls cpu power management (default: off)\n",
3676 ``-overcommit mem-lock=on|off``
3678 ``-overcommit cpu-pm=on|off``
3679 Run qemu with hints about host resource overcommit. The default is
3680 to assume that host overcommits all resources.
3682 Locking qemu and guest memory can be enabled via ``mem-lock=on``
3683 (disabled by default). This works when host memory is not
3684 overcommitted and reduces the worst-case latency for guest. This is
3685 equivalent to ``realtime``.
3687 Guest ability to manage power state of host cpus (increasing latency
3688 for other processes on the same host cpu, but decreasing latency for
3689 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This
3690 works best when host CPU is not overcommitted. When used, host
3691 estimates of CPU cycle and power utilization will be incorrect, not
3692 taking into account guest idle time.
3695 DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \
3696 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n"
3697 " the guest without waiting for gdb to connect; use -S too\n"
3698 " if you want it to not start execution.)\n",
3702 Accept a gdb connection on device dev (see
3703 :ref:`gdb_005fusage`). Note that this option does not pause QEMU
3704 execution -- if you want QEMU to not start the guest until you
3705 connect with gdb and issue a ``continue`` command, you will need to
3706 also pass the ``-S`` option to QEMU.
3708 The most usual configuration is to listen on a local TCP socket::
3712 but you can specify other backends; UDP, pseudo TTY, or even stdio
3713 are all reasonable use cases. For example, a stdio connection
3714 allows you to start QEMU from within gdb and establish the
3715 connection via a pipe:
3719 (gdb) target remote | exec |qemu_system| -gdb stdio ...
3722 DEF("s", 0, QEMU_OPTION_s, \
3723 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n",
3727 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234
3728 (see :ref:`gdb_005fusage`).
3731 DEF("d", HAS_ARG, QEMU_OPTION_d, \
3732 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n",
3736 Enable logging of specified items. Use '-d help' for a list of log
3740 DEF("D", HAS_ARG, QEMU_OPTION_D, \
3741 "-D logfile output log to logfile (default stderr)\n",
3745 Output log in logfile instead of to stderr
3748 DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \
3749 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n",
3752 ``-dfilter range1[,...]``
3753 Filter debug output to that relevant to a range of target addresses.
3754 The filter spec can be either start+size, start-size or start..end
3755 where start end and size are the addresses and sizes required. For
3760 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000
3762 Will dump output for any code in the 0x1000 sized block starting at
3763 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and
3764 another 0x1000 sized block starting at 0xffffffc00005f000.
3767 DEF("seed", HAS_ARG, QEMU_OPTION_seed, \
3768 "-seed number seed the pseudo-random number generator\n",
3772 Force the guest to use a deterministic pseudo-random number
3773 generator, seeded with number. This does not affect crypto routines
3777 DEF("L", HAS_ARG, QEMU_OPTION_L, \
3778 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n",
3782 Set the directory for the BIOS, VGA BIOS and keymaps.
3784 To list all the data directories, use ``-L help``.
3787 DEF("bios", HAS_ARG, QEMU_OPTION_bios, \
3788 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL)
3791 Set the filename for the BIOS.
3794 DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \
3795 "-enable-kvm enable KVM full virtualization support\n", QEMU_ARCH_ALL)
3798 Enable KVM full virtualization support. This option is only
3799 available if KVM support is enabled when compiling.
3802 DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid,
3803 "-xen-domid id specify xen guest domain id\n", QEMU_ARCH_ALL)
3804 DEF("xen-attach", 0, QEMU_OPTION_xen_attach,
3805 "-xen-attach attach to existing xen domain\n"
3806 " libxl will use this when starting QEMU\n",
3808 DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict,
3809 "-xen-domid-restrict restrict set of available xen operations\n"
3810 " to specified domain id. (Does not affect\n"
3811 " xenpv machine type).\n",
3815 Specify xen guest domain id (XEN only).
3818 Attach to existing xen domain. libxl will use this when starting
3819 QEMU (XEN only). Restrict set of available xen operations to
3820 specified domain id (XEN only).
3823 DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \
3824 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL)
3827 Exit instead of rebooting.
3830 DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \
3831 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL)
3834 Don't exit QEMU on guest shutdown, but instead only stop the
3835 emulation. This allows for instance switching to monitor to commit
3836 changes to the disk image.
3839 DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \
3840 "-loadvm [tag|id]\n" \
3841 " start right away with a saved state (loadvm in monitor)\n",
3845 Start right away with a saved state (``loadvm`` in monitor)
3849 DEF("daemonize", 0, QEMU_OPTION_daemonize, \
3850 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL)
3854 Daemonize the QEMU process after initialization. QEMU will not
3855 detach from standard IO until it is ready to receive connections on
3856 any of its devices. This option is a useful way for external
3857 programs to launch QEMU without having to cope with initialization
3861 DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \
3862 "-option-rom rom load a file, rom, into the option ROM space\n",
3865 ``-option-rom file``
3866 Load the contents of file as an option ROM. This option is useful to
3867 load things like EtherBoot.
3870 DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \
3871 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \
3872 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n",
3876 ``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]``
3877 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at
3878 the current UTC or local time, respectively. ``localtime`` is
3879 required for correct date in MS-DOS or Windows. To start at a
3880 specific point in time, provide datetime in the format
3881 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC.
3883 By default the RTC is driven by the host system time. This allows
3884 using of the RTC as accurate reference clock inside the guest,
3885 specifically if the host time is smoothly following an accurate
3886 external reference clock, e.g. via NTP. If you want to isolate the
3887 guest time from the host, you can set ``clock`` to ``rt`` instead,
3888 which provides a host monotonic clock if host support it. To even
3889 prevent the RTC from progressing during suspension, you can set
3890 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is
3891 recommended especially in icount mode in order to preserve
3892 determinism; however, note that in icount mode the speed of the
3893 virtual clock is variable and can in general differ from the host
3896 Enable ``driftfix`` (i386 targets only) if you experience time drift
3897 problems, specifically with Windows' ACPI HAL. This option will try
3898 to figure out how many timer interrupts were not processed by the
3899 Windows guest and will re-inject them.
3902 DEF("icount", HAS_ARG, QEMU_OPTION_icount, \
3903 "-icount [shift=N|auto][,align=on|off][,sleep=on|off,rr=record|replay,rrfile=<filename>,rrsnapshot=<snapshot>]\n" \
3904 " enable virtual instruction counter with 2^N clock ticks per\n" \
3905 " instruction, enable aligning the host and virtual clocks\n" \
3906 " or disable real time cpu sleeping\n", QEMU_ARCH_ALL)
3908 ``-icount [shift=N|auto][,rr=record|replay,rrfile=filename,rrsnapshot=snapshot]``
3909 Enable virtual instruction counter. The virtual cpu will execute one
3910 instruction every 2^N ns of virtual time. If ``auto`` is specified
3911 then the virtual cpu speed will be automatically adjusted to keep
3912 virtual time within a few seconds of real time.
3914 When the virtual cpu is sleeping, the virtual time will advance at
3915 default speed unless ``sleep=on|off`` is specified. With
3916 ``sleep=on|off``, the virtual time will jump to the next timer
3917 deadline instantly whenever the virtual cpu goes to sleep mode and
3918 will not advance if no timer is enabled. This behavior give
3919 deterministic execution times from the guest point of view.
3921 Note that while this option can give deterministic behavior, it does
3922 not provide cycle accurate emulation. Modern CPUs contain
3923 superscalar out of order cores with complex cache hierarchies. The
3924 number of instructions executed often has little or no correlation
3925 with actual performance.
3927 ``align=on`` will activate the delay algorithm which will try to
3928 synchronise the host clock and the virtual clock. The goal is to
3929 have a guest running at the real frequency imposed by the shift
3930 option. Whenever the guest clock is behind the host clock and if
3931 ``align=on`` is specified then we print a message to the user to
3932 inform about the delay. Currently this option does not work when
3933 ``shift`` is ``auto``. Note: The sync algorithm will work for those
3934 shift values for which the guest clock runs ahead of the host clock.
3935 Typically this happens when the shift value is high (how high
3936 depends on the host machine).
3938 When ``rr`` option is specified deterministic record/replay is
3939 enabled. Replay log is written into filename file in record mode and
3940 read from this file in replay mode.
3942 Option rrsnapshot is used to create new vm snapshot named snapshot
3943 at the start of execution recording. In replay mode this option is
3944 used to load the initial VM state.
3947 DEF("watchdog", HAS_ARG, QEMU_OPTION_watchdog, \
3948 "-watchdog model\n" \
3949 " enable virtual hardware watchdog [default=none]\n",
3953 Create a virtual hardware watchdog device. Once enabled (by a guest
3954 action), the watchdog must be periodically polled by an agent inside
3955 the guest or else the guest will be restarted. Choose a model for
3956 which your guest has drivers.
3958 The model is the model of hardware watchdog to emulate. Use
3959 ``-watchdog help`` to list available hardware models. Only one
3960 watchdog can be enabled for a guest.
3962 The following models may be available:
3965 iBASE 700 is a very simple ISA watchdog with a single timer.
3968 Intel 6300ESB I/O controller hub is a much more featureful
3969 PCI-based dual-timer watchdog.
3972 A virtual watchdog for s390x backed by the diagnose 288
3973 hypercall (currently KVM only).
3976 DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \
3977 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \
3978 " action when watchdog fires [default=reset]\n",
3981 ``-watchdog-action action``
3982 The action controls what QEMU will do when the watchdog timer
3983 expires. The default is ``reset`` (forcefully reset the guest).
3984 Other possible actions are: ``shutdown`` (attempt to gracefully
3985 shutdown the guest), ``poweroff`` (forcefully poweroff the guest),
3986 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the
3987 guest), ``debug`` (print a debug message and continue), or ``none``
3990 Note that the ``shutdown`` action requires that the guest responds
3991 to ACPI signals, which it may not be able to do in the sort of
3992 situations where the watchdog would have expired, and thus
3993 ``-watchdog-action shutdown`` is not recommended for production use.
3997 ``-watchdog i6300esb -watchdog-action pause``; \ ``-watchdog ib700``
4001 DEF("echr", HAS_ARG, QEMU_OPTION_echr, \
4002 "-echr chr set terminal escape character instead of ctrl-a\n",
4005 ``-echr numeric_ascii_value``
4006 Change the escape character used for switching to the monitor when
4007 using monitor and serial sharing. The default is ``0x01`` when using
4008 the ``-nographic`` option. ``0x01`` is equal to pressing
4009 ``Control-a``. You can select a different character from the ascii
4010 control keys where 1 through 26 map to Control-a through Control-z.
4011 For instance you could use the either of the following to change the
4012 escape character to Control-t.
4014 ``-echr 0x14``; \ ``-echr 20``
4018 DEF("show-cursor", 0, QEMU_OPTION_show_cursor, \
4019 "-show-cursor show cursor\n", QEMU_ARCH_ALL)
4025 DEF("tb-size", HAS_ARG, QEMU_OPTION_tb_size, \
4026 "-tb-size n set TB size\n", QEMU_ARCH_ALL)
4029 Set TCG translation block cache size. Deprecated, use
4030 '\ ``-accel tcg,tb-size=n``\ ' instead.
4033 DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \
4034 "-incoming tcp:[host]:port[,to=maxport][,ipv4][,ipv6]\n" \
4035 "-incoming rdma:host:port[,ipv4][,ipv6]\n" \
4036 "-incoming unix:socketpath\n" \
4037 " prepare for incoming migration, listen on\n" \
4038 " specified protocol and socket address\n" \
4039 "-incoming fd:fd\n" \
4040 "-incoming exec:cmdline\n" \
4041 " accept incoming migration on given file descriptor\n" \
4042 " or from given external command\n" \
4043 "-incoming defer\n" \
4044 " wait for the URI to be specified via migrate_incoming\n",
4047 ``-incoming tcp:[host]:port[,to=maxport][,ipv4][,ipv6]``
4049 ``-incoming rdma:host:port[,ipv4][,ipv6]``
4050 Prepare for incoming migration, listen on a given tcp port.
4052 ``-incoming unix:socketpath``
4053 Prepare for incoming migration, listen on a given unix socket.
4056 Accept incoming migration from a given filedescriptor.
4058 ``-incoming exec:cmdline``
4059 Accept incoming migration as an output from specified external
4063 Wait for the URI to be specified via migrate\_incoming. The monitor
4064 can be used to change settings (such as migration parameters) prior
4065 to issuing the migrate\_incoming to allow the migration to begin.
4068 DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \
4069 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL)
4071 ``-only-migratable``
4072 Only allow migratable devices. Devices will not be allowed to enter
4073 an unmigratable state.
4076 DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \
4077 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL)
4080 Don't create default devices. Normally, QEMU sets the default
4081 devices like serial port, parallel port, virtual console, monitor
4082 device, VGA adapter, floppy and CD-ROM drive and others. The
4083 ``-nodefaults`` option will disable all those default devices.
4087 DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \
4088 "-chroot dir chroot to dir just before starting the VM\n",
4093 Immediately before starting guest execution, chroot to the specified
4094 directory. Especially useful in combination with -runas.
4098 DEF("runas", HAS_ARG, QEMU_OPTION_runas, \
4099 "-runas user change to user id user just before starting the VM\n" \
4100 " user can be numeric uid:gid instead\n",
4105 Immediately before starting guest execution, drop root privileges,
4106 switching to the specified user.
4109 DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env,
4110 "-prom-env variable=value\n"
4111 " set OpenBIOS nvram variables\n",
4112 QEMU_ARCH_PPC | QEMU_ARCH_SPARC)
4114 ``-prom-env variable=value``
4115 Set OpenBIOS nvram variable to given value (PPC, SPARC only).
4119 qemu-system-sparc -prom-env 'auto-boot?=false' \
4120 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single'
4124 qemu-system-ppc -prom-env 'auto-boot?=false' \
4125 -prom-env 'boot-device=hd:2,\yaboot' \
4126 -prom-env 'boot-args=conf=hd:2,\yaboot.conf'
4128 DEF("semihosting", 0, QEMU_OPTION_semihosting,
4129 "-semihosting semihosting mode\n",
4130 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | QEMU_ARCH_LM32 |
4131 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2)
4134 Enable semihosting mode (ARM, M68K, Xtensa, MIPS, Nios II only).
4136 Note that this allows guest direct access to the host filesystem, so
4137 should only be used with a trusted guest OS.
4139 See the -semihosting-config option documentation for further
4140 information about the facilities this enables.
4142 DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config,
4143 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]\n" \
4144 " semihosting configuration\n",
4145 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | QEMU_ARCH_LM32 |
4146 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2)
4148 ``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]``
4149 Enable and configure semihosting (ARM, M68K, Xtensa, MIPS, Nios II
4152 Note that this allows guest direct access to the host filesystem, so
4153 should only be used with a trusted guest OS.
4155 On Arm this implements the standard semihosting API, version 2.0.
4157 On M68K this implements the "ColdFire GDB" interface used by
4160 Xtensa semihosting provides basic file IO calls, such as
4161 open/read/write/seek/select. Tensilica baremetal libc for ISS and
4162 linux platform "sim" use this interface.
4164 ``target=native|gdb|auto``
4165 Defines where the semihosting calls will be addressed, to QEMU
4166 (``native``) or to GDB (``gdb``). The default is ``auto``, which
4167 means ``gdb`` during debug sessions and ``native`` otherwise.
4170 Send the output to a chardev backend output for native or auto
4171 output when not in gdb
4173 ``arg=str1,arg=str2,...``
4174 Allows the user to pass input arguments, and can be used
4175 multiple times to build up a list. The old-style
4176 ``-kernel``/``-append`` method of passing a command line is
4177 still supported for backward compatibility. If both the
4178 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are
4179 specified, the former is passed to semihosting as it always
4182 DEF("old-param", 0, QEMU_OPTION_old_param,
4183 "-old-param old param mode\n", QEMU_ARCH_ARM)
4186 Old param mode (ARM only).
4189 DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \
4190 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \
4191 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \
4192 " Enable seccomp mode 2 system call filter (default 'off').\n" \
4193 " use 'obsolete' to allow obsolete system calls that are provided\n" \
4194 " by the kernel, but typically no longer used by modern\n" \
4195 " C library implementations.\n" \
4196 " use 'elevateprivileges' to allow or deny QEMU process to elevate\n" \
4197 " its privileges by blacklisting all set*uid|gid system calls.\n" \
4198 " The value 'children' will deny set*uid|gid system calls for\n" \
4199 " main QEMU process but will allow forks and execves to run unprivileged\n" \
4200 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \
4201 " blacklisting *fork and execve\n" \
4202 " use 'resourcecontrol' to disable process affinity and schedular priority\n",
4205 ``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]``
4206 Enable Seccomp mode 2 system call filter. 'on' will enable syscall
4207 filtering and 'off' will disable it. The default is 'off'.
4210 Enable Obsolete system calls
4212 ``elevateprivileges=string``
4213 Disable set\*uid\|gid system calls
4216 Disable \*fork and execve
4218 ``resourcecontrol=string``
4219 Disable process affinity and schedular priority
4222 DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig,
4223 "-readconfig <file>\n", QEMU_ARCH_ALL)
4225 ``-readconfig file``
4226 Read device configuration from file. This approach is useful when
4227 you want to spawn QEMU process with many command line options but
4228 you don't want to exceed the command line character limit.
4230 DEF("writeconfig", HAS_ARG, QEMU_OPTION_writeconfig,
4231 "-writeconfig <file>\n"
4232 " read/write config file\n", QEMU_ARCH_ALL)
4234 ``-writeconfig file``
4235 Write device configuration to file. The file can be either filename
4236 to save command line and device configuration into file or dash
4237 ``-``) character to print the output to stdout. This can be later
4238 used as input file for ``-readconfig`` option.
4241 DEF("no-user-config", 0, QEMU_OPTION_nouserconfig,
4243 " do not load default user-provided config files at startup\n",
4247 The ``-no-user-config`` option makes QEMU not load any of the
4248 user-provided config files on sysconfdir.
4251 DEF("trace", HAS_ARG, QEMU_OPTION_trace,
4252 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n"
4253 " specify tracing options\n",
4256 ``-trace [[enable=]pattern][,events=file][,file=file]``
4257 .. include:: ../qemu-option-trace.rst.inc
4260 DEF("plugin", HAS_ARG, QEMU_OPTION_plugin,
4261 "-plugin [file=]<file>[,arg=<string>]\n"
4265 ``-plugin file=file[,arg=string]``
4269 Load the given plugin from a shared library file.
4272 Argument string passed to the plugin. (Can be given multiple
4277 DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL)
4278 DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL)
4281 DEF("enable-fips", 0, QEMU_OPTION_enablefips,
4282 "-enable-fips enable FIPS 140-2 compliance\n",
4287 Enable FIPS 140-2 compliance mode.
4290 HXCOMM Deprecated by -accel tcg
4291 DEF("no-kvm", 0, QEMU_OPTION_no_kvm, "", QEMU_ARCH_I386)
4293 DEF("msg", HAS_ARG, QEMU_OPTION_msg,
4294 "-msg timestamp[=on|off]\n"
4295 " control error message format\n"
4296 " timestamp=on enables timestamps (default: off)\n",
4299 ``-msg timestamp[=on|off]``
4300 Control error message format.
4302 ``timestamp=on|off``
4303 Prefix messages with a timestamp. Default is off.
4306 DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate,
4307 "-dump-vmstate <file>\n"
4308 " Output vmstate information in JSON format to file.\n"
4309 " Use the scripts/vmstate-static-checker.py file to\n"
4310 " check for possible regressions in migration code\n"
4311 " by comparing two such vmstate dumps.\n",
4314 ``-dump-vmstate file``
4315 Dump json-encoded vmstate information for current machine type to
4319 DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile,
4320 "-enable-sync-profile\n"
4321 " enable synchronization profiling\n",
4324 ``-enable-sync-profile``
4325 Enable synchronization profiling.
4330 DEFHEADING(Generic object creation:)
4332 DEF("object", HAS_ARG, QEMU_OPTION_object,
4333 "-object TYPENAME[,PROP1=VALUE1,...]\n"
4334 " create a new object of type TYPENAME setting properties\n"
4335 " in the order they are specified. Note that the 'id'\n"
4336 " property must be set. These objects are placed in the\n"
4337 " '/objects' path.\n",
4340 ``-object typename[,prop1=value1,...]``
4341 Create a new object of type typename setting properties in the order
4342 they are specified. Note that the 'id' property must be set. These
4343 objects are placed in the '/objects' path.
4345 ``-object memory-backend-file,id=id,size=size,mem-path=dir,share=on|off,discard-data=on|off,merge=on|off,dump=on|off,prealloc=on|off,host-nodes=host-nodes,policy=default|preferred|bind|interleave,align=align``
4346 Creates a memory file backend object, which can be used to back
4347 the guest RAM with huge pages.
4349 The ``id`` parameter is a unique ID that will be used to
4350 reference this memory region when configuring the ``-numa``
4353 The ``size`` option provides the size of the memory region, and
4354 accepts common suffixes, eg ``500M``.
4356 The ``mem-path`` provides the path to either a shared memory or
4357 huge page filesystem mount.
4359 The ``share`` boolean option determines whether the memory
4360 region is marked as private to QEMU, or shared. The latter
4361 allows a co-operating external process to access the QEMU memory
4364 The ``share`` is also required for pvrdma devices due to
4365 limitations in the RDMA API provided by Linux.
4367 Setting share=on might affect the ability to configure NUMA
4368 bindings for the memory backend under some circumstances, see
4369 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel
4370 source tree for additional details.
4372 Setting the ``discard-data`` boolean option to on indicates that
4373 file contents can be destroyed when QEMU exits, to avoid
4374 unnecessarily flushing data to the backing file. Note that
4375 ``discard-data`` is only an optimization, and QEMU might not
4376 discard file contents if it aborts unexpectedly or is terminated
4379 The ``merge`` boolean option enables memory merge, also known as
4380 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider
4381 the pages for memory deduplication.
4383 Setting the ``dump`` boolean option to off excludes the memory
4384 from core dumps. This feature is also known as MADV\_DONTDUMP.
4386 The ``prealloc`` boolean option enables memory preallocation.
4388 The ``host-nodes`` option binds the memory range to a list of
4391 The ``policy`` option sets the NUMA policy to one of the
4398 prefer the given host node list for allocation
4401 restrict memory allocation to the given host node list
4404 interleave memory allocations across the given host node
4407 The ``align`` option specifies the base address alignment when
4408 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg
4409 ``2M``. Some backend store specified by ``mem-path`` requires an
4410 alignment different than the default one used by QEMU, eg the
4411 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
4412 such cases, users can specify the required alignment via this
4415 The ``pmem`` option specifies whether the backing file specified
4416 by ``mem-path`` is in host persistent memory that can be
4417 accessed using the SNIA NVM programming model (e.g. Intel
4418 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary
4419 operations to guarantee the persistence of its own writes to
4420 ``mem-path`` (e.g. in vNVDIMM label emulation and live
4421 migration). Also, we will map the backend-file with MAP\_SYNC
4422 flag, which ensures the file metadata is in sync for
4423 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC
4424 requires support from both the host kernel (since Linux kernel
4425 4.15) and the filesystem of ``mem-path`` mounted with DAX
4428 ``-object memory-backend-ram,id=id,merge=on|off,dump=on|off,share=on|off,prealloc=on|off,size=size,host-nodes=host-nodes,policy=default|preferred|bind|interleave``
4429 Creates a memory backend object, which can be used to back the
4430 guest RAM. Memory backend objects offer more control than the
4431 ``-m`` option that is traditionally used to define guest RAM.
4432 Please refer to ``memory-backend-file`` for a description of the
4435 ``-object memory-backend-memfd,id=id,merge=on|off,dump=on|off,share=on|off,prealloc=on|off,size=size,host-nodes=host-nodes,policy=default|preferred|bind|interleave,seal=on|off,hugetlb=on|off,hugetlbsize=size``
4436 Creates an anonymous memory file backend object, which allows
4437 QEMU to share the memory with an external process (e.g. when
4438 using vhost-user). The memory is allocated with memfd and
4439 optional sealing. (Linux only)
4441 The ``seal`` option creates a sealed-file, that will block
4442 further resizing the memory ('on' by default).
4444 The ``hugetlb`` option specify the file to be created resides in
4445 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction
4446 with the ``hugetlb`` option, the ``hugetlbsize`` option specify
4447 the hugetlb page size on systems that support multiple hugetlb
4448 page sizes (it must be a power of 2 value supported by the
4451 In some versions of Linux, the ``hugetlb`` option is
4452 incompatible with the ``seal`` option (requires at least Linux
4455 Please refer to ``memory-backend-file`` for a description of the
4458 The ``share`` boolean option is on by default with memfd.
4460 ``-object rng-builtin,id=id``
4461 Creates a random number generator backend which obtains entropy
4462 from QEMU builtin functions. The ``id`` parameter is a unique ID
4463 that will be used to reference this entropy backend from the
4464 ``virtio-rng`` device. By default, the ``virtio-rng`` device
4465 uses this RNG backend.
4467 ``-object rng-random,id=id,filename=/dev/random``
4468 Creates a random number generator backend which obtains entropy
4469 from a device on the host. The ``id`` parameter is a unique ID
4470 that will be used to reference this entropy backend from the
4471 ``virtio-rng`` device. The ``filename`` parameter specifies
4472 which file to obtain entropy from and if omitted defaults to
4475 ``-object rng-egd,id=id,chardev=chardevid``
4476 Creates a random number generator backend which obtains entropy
4477 from an external daemon running on the host. The ``id``
4478 parameter is a unique ID that will be used to reference this
4479 entropy backend from the ``virtio-rng`` device. The ``chardev``
4480 parameter is the unique ID of a character device backend that
4481 provides the connection to the RNG daemon.
4483 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off``
4484 Creates a TLS anonymous credentials object, which can be used to
4485 provide TLS support on network backends. The ``id`` parameter is
4486 a unique ID which network backends will use to access the
4487 credentials. The ``endpoint`` is either ``server`` or ``client``
4488 depending on whether the QEMU network backend that uses the
4489 credentials will be acting as a client or as a server. If
4490 ``verify-peer`` is enabled (the default) then once the handshake
4491 is completed, the peer credentials will be verified, though this
4492 is a no-op for anonymous credentials.
4494 The dir parameter tells QEMU where to find the credential files.
4495 For server endpoints, this directory may contain a file
4496 dh-params.pem providing diffie-hellman parameters to use for the
4497 TLS server. If the file is missing, QEMU will generate a set of
4498 DH parameters at startup. This is a computationally expensive
4499 operation that consumes random pool entropy, so it is
4500 recommended that a persistent set of parameters be generated
4503 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]``
4504 Creates a TLS Pre-Shared Keys (PSK) credentials object, which
4505 can be used to provide TLS support on network backends. The
4506 ``id`` parameter is a unique ID which network backends will use
4507 to access the credentials. The ``endpoint`` is either ``server``
4508 or ``client`` depending on whether the QEMU network backend that
4509 uses the credentials will be acting as a client or as a server.
4510 For clients only, ``username`` is the username which will be
4511 sent to the server. If omitted it defaults to "qemu".
4513 The dir parameter tells QEMU where to find the keys file. It is
4514 called "dir/keys.psk" and contains "username:key" pairs. This
4515 file can most easily be created using the GnuTLS ``psktool``
4518 For server endpoints, dir may also contain a file dh-params.pem
4519 providing diffie-hellman parameters to use for the TLS server.
4520 If the file is missing, QEMU will generate a set of DH
4521 parameters at startup. This is a computationally expensive
4522 operation that consumes random pool entropy, so it is
4523 recommended that a persistent set of parameters be generated up
4526 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id``
4527 Creates a TLS anonymous credentials object, which can be used to
4528 provide TLS support on network backends. The ``id`` parameter is
4529 a unique ID which network backends will use to access the
4530 credentials. The ``endpoint`` is either ``server`` or ``client``
4531 depending on whether the QEMU network backend that uses the
4532 credentials will be acting as a client or as a server. If
4533 ``verify-peer`` is enabled (the default) then once the handshake
4534 is completed, the peer credentials will be verified. With x509
4535 certificates, this implies that the clients must be provided
4536 with valid client certificates too.
4538 The dir parameter tells QEMU where to find the credential files.
4539 For server endpoints, this directory may contain a file
4540 dh-params.pem providing diffie-hellman parameters to use for the
4541 TLS server. If the file is missing, QEMU will generate a set of
4542 DH parameters at startup. This is a computationally expensive
4543 operation that consumes random pool entropy, so it is
4544 recommended that a persistent set of parameters be generated
4547 For x509 certificate credentials the directory will contain
4548 further files providing the x509 certificates. The certificates
4549 must be stored in PEM format, in filenames ca-cert.pem,
4550 ca-crl.pem (optional), server-cert.pem (only servers),
4551 server-key.pem (only servers), client-cert.pem (only clients),
4552 and client-key.pem (only clients).
4554 For the server-key.pem and client-key.pem files which contain
4555 sensitive private keys, it is possible to use an encrypted
4556 version by providing the passwordid parameter. This provides the
4557 ID of a previously created ``secret`` object containing the
4558 password for decryption.
4560 The priority parameter allows to override the global default
4561 priority used by gnutls. This can be useful if the system
4562 administrator needs to use a weaker set of crypto priorities for
4563 QEMU without potentially forcing the weakness onto all
4564 applications. Or conversely if one wants wants a stronger
4565 default for QEMU than for all other applications, they can do
4566 this through this parameter. Its format is a gnutls priority
4567 string as described at
4568 https://gnutls.org/manual/html_node/Priority-Strings.html.
4570 ``-object tls-cipher-suites,id=id,priority=priority``
4571 Creates a TLS cipher suites object, which can be used to control
4572 the TLS cipher/protocol algorithms that applications are permitted
4575 The ``id`` parameter is a unique ID which frontends will use to
4576 access the ordered list of permitted TLS cipher suites from the
4579 The ``priority`` parameter allows to override the global default
4580 priority used by gnutls. This can be useful if the system
4581 administrator needs to use a weaker set of crypto priorities for
4582 QEMU without potentially forcing the weakness onto all
4583 applications. Or conversely if one wants wants a stronger
4584 default for QEMU than for all other applications, they can do
4585 this through this parameter. Its format is a gnutls priority
4586 string as described at
4587 https://gnutls.org/manual/html_node/Priority-Strings.html.
4589 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]``
4590 Interval t can't be 0, this filter batches the packet delivery:
4591 all packets arriving in a given interval on netdev netdevid are
4592 delayed until the end of the interval. Interval is in
4593 microseconds. ``status`` is optional that indicate whether the
4594 netfilter is on (enabled) or off (disabled), the default status
4595 for netfilter will be 'on'.
4597 queue all\|rx\|tx is an option that can be applied to any
4600 ``all``: the filter is attached both to the receive and the
4601 transmit queue of the netdev (default).
4603 ``rx``: the filter is attached to the receive queue of the
4604 netdev, where it will receive packets sent to the netdev.
4606 ``tx``: the filter is attached to the transmit queue of the
4607 netdev, where it will receive packets sent by the netdev.
4609 position head\|tail\|id=<id> is an option to specify where the
4610 filter should be inserted in the filter list. It can be applied
4613 ``head``: the filter is inserted at the head of the filter list,
4614 before any existing filters.
4616 ``tail``: the filter is inserted at the tail of the filter list,
4617 behind any existing filters (default).
4619 ``id=<id>``: the filter is inserted before or behind the filter
4620 specified by <id>, see the insert option below.
4622 insert behind\|before is an option to specify where to insert
4623 the new filter relative to the one specified with
4624 position=id=<id>. It can be applied to any netfilter.
4626 ``before``: insert before the specified filter.
4628 ``behind``: insert behind the specified filter (default).
4630 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
4631 filter-mirror on netdev netdevid,mirror net packet to
4632 chardevchardevid, if it has the vnet\_hdr\_support flag,
4633 filter-mirror will mirror packet with vnet\_hdr\_len.
4635 ``-object filter-redirector,id=id,netdev=netdevid,indev=chardevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
4636 filter-redirector on netdev netdevid,redirect filter's net
4637 packet to chardev chardevid,and redirect indev's packet to
4638 filter.if it has the vnet\_hdr\_support flag, filter-redirector
4639 will redirect packet with vnet\_hdr\_len. Create a
4640 filter-redirector we need to differ outdev id from indev id, id
4641 can not be the same. we can just use indev or outdev, but at
4642 least one of indev or outdev need to be specified.
4644 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
4645 Filter-rewriter is a part of COLO project.It will rewrite tcp
4646 packet to secondary from primary to keep secondary tcp
4647 connection,and rewrite tcp packet to primary from secondary make
4648 tcp packet can be handled by client.if it has the
4649 vnet\_hdr\_support flag, we can parse packet with vnet header.
4651 usage: colo secondary: -object
4652 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object
4653 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object
4654 filter-rewriter,id=rew0,netdev=hn0,queue=all
4656 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]``
4657 Dump the network traffic on netdev dev to the file specified by
4658 filename. At most len bytes (64k by default) per packet are
4659 stored. The file format is libpcap, so it can be analyzed with
4660 tools such as tcpdump or Wireshark.
4662 ``-object colo-compare,id=id,primary_in=chardevid,secondary_in=chardevid,outdev=chardevid,iothread=id[,vnet_hdr_support][,notify_dev=id][,compare_timeout=@var{ms}][,expired_scan_cycle=@var{ms}``
4663 Colo-compare gets packet from primary\_inchardevid and
4664 secondary\_inchardevid, than compare primary packet with
4665 secondary packet. If the packets are same, we will output
4666 primary packet to outdevchardevid, else we will notify
4667 colo-frame do checkpoint and send primary packet to
4668 outdevchardevid. In order to improve efficiency, we need to put
4669 the task of comparison in another thread. If it has the
4670 vnet\_hdr\_support flag, colo compare will send/recv packet with
4671 vnet\_hdr\_len. Then compare\_timeout=@var{ms} determines the
4672 maximum delay colo-compare wait for the packet.
4673 The expired\_scan\_cycle=@var{ms} to set the period of scanning
4674 expired primary node network packets.
4675 If you want to use Xen COLO, will need the notify\_dev to
4676 notify Xen colo-frame to do checkpoint.
4678 we must use it with the help of filter-mirror and
4686 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
4687 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
4688 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server,nowait
4689 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server,nowait
4690 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server,nowait
4691 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
4692 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server,nowait
4693 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
4694 -object iothread,id=iothread1
4695 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
4696 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
4697 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
4698 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1
4701 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
4702 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
4703 -chardev socket,id=red0,host=3.3.3.3,port=9003
4704 -chardev socket,id=red1,host=3.3.3.3,port=9004
4705 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
4706 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
4712 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
4713 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
4714 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server,nowait
4715 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server,nowait
4716 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server,nowait
4717 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
4718 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server,nowait
4719 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
4720 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server,nowait
4721 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
4722 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
4723 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
4724 -object iothread,id=iothread1
4725 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1
4728 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
4729 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
4730 -chardev socket,id=red0,host=3.3.3.3,port=9003
4731 -chardev socket,id=red1,host=3.3.3.3,port=9004
4732 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
4733 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
4735 If you want to know the detail of above command line, you can
4736 read the colo-compare git log.
4738 ``-object cryptodev-backend-builtin,id=id[,queues=queues]``
4739 Creates a cryptodev backend which executes crypto opreation from
4740 the QEMU cipher APIS. The id parameter is a unique ID that will
4741 be used to reference this cryptodev backend from the
4742 ``virtio-crypto`` device. The queues parameter is optional,
4743 which specify the queue number of cryptodev backend, the default
4750 -object cryptodev-backend-builtin,id=cryptodev0 \
4751 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \
4754 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]``
4755 Creates a vhost-user cryptodev backend, backed by a chardev
4756 chardevid. The id parameter is a unique ID that will be used to
4757 reference this cryptodev backend from the ``virtio-crypto``
4758 device. The chardev should be a unix domain socket backed one.
4759 The vhost-user uses a specifically defined protocol to pass
4760 vhost ioctl replacement messages to an application on the other
4761 end of the socket. The queues parameter is optional, which
4762 specify the queue number of cryptodev backend for multiqueue
4763 vhost-user, the default of queues is 1.
4769 -chardev socket,id=chardev0,path=/path/to/socket \
4770 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \
4771 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \
4774 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]``
4776 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]``
4777 Defines a secret to store a password, encryption key, or some
4778 other sensitive data. The sensitive data can either be passed
4779 directly via the data parameter, or indirectly via the file
4780 parameter. Using the data parameter is insecure unless the
4781 sensitive data is encrypted.
4783 The sensitive data can be provided in raw format (the default),
4784 or base64. When encoded as JSON, the raw format only supports
4785 valid UTF-8 characters, so base64 is recommended for sending
4786 binary data. QEMU will convert from which ever format is
4787 provided to the format it needs internally. eg, an RBD password
4788 can be provided in raw format, even though it will be base64
4789 encoded when passed onto the RBD sever.
4791 For added protection, it is possible to encrypt the data
4792 associated with a secret using the AES-256-CBC cipher. Use of
4793 encryption is indicated by providing the keyid and iv
4794 parameters. The keyid parameter provides the ID of a previously
4795 defined secret that contains the AES-256 decryption key. This
4796 key should be 32-bytes long and be base64 encoded. The iv
4797 parameter provides the random initialization vector used for
4798 encryption of this particular secret and should be a base64
4799 encrypted string of the 16-byte IV.
4801 The simplest (insecure) usage is to provide the secret inline
4805 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw
4807 The simplest secure usage is to provide the secret via a file
4809 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object
4810 secret,id=sec0,file=mypasswd.txt,format=raw
4812 For greater security, AES-256-CBC should be used. To illustrate
4813 usage, consider the openssl command line tool which can encrypt
4814 the data. Note that when encrypting, the plaintext must be
4815 padded to the cipher block size (32 bytes) using the standard
4816 PKCS#5/6 compatible padding algorithm.
4818 First a master key needs to be created in base64 encoding:
4822 # openssl rand -base64 32 > key.b64
4823 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"')
4825 Each secret to be encrypted needs to have a random
4826 initialization vector generated. These do not need to be kept
4831 # openssl rand -base64 16 > iv.b64
4832 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"')
4834 The secret to be defined can now be encrypted, in this case
4835 we're telling openssl to base64 encode the result, but it could
4836 be left as raw bytes if desired.
4840 # SECRET=$(printf "letmein" |
4841 openssl enc -aes-256-cbc -a -K $KEY -iv $IV)
4843 When launching QEMU, create a master secret pointing to
4844 ``key.b64`` and specify that to be used to decrypt the user
4845 password. Pass the contents of ``iv.b64`` to the second secret
4850 -object secret,id=secmaster0,format=base64,file=key.b64 \
4851 -object secret,id=sec0,keyid=secmaster0,format=base64,\
4852 data=$SECRET,iv=$(<iv.b64)
4854 ``-object sev-guest,id=id,cbitpos=cbitpos,reduced-phys-bits=val,[sev-device=string,policy=policy,handle=handle,dh-cert-file=file,session-file=file]``
4855 Create a Secure Encrypted Virtualization (SEV) guest object,
4856 which can be used to provide the guest memory encryption support
4859 When memory encryption is enabled, one of the physical address
4860 bit (aka the C-bit) is utilized to mark if a memory page is
4861 protected. The ``cbitpos`` is used to provide the C-bit
4862 position. The C-bit position is Host family dependent hence user
4863 must provide this value. On EPYC, the value should be 47.
4865 When memory encryption is enabled, we loose certain bits in
4866 physical address space. The ``reduced-phys-bits`` is used to
4867 provide the number of bits we loose in physical address space.
4868 Similar to C-bit, the value is Host family dependent. On EPYC,
4869 the value should be 5.
4871 The ``sev-device`` provides the device file to use for
4872 communicating with the SEV firmware running inside AMD Secure
4873 Processor. The default device is '/dev/sev'. If hardware
4874 supports memory encryption then /dev/sev devices are created by
4877 The ``policy`` provides the guest policy to be enforced by the
4878 SEV firmware and restrict what configuration and operational
4879 commands can be performed on this guest by the hypervisor. The
4880 policy should be provided by the guest owner and is bound to the
4881 guest and cannot be changed throughout the lifetime of the
4882 guest. The default is 0.
4884 If guest ``policy`` allows sharing the key with another SEV
4885 guest then ``handle`` can be use to provide handle of the guest
4886 from which to share the key.
4888 The ``dh-cert-file`` and ``session-file`` provides the guest
4889 owner's Public Diffie-Hillman key defined in SEV spec. The PDH
4890 and session parameters are used for establishing a cryptographic
4891 session with the guest owner to negotiate keys used for
4892 attestation. The file must be encoded in base64.
4894 e.g to launch a SEV guest
4898 # |qemu_system_x86| \
4900 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \
4901 -machine ...,memory-encryption=sev0
4904 ``-object authz-simple,id=id,identity=string``
4905 Create an authorization object that will control access to
4908 The ``identity`` parameter is identifies the user and its format
4909 depends on the network service that authorization object is
4910 associated with. For authorizing based on TLS x509 certificates,
4911 the identity must be the x509 distinguished name. Note that care
4912 must be taken to escape any commas in the distinguished name.
4914 An example authorization object to validate a x509 distinguished
4915 name would look like:
4921 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \
4924 Note the use of quotes due to the x509 distinguished name
4925 containing whitespace, and escaping of ','.
4927 ``-object authz-listfile,id=id,filename=path,refresh=yes|no``
4928 Create an authorization object that will control access to
4931 The ``filename`` parameter is the fully qualified path to a file
4932 containing the access control list rules in JSON format.
4934 An example set of rules that match against SASL usernames might
4941 { "match": "fred", "policy": "allow", "format": "exact" },
4942 { "match": "bob", "policy": "allow", "format": "exact" },
4943 { "match": "danb", "policy": "deny", "format": "glob" },
4944 { "match": "dan*", "policy": "allow", "format": "exact" },
4949 When checking access the object will iterate over all the rules
4950 and the first rule to match will have its ``policy`` value
4951 returned as the result. If no rules match, then the default
4952 ``policy`` value is returned.
4954 The rules can either be an exact string match, or they can use
4955 the simple UNIX glob pattern matching to allow wildcards to be
4958 If ``refresh`` is set to true the file will be monitored and
4959 automatically reloaded whenever its content changes.
4961 As with the ``authz-simple`` object, the format of the identity
4962 strings being matched depends on the network service, but is
4963 usually a TLS x509 distinguished name, or a SASL username.
4965 An example authorization object to validate a SASL username
4972 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=yes
4975 ``-object authz-pam,id=id,service=string``
4976 Create an authorization object that will control access to
4979 The ``service`` parameter provides the name of a PAM service to
4980 use for authorization. It requires that a file
4981 ``/etc/pam.d/service`` exist to provide the configuration for
4982 the ``account`` subsystem.
4984 An example authorization object to validate a TLS x509
4985 distinguished name would look like:
4991 -object authz-pam,id=auth0,service=qemu-vnc
4994 There would then be a corresponding config file for PAM at
4995 ``/etc/pam.d/qemu-vnc`` that contains:
4999 account requisite pam_listfile.so item=user sense=allow \
5000 file=/etc/qemu/vnc.allow
5002 Finally the ``/etc/qemu/vnc.allow`` file would contain the list
5003 of x509 distingished names that are permitted access
5007 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB
5009 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink``
5010 Creates a dedicated event loop thread that devices can be
5011 assigned to. This is known as an IOThread. By default device
5012 emulation happens in vCPU threads or the main event loop thread.
5013 This can become a scalability bottleneck. IOThreads allow device
5014 emulation and I/O to run on other host CPUs.
5016 The ``id`` parameter is a unique ID that will be used to
5017 reference this IOThread from ``-device ...,iothread=id``.
5018 Multiple devices can be assigned to an IOThread. Note that not
5019 all devices support an ``iothread`` parameter.
5021 The ``query-iothreads`` QMP command lists IOThreads and reports
5022 their thread IDs so that the user can configure host CPU
5025 IOThreads use an adaptive polling algorithm to reduce event loop
5026 latency. Instead of entering a blocking system call to monitor
5027 file descriptors and then pay the cost of being woken up when an
5028 event occurs, the polling algorithm spins waiting for events for
5029 a short time. The algorithm's default parameters are suitable
5030 for many cases but can be adjusted based on knowledge of the
5031 workload and/or host device latency.
5033 The ``poll-max-ns`` parameter is the maximum number of
5034 nanoseconds to busy wait for events. Polling can be disabled by
5035 setting this value to 0.
5037 The ``poll-grow`` parameter is the multiplier used to increase
5038 the polling time when the algorithm detects it is missing events
5039 due to not polling long enough.
5041 The ``poll-shrink`` parameter is the divisor used to decrease
5042 the polling time when the algorithm detects it is spending too
5043 long polling without encountering events.
5045 The polling parameters can be modified at run-time using the
5046 ``qom-set`` command (where ``iothread1`` is the IOThread's
5051 (qemu) qom-set /objects/iothread1 poll-max-ns 100000
5055 HXCOMM This is the last statement. Insert new options before this line!