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, nvmm, whpx or tcg (default: tcg)\n"
30 " vmport=on|off|auto controls emulation of vmport (default: auto)\n"
31 " dump-guest-core=on|off include guest memory in a core dump (default=on)\n"
32 " mem-merge=on|off controls memory merge support (default: on)\n"
33 " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n"
34 " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n"
35 " suppress-vmdesc=on|off disables self-describing migration (default=off)\n"
36 " nvdimm=on|off controls NVDIMM support (default=off)\n"
37 " memory-encryption=@var{} memory encryption object to use (default=none)\n"
38 " hmat=on|off controls ACPI HMAT support (default=off)\n"
39 " memory-backend='backend-id' specifies explicitly provided backend for main RAM (default=none)\n",
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, nvmm, whpx or tcg can be available.
62 By default, tcg is used. If there is more than one accelerator
63 specified, the next one is used if the previous one fails to
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 ``memory-encryption=``
95 Memory encryption object to use. The default is none.
98 Enables or disables ACPI Heterogeneous Memory Attribute Table
99 (HMAT) support. The default is off.
101 ``memory-backend='id'``
102 An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options.
103 Allows to use a memory backend as main RAM.
107 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on
108 -machine memory-backend=pc.ram
111 Migration compatibility note:
112 a) as backend id one shall use value of 'default-ram-id', advertised by
113 machine type (available via ``query-machines`` QMP command), if migration
114 to/from old QEMU (<5.0) is expected.
115 b) for machine types 4.0 and older, user shall
116 use ``x-use-canonical-path-for-ramblock-id=off`` backend option
117 if migration to/from old QEMU (<5.0) is expected.
120 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off
121 -machine memory-backend=pc.ram
125 HXCOMM Deprecated by -machine
126 DEF("M", HAS_ARG, QEMU_OPTION_M, "", QEMU_ARCH_ALL)
128 DEF("cpu", HAS_ARG, QEMU_OPTION_cpu,
129 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL)
132 Select CPU model (``-cpu help`` for list and additional feature
136 DEF("accel", HAS_ARG, QEMU_OPTION_accel,
137 "-accel [accel=]accelerator[,prop[=value][,...]]\n"
138 " select accelerator (kvm, xen, hax, hvf, nvmm, whpx or tcg; use 'help' for a list)\n"
139 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n"
140 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n"
141 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n"
142 " split-wx=on|off (enable TCG split w^x mapping)\n"
143 " tb-size=n (TCG translation block cache size)\n"
144 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL)
146 ``-accel name[,prop=value[,...]]``
147 This is used to enable an accelerator. Depending on the target
148 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available. By
149 default, tcg is used. If there is more than one accelerator
150 specified, the next one is used if the previous one fails to
153 ``igd-passthru=on|off``
154 When Xen is in use, this option controls whether Intel
155 integrated graphics devices can be passed through to the guest
158 ``kernel-irqchip=on|off|split``
159 Controls KVM in-kernel irqchip support. The default is full
160 acceleration of the interrupt controllers. On x86, split irqchip
161 reduces the kernel attack surface, at a performance cost for
162 non-MSI interrupts. Disabling the in-kernel irqchip completely
163 is not recommended except for debugging purposes.
165 ``kvm-shadow-mem=size``
166 Defines the size of the KVM shadow MMU.
169 Controls the use of split w^x mapping for the TCG code generation
170 buffer. Some operating systems require this to be enabled, and in
171 such a case this will default on. On other operating systems, this
172 will default off, but one may enable this for testing or debugging.
175 Controls the size (in MiB) of the TCG translation block cache.
177 ``thread=single|multi``
178 Controls number of TCG threads. When the TCG is multi-threaded
179 there will be one thread per vCPU therefore taking advantage of
180 additional host cores. The default is to enable multi-threading
181 where both the back-end and front-ends support it and no
182 incompatible TCG features have been enabled (e.g.
186 DEF("smp", HAS_ARG, QEMU_OPTION_smp,
187 "-smp [cpus=]n[,maxcpus=cpus][,cores=cores][,threads=threads][,dies=dies][,sockets=sockets]\n"
188 " set the number of CPUs to 'n' [default=1]\n"
189 " maxcpus= maximum number of total cpus, including\n"
190 " offline CPUs for hotplug, etc\n"
191 " cores= number of CPU cores on one socket (for PC, it's on one die)\n"
192 " threads= number of threads on one CPU core\n"
193 " dies= number of CPU dies on one socket (for PC only)\n"
194 " sockets= number of discrete sockets in the system\n",
197 ``-smp [cpus=]n[,cores=cores][,threads=threads][,dies=dies][,sockets=sockets][,maxcpus=maxcpus]``
198 Simulate an SMP system with n CPUs. On the PC target, up to 255 CPUs
199 are supported. On Sparc32 target, Linux limits the number of usable
200 CPUs to 4. For the PC target, the number of cores per die, the
201 number of threads per cores, the number of dies per packages and the
202 total number of sockets can be specified. Missing values will be
203 computed. If any on the three values is given, the total number of
204 CPUs n can be omitted. maxcpus specifies the maximum number of
208 DEF("numa", HAS_ARG, QEMU_OPTION_numa,
209 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
210 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
211 "-numa dist,src=source,dst=destination,val=distance\n"
212 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n"
213 "-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"
214 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n",
217 ``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
219 ``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
221 ``-numa dist,src=source,dst=destination,val=distance``
223 ``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]``
225 ``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=tpye[,latency=lat][,bandwidth=bw]``
227 ``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]``
228 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA
229 distance from a source node to a destination node. Set the ACPI
230 Heterogeneous Memory Attributes for the given nodes.
232 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and
233 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a
234 contiguous range of CPU indexes (or a single VCPU if lastcpu is
235 omitted). A non-contiguous set of VCPUs can be represented by
236 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is
237 omitted on all nodes, VCPUs are automatically split between them.
239 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a
244 -numa node,cpus=0-2,cpus=5
246 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option
247 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to
248 assign CPU objects to a node using topology layout properties of
249 CPU. The set of properties is machine specific, and depends on used
250 machine type/'\ ``smp``\ ' options. It could be queried with
251 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ '
252 property specifies node to which CPU object will be assigned, it's
253 required for node to be declared with '\ ``node``\ ' option before
254 it's used with '\ ``cpu``\ ' option.
261 -smp 1,sockets=2,maxcpus=2 \
262 -numa node,nodeid=0 -numa node,nodeid=1 \
263 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1
265 Legacy '\ ``mem``\ ' assigns a given RAM amount to a node (not supported
266 for 5.1 and newer machine types). '\ ``memdev``\ ' assigns RAM from
267 a given memory backend device to a node. If '\ ``mem``\ ' and
268 '\ ``memdev``\ ' are omitted in all nodes, RAM is split equally between them.
271 '\ ``mem``\ ' and '\ ``memdev``\ ' are mutually exclusive.
272 Furthermore, if one node uses '\ ``memdev``\ ', all of them have to
275 '\ ``initiator``\ ' is an additional option that points to an
276 initiator NUMA node that has best performance (the lowest latency or
277 largest bandwidth) to this NUMA node. Note that this option can be
278 set only when the machine property 'hmat' is set to 'on'.
280 Following example creates a machine with 2 NUMA nodes, node 0 has
281 CPU. node 1 has only memory, and its initiator is node 0. Note that
282 because node 0 has CPU, by default the initiator of node 0 is itself
288 -m 2G,slots=2,maxmem=4G \
289 -object memory-backend-ram,size=1G,id=m0 \
290 -object memory-backend-ram,size=1G,id=m1 \
291 -numa node,nodeid=0,memdev=m0 \
292 -numa node,nodeid=1,memdev=m1,initiator=0 \
293 -smp 2,sockets=2,maxcpus=2 \
294 -numa cpu,node-id=0,socket-id=0 \
295 -numa cpu,node-id=0,socket-id=1
297 source and destination are NUMA node IDs. distance is the NUMA
298 distance from source to destination. The distance from a node to
299 itself is always 10. If any pair of nodes is given a distance, then
300 all pairs must be given distances. Although, when distances are only
301 given in one direction for each pair of nodes, then the distances in
302 the opposite directions are assumed to be the same. If, however, an
303 asymmetrical pair of distances is given for even one node pair, then
304 all node pairs must be provided distance values for both directions,
305 even when they are symmetrical. When a node is unreachable from
306 another node, set the pair's distance to 255.
308 Note that the -``numa`` option doesn't allocate any of the specified
309 resources, it just assigns existing resources to NUMA nodes. This
310 means that one still has to use the ``-m``, ``-smp`` options to
311 allocate RAM and VCPUs respectively.
313 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth
314 Information between initiator and target NUMA nodes in ACPI
315 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can
316 create memory requests, usually it has one or more processors.
317 Target NUMA node contains addressable memory.
319 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is
320 the memory hierarchy of the target NUMA node: if hierarchy is
321 'memory', the structure represents the memory performance; if
322 hierarchy is 'first-level\|second-level\|third-level', this
323 structure represents aggregated performance of memory side caches
324 for each domain. type of 'data-type' is type of data represented by
325 this structure instance: if 'hierarchy' is 'memory', 'data-type' is
326 'access\|read\|write' latency or 'access\|read\|write' bandwidth of
327 the target memory; if 'hierarchy' is
328 'first-level\|second-level\|third-level', 'data-type' is
329 'access\|read\|write' hit latency or 'access\|read\|write' hit
330 bandwidth of the target memory side cache.
332 lat is latency value in nanoseconds. bw is bandwidth value, the
333 possible value and units are NUM[M\|G\|T], mean that the bandwidth
334 value are NUM byte per second (or MB/s, GB/s or TB/s depending on
335 used suffix). Note that if latency or bandwidth value is 0, means
336 the corresponding latency or bandwidth information is not provided.
338 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory
339 belongs. size is the size of memory side cache in bytes. level is
340 the cache level described in this structure, note that the cache
341 level 0 should not be used with '\ ``hmat-cache``\ ' option.
342 associativity is the cache associativity, the possible value is
343 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy
344 is the write policy. line is the cache Line size in bytes.
346 For example, the following options describe 2 NUMA nodes. Node 0 has
347 2 cpus and a ram, node 1 has only a ram. The processors in node 0
348 access memory in node 0 with access-latency 5 nanoseconds,
349 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access
350 memory in NUMA node 1 with access-latency 10 nanoseconds,
351 access-bandwidth is 100 MB/s. And for memory side cache information,
352 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB,
353 policy is write-back, the cache Line size is 8 bytes:
359 -object memory-backend-ram,size=1G,id=m0 \
360 -object memory-backend-ram,size=1G,id=m1 \
362 -numa node,nodeid=0,memdev=m0 \
363 -numa node,nodeid=1,memdev=m1,initiator=0 \
364 -numa cpu,node-id=0,socket-id=0 \
365 -numa cpu,node-id=0,socket-id=1 \
366 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \
367 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \
368 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \
369 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \
370 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \
371 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8
374 DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd,
375 "-add-fd fd=fd,set=set[,opaque=opaque]\n"
376 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL)
378 ``-add-fd fd=fd,set=set[,opaque=opaque]``
379 Add a file descriptor to an fd set. Valid options are:
382 This option defines the file descriptor of which a duplicate is
383 added to fd set. The file descriptor cannot be stdin, stdout, or
387 This option defines the ID of the fd set to add the file
391 This option defines a free-form string that can be used to
394 You can open an image using pre-opened file descriptors from an fd
400 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
401 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
402 -drive file=/dev/fdset/2,index=0,media=disk
405 DEF("set", HAS_ARG, QEMU_OPTION_set,
406 "-set group.id.arg=value\n"
407 " set <arg> parameter for item <id> of type <group>\n"
408 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL)
410 ``-set group.id.arg=value``
411 Set parameter arg for item id of type group
414 DEF("global", HAS_ARG, QEMU_OPTION_global,
415 "-global driver.property=value\n"
416 "-global driver=driver,property=property,value=value\n"
417 " set a global default for a driver property\n",
420 ``-global driver.prop=value``
422 ``-global driver=driver,property=property,value=value``
423 Set default value of driver's property prop to value, e.g.:
427 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img
429 In particular, you can use this to set driver properties for devices
430 which are created automatically by the machine model. To create a
431 device which is not created automatically and set properties on it,
434 -global driver.prop=value is shorthand for -global
435 driver=driver,property=prop,value=value. The longhand syntax works
436 even when driver contains a dot.
439 DEF("boot", HAS_ARG, QEMU_OPTION_boot,
440 "-boot [order=drives][,once=drives][,menu=on|off]\n"
441 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n"
442 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n"
443 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n"
444 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n"
445 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n",
448 ``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]``
449 Specify boot order drives as a string of drive letters. Valid drive
450 letters depend on the target architecture. The x86 PC uses: a, b
451 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p
452 (Etherboot from network adapter 1-4), hard disk boot is the default.
453 To apply a particular boot order only on the first startup, specify
454 it via ``once``. Note that the ``order`` or ``once`` parameter
455 should not be used together with the ``bootindex`` property of
456 devices, since the firmware implementations normally do not support
457 both at the same time.
459 Interactive boot menus/prompts can be enabled via ``menu=on`` as far
460 as firmware/BIOS supports them. The default is non-interactive boot.
462 A splash picture could be passed to bios, enabling user to show it
463 as logo, when option splash=sp\_name is given and menu=on, If
464 firmware/BIOS supports them. Currently Seabios for X86 system
465 support it. limitation: The splash file could be a jpeg file or a
466 BMP file in 24 BPP format(true color). The resolution should be
467 supported by the SVGA mode, so the recommended is 320x240, 640x480,
470 A timeout could be passed to bios, guest will pause for rb\_timeout
471 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will
472 not reboot, qemu passes '-1' to bios by default. Currently Seabios
473 for X86 system support it.
475 Do strict boot via ``strict=on`` as far as firmware/BIOS supports
476 it. This only effects when boot priority is changed by bootindex
477 options. The default is non-strict boot.
481 # try to boot from network first, then from hard disk
482 |qemu_system_x86| -boot order=nc
483 # boot from CD-ROM first, switch back to default order after reboot
484 |qemu_system_x86| -boot once=d
485 # boot with a splash picture for 5 seconds.
486 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000
488 Note: The legacy format '-boot drives' is still supported but its
489 use is discouraged as it may be removed from future versions.
492 DEF("m", HAS_ARG, QEMU_OPTION_m,
493 "-m [size=]megs[,slots=n,maxmem=size]\n"
494 " configure guest RAM\n"
495 " size: initial amount of guest memory\n"
496 " slots: number of hotplug slots (default: none)\n"
497 " maxmem: maximum amount of guest memory (default: none)\n"
498 "NOTE: Some architectures might enforce a specific granularity\n",
501 ``-m [size=]megs[,slots=n,maxmem=size]``
502 Sets guest startup RAM size to megs megabytes. Default is 128 MiB.
503 Optionally, a suffix of "M" or "G" can be used to signify a value in
504 megabytes or gigabytes respectively. Optional pair slots, maxmem
505 could be used to set amount of hotpluggable memory slots and maximum
506 amount of memory. Note that maxmem must be aligned to the page size.
508 For example, the following command-line sets the guest startup RAM
509 size to 1GB, creates 3 slots to hotplug additional memory and sets
510 the maximum memory the guest can reach to 4GB:
514 |qemu_system| -m 1G,slots=3,maxmem=4G
516 If slots and maxmem are not specified, memory hotplug won't be
517 enabled and the guest startup RAM will never increase.
520 DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath,
521 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL)
524 Allocate guest RAM from a temporarily created file in path.
527 DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc,
528 "-mem-prealloc preallocate guest memory (use with -mem-path)\n",
532 Preallocate memory when using -mem-path.
535 DEF("k", HAS_ARG, QEMU_OPTION_k,
536 "-k language use keyboard layout (for example 'fr' for French)\n",
540 Use keyboard layout language (for example ``fr`` for French). This
541 option is only needed where it is not easy to get raw PC keycodes
542 (e.g. on Macs, with some X11 servers or with a VNC or curses
543 display). You don't normally need to use it on PC/Linux or
546 The available layouts are:
550 ar de-ch es fo fr-ca hu ja mk no pt-br sv
551 da en-gb et fr fr-ch is lt nl pl ru th
552 de en-us fi fr-be hr it lv nl-be pt sl tr
554 The default is ``en-us``.
558 HXCOMM Deprecated by -audiodev
559 DEF("audio-help", 0, QEMU_OPTION_audio_help,
560 "-audio-help show -audiodev equivalent of the currently specified audio settings\n",
564 Will show the -audiodev equivalent of the currently specified
565 (deprecated) environment variables.
568 DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev,
569 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n"
570 " specifies the audio backend to use\n"
571 " id= identifier of the backend\n"
572 " timer-period= timer period in microseconds\n"
573 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n"
574 " in|out.fixed-settings= use fixed settings for host audio\n"
575 " in|out.frequency= frequency to use with fixed settings\n"
576 " in|out.channels= number of channels to use with fixed settings\n"
577 " in|out.format= sample format to use with fixed settings\n"
578 " valid values: s8, s16, s32, u8, u16, u32, f32\n"
579 " in|out.voices= number of voices to use\n"
580 " in|out.buffer-length= length of buffer in microseconds\n"
581 "-audiodev none,id=id,[,prop[=value][,...]]\n"
582 " dummy driver that discards all output\n"
583 #ifdef CONFIG_AUDIO_ALSA
584 "-audiodev alsa,id=id[,prop[=value][,...]]\n"
585 " in|out.dev= name of the audio device to use\n"
586 " in|out.period-length= length of period in microseconds\n"
587 " in|out.try-poll= attempt to use poll mode\n"
588 " threshold= threshold (in microseconds) when playback starts\n"
590 #ifdef CONFIG_AUDIO_COREAUDIO
591 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n"
592 " in|out.buffer-count= number of buffers\n"
594 #ifdef CONFIG_AUDIO_DSOUND
595 "-audiodev dsound,id=id[,prop[=value][,...]]\n"
596 " latency= add extra latency to playback in microseconds\n"
598 #ifdef CONFIG_AUDIO_OSS
599 "-audiodev oss,id=id[,prop[=value][,...]]\n"
600 " in|out.dev= path of the audio device to use\n"
601 " in|out.buffer-count= number of buffers\n"
602 " in|out.try-poll= attempt to use poll mode\n"
603 " try-mmap= try using memory mapped access\n"
604 " exclusive= open device in exclusive mode\n"
605 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n"
607 #ifdef CONFIG_AUDIO_PA
608 "-audiodev pa,id=id[,prop[=value][,...]]\n"
609 " server= PulseAudio server address\n"
610 " in|out.name= source/sink device name\n"
611 " in|out.latency= desired latency in microseconds\n"
613 #ifdef CONFIG_AUDIO_SDL
614 "-audiodev sdl,id=id[,prop[=value][,...]]\n"
615 " in|out.buffer-count= number of buffers\n"
618 "-audiodev spice,id=id[,prop[=value][,...]]\n"
620 "-audiodev wav,id=id[,prop[=value][,...]]\n"
621 " path= path of wav file to record\n",
624 ``-audiodev [driver=]driver,id=id[,prop[=value][,...]]``
625 Adds a new audio backend driver identified by id. There are global
626 and driver specific properties. Some values can be set differently
627 for input and output, they're marked with ``in|out.``. You can set
628 the input's property with ``in.prop`` and the output's property with
629 ``out.prop``. For example:
633 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000
634 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified
636 NOTE: parameter validation is known to be incomplete, in many cases
637 specifying an invalid option causes QEMU to print an error message
638 and continue emulation without sound.
640 Valid global options are:
643 Identifies the audio backend.
645 ``timer-period=period``
646 Sets the timer period used by the audio subsystem in
647 microseconds. Default is 10000 (10 ms).
649 ``in|out.mixing-engine=on|off``
650 Use QEMU's mixing engine to mix all streams inside QEMU and
651 convert audio formats when not supported by the backend. When
652 off, fixed-settings must be off too. Note that disabling this
653 option means that the selected backend must support multiple
654 streams and the audio formats used by the virtual cards,
655 otherwise you'll get no sound. It's not recommended to disable
656 this option unless you want to use 5.1 or 7.1 audio, as mixing
657 engine only supports mono and stereo audio. Default is on.
659 ``in|out.fixed-settings=on|off``
660 Use fixed settings for host audio. When off, it will change
661 based on how the guest opens the sound card. In this case you
662 must not specify frequency, channels or format. Default is on.
664 ``in|out.frequency=frequency``
665 Specify the frequency to use when using fixed-settings. Default
668 ``in|out.channels=channels``
669 Specify the number of channels to use when using fixed-settings.
670 Default is 2 (stereo).
672 ``in|out.format=format``
673 Specify the sample format to use when using fixed-settings.
674 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``,
675 ``u32``, ``f32``. Default is ``s16``.
677 ``in|out.voices=voices``
678 Specify the number of voices to use. Default is 1.
680 ``in|out.buffer-length=usecs``
681 Sets the size of the buffer in microseconds.
683 ``-audiodev none,id=id[,prop[=value][,...]]``
684 Creates a dummy backend that discards all outputs. This backend has
685 no backend specific properties.
687 ``-audiodev alsa,id=id[,prop[=value][,...]]``
688 Creates backend using the ALSA. This backend is only available on
691 ALSA specific options are:
693 ``in|out.dev=device``
694 Specify the ALSA device to use for input and/or output. Default
697 ``in|out.period-length=usecs``
698 Sets the period length in microseconds.
700 ``in|out.try-poll=on|off``
701 Attempt to use poll mode with the device. Default is on.
703 ``threshold=threshold``
704 Threshold (in microseconds) when playback starts. Default is 0.
706 ``-audiodev coreaudio,id=id[,prop[=value][,...]]``
707 Creates a backend using Apple's Core Audio. This backend is only
708 available on Mac OS and only supports playback.
710 Core Audio specific options are:
712 ``in|out.buffer-count=count``
713 Sets the count of the buffers.
715 ``-audiodev dsound,id=id[,prop[=value][,...]]``
716 Creates a backend using Microsoft's DirectSound. This backend is
717 only available on Windows and only supports playback.
719 DirectSound specific options are:
722 Add extra usecs microseconds latency to playback. Default is
725 ``-audiodev oss,id=id[,prop[=value][,...]]``
726 Creates a backend using OSS. This backend is available on most
729 OSS specific options are:
731 ``in|out.dev=device``
732 Specify the file name of the OSS device to use. Default is
735 ``in|out.buffer-count=count``
736 Sets the count of the buffers.
738 ``in|out.try-poll=on|of``
739 Attempt to use poll mode with the device. Default is on.
742 Try using memory mapped device access. Default is off.
745 Open the device in exclusive mode (vmix won't work in this
746 case). Default is off.
748 ``dsp-policy=policy``
749 Sets the timing policy (between 0 and 10, where smaller number
750 means smaller latency but higher CPU usage). Use -1 to use
751 buffer sizes specified by ``buffer`` and ``buffer-count``. This
752 option is ignored if you do not have OSS 4. Default is 5.
754 ``-audiodev pa,id=id[,prop[=value][,...]]``
755 Creates a backend using PulseAudio. This backend is available on
758 PulseAudio specific options are:
761 Sets the PulseAudio server to connect to.
764 Use the specified source/sink for recording/playback.
766 ``in|out.latency=usecs``
767 Desired latency in microseconds. The PulseAudio server will try
768 to honor this value but actual latencies may be lower or higher.
770 ``-audiodev sdl,id=id[,prop[=value][,...]]``
771 Creates a backend using SDL. This backend is available on most
772 systems, but you should use your platform's native backend if
775 SDL specific options are:
777 ``in|out.buffer-count=count``
778 Sets the count of the buffers.
780 ``-audiodev spice,id=id[,prop[=value][,...]]``
781 Creates a backend that sends audio through SPICE. This backend
782 requires ``-spice`` and automatically selected in that case, so
783 usually you can ignore this option. This backend has no backend
786 ``-audiodev wav,id=id[,prop[=value][,...]]``
787 Creates a backend that writes audio to a WAV file.
789 Backend specific options are:
792 Write recorded audio into the specified file. Default is
796 DEF("soundhw", HAS_ARG, QEMU_OPTION_soundhw,
797 "-soundhw c1,... enable audio support\n"
798 " and only specified sound cards (comma separated list)\n"
799 " use '-soundhw help' to get the list of supported cards\n"
800 " use '-soundhw all' to enable all of them\n", QEMU_ARCH_ALL)
802 ``-soundhw card1[,card2,...] or -soundhw all``
803 Enable audio and selected sound hardware. Use 'help' to print all
804 available sound hardware. For example:
808 |qemu_system_x86| -soundhw sb16,adlib disk.img
809 |qemu_system_x86| -soundhw es1370 disk.img
810 |qemu_system_x86| -soundhw ac97 disk.img
811 |qemu_system_x86| -soundhw hda disk.img
812 |qemu_system_x86| -soundhw all disk.img
813 |qemu_system_x86| -soundhw help
815 Note that Linux's i810\_audio OSS kernel (for AC97) module might
816 require manually specifying clocking.
820 modprobe i810_audio clocking=48000
823 DEF("device", HAS_ARG, QEMU_OPTION_device,
824 "-device driver[,prop[=value][,...]]\n"
825 " add device (based on driver)\n"
826 " prop=value,... sets driver properties\n"
827 " use '-device help' to print all possible drivers\n"
828 " use '-device driver,help' to print all possible properties\n",
831 ``-device driver[,prop[=value][,...]]``
832 Add device driver. prop=value sets driver properties. Valid
833 properties depend on the driver. To get help on possible drivers and
834 properties, use ``-device help`` and ``-device driver,help``.
838 ``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]``
839 Add an IPMI BMC. This is a simulation of a hardware management
840 interface processor that normally sits on a system. It provides a
841 watchdog and the ability to reset and power control the system. You
842 need to connect this to an IPMI interface to make it useful
844 The IPMI slave address to use for the BMC. The default is 0x20. This
845 address is the BMC's address on the I2C network of management
846 controllers. If you don't know what this means, it is safe to ignore
850 The BMC id for interfaces to use this device.
853 Define slave address to use for the BMC. The default is 0x20.
856 file containing raw Sensor Data Records (SDR) data. The default
860 size of a Field Replaceable Unit (FRU) area. The default is
864 file containing raw Field Replaceable Unit (FRU) inventory data.
868 value for the GUID for the BMC, in standard UUID format. If this
869 is set, get "Get GUID" command to the BMC will return it.
870 Otherwise "Get GUID" will return an error.
872 ``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]``
873 Add a connection to an external IPMI BMC simulator. Instead of
874 locally emulating the BMC like the above item, instead connect to an
875 external entity that provides the IPMI services.
877 A connection is made to an external BMC simulator. If you do this,
878 it is strongly recommended that you use the "reconnect=" chardev
879 option to reconnect to the simulator if the connection is lost. Note
880 that if this is not used carefully, it can be a security issue, as
881 the interface has the ability to send resets, NMIs, and power off
882 the VM. It's best if QEMU makes a connection to an external
883 simulator running on a secure port on localhost, so neither the
884 simulator nor QEMU is exposed to any outside network.
886 See the "lanserv/README.vm" file in the OpenIPMI library for more
887 details on the external interface.
889 ``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]``
890 Add a KCS IPMI interafce on the ISA bus. This also adds a
891 corresponding ACPI and SMBIOS entries, if appropriate.
894 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern
898 Define the I/O address of the interface. The default is 0xca0
902 Define the interrupt to use. The default is 5. To disable
903 interrupts, set this to 0.
905 ``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]``
906 Like the KCS interface, but defines a BT interface. The default port
907 is 0xe4 and the default interrupt is 5.
909 ``-device pci-ipmi-kcs,bmc=id``
910 Add a KCS IPMI interafce on the PCI bus.
913 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above.
915 ``-device pci-ipmi-bt,bmc=id``
916 Like the KCS interface, but defines a BT interface on the PCI bus.
919 DEF("name", HAS_ARG, QEMU_OPTION_name,
920 "-name string1[,process=string2][,debug-threads=on|off]\n"
921 " set the name of the guest\n"
922 " string1 sets the window title and string2 the process name\n"
923 " When debug-threads is enabled, individual threads are given a separate name\n"
924 " NOTE: The thread names are for debugging and not a stable API.\n",
928 Sets the name of the guest. This name will be displayed in the SDL
929 window caption. The name will also be used for the VNC server. Also
930 optionally set the top visible process name in Linux. Naming of
931 individual threads can also be enabled on Linux to aid debugging.
934 DEF("uuid", HAS_ARG, QEMU_OPTION_uuid,
935 "-uuid %08x-%04x-%04x-%04x-%012x\n"
936 " specify machine UUID\n", QEMU_ARCH_ALL)
944 DEFHEADING(Block device options:)
946 DEF("fda", HAS_ARG, QEMU_OPTION_fda,
947 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
948 DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
953 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in
954 the System Emulation Users Guide).
957 DEF("hda", HAS_ARG, QEMU_OPTION_hda,
958 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL)
959 DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL)
960 DEF("hdc", HAS_ARG, QEMU_OPTION_hdc,
961 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL)
962 DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL)
971 Use file as hard disk 0, 1, 2 or 3 image (see the :ref:`disk images`
972 chapter in the System Emulation Users Guide).
975 DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom,
976 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n",
980 Use file as CD-ROM image (you cannot use ``-hdc`` and ``-cdrom`` at
981 the same time). You can use the host CD-ROM by using ``/dev/cdrom``
985 DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev,
986 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n"
987 " [,cache.direct=on|off][,cache.no-flush=on|off]\n"
988 " [,read-only=on|off][,auto-read-only=on|off]\n"
989 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n"
990 " [,driver specific parameters...]\n"
991 " configure a block backend\n", QEMU_ARCH_ALL)
993 ``-blockdev option[,option[,option[,...]]]``
994 Define a new block driver node. Some of the options apply to all
995 block drivers, other options are only accepted for a specific block
996 driver. See below for a list of generic options and options for the
997 most common block drivers.
999 Options that expect a reference to another node (e.g. ``file``) can
1000 be given in two ways. Either you specify the node name of an already
1001 existing node (file=node-name), or you define a new node inline,
1002 adding options for the referenced node after a dot
1003 (file.filename=path,file.aio=native).
1005 A block driver node created with ``-blockdev`` can be used for a
1006 guest device by specifying its node name for the ``drive`` property
1007 in a ``-device`` argument that defines a block device.
1009 ``Valid options for any block driver node:``
1011 Specifies the block driver to use for the given node.
1014 This defines the name of the block driver node by which it
1015 will be referenced later. The name must be unique, i.e. it
1016 must not match the name of a different block driver node, or
1017 (if you use ``-drive`` as well) the ID of a drive.
1019 If no node name is specified, it is automatically generated.
1020 The generated node name is not intended to be predictable
1021 and changes between QEMU invocations. For the top level, an
1022 explicit node name must be specified.
1025 Open the node read-only. Guest write attempts will fail.
1027 Note that some block drivers support only read-only access,
1028 either generally or in certain configurations. In this case,
1029 the default value ``read-only=off`` does not work and the
1030 option must be specified explicitly.
1033 If ``auto-read-only=on`` is set, QEMU may fall back to
1034 read-only usage even when ``read-only=off`` is requested, or
1035 even switch between modes as needed, e.g. depending on
1036 whether the image file is writable or whether a writing user
1037 is attached to the node.
1040 Override the image locking system of QEMU by forcing the
1041 node to utilize weaker shared access for permissions where
1042 it would normally request exclusive access. When there is
1043 the potential for multiple instances to have the same file
1044 open (whether this invocation of QEMU is the first or the
1045 second instance), both instances must permit shared access
1046 for the second instance to succeed at opening the file.
1048 Enabling ``force-share=on`` requires ``read-only=on``.
1051 The host page cache can be avoided with ``cache.direct=on``.
1052 This will attempt to do disk IO directly to the guest's
1053 memory. QEMU may still perform an internal copy of the data.
1056 In case you don't care about data integrity over host
1057 failures, you can use ``cache.no-flush=on``. This option
1058 tells QEMU that it never needs to write any data to the disk
1059 but can instead keep things in cache. If anything goes
1060 wrong, like your host losing power, the disk storage getting
1061 disconnected accidentally, etc. your image will most
1062 probably be rendered unusable.
1065 discard is one of "ignore" (or "off") or "unmap" (or "on")
1066 and controls whether ``discard`` (also known as ``trim`` or
1067 ``unmap``) requests are ignored or passed to the filesystem.
1068 Some machine types may not support discard requests.
1070 ``detect-zeroes=detect-zeroes``
1071 detect-zeroes is "off", "on" or "unmap" and enables the
1072 automatic conversion of plain zero writes by the OS to
1073 driver specific optimized zero write commands. You may even
1074 choose "unmap" if discard is set to "unmap" to allow a zero
1075 write to be converted to an ``unmap`` operation.
1077 ``Driver-specific options for file``
1078 This is the protocol-level block driver for accessing regular
1082 The path to the image file in the local filesystem
1085 Specifies the AIO backend (threads/native/io_uring,
1089 Specifies whether the image file is protected with Linux OFD
1090 / POSIX locks. The default is to use the Linux Open File
1091 Descriptor API if available, otherwise no lock is applied.
1092 (auto/on/off, default: auto)
1098 -blockdev driver=file,node-name=disk,filename=disk.img
1100 ``Driver-specific options for raw``
1101 This is the image format block driver for raw images. It is
1102 usually stacked on top of a protocol level block driver such as
1106 Reference to or definition of the data source block driver
1107 node (e.g. a ``file`` driver node)
1113 -blockdev driver=file,node-name=disk_file,filename=disk.img
1114 -blockdev driver=raw,node-name=disk,file=disk_file
1120 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img
1122 ``Driver-specific options for qcow2``
1123 This is the image format block driver for qcow2 images. It is
1124 usually stacked on top of a protocol level block driver such as
1128 Reference to or definition of the data source block driver
1129 node (e.g. a ``file`` driver node)
1132 Reference to or definition of the backing file block device
1133 (default is taken from the image file). It is allowed to
1134 pass ``null`` here in order to disable the default backing
1138 Whether to enable the lazy refcounts feature (on/off;
1139 default is taken from the image file)
1142 The maximum total size of the L2 table and refcount block
1143 caches in bytes (default: the sum of l2-cache-size and
1144 refcount-cache-size)
1147 The maximum size of the L2 table cache in bytes (default: if
1148 cache-size is not specified - 32M on Linux platforms, and 8M
1149 on non-Linux platforms; otherwise, as large as possible
1150 within the cache-size, while permitting the requested or the
1151 minimal refcount cache size)
1153 ``refcount-cache-size``
1154 The maximum size of the refcount block cache in bytes
1155 (default: 4 times the cluster size; or if cache-size is
1156 specified, the part of it which is not used for the L2
1159 ``cache-clean-interval``
1160 Clean unused entries in the L2 and refcount caches. The
1161 interval is in seconds. The default value is 600 on
1162 supporting platforms, and 0 on other platforms. Setting it
1163 to 0 disables this feature.
1165 ``pass-discard-request``
1166 Whether discard requests to the qcow2 device should be
1167 forwarded to the data source (on/off; default: on if
1168 discard=unmap is specified, off otherwise)
1170 ``pass-discard-snapshot``
1171 Whether discard requests for the data source should be
1172 issued when a snapshot operation (e.g. deleting a snapshot)
1173 frees clusters in the qcow2 file (on/off; default: on)
1175 ``pass-discard-other``
1176 Whether discard requests for the data source should be
1177 issued on other occasions where a cluster gets freed
1178 (on/off; default: off)
1181 Which overlap checks to perform for writes to the image
1182 (none/constant/cached/all; default: cached). For details or
1183 finer granularity control refer to the QAPI documentation of
1190 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2
1191 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216
1197 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2
1199 ``Driver-specific options for other drivers``
1200 Please refer to the QAPI documentation of the ``blockdev-add``
1204 DEF("drive", HAS_ARG, QEMU_OPTION_drive,
1205 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n"
1206 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n"
1207 " [,snapshot=on|off][,rerror=ignore|stop|report]\n"
1208 " [,werror=ignore|stop|report|enospc][,id=name]\n"
1209 " [,aio=threads|native|io_uring]\n"
1210 " [,readonly=on|off][,copy-on-read=on|off]\n"
1211 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n"
1212 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n"
1213 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n"
1214 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n"
1215 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n"
1216 " [[,iops_size=is]]\n"
1218 " use 'file' as a drive image\n", QEMU_ARCH_ALL)
1220 ``-drive option[,option[,option[,...]]]``
1221 Define a new drive. This includes creating a block driver node (the
1222 backend) as well as a guest device, and is mostly a shortcut for
1223 defining the corresponding ``-blockdev`` and ``-device`` options.
1225 ``-drive`` accepts all options that are accepted by ``-blockdev``.
1226 In addition, it knows the following options:
1229 This option defines which disk image (see the :ref:`disk images`
1230 chapter in the System Emulation Users Guide) to use with this drive.
1231 If the filename contains comma, you must double it (for instance,
1232 "file=my,,file" to use file "my,file").
1234 Special files such as iSCSI devices can be specified using
1235 protocol specific URLs. See the section for "Device URL Syntax"
1236 for more information.
1239 This option defines on which type on interface the drive is
1240 connected. Available types are: ide, scsi, sd, mtd, floppy,
1241 pflash, virtio, none.
1243 ``bus=bus,unit=unit``
1244 These options define where is connected the drive by defining
1245 the bus number and the unit id.
1248 This option defines where is connected the drive by using an
1249 index in the list of available connectors of a given interface
1253 This option defines the type of the media: disk or cdrom.
1255 ``snapshot=snapshot``
1256 snapshot is "on" or "off" and controls snapshot mode for the
1257 given drive (see ``-snapshot``).
1260 cache is "none", "writeback", "unsafe", "directsync" or
1261 "writethrough" and controls how the host cache is used to access
1262 block data. This is a shortcut that sets the ``cache.direct``
1263 and ``cache.no-flush`` options (as in ``-blockdev``), and
1264 additionally ``cache.writeback``, which provides a default for
1265 the ``write-cache`` option of block guest devices (as in
1266 ``-device``). The modes correspond to the following settings:
1268 ============= =============== ============ ==============
1269 \ cache.writeback cache.direct cache.no-flush
1270 ============= =============== ============ ==============
1271 writeback on off off
1273 writethrough off off off
1274 directsync off on off
1276 ============= =============== ============ ==============
1278 The default mode is ``cache=writeback``.
1281 aio is "threads", "native", or "io_uring" and selects between pthread
1282 based disk I/O, native Linux AIO, or Linux io_uring API.
1285 Specify which disk format will be used rather than detecting the
1286 format. Can be used to specify format=raw to avoid interpreting
1287 an untrusted format header.
1289 ``werror=action,rerror=action``
1290 Specify which action to take on write and read errors. Valid
1291 actions are: "ignore" (ignore the error and try to continue),
1292 "stop" (pause QEMU), "report" (report the error to the guest),
1293 "enospc" (pause QEMU only if the host disk is full; report the
1294 error to the guest otherwise). The default setting is
1295 ``werror=enospc`` and ``rerror=report``.
1297 ``copy-on-read=copy-on-read``
1298 copy-on-read is "on" or "off" and enables whether to copy read
1299 backing file sectors into the image file.
1301 ``bps=b,bps_rd=r,bps_wr=w``
1302 Specify bandwidth throttling limits in bytes per second, either
1303 for all request types or for reads or writes only. Small values
1304 can lead to timeouts or hangs inside the guest. A safe minimum
1305 for disks is 2 MB/s.
1307 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm``
1308 Specify bursts in bytes per second, either for all request types
1309 or for reads or writes only. Bursts allow the guest I/O to spike
1310 above the limit temporarily.
1312 ``iops=i,iops_rd=r,iops_wr=w``
1313 Specify request rate limits in requests per second, either for
1314 all request types or for reads or writes only.
1316 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm``
1317 Specify bursts in requests per second, either for all request
1318 types or for reads or writes only. Bursts allow the guest I/O to
1319 spike above the limit temporarily.
1322 Let every is bytes of a request count as a new request for iops
1323 throttling purposes. Use this option to prevent guests from
1324 circumventing iops limits by sending fewer but larger requests.
1327 Join a throttling quota group with given name g. All drives that
1328 are members of the same group are accounted for together. Use
1329 this option to prevent guests from circumventing throttling
1330 limits by using many small disks instead of a single larger
1333 By default, the ``cache.writeback=on`` mode is used. It will report
1334 data writes as completed as soon as the data is present in the host
1335 page cache. This is safe as long as your guest OS makes sure to
1336 correctly flush disk caches where needed. If your guest OS does not
1337 handle volatile disk write caches correctly and your host crashes or
1338 loses power, then the guest may experience data corruption.
1340 For such guests, you should consider using ``cache.writeback=off``.
1341 This means that the host page cache will be used to read and write
1342 data, but write notification will be sent to the guest only after
1343 QEMU has made sure to flush each write to the disk. Be aware that
1344 this has a major impact on performance.
1346 When using the ``-snapshot`` option, unsafe caching is always used.
1348 Copy-on-read avoids accessing the same backing file sectors
1349 repeatedly and is useful when the backing file is over a slow
1350 network. By default copy-on-read is off.
1352 Instead of ``-cdrom`` you can use:
1356 |qemu_system| -drive file=file,index=2,media=cdrom
1358 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use:
1362 |qemu_system| -drive file=file,index=0,media=disk
1363 |qemu_system| -drive file=file,index=1,media=disk
1364 |qemu_system| -drive file=file,index=2,media=disk
1365 |qemu_system| -drive file=file,index=3,media=disk
1367 You can open an image using pre-opened file descriptors from an fd
1373 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
1374 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
1375 -drive file=/dev/fdset/2,index=0,media=disk
1377 You can connect a CDROM to the slave of ide0:
1381 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom
1383 If you don't specify the "file=" argument, you define an empty
1388 |qemu_system_x86| -drive if=ide,index=1,media=cdrom
1390 Instead of ``-fda``, ``-fdb``, you can use:
1394 |qemu_system_x86| -drive file=file,index=0,if=floppy
1395 |qemu_system_x86| -drive file=file,index=1,if=floppy
1397 By default, interface is "ide" and index is automatically
1402 |qemu_system_x86| -drive file=a -drive file=b"
1404 is interpreted like:
1408 |qemu_system_x86| -hda a -hdb b
1411 DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock,
1412 "-mtdblock file use 'file' as on-board Flash memory image\n",
1416 Use file as on-board Flash memory image.
1419 DEF("sd", HAS_ARG, QEMU_OPTION_sd,
1420 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL)
1423 Use file as SecureDigital card image.
1426 DEF("pflash", HAS_ARG, QEMU_OPTION_pflash,
1427 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL)
1430 Use file as a parallel flash image.
1433 DEF("snapshot", 0, QEMU_OPTION_snapshot,
1434 "-snapshot write to temporary files instead of disk image files\n",
1438 Write to temporary files instead of disk image files. In this case,
1439 the raw disk image you use is not written back. You can however
1440 force the write back by pressing C-a s (see the :ref:`disk images`
1441 chapter in the System Emulation Users Guide).
1444 DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev,
1445 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1446 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n"
1447 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n"
1448 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n"
1449 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n"
1450 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n"
1451 " [[,throttling.iops-size=is]]\n"
1452 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n"
1453 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n"
1454 "-fsdev synth,id=id\n",
1458 ``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]``
1460 ``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]``
1462 ``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]``
1464 ``-fsdev synth,id=id[,readonly=on]``
1465 Define a new file system device. Valid options are:
1468 Accesses to the filesystem are done by QEMU.
1471 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1474 Synthetic filesystem, only used by QTests.
1477 Specifies identifier for this device.
1480 Specifies the export path for the file system device. Files
1481 under this path will be available to the 9p client on the guest.
1483 ``security_model=security_model``
1484 Specifies the security model to be used for this export path.
1485 Supported security models are "passthrough", "mapped-xattr",
1486 "mapped-file" and "none". In "passthrough" security model, files
1487 are stored using the same credentials as they are created on the
1488 guest. This requires QEMU to run as root. In "mapped-xattr"
1489 security model, some of the file attributes like uid, gid, mode
1490 bits and link target are stored as file attributes. For
1491 "mapped-file" these attributes are stored in the hidden
1492 .virtfs\_metadata directory. Directories exported by this
1493 security model cannot interact with other unix tools. "none"
1494 security model is same as passthrough except the sever won't
1495 report failures if it fails to set file attributes like
1496 ownership. Security model is mandatory only for local fsdriver.
1497 Other fsdrivers (like proxy) don't take security model as a
1500 ``writeout=writeout``
1501 This is an optional argument. The only supported value is
1502 "immediate". This means that host page cache will be used to
1503 read and write data but write notification will be sent to the
1504 guest only when the data has been reported as written by the
1508 Enables exporting 9p share as a readonly mount for guests. By
1509 default read-write access is given.
1512 Enables proxy filesystem driver to use passed socket file for
1513 communicating with virtfs-proxy-helper(1).
1516 Enables proxy filesystem driver to use passed socket descriptor
1517 for communicating with virtfs-proxy-helper(1). Usually a helper
1518 like libvirt will create socketpair and pass one of the fds as
1522 Specifies the default mode for newly created files on the host.
1523 Works only with security models "mapped-xattr" and
1527 Specifies the default mode for newly created directories on the
1528 host. Works only with security models "mapped-xattr" and
1531 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w``
1532 Specify bandwidth throttling limits in bytes per second, either
1533 for all request types or for reads or writes only.
1535 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm``
1536 Specify bursts in bytes per second, either for all request types
1537 or for reads or writes only. Bursts allow the guest I/O to spike
1538 above the limit temporarily.
1540 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w``
1541 Specify request rate limits in requests per second, either for
1542 all request types or for reads or writes only.
1544 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm``
1545 Specify bursts in requests per second, either for all request
1546 types or for reads or writes only. Bursts allow the guest I/O to
1547 spike above the limit temporarily.
1549 ``throttling.iops-size=is``
1550 Let every is bytes of a request count as a new request for iops
1551 throttling purposes.
1553 -fsdev option is used along with -device driver "virtio-9p-...".
1555 ``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag``
1556 Options for virtio-9p-... driver are:
1559 Specifies the variant to be used. Supported values are "pci",
1560 "ccw" or "device", depending on the machine type.
1563 Specifies the id value specified along with -fsdev option.
1565 ``mount_tag=mount_tag``
1566 Specifies the tag name to be used by the guest to mount this
1570 DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs,
1571 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1572 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n"
1573 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n"
1574 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n"
1575 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n",
1579 ``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]``
1581 ``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1583 ``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1585 ``-virtfs synth,mount_tag=mount_tag``
1586 Define a new virtual filesystem device and expose it to the guest using
1587 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain
1588 directory on host is made directly accessible by guest as a pass-through
1589 file system by using the 9P network protocol for communication between
1590 host and guests, if desired even accessible, shared by several guests
1593 Note that ``-virtfs`` is actually just a convenience shortcut for its
1594 generalized form ``-fsdev -device virtio-9p-pci``.
1596 The general form of pass-through file system options are:
1599 Accesses to the filesystem are done by QEMU.
1602 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1605 Synthetic filesystem, only used by QTests.
1608 Specifies identifier for the filesystem device
1611 Specifies the export path for the file system device. Files
1612 under this path will be available to the 9p client on the guest.
1614 ``security_model=security_model``
1615 Specifies the security model to be used for this export path.
1616 Supported security models are "passthrough", "mapped-xattr",
1617 "mapped-file" and "none". In "passthrough" security model, files
1618 are stored using the same credentials as they are created on the
1619 guest. This requires QEMU to run as root. In "mapped-xattr"
1620 security model, some of the file attributes like uid, gid, mode
1621 bits and link target are stored as file attributes. For
1622 "mapped-file" these attributes are stored in the hidden
1623 .virtfs\_metadata directory. Directories exported by this
1624 security model cannot interact with other unix tools. "none"
1625 security model is same as passthrough except the sever won't
1626 report failures if it fails to set file attributes like
1627 ownership. Security model is mandatory only for local fsdriver.
1628 Other fsdrivers (like proxy) don't take security model as a
1631 ``writeout=writeout``
1632 This is an optional argument. The only supported value is
1633 "immediate". This means that host page cache will be used to
1634 read and write data but write notification will be sent to the
1635 guest only when the data has been reported as written by the
1639 Enables exporting 9p share as a readonly mount for guests. By
1640 default read-write access is given.
1643 Enables proxy filesystem driver to use passed socket file for
1644 communicating with virtfs-proxy-helper(1). Usually a helper like
1645 libvirt will create socketpair and pass one of the fds as
1649 Enables proxy filesystem driver to use passed 'sock\_fd' as the
1650 socket descriptor for interfacing with virtfs-proxy-helper(1).
1653 Specifies the default mode for newly created files on the host.
1654 Works only with security models "mapped-xattr" and
1658 Specifies the default mode for newly created directories on the
1659 host. Works only with security models "mapped-xattr" and
1662 ``mount_tag=mount_tag``
1663 Specifies the tag name to be used by the guest to mount this
1666 ``multidevs=multidevs``
1667 Specifies how to deal with multiple devices being shared with a
1668 9p export. Supported behaviours are either "remap", "forbid" or
1669 "warn". The latter is the default behaviour on which virtfs 9p
1670 expects only one device to be shared with the same export, and
1671 if more than one device is shared and accessed via the same 9p
1672 export then only a warning message is logged (once) by qemu on
1673 host side. In order to avoid file ID collisions on guest you
1674 should either create a separate virtfs export for each device to
1675 be shared with guests (recommended way) or you might use "remap"
1676 instead which allows you to share multiple devices with only one
1677 export instead, which is achieved by remapping the original
1678 inode numbers from host to guest in a way that would prevent
1679 such collisions. Remapping inodes in such use cases is required
1680 because the original device IDs from host are never passed and
1681 exposed on guest. Instead all files of an export shared with
1682 virtfs always share the same device id on guest. So two files
1683 with identical inode numbers but from actually different devices
1684 on host would otherwise cause a file ID collision and hence
1685 potential misbehaviours on guest. "forbid" on the other hand
1686 assumes like "warn" that only one device is shared by the same
1687 export, however it will not only log a warning message but also
1688 deny access to additional devices on guest. Note though that
1689 "forbid" does currently not block all possible file access
1690 operations (e.g. readdir() would still return entries from other
1694 DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi,
1695 "-iscsi [user=user][,password=password]\n"
1696 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE\n"
1697 " [,initiator-name=initiator-iqn][,id=target-iqn]\n"
1698 " [,timeout=timeout]\n"
1699 " iSCSI session parameters\n", QEMU_ARCH_ALL)
1703 Configure iSCSI session parameters.
1708 DEFHEADING(USB convenience options:)
1710 DEF("usb", 0, QEMU_OPTION_usb,
1711 "-usb enable on-board USB host controller (if not enabled by default)\n",
1715 Enable USB emulation on machine types with an on-board USB host
1716 controller (if not enabled by default). Note that on-board USB host
1717 controllers may not support USB 3.0. In this case
1718 ``-device qemu-xhci`` can be used instead on machines with PCI.
1721 DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice,
1722 "-usbdevice name add the host or guest USB device 'name'\n",
1725 ``-usbdevice devname``
1726 Add the USB device devname, and enable an on-board USB controller
1727 if possible and necessary (just like it can be done via
1728 ``-machine usb=on``). Note that this option is mainly intended for
1729 the user's convenience only. More fine-grained control can be
1730 achieved by selecting a USB host controller (if necessary) and the
1731 desired USB device via the ``-device`` option instead. For example,
1732 instead of using ``-usbdevice mouse`` it is possible to use
1733 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse
1734 to a USB 3.0 controller instead (at least on machines that support
1735 PCI and do not have an USB controller enabled by default yet).
1736 For more details, see the chapter about
1737 :ref:`Connecting USB devices` in the System Emulation Users Guide.
1738 Possible devices for devname are:
1741 Braille device. This will use BrlAPI to display the braille
1742 output on a real or fake device (i.e. it also creates a
1743 corresponding ``braille`` chardev automatically beside the
1744 ``usb-braille`` USB device).
1747 Standard USB keyboard. Will override the PS/2 keyboard (if present).
1750 Virtual Mouse. This will override the PS/2 mouse emulation when
1754 Pointer device that uses absolute coordinates (like a
1755 touchscreen). This means QEMU is able to report the mouse
1756 position without having to grab the mouse. Also overrides the
1757 PS/2 mouse emulation when activated.
1760 Wacom PenPartner USB tablet.
1767 DEFHEADING(Display options:)
1769 DEF("display", HAS_ARG, QEMU_OPTION_display,
1770 #if defined(CONFIG_SPICE)
1771 "-display spice-app[,gl=on|off]\n"
1773 #if defined(CONFIG_SDL)
1774 "-display sdl[,alt_grab=on|off][,ctrl_grab=on|off]\n"
1775 " [,window_close=on|off][,gl=on|core|es|off]\n"
1777 #if defined(CONFIG_GTK)
1778 "-display gtk[,grab_on_hover=on|off][,gl=on|off]|\n"
1780 #if defined(CONFIG_VNC)
1781 "-display vnc=<display>[,<optargs>]\n"
1783 #if defined(CONFIG_CURSES)
1784 "-display curses[,charset=<encoding>]\n"
1786 #if defined(CONFIG_OPENGL)
1787 "-display egl-headless[,rendernode=<file>]\n"
1790 " select display backend type\n"
1791 " The default display is equivalent to\n "
1792 #if defined(CONFIG_GTK)
1793 "\"-display gtk\"\n"
1794 #elif defined(CONFIG_SDL)
1795 "\"-display sdl\"\n"
1796 #elif defined(CONFIG_COCOA)
1797 "\"-display cocoa\"\n"
1798 #elif defined(CONFIG_VNC)
1799 "\"-vnc localhost:0,to=99,id=default\"\n"
1801 "\"-display none\"\n"
1806 Select type of display to use. This option is a replacement for the
1807 old style -sdl/-curses/... options. Use ``-display help`` to list
1808 the available display types. Valid values for type are
1811 Display video output via SDL (usually in a separate graphics
1812 window; see the SDL documentation for other possibilities).
1815 Display video output via curses. For graphics device models
1816 which support a text mode, QEMU can display this output using a
1817 curses/ncurses interface. Nothing is displayed when the graphics
1818 device is in graphical mode or if the graphics device does not
1819 support a text mode. Generally only the VGA device models
1820 support text mode. The font charset used by the guest can be
1821 specified with the ``charset`` option, for example
1822 ``charset=CP850`` for IBM CP850 encoding. The default is
1826 Do not display video output. The guest will still see an
1827 emulated graphics card, but its output will not be displayed to
1828 the QEMU user. This option differs from the -nographic option in
1829 that it only affects what is done with video output; -nographic
1830 also changes the destination of the serial and parallel port
1834 Display video output in a GTK window. This interface provides
1835 drop-down menus and other UI elements to configure and control
1836 the VM during runtime.
1839 Start a VNC server on display <arg>
1842 Offload all OpenGL operations to a local DRI device. For any
1843 graphical display, this display needs to be paired with either
1844 VNC or SPICE displays.
1847 Start QEMU as a Spice server and launch the default Spice client
1848 application. The Spice server will redirect the serial consoles
1849 and QEMU monitors. (Since 4.0)
1852 DEF("nographic", 0, QEMU_OPTION_nographic,
1853 "-nographic disable graphical output and redirect serial I/Os to console\n",
1857 Normally, if QEMU is compiled with graphical window support, it
1858 displays output such as guest graphics, guest console, and the QEMU
1859 monitor in a window. With this option, you can totally disable
1860 graphical output so that QEMU is a simple command line application.
1861 The emulated serial port is redirected on the console and muxed with
1862 the monitor (unless redirected elsewhere explicitly). Therefore, you
1863 can still use QEMU to debug a Linux kernel with a serial console.
1864 Use C-a h for help on switching between the console and monitor.
1867 DEF("curses", 0, QEMU_OPTION_curses,
1868 "-curses shorthand for -display curses\n",
1872 Normally, if QEMU is compiled with graphical window support, it
1873 displays output such as guest graphics, guest console, and the QEMU
1874 monitor in a window. With this option, QEMU can display the VGA
1875 output when in text mode using a curses/ncurses interface. Nothing
1876 is displayed in graphical mode.
1879 DEF("alt-grab", 0, QEMU_OPTION_alt_grab,
1880 "-alt-grab use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt)\n",
1884 Use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt). Note that
1885 this also affects the special keys (for fullscreen, monitor-mode
1889 DEF("ctrl-grab", 0, QEMU_OPTION_ctrl_grab,
1890 "-ctrl-grab use Right-Ctrl to grab mouse (instead of Ctrl-Alt)\n",
1894 Use Right-Ctrl to grab mouse (instead of Ctrl-Alt). Note that this
1895 also affects the special keys (for fullscreen, monitor-mode
1899 DEF("no-quit", 0, QEMU_OPTION_no_quit,
1900 "-no-quit disable SDL window close capability\n", QEMU_ARCH_ALL)
1903 Disable SDL window close capability.
1906 DEF("sdl", 0, QEMU_OPTION_sdl,
1907 "-sdl shorthand for -display sdl\n", QEMU_ARCH_ALL)
1913 DEF("spice", HAS_ARG, QEMU_OPTION_spice,
1914 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n"
1915 " [,x509-key-file=<file>][,x509-key-password=<file>]\n"
1916 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n"
1917 " [,x509-dh-key-file=<file>][,addr=addr]\n"
1918 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n"
1919 " [,tls-ciphers=<list>]\n"
1920 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n"
1921 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n"
1922 " [,sasl=on|off][,disable-ticketing=on|off]\n"
1923 " [,password=<string>][,password-secret=<secret-id>]\n"
1924 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n"
1925 " [,jpeg-wan-compression=[auto|never|always]]\n"
1926 " [,zlib-glz-wan-compression=[auto|never|always]]\n"
1927 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n"
1928 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n"
1929 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n"
1930 " [,gl=[on|off]][,rendernode=<file>]\n"
1932 " at least one of {port, tls-port} is mandatory\n",
1935 ``-spice option[,option[,...]]``
1936 Enable the spice remote desktop protocol. Valid options are
1939 Set the TCP port spice is listening on for plaintext channels.
1942 Set the IP address spice is listening on. Default is any
1945 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off``
1946 Force using the specified IP version.
1948 ``password=<string>``
1949 Set the password you need to authenticate.
1951 This option is deprecated and insecure because it leaves the
1952 password visible in the process listing. Use ``password-secret``
1955 ``password-secret=<secret-id>``
1956 Set the ID of the ``secret`` object containing the password
1957 you need to authenticate.
1960 Require that the client use SASL to authenticate with the spice.
1961 The exact choice of authentication method used is controlled
1962 from the system / user's SASL configuration file for the 'qemu'
1963 service. This is typically found in /etc/sasl2/qemu.conf. If
1964 running QEMU as an unprivileged user, an environment variable
1965 SASL\_CONF\_PATH can be used to make it search alternate
1966 locations for the service config. While some SASL auth methods
1967 can also provide data encryption (eg GSSAPI), it is recommended
1968 that SASL always be combined with the 'tls' and 'x509' settings
1969 to enable use of SSL and server certificates. This ensures a
1970 data encryption preventing compromise of authentication
1973 ``disable-ticketing=on|off``
1974 Allow client connects without authentication.
1976 ``disable-copy-paste=on|off``
1977 Disable copy paste between the client and the guest.
1979 ``disable-agent-file-xfer=on|off``
1980 Disable spice-vdagent based file-xfer between the client and the
1984 Set the TCP port spice is listening on for encrypted channels.
1987 Set the x509 file directory. Expects same filenames as -vnc
1990 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>``
1991 The x509 file names can also be configured individually.
1993 ``tls-ciphers=<list>``
1994 Specify which ciphers to use.
1996 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]``
1997 Force specific channel to be used with or without TLS
1998 encryption. The options can be specified multiple times to
1999 configure multiple channels. The special name "default" can be
2000 used to set the default mode. For channels which are not
2001 explicitly forced into one mode the spice client is allowed to
2002 pick tls/plaintext as he pleases.
2004 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]``
2005 Configure image compression (lossless). Default is auto\_glz.
2007 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]``
2008 Configure wan image compression (lossy for slow links). Default
2011 ``streaming-video=[off|all|filter]``
2012 Configure video stream detection. Default is off.
2014 ``agent-mouse=[on|off]``
2015 Enable/disable passing mouse events via vdagent. Default is on.
2017 ``playback-compression=[on|off]``
2018 Enable/disable audio stream compression (using celt 0.5.1).
2021 ``seamless-migration=[on|off]``
2022 Enable/disable spice seamless migration. Default is off.
2025 Enable/disable OpenGL context. Default is off.
2027 ``rendernode=<file>``
2028 DRM render node for OpenGL rendering. If not specified, it will
2029 pick the first available. (Since 2.9)
2032 DEF("portrait", 0, QEMU_OPTION_portrait,
2033 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n",
2037 Rotate graphical output 90 deg left (only PXA LCD).
2040 DEF("rotate", HAS_ARG, QEMU_OPTION_rotate,
2041 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n",
2045 Rotate graphical output some deg left (only PXA LCD).
2048 DEF("vga", HAS_ARG, QEMU_OPTION_vga,
2049 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n"
2050 " select video card type\n", QEMU_ARCH_ALL)
2053 Select type of VGA card to emulate. Valid values for type are
2056 Cirrus Logic GD5446 Video card. All Windows versions starting
2057 from Windows 95 should recognize and use this graphic card. For
2058 optimal performances, use 16 bit color depth in the guest and
2059 the host OS. (This card was the default before QEMU 2.2)
2062 Standard VGA card with Bochs VBE extensions. If your guest OS
2063 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if
2064 you want to use high resolution modes (>= 1280x1024x16) then you
2065 should use this option. (This card is the default since QEMU
2069 VMWare SVGA-II compatible adapter. Use it if you have
2070 sufficiently recent XFree86/XOrg server or Windows guest with a
2071 driver for this card.
2074 QXL paravirtual graphic card. It is VGA compatible (including
2075 VESA 2.0 VBE support). Works best with qxl guest drivers
2076 installed though. Recommended choice when using the spice
2080 (sun4m only) Sun TCX framebuffer. This is the default
2081 framebuffer for sun4m machines and offers both 8-bit and 24-bit
2082 colour depths at a fixed resolution of 1024x768.
2085 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit
2086 framebuffer for sun4m machines available in both 1024x768
2087 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people
2088 wishing to run older Solaris versions.
2097 DEF("full-screen", 0, QEMU_OPTION_full_screen,
2098 "-full-screen start in full screen\n", QEMU_ARCH_ALL)
2101 Start in full screen.
2104 DEF("g", HAS_ARG, QEMU_OPTION_g ,
2105 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n",
2106 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K)
2108 ``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]``
2109 Set the initial graphical resolution and depth (PPC, SPARC only).
2111 For PPC the default is 800x600x32.
2113 For SPARC with the TCX graphics device, the default is 1024x768x8
2114 with the option of 1024x768x24. For cgthree, the default is
2115 1024x768x8 with the option of 1152x900x8 for people who wish to use
2119 DEF("vnc", HAS_ARG, QEMU_OPTION_vnc ,
2120 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL)
2122 ``-vnc display[,option[,option[,...]]]``
2123 Normally, if QEMU is compiled with graphical window support, it
2124 displays output such as guest graphics, guest console, and the QEMU
2125 monitor in a window. With this option, you can have QEMU listen on
2126 VNC display display and redirect the VGA display over the VNC
2127 session. It is very useful to enable the usb tablet device when
2128 using this option (option ``-device usb-tablet``). When using the
2129 VNC display, you must use the ``-k`` parameter to set the keyboard
2130 layout if you are not using en-us. Valid syntax for the display is
2133 With this option, QEMU will try next available VNC displays,
2134 until the number L, if the origianlly defined "-vnc display" is
2135 not available, e.g. port 5900+display is already used by another
2136 application. By default, to=0.
2139 TCP connections will only be allowed from host on display d. By
2140 convention the TCP port is 5900+d. Optionally, host can be
2141 omitted in which case the server will accept connections from
2145 Connections will be allowed over UNIX domain sockets where path
2146 is the location of a unix socket to listen for connections on.
2149 VNC is initialized but not started. The monitor ``change``
2150 command can be used to later start the VNC server.
2152 Following the display value there may be one or more option flags
2153 separated by commas. Valid options are
2156 Connect to a listening VNC client via a "reverse" connection.
2157 The client is specified by the display. For reverse network
2158 connections (host:d,``reverse``), the d argument is a TCP port
2159 number, not a display number.
2161 ``websocket=on|off``
2162 Opens an additional TCP listening port dedicated to VNC
2163 Websocket connections. If a bare websocket option is given, the
2164 Websocket port is 5700+display. An alternative port can be
2165 specified with the syntax ``websocket``\ =port.
2167 If host is specified connections will only be allowed from this
2168 host. It is possible to control the websocket listen address
2169 independently, using the syntax ``websocket``\ =host:port.
2171 If no TLS credentials are provided, the websocket connection
2172 runs in unencrypted mode. If TLS credentials are provided, the
2173 websocket connection requires encrypted client connections.
2176 Require that password based authentication is used for client
2179 The password must be set separately using the ``set_password``
2180 command in the :ref:`QEMU monitor`. The
2181 syntax to change your password is:
2182 ``set_password <protocol> <password>`` where <protocol> could be
2183 either "vnc" or "spice".
2185 If you would like to change <protocol> password expiration, you
2186 should use ``expire_password <protocol> <expiration-time>``
2187 where expiration time could be one of the following options:
2188 now, never, +seconds or UNIX time of expiration, e.g. +60 to
2189 make password expire in 60 seconds, or 1335196800 to make
2190 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for
2191 this date and time).
2193 You can also use keywords "now" or "never" for the expiration
2194 time to allow <protocol> password to expire immediately or never
2197 ``password-secret=<secret-id>``
2198 Require that password based authentication is used for client
2199 connections, using the password provided by the ``secret``
2200 object identified by ``secret-id``.
2203 Provides the ID of a set of TLS credentials to use to secure the
2204 VNC server. They will apply to both the normal VNC server socket
2205 and the websocket socket (if enabled). Setting TLS credentials
2206 will cause the VNC server socket to enable the VeNCrypt auth
2207 mechanism. The credentials should have been previously created
2208 using the ``-object tls-creds`` argument.
2211 Provides the ID of the QAuthZ authorization object against which
2212 the client's x509 distinguished name will validated. This object
2213 is only resolved at time of use, so can be deleted and recreated
2214 on the fly while the VNC server is active. If missing, it will
2215 default to denying access.
2218 Require that the client use SASL to authenticate with the VNC
2219 server. The exact choice of authentication method used is
2220 controlled from the system / user's SASL configuration file for
2221 the 'qemu' service. This is typically found in
2222 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user,
2223 an environment variable SASL\_CONF\_PATH can be used to make it
2224 search alternate locations for the service config. While some
2225 SASL auth methods can also provide data encryption (eg GSSAPI),
2226 it is recommended that SASL always be combined with the 'tls'
2227 and 'x509' settings to enable use of SSL and server
2228 certificates. This ensures a data encryption preventing
2229 compromise of authentication credentials. See the
2230 :ref:`VNC security` section in the System Emulation Users Guide
2231 for details on using SASL authentication.
2234 Provides the ID of the QAuthZ authorization object against which
2235 the client's SASL username will validated. This object is only
2236 resolved at time of use, so can be deleted and recreated on the
2237 fly while the VNC server is active. If missing, it will default
2241 Legacy method for enabling authorization of clients against the
2242 x509 distinguished name and SASL username. It results in the
2243 creation of two ``authz-list`` objects with IDs of
2244 ``vnc.username`` and ``vnc.x509dname``. The rules for these
2245 objects must be configured with the HMP ACL commands.
2247 This option is deprecated and should no longer be used. The new
2248 ``sasl-authz`` and ``tls-authz`` options are a replacement.
2251 Enable lossy compression methods (gradient, JPEG, ...). If this
2252 option is set, VNC client may receive lossy framebuffer updates
2253 depending on its encoding settings. Enabling this option can
2254 save a lot of bandwidth at the expense of quality.
2256 ``non-adaptive=on|off``
2257 Disable adaptive encodings. Adaptive encodings are enabled by
2258 default. An adaptive encoding will try to detect frequently
2259 updated screen regions, and send updates in these regions using
2260 a lossy encoding (like JPEG). This can be really helpful to save
2261 bandwidth when playing videos. Disabling adaptive encodings
2262 restores the original static behavior of encodings like Tight.
2264 ``share=[allow-exclusive|force-shared|ignore]``
2265 Set display sharing policy. 'allow-exclusive' allows clients to
2266 ask for exclusive access. As suggested by the rfb spec this is
2267 implemented by dropping other connections. Connecting multiple
2268 clients in parallel requires all clients asking for a shared
2269 session (vncviewer: -shared switch). This is the default.
2270 'force-shared' disables exclusive client access. Useful for
2271 shared desktop sessions, where you don't want someone forgetting
2272 specify -shared disconnect everybody else. 'ignore' completely
2273 ignores the shared flag and allows everybody connect
2274 unconditionally. Doesn't conform to the rfb spec but is
2275 traditional QEMU behavior.
2278 Set keyboard delay, for key down and key up events, in
2279 milliseconds. Default is 10. Keyboards are low-bandwidth
2280 devices, so this slowdown can help the device and guest to keep
2281 up and not lose events in case events are arriving in bulk.
2282 Possible causes for the latter are flaky network connections, or
2283 scripts for automated testing.
2285 ``audiodev=audiodev``
2286 Use the specified audiodev when the VNC client requests audio
2287 transmission. When not using an -audiodev argument, this option
2288 must be omitted, otherwise is must be present and specify a
2291 ``power-control=on|off``
2292 Permit the remote client to issue shutdown, reboot or reset power
2296 ARCHHEADING(, QEMU_ARCH_I386)
2298 ARCHHEADING(i386 target only:, QEMU_ARCH_I386)
2300 DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack,
2301 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n",
2305 Use it when installing Windows 2000 to avoid a disk full bug. After
2306 Windows 2000 is installed, you no longer need this option (this
2307 option slows down the IDE transfers).
2310 DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk,
2311 "-no-fd-bootchk disable boot signature checking for floppy disks\n",
2315 Disable boot signature checking for floppy disks in BIOS. May be
2316 needed to boot from old floppy disks.
2319 DEF("no-acpi", 0, QEMU_OPTION_no_acpi,
2320 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM)
2323 Disable ACPI (Advanced Configuration and Power Interface) support.
2324 Use it if your guest OS complains about ACPI problems (PC target
2328 DEF("no-hpet", 0, QEMU_OPTION_no_hpet,
2329 "-no-hpet disable HPET\n", QEMU_ARCH_I386)
2332 Disable HPET support.
2335 DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable,
2336 "-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"
2337 " ACPI table description\n", QEMU_ARCH_I386)
2339 ``-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]...]``
2340 Add ACPI table with specified header fields and context from
2341 specified files. For file=, take whole ACPI table from the specified
2342 files, including all ACPI headers (possible overridden by other
2343 options). For data=, only data portion of the table is used, all
2344 header information is specified in the command line. If a SLIC table
2345 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id
2346 fields will override the same in the RSDT and the FADT (a.k.a.
2347 FACP), in order to ensure the field matches required by the
2348 Microsoft SLIC spec and the ACPI spec.
2351 DEF("smbios", HAS_ARG, QEMU_OPTION_smbios,
2352 "-smbios file=binary\n"
2353 " load SMBIOS entry from binary file\n"
2354 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n"
2356 " specify SMBIOS type 0 fields\n"
2357 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2358 " [,uuid=uuid][,sku=str][,family=str]\n"
2359 " specify SMBIOS type 1 fields\n"
2360 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2361 " [,asset=str][,location=str]\n"
2362 " specify SMBIOS type 2 fields\n"
2363 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n"
2365 " specify SMBIOS type 3 fields\n"
2366 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n"
2367 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n"
2368 " specify SMBIOS type 4 fields\n"
2369 "-smbios type=11[,value=str][,path=filename]\n"
2370 " specify SMBIOS type 11 fields\n"
2371 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n"
2372 " [,asset=str][,part=str][,speed=%d]\n"
2373 " specify SMBIOS type 17 fields\n",
2374 QEMU_ARCH_I386 | QEMU_ARCH_ARM)
2376 ``-smbios file=binary``
2377 Load SMBIOS entry from binary file.
2379 ``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]``
2380 Specify SMBIOS type 0 fields
2382 ``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]``
2383 Specify SMBIOS type 1 fields
2385 ``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]``
2386 Specify SMBIOS type 2 fields
2388 ``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]``
2389 Specify SMBIOS type 3 fields
2391 ``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str]``
2392 Specify SMBIOS type 4 fields
2394 ``-smbios type=11[,value=str][,path=filename]``
2395 Specify SMBIOS type 11 fields
2397 This argument can be repeated multiple times, and values are added in the order they are parsed.
2398 Applications intending to use OEM strings data are encouraged to use their application name as
2399 a prefix for the value string. This facilitates passing information for multiple applications
2402 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax
2403 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes.
2405 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to
2406 the SMBIOS table in the order in which they appear.
2408 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535
2409 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the
2410 guest. Instead it should be used as a indicator to inform the guest where to locate the real
2411 data set, for example, by specifying the serial ID of a block device.
2413 An example passing three strings is
2417 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\
2418 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\
2419 path=/some/file/with/oemstringsdata.txt
2421 In the guest OS this is visible with the ``dmidecode`` command
2426 Handle 0x0E00, DMI type 11, 5 bytes
2428 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/
2429 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os
2430 String 3: myapp:some extra data
2433 ``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]``
2434 Specify SMBIOS type 17 fields
2439 DEFHEADING(Network options:)
2441 DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
2443 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n"
2444 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n"
2445 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n"
2446 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n"
2447 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]"
2449 "[,smb=dir[,smbserver=addr]]\n"
2451 " configure a user mode network backend with ID 'str',\n"
2452 " its DHCP server and optional services\n"
2455 "-netdev tap,id=str,ifname=name\n"
2456 " configure a host TAP network backend with ID 'str'\n"
2458 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n"
2459 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n"
2460 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n"
2462 " configure a host TAP network backend with ID 'str'\n"
2463 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2464 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n"
2465 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n"
2466 " to deconfigure it\n"
2467 " use '[down]script=no' to disable script execution\n"
2468 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n"
2470 " use 'fd=h' to connect to an already opened TAP interface\n"
2471 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n"
2472 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n"
2473 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n"
2474 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n"
2475 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n"
2476 " use vhost=on to enable experimental in kernel accelerator\n"
2477 " (only has effect for virtio guests which use MSIX)\n"
2478 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n"
2479 " use 'vhostfd=h' to connect to an already opened vhost net device\n"
2480 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n"
2481 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n"
2482 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n"
2483 " spent on busy polling for vhost net\n"
2484 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n"
2485 " configure a host TAP network backend with ID 'str' that is\n"
2486 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2487 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n"
2490 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n"
2491 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n"
2492 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n"
2493 " [,rxcookie=rxcookie][,offset=offset]\n"
2494 " configure a network backend with ID 'str' connected to\n"
2495 " an Ethernet over L2TPv3 pseudowire.\n"
2496 " Linux kernel 3.3+ as well as most routers can talk\n"
2497 " L2TPv3. This transport allows connecting a VM to a VM,\n"
2498 " VM to a router and even VM to Host. It is a nearly-universal\n"
2499 " standard (RFC3931). Note - this implementation uses static\n"
2500 " pre-configured tunnels (same as the Linux kernel).\n"
2501 " use 'src=' to specify source address\n"
2502 " use 'dst=' to specify destination address\n"
2503 " use 'udp=on' to specify udp encapsulation\n"
2504 " use 'srcport=' to specify source udp port\n"
2505 " use 'dstport=' to specify destination udp port\n"
2506 " use 'ipv6=on' to force v6\n"
2507 " L2TPv3 uses cookies to prevent misconfiguration as\n"
2508 " well as a weak security measure\n"
2509 " use 'rxcookie=0x012345678' to specify a rxcookie\n"
2510 " use 'txcookie=0x012345678' to specify a txcookie\n"
2511 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n"
2512 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n"
2513 " use 'pincounter=on' to work around broken counter handling in peer\n"
2514 " use 'offset=X' to add an extra offset between header and data\n"
2516 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n"
2517 " configure a network backend to connect to another network\n"
2518 " using a socket connection\n"
2519 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n"
2520 " configure a network backend to connect to a multicast maddr and port\n"
2521 " use 'localaddr=addr' to specify the host address to send packets from\n"
2522 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n"
2523 " configure a network backend to connect to another network\n"
2524 " using an UDP tunnel\n"
2526 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n"
2527 " configure a network backend to connect to port 'n' of a vde switch\n"
2528 " running on host and listening for incoming connections on 'socketpath'.\n"
2529 " Use group 'groupname' and mode 'octalmode' to change default\n"
2530 " ownership and permissions for communication port.\n"
2532 #ifdef CONFIG_NETMAP
2533 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n"
2534 " attach to the existing netmap-enabled network interface 'name', or to a\n"
2535 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n"
2536 " netmap device, defaults to '/dev/netmap')\n"
2539 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n"
2540 " configure a vhost-user network, backed by a chardev 'dev'\n"
2543 "-netdev vhost-vdpa,id=str,vhostdev=/path/to/dev\n"
2544 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n"
2546 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n"
2547 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL)
2548 DEF("nic", HAS_ARG, QEMU_OPTION_nic,
2559 #ifdef CONFIG_NETMAP
2565 "socket][,option][,...][mac=macaddr]\n"
2566 " initialize an on-board / default host NIC (using MAC address\n"
2567 " macaddr) and connect it to the given host network backend\n"
2568 "-nic none use it alone to have zero network devices (the default is to\n"
2569 " provided a 'user' network connection)\n",
2571 DEF("net", HAS_ARG, QEMU_OPTION_net,
2572 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n"
2573 " configure or create an on-board (or machine default) NIC and\n"
2574 " connect it to hub 0 (please use -nic unless you need a hub)\n"
2584 #ifdef CONFIG_NETMAP
2587 "socket][,option][,option][,...]\n"
2588 " old way to initialize a host network interface\n"
2589 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL)
2591 ``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]``
2592 This option is a shortcut for configuring both the on-board
2593 (default) guest NIC hardware and the host network backend in one go.
2594 The host backend options are the same as with the corresponding
2595 ``-netdev`` options below. The guest NIC model can be set with
2596 ``model=modelname``. Use ``model=help`` to list the available device
2597 types. The hardware MAC address can be set with ``mac=macaddr``.
2599 The following two example do exactly the same, to show how ``-nic``
2600 can be used to shorten the command line length:
2604 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32
2605 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32
2608 Indicate that no network devices should be configured. It is used to
2609 override the default configuration (default NIC with "user" host
2610 network backend) which is activated if no other networking options
2613 ``-netdev user,id=id[,option][,option][,...]``
2614 Configure user mode host network backend which requires no
2615 administrator privilege to run. Valid options are:
2618 Assign symbolic name for use in monitor commands.
2620 ``ipv4=on|off and ipv6=on|off``
2621 Specify that either IPv4 or IPv6 must be enabled. If neither is
2622 specified both protocols are enabled.
2625 Set IP network address the guest will see. Optionally specify
2626 the netmask, either in the form a.b.c.d or as number of valid
2627 top-most bits. Default is 10.0.2.0/24.
2630 Specify the guest-visible address of the host. Default is the
2631 2nd IP in the guest network, i.e. x.x.x.2.
2633 ``ipv6-net=addr[/int]``
2634 Set IPv6 network address the guest will see (default is
2635 fec0::/64). The network prefix is given in the usual hexadecimal
2636 IPv6 address notation. The prefix size is optional, and is given
2637 as the number of valid top-most bits (default is 64).
2640 Specify the guest-visible IPv6 address of the host. Default is
2641 the 2nd IPv6 in the guest network, i.e. xxxx::2.
2644 If this option is enabled, the guest will be isolated, i.e. it
2645 will not be able to contact the host and no guest IP packets
2646 will be routed over the host to the outside. This option does
2647 not affect any explicitly set forwarding rules.
2650 Specifies the client hostname reported by the built-in DHCP
2654 Specify the first of the 16 IPs the built-in DHCP server can
2655 assign. Default is the 15th to 31st IP in the guest network,
2656 i.e. x.x.x.15 to x.x.x.31.
2659 Specify the guest-visible address of the virtual nameserver. The
2660 address must be different from the host address. Default is the
2661 3rd IP in the guest network, i.e. x.x.x.3.
2664 Specify the guest-visible address of the IPv6 virtual
2665 nameserver. The address must be different from the host address.
2666 Default is the 3rd IP in the guest network, i.e. xxxx::3.
2668 ``dnssearch=domain``
2669 Provides an entry for the domain-search list sent by the
2670 built-in DHCP server. More than one domain suffix can be
2671 transmitted by specifying this option multiple times. If
2672 supported, this will cause the guest to automatically try to
2673 append the given domain suffix(es) in case a domain name can not
2680 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org
2682 ``domainname=domain``
2683 Specifies the client domain name reported by the built-in DHCP
2687 When using the user mode network stack, activate a built-in TFTP
2688 server. The files in dir will be exposed as the root of a TFTP
2689 server. The TFTP client on the guest must be configured in
2690 binary mode (use the command ``bin`` of the Unix TFTP client).
2692 ``tftp-server-name=name``
2693 In BOOTP reply, broadcast name as the "TFTP server name"
2694 (RFC2132 option 66). This can be used to advise the guest to
2695 load boot files or configurations from a different server than
2699 When using the user mode network stack, broadcast file as the
2700 BOOTP filename. In conjunction with ``tftp``, this can be used
2701 to network boot a guest from a local directory.
2703 Example (using pxelinux):
2707 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\
2708 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0
2710 ``smb=dir[,smbserver=addr]``
2711 When using the user mode network stack, activate a built-in SMB
2712 server so that Windows OSes can access to the host files in
2713 ``dir`` transparently. The IP address of the SMB server can be
2714 set to addr. By default the 4th IP in the guest network is used,
2717 In the guest Windows OS, the line:
2723 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows
2724 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows
2727 Then ``dir`` can be accessed in ``\\smbserver\qemu``.
2729 Note that a SAMBA server must be installed on the host OS.
2731 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport``
2732 Redirect incoming TCP or UDP connections to the host port
2733 hostport to the guest IP address guestaddr on guest port
2734 guestport. If guestaddr is not specified, its value is x.x.x.15
2735 (default first address given by the built-in DHCP server). By
2736 specifying hostaddr, the rule can be bound to a specific host
2737 interface. If no connection type is set, TCP is used. This
2738 option can be given multiple times.
2740 For example, to redirect host X11 connection from screen 1 to
2741 guest screen 0, use the following:
2746 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000
2747 # this host xterm should open in the guest X11 server
2750 To redirect telnet connections from host port 5555 to telnet
2751 port on the guest, use the following:
2756 |qemu_system| -nic user,hostfwd=tcp::5555-:23
2757 telnet localhost 5555
2759 Then when you use on the host ``telnet localhost 5555``, you
2760 connect to the guest telnet server.
2762 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command``
2763 Forward guest TCP connections to the IP address server on port
2764 port to the character device dev or to a program executed by
2765 cmd:command which gets spawned for each connection. This option
2766 can be given multiple times.
2768 You can either use a chardev directly and have that one used
2769 throughout QEMU's lifetime, like in the following example:
2773 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever
2774 # the guest accesses it
2775 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321
2777 Or you can execute a command on every TCP connection established
2778 by the guest, so that QEMU behaves similar to an inetd process
2779 for that virtual server:
2783 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234
2784 # and connect the TCP stream to its stdin/stdout
2785 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321'
2787 ``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]``
2788 Configure a host TAP network backend with ID id.
2790 Use the network script file to configure it and the network script
2791 dfile to deconfigure it. If name is not provided, the OS
2792 automatically provides one. The default network configure script is
2793 ``/etc/qemu-ifup`` and the default network deconfigure script is
2794 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to
2795 disable script execution.
2797 If running QEMU as an unprivileged user, use the network helper
2798 to configure the TAP interface and attach it to the bridge.
2799 The default network helper executable is
2800 ``/path/to/qemu-bridge-helper`` and the default bridge device is
2803 ``fd``\ =h can be used to specify the handle of an already opened
2810 #launch a QEMU instance with the default network script
2811 |qemu_system| linux.img -nic tap
2815 #launch a QEMU instance with two NICs, each one connected
2817 |qemu_system| linux.img \\
2818 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\
2819 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1
2823 #launch a QEMU instance with the default network helper to
2824 #connect a TAP device to bridge br0
2825 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\
2826 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper"
2828 ``-netdev bridge,id=id[,br=bridge][,helper=helper]``
2829 Connect a host TAP network interface to a host bridge device.
2831 Use the network helper helper to configure the TAP interface and
2832 attach it to the bridge. The default network helper executable is
2833 ``/path/to/qemu-bridge-helper`` and the default bridge device is
2840 #launch a QEMU instance with the default network helper to
2841 #connect a TAP device to bridge br0
2842 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1
2846 #launch a QEMU instance with the default network helper to
2847 #connect a TAP device to bridge qemubr0
2848 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1
2850 ``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]``
2851 This host network backend can be used to connect the guest's network
2852 to another QEMU virtual machine using a TCP socket connection. If
2853 ``listen`` is specified, QEMU waits for incoming connections on port
2854 (host is optional). ``connect`` is used to connect to another QEMU
2855 instance using the ``listen`` option. ``fd``\ =h specifies an
2856 already opened TCP socket.
2862 # launch a first QEMU instance
2863 |qemu_system| linux.img \\
2864 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
2865 -netdev socket,id=n1,listen=:1234
2866 # connect the network of this instance to the network of the first instance
2867 |qemu_system| linux.img \\
2868 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
2869 -netdev socket,id=n2,connect=127.0.0.1:1234
2871 ``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]``
2872 Configure a socket host network backend to share the guest's network
2873 traffic with another QEMU virtual machines using a UDP multicast
2874 socket, effectively making a bus for every QEMU with same multicast
2875 address maddr and port. NOTES:
2877 1. Several QEMU can be running on different hosts and share same bus
2878 (assuming correct multicast setup for these hosts).
2880 2. mcast support is compatible with User Mode Linux (argument
2881 ``ethN=mcast``), see http://user-mode-linux.sf.net.
2883 3. Use ``fd=h`` to specify an already opened UDP multicast socket.
2889 # launch one QEMU instance
2890 |qemu_system| linux.img \\
2891 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
2892 -netdev socket,id=n1,mcast=230.0.0.1:1234
2893 # launch another QEMU instance on same "bus"
2894 |qemu_system| linux.img \\
2895 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
2896 -netdev socket,id=n2,mcast=230.0.0.1:1234
2897 # launch yet another QEMU instance on same "bus"
2898 |qemu_system| linux.img \\
2899 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\
2900 -netdev socket,id=n3,mcast=230.0.0.1:1234
2902 Example (User Mode Linux compat.):
2906 # launch QEMU instance (note mcast address selected is UML's default)
2907 |qemu_system| linux.img \\
2908 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
2909 -netdev socket,id=n1,mcast=239.192.168.1:1102
2911 /path/to/linux ubd0=/path/to/root_fs eth0=mcast
2913 Example (send packets from host's 1.2.3.4):
2917 |qemu_system| linux.img \\
2918 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
2919 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4
2921 ``-netdev l2tpv3,id=id,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport],txsession=txsession[,rxsession=rxsession][,ipv6=on|off][,udp=on|off][,cookie64][,counter][,pincounter][,txcookie=txcookie][,rxcookie=rxcookie][,offset=offset]``
2922 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931)
2923 is a popular protocol to transport Ethernet (and other Layer 2) data
2924 frames between two systems. It is present in routers, firewalls and
2925 the Linux kernel (from version 3.3 onwards).
2927 This transport allows a VM to communicate to another VM, router or
2931 source address (mandatory)
2934 destination address (mandatory)
2937 select udp encapsulation (default is ip).
2943 destination udp port.
2946 force v6, otherwise defaults to v4.
2948 ``rxcookie=rxcookie``; \ ``txcookie=txcookie``
2949 Cookies are a weak form of security in the l2tpv3 specification.
2950 Their function is mostly to prevent misconfiguration. By default
2954 Set cookie size to 64 bit instead of the default 32
2957 Force a 'cut-down' L2TPv3 with no counter as in
2958 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00
2961 Work around broken counter handling in peer. This may also help
2962 on networks which have packet reorder.
2965 Add an extra offset between header and data
2967 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to
2968 the bridge br-lan on the remote Linux host 1.2.3.4:
2972 # Setup tunnel on linux host using raw ip as encapsulation
2974 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\
2975 encap udp udp_sport 16384 udp_dport 16384
2976 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\
2977 0xFFFFFFFF peer_session_id 0xFFFFFFFF
2978 ifconfig vmtunnel0 mtu 1500
2979 ifconfig vmtunnel0 up
2980 brctl addif br-lan vmtunnel0
2984 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter
2986 |qemu_system| linux.img -device e1000,netdev=n1 \\
2987 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter
2989 ``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]``
2990 Configure VDE backend to connect to PORT n of a vde switch running
2991 on host and listening for incoming connections on socketpath. Use
2992 GROUP groupname and MODE octalmode to change default ownership and
2993 permissions for communication port. This option is only available if
2994 QEMU has been compiled with vde support enabled.
3001 vde_switch -F -sock /tmp/myswitch
3002 # launch QEMU instance
3003 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch
3005 ``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]``
3006 Establish a vhost-user netdev, backed by a chardev id. The chardev
3007 should be a unix domain socket backed one. The vhost-user uses a
3008 specifically defined protocol to pass vhost ioctl replacement
3009 messages to an application on the other end of the socket. On
3010 non-MSIX guests, the feature can be forced with vhostforce. Use
3011 'queues=n' to specify the number of queues to be created for
3012 multiqueue vhost-user.
3018 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
3019 -numa node,memdev=mem \
3020 -chardev socket,id=chr0,path=/path/to/socket \
3021 -netdev type=vhost-user,id=net0,chardev=chr0 \
3022 -device virtio-net-pci,netdev=net0
3024 ``-netdev vhost-vdpa,vhostdev=/path/to/dev``
3025 Establish a vhost-vdpa netdev.
3027 vDPA device is a device that uses a datapath which complies with
3028 the virtio specifications with a vendor specific control path.
3029 vDPA devices can be both physically located on the hardware or
3030 emulated by software.
3032 ``-netdev hubport,id=id,hubid=hubid[,netdev=nd]``
3033 Create a hub port on the emulated hub with ID hubid.
3035 The hubport netdev lets you connect a NIC to a QEMU emulated hub
3036 instead of a single netdev. Alternatively, you can also connect the
3037 hubport to another netdev with ID nd by using the ``netdev=nd``
3040 ``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]``
3041 Legacy option to configure or create an on-board (or machine
3042 default) Network Interface Card(NIC) and connect it either to the
3043 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd.
3044 If model is omitted, then the default NIC model associated with the
3045 machine type is used. Note that the default NIC model may change in
3046 future QEMU releases, so it is highly recommended to always specify
3047 a model. Optionally, the MAC address can be changed to mac, the
3048 device address set to addr (PCI cards only), and a name can be
3049 assigned for use in monitor commands. Optionally, for PCI cards, you
3050 can specify the number v of MSI-X vectors that the card should have;
3051 this option currently only affects virtio cards; set v = 0 to
3052 disable MSI-X. If no ``-net`` option is specified, a single NIC is
3053 created. QEMU can emulate several different models of network card.
3054 Use ``-net nic,model=help`` for a list of available devices for your
3057 ``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]``
3058 Configure a host network backend (with the options corresponding to
3059 the same ``-netdev`` option) and connect it to the emulated hub 0
3060 (the default hub). Use name to specify the name of the hub port.
3065 DEFHEADING(Character device options:)
3067 DEF("chardev", HAS_ARG, QEMU_OPTION_chardev,
3069 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3070 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off][,reconnect=seconds]\n"
3071 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n"
3072 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n"
3073 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n"
3074 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n"
3075 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n"
3076 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n"
3077 " [,logfile=PATH][,logappend=on|off]\n"
3078 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3079 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n"
3080 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3081 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n"
3082 "-chardev file,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3083 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3085 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3086 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3088 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3089 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n"
3091 #ifdef CONFIG_BRLAPI
3092 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3094 #if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \
3095 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__)
3096 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3097 "-chardev tty,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3099 #if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__)
3100 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3101 "-chardev parport,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3103 #if defined(CONFIG_SPICE)
3104 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3105 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3111 The general form of a character device option is:
3113 ``-chardev backend,id=id[,mux=on|off][,options]``
3114 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``,
3115 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``,
3116 ``pty``, ``stdio``, ``braille``, ``tty``, ``parallel``, ``parport``,
3117 ``spicevmc``, ``spiceport``. The specific backend will determine the
3120 Use ``-chardev help`` to print all available chardev backend types.
3122 All devices must have an id, which can be any string up to 127
3123 characters long. It is used to uniquely identify this device in
3124 other command line directives.
3126 A character device may be used in multiplexing mode by multiple
3127 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is
3128 a "1:N" device, and here the "1" end is your specified chardev
3129 backend, and the "N" end is the various parts of QEMU that can talk
3130 to a chardev. If you create a chardev with ``id=myid`` and
3131 ``mux=on``, QEMU will create a multiplexer with your specified ID,
3132 and you can then configure multiple front ends to use that chardev
3133 ID for their input/output. Up to four different front ends can be
3134 connected to a single multiplexed chardev. (Without multiplexing
3135 enabled, a chardev can only be used by a single front end.) For
3136 instance you could use this to allow a single stdio chardev to be
3137 used by two serial ports and the QEMU monitor:
3141 -chardev stdio,mux=on,id=char0 \
3142 -mon chardev=char0,mode=readline \
3143 -serial chardev:char0 \
3144 -serial chardev:char0
3146 You can have more than one multiplexer in a system configuration;
3147 for instance you could have a TCP port multiplexed between UART 0
3148 and UART 1, and stdio multiplexed between the QEMU monitor and a
3153 -chardev stdio,mux=on,id=char0 \
3154 -mon chardev=char0,mode=readline \
3155 -parallel chardev:char0 \
3156 -chardev tcp,...,mux=on,id=char1 \
3157 -serial chardev:char1 \
3158 -serial chardev:char1
3160 When you're using a multiplexed character device, some escape
3161 sequences are interpreted in the input. See the chapter about
3162 :ref:`keys in the character backend multiplexer` in the
3163 System Emulation Users Guide for more details.
3165 Note that some other command line options may implicitly create
3166 multiplexed character backends; for instance ``-serial mon:stdio``
3167 creates a multiplexed stdio backend connected to the serial port and
3168 the QEMU monitor, and ``-nographic`` also multiplexes the console
3169 and the monitor to stdio.
3171 There is currently no support for multiplexing in the other
3172 direction (where a single QEMU front end takes input and output from
3175 Every backend supports the ``logfile`` option, which supplies the
3176 path to a file to record all data transmitted via the backend. The
3177 ``logappend`` option controls whether the log file will be truncated
3178 or appended to when opened.
3180 The available backends are:
3182 ``-chardev null,id=id``
3183 A void device. This device will not emit any data, and will drop any
3184 data it receives. The null backend does not take any options.
3186 ``-chardev socket,id=id[,TCP options or unix options][,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,tls-creds=id][,tls-authz=id]``
3187 Create a two-way stream socket, which can be either a TCP or a unix
3188 socket. A unix socket will be created if ``path`` is specified.
3189 Behaviour is undefined if TCP options are specified for a unix
3192 ``server=on|off`` specifies that the socket shall be a listening socket.
3194 ``wait=on|off`` specifies that QEMU should not block waiting for a client
3195 to connect to a listening socket.
3197 ``telnet=on|off`` specifies that traffic on the socket should interpret
3198 telnet escape sequences.
3200 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for
3203 ``reconnect`` sets the timeout for reconnecting on non-server
3204 sockets when the remote end goes away. qemu will delay this many
3205 seconds and then attempt to reconnect. Zero disables reconnecting,
3208 ``tls-creds`` requests enablement of the TLS protocol for
3209 encryption, and specifies the id of the TLS credentials to use for
3210 the handshake. The credentials must be previously created with the
3211 ``-object tls-creds`` argument.
3213 ``tls-auth`` provides the ID of the QAuthZ authorization object
3214 against which the client's x509 distinguished name will be
3215 validated. This object is only resolved at time of use, so can be
3216 deleted and recreated on the fly while the chardev server is active.
3217 If missing, it will default to denying access.
3219 TCP and unix socket options are given below:
3221 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]``
3222 ``host`` for a listening socket specifies the local address to
3223 be bound. For a connecting socket species the remote host to
3224 connect to. ``host`` is optional for listening sockets. If not
3225 specified it defaults to ``0.0.0.0``.
3227 ``port`` for a listening socket specifies the local port to be
3228 bound. For a connecting socket specifies the port on the remote
3229 host to connect to. ``port`` can be given as either a port
3230 number or a service name. ``port`` is required.
3232 ``to`` is only relevant to listening sockets. If it is
3233 specified, and ``port`` cannot be bound, QEMU will attempt to
3234 bind to subsequent ports up to and including ``to`` until it
3235 succeeds. ``to`` must be specified as a port number.
3237 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4
3238 or IPv6 must be used. If neither is specified the socket may
3239 use either protocol.
3241 ``nodelay=on|off`` disables the Nagle algorithm.
3243 ``unix options: path=path[,abstract=on|off][,tight=on|off]``
3244 ``path`` specifies the local path of the unix socket. ``path``
3246 ``abstract=on|off`` specifies the use of the abstract socket namespace,
3247 rather than the filesystem. Optional, defaults to false.
3248 ``tight=on|off`` sets the socket length of abstract sockets to their minimum,
3249 rather than the full sun_path length. Optional, defaults to true.
3251 ``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]``
3252 Sends all traffic from the guest to a remote host over UDP.
3254 ``host`` specifies the remote host to connect to. If not specified
3255 it defaults to ``localhost``.
3257 ``port`` specifies the port on the remote host to connect to.
3258 ``port`` is required.
3260 ``localaddr`` specifies the local address to bind to. If not
3261 specified it defaults to ``0.0.0.0``.
3263 ``localport`` specifies the local port to bind to. If not specified
3264 any available local port will be used.
3266 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used.
3267 If neither is specified the device may use either protocol.
3269 ``-chardev msmouse,id=id``
3270 Forward QEMU's emulated msmouse events to the guest. ``msmouse``
3271 does not take any options.
3273 ``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]``
3274 Connect to a QEMU text console. ``vc`` may optionally be given a
3277 ``width`` and ``height`` specify the width and height respectively
3278 of the console, in pixels.
3280 ``cols`` and ``rows`` specify that the console be sized to fit a
3281 text console with the given dimensions.
3283 ``-chardev ringbuf,id=id[,size=size]``
3284 Create a ring buffer with fixed size ``size``. size must be a power
3285 of two and defaults to ``64K``.
3287 ``-chardev file,id=id,path=path``
3288 Log all traffic received from the guest to a file.
3290 ``path`` specifies the path of the file to be opened. This file will
3291 be created if it does not already exist, and overwritten if it does.
3292 ``path`` is required.
3294 ``-chardev pipe,id=id,path=path``
3295 Create a two-way connection to the guest. The behaviour differs
3296 slightly between Windows hosts and other hosts:
3298 On Windows, a single duplex pipe will be created at
3301 On other hosts, 2 pipes will be created called ``path.in`` and
3302 ``path.out``. Data written to ``path.in`` will be received by the
3303 guest. Data written by the guest can be read from ``path.out``. QEMU
3304 will not create these fifos, and requires them to be present.
3306 ``path`` forms part of the pipe path as described above. ``path`` is
3309 ``-chardev console,id=id``
3310 Send traffic from the guest to QEMU's standard output. ``console``
3311 does not take any options.
3313 ``console`` is only available on Windows hosts.
3315 ``-chardev serial,id=id,path=path``
3316 Send traffic from the guest to a serial device on the host.
3318 On Unix hosts serial will actually accept any tty device, not only
3321 ``path`` specifies the name of the serial device to open.
3323 ``-chardev pty,id=id``
3324 Create a new pseudo-terminal on the host and connect to it. ``pty``
3325 does not take any options.
3327 ``pty`` is not available on Windows hosts.
3329 ``-chardev stdio,id=id[,signal=on|off]``
3330 Connect to standard input and standard output of the QEMU process.
3332 ``signal`` controls if signals are enabled on the terminal, that
3333 includes exiting QEMU with the key sequence Control-c. This option
3334 is enabled by default, use ``signal=off`` to disable it.
3336 ``-chardev braille,id=id``
3337 Connect to a local BrlAPI server. ``braille`` does not take any
3340 ``-chardev tty,id=id,path=path``
3341 ``tty`` is only available on Linux, Sun, FreeBSD, NetBSD, OpenBSD
3342 and DragonFlyBSD hosts. It is an alias for ``serial``.
3344 ``path`` specifies the path to the tty. ``path`` is required.
3346 ``-chardev parallel,id=id,path=path``
3348 ``-chardev parport,id=id,path=path``
3349 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD
3352 Connect to a local parallel port.
3354 ``path`` specifies the path to the parallel port device. ``path`` is
3357 ``-chardev spicevmc,id=id,debug=debug,name=name``
3358 ``spicevmc`` is only available when spice support is built in.
3360 ``debug`` debug level for spicevmc
3362 ``name`` name of spice channel to connect to
3364 Connect to a spice virtual machine channel, such as vdiport.
3366 ``-chardev spiceport,id=id,debug=debug,name=name``
3367 ``spiceport`` is only available when spice support is built in.
3369 ``debug`` debug level for spicevmc
3371 ``name`` name of spice port to connect to
3373 Connect to a spice port, allowing a Spice client to handle the
3374 traffic identified by a name (preferably a fqdn).
3380 DEFHEADING(TPM device options:)
3382 DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \
3383 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n"
3384 " use path to provide path to a character device; default is /dev/tpm0\n"
3385 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n"
3386 " not provided it will be searched for in /sys/class/misc/tpm?/device\n"
3387 "-tpmdev emulator,id=id,chardev=dev\n"
3388 " configure the TPM device using chardev backend\n",
3391 The general form of a TPM device option is:
3393 ``-tpmdev backend,id=id[,options]``
3394 The specific backend type will determine the applicable options. The
3395 ``-tpmdev`` option creates the TPM backend and requires a
3396 ``-device`` option that specifies the TPM frontend interface model.
3398 Use ``-tpmdev help`` to print all available TPM backend types.
3400 The available backends are:
3402 ``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path``
3403 (Linux-host only) Enable access to the host's TPM using the
3406 ``path`` specifies the path to the host's TPM device, i.e., on a
3407 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by
3408 default ``/dev/tpm0`` is used.
3410 ``cancel-path`` specifies the path to the host TPM device's sysfs
3411 entry allowing for cancellation of an ongoing TPM command.
3412 ``cancel-path`` is optional and by default QEMU will search for the
3415 Some notes about using the host's TPM with the passthrough driver:
3417 The TPM device accessed by the passthrough driver must not be used
3418 by any other application on the host.
3420 Since the host's firmware (BIOS/UEFI) has already initialized the
3421 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize
3422 the TPM again and may therefore not show a TPM-specific menu that
3423 would otherwise allow the user to configure the TPM, e.g., allow the
3424 user to enable/disable or activate/deactivate the TPM. Further, if
3425 TPM ownership is released from within a VM then the host's TPM will
3426 get disabled and deactivated. To enable and activate the TPM again
3427 afterwards, the host has to be rebooted and the user is required to
3428 enter the firmware's menu to enable and activate the TPM. If the TPM
3429 is left disabled and/or deactivated most TPM commands will fail.
3431 To create a passthrough TPM use the following two options:
3435 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0
3437 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by
3438 ``tpmdev=tpm0`` in the device option.
3440 ``-tpmdev emulator,id=id,chardev=dev``
3441 (Linux-host only) Enable access to a TPM emulator using Unix domain
3442 socket based chardev backend.
3444 ``chardev`` specifies the unique ID of a character device backend
3445 that provides connection to the software TPM server.
3447 To create a TPM emulator backend device with chardev socket backend:
3451 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0
3458 DEFHEADING(Linux/Multiboot boot specific:)
3460 When using these options, you can use a given Linux or Multiboot kernel
3461 without installing it in the disk image. It can be useful for easier
3462 testing of various kernels.
3467 DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \
3468 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL)
3471 Use bzImage as kernel image. The kernel can be either a Linux kernel
3472 or in multiboot format.
3475 DEF("append", HAS_ARG, QEMU_OPTION_append, \
3476 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL)
3479 Use cmdline as kernel command line
3482 DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \
3483 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL)
3486 Use file as initial ram disk.
3488 ``-initrd "file1 arg=foo,file2"``
3489 This syntax is only available with multiboot.
3491 Use file1 and file2 as modules and pass arg=foo as parameter to the
3495 DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \
3496 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL)
3499 Use file as a device tree binary (dtb) image and pass it to the
3505 DEFHEADING(Debug/Expert options:)
3507 DEF("compat", HAS_ARG, QEMU_OPTION_compat,
3508 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n"
3509 " Policy for handling deprecated management interfaces\n",
3512 ``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]``
3513 Set policy for handling deprecated management interfaces (experimental):
3515 ``deprecated-input=accept`` (default)
3516 Accept deprecated commands and arguments
3517 ``deprecated-input=reject``
3518 Reject deprecated commands and arguments
3519 ``deprecated-input=crash``
3520 Crash on deprecated commands and arguments
3521 ``deprecated-output=accept`` (default)
3522 Emit deprecated command results and events
3523 ``deprecated-output=hide``
3524 Suppress deprecated command results and events
3526 Limitation: covers only syntactic aspects of QMP.
3529 DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg,
3530 "-fw_cfg [name=]<name>,file=<file>\n"
3531 " add named fw_cfg entry with contents from file\n"
3532 "-fw_cfg [name=]<name>,string=<str>\n"
3533 " add named fw_cfg entry with contents from string\n",
3536 ``-fw_cfg [name=]name,file=file``
3537 Add named fw\_cfg entry with contents from file file.
3539 ``-fw_cfg [name=]name,string=str``
3540 Add named fw\_cfg entry with contents from string str.
3542 The terminating NUL character of the contents of str will not be
3543 included as part of the fw\_cfg item data. To insert contents with
3544 embedded NUL characters, you have to use the file parameter.
3546 The fw\_cfg entries are passed by QEMU through to the guest.
3552 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin
3554 creates an fw\_cfg entry named opt/com.mycompany/blob with contents
3555 from ./my\_blob.bin.
3558 DEF("serial", HAS_ARG, QEMU_OPTION_serial, \
3559 "-serial dev redirect the serial port to char device 'dev'\n",
3563 Redirect the virtual serial port to host character device dev. The
3564 default device is ``vc`` in graphical mode and ``stdio`` in non
3567 This option can be used several times to simulate up to 4 serial
3570 Use ``-serial none`` to disable all serial ports.
3572 Available character devices are:
3575 Virtual console. Optionally, a width and height can be given in
3582 It is also possible to specify width or height in characters:
3589 [Linux only] Pseudo TTY (a new PTY is automatically allocated)
3592 No device is allocated.
3598 Use a named character device defined with the ``-chardev``
3602 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial
3603 port parameters are set according to the emulated ones.
3606 [Linux only, parallel port only] Use host parallel port N.
3607 Currently SPP and EPP parallel port features can be used.
3610 Write output to filename. No character can be read.
3613 [Unix only] standard input/output
3619 [Windows only] Use host serial port n
3621 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]``
3622 This implements UDP Net Console. When remote\_host or src\_ip
3623 are not specified they default to ``0.0.0.0``. When not using a
3624 specified src\_port a random port is automatically chosen.
3626 If you just want a simple readonly console you can use
3627 ``netcat`` or ``nc``, by starting QEMU with:
3628 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time
3629 QEMU writes something to that port it will appear in the
3632 If you plan to send characters back via netconsole or you want
3633 to stop and start QEMU a lot of times, you should have QEMU use
3634 the same source port each time by using something like ``-serial
3635 udp::4555@:4556`` to QEMU. Another approach is to use a patched
3636 version of netcat which can listen to a TCP port and send and
3637 receive characters via udp. If you have a patched version of
3638 netcat which activates telnet remote echo and single char
3639 transfer, then you can use the following options to set up a
3640 netcat redirector to allow telnet on port 5555 to access the
3644 -serial udp::4555@:4556
3647 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T
3652 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]``
3653 The TCP Net Console has two modes of operation. It can send the
3654 serial I/O to a location or wait for a connection from a
3655 location. By default the TCP Net Console is sent to host at the
3656 port. If you use the ``server=on`` option QEMU will wait for a client
3657 socket application to connect to the port before continuing,
3658 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off``
3659 option disables the Nagle buffering algorithm. The ``reconnect=on``
3660 option only applies if ``server=no`` is set, if the connection goes
3661 down it will attempt to reconnect at the given interval. If host
3662 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a
3663 time is accepted. You can use ``telnet=on`` to connect to the
3664 corresponding character device.
3666 ``Example to send tcp console to 192.168.0.2 port 4444``
3667 -serial tcp:192.168.0.2:4444
3669 ``Example to listen and wait on port 4444 for connection``
3670 -serial tcp::4444,server=on
3672 ``Example to not wait and listen on ip 192.168.0.100 port 4444``
3673 -serial tcp:192.168.0.100:4444,server=on,wait=off
3675 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]``
3676 The telnet protocol is used instead of raw tcp sockets. The
3677 options work the same as if you had specified ``-serial tcp``.
3678 The difference is that the port acts like a telnet server or
3679 client using telnet option negotiation. This will also allow you
3680 to send the MAGIC\_SYSRQ sequence if you use a telnet that
3681 supports sending the break sequence. Typically in unix telnet
3682 you do it with Control-] and then type "send break" followed by
3683 pressing the enter key.
3685 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]``
3686 The WebSocket protocol is used instead of raw tcp socket. The
3687 port acts as a WebSocket server. Client mode is not supported.
3689 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]``
3690 A unix domain socket is used instead of a tcp socket. The option
3691 works the same as if you had specified ``-serial tcp`` except
3692 the unix domain socket path is used for connections.
3695 This is a special option to allow the monitor to be multiplexed
3696 onto another serial port. The monitor is accessed with key
3697 sequence of Control-a and then pressing c. dev\_string should be
3698 any one of the serial devices specified above. An example to
3699 multiplex the monitor onto a telnet server listening on port
3702 ``-serial mon:telnet::4444,server=on,wait=off``
3704 When the monitor is multiplexed to stdio in this way, Ctrl+C
3705 will not terminate QEMU any more but will be passed to the guest
3709 Braille device. This will use BrlAPI to display the braille
3710 output on a real or fake device.
3713 Three button serial mouse. Configure the guest to use Microsoft
3717 DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \
3718 "-parallel dev redirect the parallel port to char device 'dev'\n",
3722 Redirect the virtual parallel port to host device dev (same devices
3723 as the serial port). On Linux hosts, ``/dev/parportN`` can be used
3724 to use hardware devices connected on the corresponding host parallel
3727 This option can be used several times to simulate up to 3 parallel
3730 Use ``-parallel none`` to disable all parallel ports.
3733 DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \
3734 "-monitor dev redirect the monitor to char device 'dev'\n",
3738 Redirect the monitor to host device dev (same devices as the serial
3739 port). The default device is ``vc`` in graphical mode and ``stdio``
3740 in non graphical mode. Use ``-monitor none`` to disable the default
3743 DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \
3744 "-qmp dev like -monitor but opens in 'control' mode\n",
3748 Like -monitor but opens in 'control' mode.
3750 DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \
3751 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n",
3755 Like -qmp but uses pretty JSON formatting.
3758 DEF("mon", HAS_ARG, QEMU_OPTION_mon, \
3759 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL)
3761 ``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]``
3762 Setup monitor on chardev name. ``pretty`` is only valid when
3763 ``mode=control``, turning on JSON pretty printing to ease
3764 human reading and debugging.
3767 DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \
3768 "-debugcon dev redirect the debug console to char device 'dev'\n",
3772 Redirect the debug console to host device dev (same devices as the
3773 serial port). The debug console is an I/O port which is typically
3774 port 0xe9; writing to that I/O port sends output to this device. The
3775 default device is ``vc`` in graphical mode and ``stdio`` in non
3779 DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \
3780 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL)
3783 Store the QEMU process PID in file. It is useful if you launch QEMU
3787 DEF("singlestep", 0, QEMU_OPTION_singlestep, \
3788 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL)
3791 Run the emulation in single step mode.
3794 DEF("preconfig", 0, QEMU_OPTION_preconfig, \
3795 "--preconfig pause QEMU before machine is initialized (experimental)\n",
3799 Pause QEMU for interactive configuration before the machine is
3800 created, which allows querying and configuring properties that will
3801 affect machine initialization. Use QMP command 'x-exit-preconfig' to
3802 exit the preconfig state and move to the next state (i.e. run guest
3803 if -S isn't used or pause the second time if -S is used). This
3804 option is experimental.
3807 DEF("S", 0, QEMU_OPTION_S, \
3808 "-S freeze CPU at startup (use 'c' to start execution)\n",
3812 Do not start CPU at startup (you must type 'c' in the monitor).
3815 DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit,
3816 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n"
3817 " run qemu with overcommit hints\n"
3818 " mem-lock=on|off controls memory lock support (default: off)\n"
3819 " cpu-pm=on|off controls cpu power management (default: off)\n",
3822 ``-overcommit mem-lock=on|off``
3824 ``-overcommit cpu-pm=on|off``
3825 Run qemu with hints about host resource overcommit. The default is
3826 to assume that host overcommits all resources.
3828 Locking qemu and guest memory can be enabled via ``mem-lock=on``
3829 (disabled by default). This works when host memory is not
3830 overcommitted and reduces the worst-case latency for guest.
3832 Guest ability to manage power state of host cpus (increasing latency
3833 for other processes on the same host cpu, but decreasing latency for
3834 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This
3835 works best when host CPU is not overcommitted. When used, host
3836 estimates of CPU cycle and power utilization will be incorrect, not
3837 taking into account guest idle time.
3840 DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \
3841 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n"
3842 " the guest without waiting for gdb to connect; use -S too\n"
3843 " if you want it to not start execution.)\n",
3847 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter
3848 in the System Emulation Users Guide). Note that this option does not pause QEMU
3849 execution -- if you want QEMU to not start the guest until you
3850 connect with gdb and issue a ``continue`` command, you will need to
3851 also pass the ``-S`` option to QEMU.
3853 The most usual configuration is to listen on a local TCP socket::
3857 but you can specify other backends; UDP, pseudo TTY, or even stdio
3858 are all reasonable use cases. For example, a stdio connection
3859 allows you to start QEMU from within gdb and establish the
3860 connection via a pipe:
3864 (gdb) target remote | exec |qemu_system| -gdb stdio ...
3867 DEF("s", 0, QEMU_OPTION_s, \
3868 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n",
3872 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234
3873 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide).
3876 DEF("d", HAS_ARG, QEMU_OPTION_d, \
3877 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n",
3881 Enable logging of specified items. Use '-d help' for a list of log
3885 DEF("D", HAS_ARG, QEMU_OPTION_D, \
3886 "-D logfile output log to logfile (default stderr)\n",
3890 Output log in logfile instead of to stderr
3893 DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \
3894 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n",
3897 ``-dfilter range1[,...]``
3898 Filter debug output to that relevant to a range of target addresses.
3899 The filter spec can be either start+size, start-size or start..end
3900 where start end and size are the addresses and sizes required. For
3905 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000
3907 Will dump output for any code in the 0x1000 sized block starting at
3908 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and
3909 another 0x1000 sized block starting at 0xffffffc00005f000.
3912 DEF("seed", HAS_ARG, QEMU_OPTION_seed, \
3913 "-seed number seed the pseudo-random number generator\n",
3917 Force the guest to use a deterministic pseudo-random number
3918 generator, seeded with number. This does not affect crypto routines
3922 DEF("L", HAS_ARG, QEMU_OPTION_L, \
3923 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n",
3927 Set the directory for the BIOS, VGA BIOS and keymaps.
3929 To list all the data directories, use ``-L help``.
3932 DEF("bios", HAS_ARG, QEMU_OPTION_bios, \
3933 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL)
3936 Set the filename for the BIOS.
3939 DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \
3940 "-enable-kvm enable KVM full virtualization support\n", QEMU_ARCH_ALL)
3943 Enable KVM full virtualization support. This option is only
3944 available if KVM support is enabled when compiling.
3947 DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid,
3948 "-xen-domid id specify xen guest domain id\n", QEMU_ARCH_ALL)
3949 DEF("xen-attach", 0, QEMU_OPTION_xen_attach,
3950 "-xen-attach attach to existing xen domain\n"
3951 " libxl will use this when starting QEMU\n",
3953 DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict,
3954 "-xen-domid-restrict restrict set of available xen operations\n"
3955 " to specified domain id. (Does not affect\n"
3956 " xenpv machine type).\n",
3960 Specify xen guest domain id (XEN only).
3963 Attach to existing xen domain. libxl will use this when starting
3964 QEMU (XEN only). Restrict set of available xen operations to
3965 specified domain id (XEN only).
3968 DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \
3969 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL)
3972 Exit instead of rebooting.
3975 DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \
3976 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL)
3979 Don't exit QEMU on guest shutdown, but instead only stop the
3980 emulation. This allows for instance switching to monitor to commit
3981 changes to the disk image.
3984 DEF("action", HAS_ARG, QEMU_OPTION_action,
3985 "-action reboot=reset|shutdown\n"
3986 " action when guest reboots [default=reset]\n"
3987 "-action shutdown=poweroff|pause\n"
3988 " action when guest shuts down [default=poweroff]\n"
3989 "-action panic=pause|shutdown|none\n"
3990 " action when guest panics [default=shutdown]\n"
3991 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n"
3992 " action when watchdog fires [default=reset]\n",
3995 ``-action event=action``
3996 The action parameter serves to modify QEMU's default behavior when
3997 certain guest events occur. It provides a generic method for specifying the
3998 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown``
4003 ``-action panic=none``
4004 ``-action reboot=shutdown,shutdown=pause``
4005 ``-watchdog i6300esb -action watchdog=pause``
4009 DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \
4010 "-loadvm [tag|id]\n" \
4011 " start right away with a saved state (loadvm in monitor)\n",
4015 Start right away with a saved state (``loadvm`` in monitor)
4019 DEF("daemonize", 0, QEMU_OPTION_daemonize, \
4020 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL)
4024 Daemonize the QEMU process after initialization. QEMU will not
4025 detach from standard IO until it is ready to receive connections on
4026 any of its devices. This option is a useful way for external
4027 programs to launch QEMU without having to cope with initialization
4031 DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \
4032 "-option-rom rom load a file, rom, into the option ROM space\n",
4035 ``-option-rom file``
4036 Load the contents of file as an option ROM. This option is useful to
4037 load things like EtherBoot.
4040 DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \
4041 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \
4042 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n",
4046 ``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]``
4047 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at
4048 the current UTC or local time, respectively. ``localtime`` is
4049 required for correct date in MS-DOS or Windows. To start at a
4050 specific point in time, provide datetime in the format
4051 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC.
4053 By default the RTC is driven by the host system time. This allows
4054 using of the RTC as accurate reference clock inside the guest,
4055 specifically if the host time is smoothly following an accurate
4056 external reference clock, e.g. via NTP. If you want to isolate the
4057 guest time from the host, you can set ``clock`` to ``rt`` instead,
4058 which provides a host monotonic clock if host support it. To even
4059 prevent the RTC from progressing during suspension, you can set
4060 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is
4061 recommended especially in icount mode in order to preserve
4062 determinism; however, note that in icount mode the speed of the
4063 virtual clock is variable and can in general differ from the host
4066 Enable ``driftfix`` (i386 targets only) if you experience time drift
4067 problems, specifically with Windows' ACPI HAL. This option will try
4068 to figure out how many timer interrupts were not processed by the
4069 Windows guest and will re-inject them.
4072 DEF("icount", HAS_ARG, QEMU_OPTION_icount, \
4073 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \
4074 " enable virtual instruction counter with 2^N clock ticks per\n" \
4075 " instruction, enable aligning the host and virtual clocks\n" \
4076 " or disable real time cpu sleeping, and optionally enable\n" \
4077 " record-and-replay mode\n", QEMU_ARCH_ALL)
4079 ``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]``
4080 Enable virtual instruction counter. The virtual cpu will execute one
4081 instruction every 2^N ns of virtual time. If ``auto`` is specified
4082 then the virtual cpu speed will be automatically adjusted to keep
4083 virtual time within a few seconds of real time.
4085 Note that while this option can give deterministic behavior, it does
4086 not provide cycle accurate emulation. Modern CPUs contain
4087 superscalar out of order cores with complex cache hierarchies. The
4088 number of instructions executed often has little or no correlation
4089 with actual performance.
4091 When the virtual cpu is sleeping, the virtual time will advance at
4092 default speed unless ``sleep=on`` is specified. With
4093 ``sleep=on``, the virtual time will jump to the next timer
4094 deadline instantly whenever the virtual cpu goes to sleep mode and
4095 will not advance if no timer is enabled. This behavior gives
4096 deterministic execution times from the guest point of view.
4097 The default if icount is enabled is ``sleep=off``.
4098 ``sleep=on`` cannot be used together with either ``shift=auto``
4101 ``align=on`` will activate the delay algorithm which will try to
4102 synchronise the host clock and the virtual clock. The goal is to
4103 have a guest running at the real frequency imposed by the shift
4104 option. Whenever the guest clock is behind the host clock and if
4105 ``align=on`` is specified then we print a message to the user to
4106 inform about the delay. Currently this option does not work when
4107 ``shift`` is ``auto``. Note: The sync algorithm will work for those
4108 shift values for which the guest clock runs ahead of the host clock.
4109 Typically this happens when the shift value is high (how high
4110 depends on the host machine). The default if icount is enabled
4113 When the ``rr`` option is specified deterministic record/replay is
4114 enabled. The ``rrfile=`` option must also be provided to
4115 specify the path to the replay log. In record mode data is written
4116 to this file, and in replay mode it is read back.
4117 If the ``rrsnapshot`` option is given then it specifies a VM snapshot
4118 name. In record mode, a new VM snapshot with the given name is created
4119 at the start of execution recording. In replay mode this option
4120 specifies the snapshot name used to load the initial VM state.
4123 DEF("watchdog", HAS_ARG, QEMU_OPTION_watchdog, \
4124 "-watchdog model\n" \
4125 " enable virtual hardware watchdog [default=none]\n",
4129 Create a virtual hardware watchdog device. Once enabled (by a guest
4130 action), the watchdog must be periodically polled by an agent inside
4131 the guest or else the guest will be restarted. Choose a model for
4132 which your guest has drivers.
4134 The model is the model of hardware watchdog to emulate. Use
4135 ``-watchdog help`` to list available hardware models. Only one
4136 watchdog can be enabled for a guest.
4138 The following models may be available:
4141 iBASE 700 is a very simple ISA watchdog with a single timer.
4144 Intel 6300ESB I/O controller hub is a much more featureful
4145 PCI-based dual-timer watchdog.
4148 A virtual watchdog for s390x backed by the diagnose 288
4149 hypercall (currently KVM only).
4152 DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \
4153 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \
4154 " action when watchdog fires [default=reset]\n",
4157 ``-watchdog-action action``
4158 The action controls what QEMU will do when the watchdog timer
4159 expires. The default is ``reset`` (forcefully reset the guest).
4160 Other possible actions are: ``shutdown`` (attempt to gracefully
4161 shutdown the guest), ``poweroff`` (forcefully poweroff the guest),
4162 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the
4163 guest), ``debug`` (print a debug message and continue), or ``none``
4166 Note that the ``shutdown`` action requires that the guest responds
4167 to ACPI signals, which it may not be able to do in the sort of
4168 situations where the watchdog would have expired, and thus
4169 ``-watchdog-action shutdown`` is not recommended for production use.
4173 ``-watchdog i6300esb -watchdog-action pause``; \ ``-watchdog ib700``
4177 DEF("echr", HAS_ARG, QEMU_OPTION_echr, \
4178 "-echr chr set terminal escape character instead of ctrl-a\n",
4181 ``-echr numeric_ascii_value``
4182 Change the escape character used for switching to the monitor when
4183 using monitor and serial sharing. The default is ``0x01`` when using
4184 the ``-nographic`` option. ``0x01`` is equal to pressing
4185 ``Control-a``. You can select a different character from the ascii
4186 control keys where 1 through 26 map to Control-a through Control-z.
4187 For instance you could use the either of the following to change the
4188 escape character to Control-t.
4190 ``-echr 0x14``; \ ``-echr 20``
4194 DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \
4195 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \
4196 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \
4197 "-incoming unix:socketpath\n" \
4198 " prepare for incoming migration, listen on\n" \
4199 " specified protocol and socket address\n" \
4200 "-incoming fd:fd\n" \
4201 "-incoming exec:cmdline\n" \
4202 " accept incoming migration on given file descriptor\n" \
4203 " or from given external command\n" \
4204 "-incoming defer\n" \
4205 " wait for the URI to be specified via migrate_incoming\n",
4208 ``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]``
4210 ``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]``
4211 Prepare for incoming migration, listen on a given tcp port.
4213 ``-incoming unix:socketpath``
4214 Prepare for incoming migration, listen on a given unix socket.
4217 Accept incoming migration from a given filedescriptor.
4219 ``-incoming exec:cmdline``
4220 Accept incoming migration as an output from specified external
4224 Wait for the URI to be specified via migrate\_incoming. The monitor
4225 can be used to change settings (such as migration parameters) prior
4226 to issuing the migrate\_incoming to allow the migration to begin.
4229 DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \
4230 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL)
4232 ``-only-migratable``
4233 Only allow migratable devices. Devices will not be allowed to enter
4234 an unmigratable state.
4237 DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \
4238 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL)
4241 Don't create default devices. Normally, QEMU sets the default
4242 devices like serial port, parallel port, virtual console, monitor
4243 device, VGA adapter, floppy and CD-ROM drive and others. The
4244 ``-nodefaults`` option will disable all those default devices.
4248 DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \
4249 "-chroot dir chroot to dir just before starting the VM\n",
4254 Immediately before starting guest execution, chroot to the specified
4255 directory. Especially useful in combination with -runas.
4259 DEF("runas", HAS_ARG, QEMU_OPTION_runas, \
4260 "-runas user change to user id user just before starting the VM\n" \
4261 " user can be numeric uid:gid instead\n",
4266 Immediately before starting guest execution, drop root privileges,
4267 switching to the specified user.
4270 DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env,
4271 "-prom-env variable=value\n"
4272 " set OpenBIOS nvram variables\n",
4273 QEMU_ARCH_PPC | QEMU_ARCH_SPARC)
4275 ``-prom-env variable=value``
4276 Set OpenBIOS nvram variable to given value (PPC, SPARC only).
4280 qemu-system-sparc -prom-env 'auto-boot?=false' \
4281 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single'
4285 qemu-system-ppc -prom-env 'auto-boot?=false' \
4286 -prom-env 'boot-device=hd:2,\yaboot' \
4287 -prom-env 'boot-args=conf=hd:2,\yaboot.conf'
4289 DEF("semihosting", 0, QEMU_OPTION_semihosting,
4290 "-semihosting semihosting mode\n",
4291 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | QEMU_ARCH_LM32 |
4292 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4295 Enable semihosting mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only).
4297 Note that this allows guest direct access to the host filesystem, so
4298 should only be used with a trusted guest OS.
4300 See the -semihosting-config option documentation for further
4301 information about the facilities this enables.
4303 DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config,
4304 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]\n" \
4305 " semihosting configuration\n",
4306 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | QEMU_ARCH_LM32 |
4307 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4309 ``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]``
4310 Enable and configure semihosting (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V
4313 Note that this allows guest direct access to the host filesystem, so
4314 should only be used with a trusted guest OS.
4316 On Arm this implements the standard semihosting API, version 2.0.
4318 On M68K this implements the "ColdFire GDB" interface used by
4321 Xtensa semihosting provides basic file IO calls, such as
4322 open/read/write/seek/select. Tensilica baremetal libc for ISS and
4323 linux platform "sim" use this interface.
4325 On RISC-V this implements the standard semihosting API, version 0.2.
4327 ``target=native|gdb|auto``
4328 Defines where the semihosting calls will be addressed, to QEMU
4329 (``native``) or to GDB (``gdb``). The default is ``auto``, which
4330 means ``gdb`` during debug sessions and ``native`` otherwise.
4333 Send the output to a chardev backend output for native or auto
4334 output when not in gdb
4336 ``arg=str1,arg=str2,...``
4337 Allows the user to pass input arguments, and can be used
4338 multiple times to build up a list. The old-style
4339 ``-kernel``/``-append`` method of passing a command line is
4340 still supported for backward compatibility. If both the
4341 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are
4342 specified, the former is passed to semihosting as it always
4345 DEF("old-param", 0, QEMU_OPTION_old_param,
4346 "-old-param old param mode\n", QEMU_ARCH_ARM)
4349 Old param mode (ARM only).
4352 DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \
4353 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \
4354 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \
4355 " Enable seccomp mode 2 system call filter (default 'off').\n" \
4356 " use 'obsolete' to allow obsolete system calls that are provided\n" \
4357 " by the kernel, but typically no longer used by modern\n" \
4358 " C library implementations.\n" \
4359 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \
4360 " to elevate privileges using set*uid|gid system calls.\n" \
4361 " The value 'children' will deny set*uid|gid system calls for\n" \
4362 " main QEMU process but will allow forks and execves to run unprivileged\n" \
4363 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \
4364 " blocking *fork and execve\n" \
4365 " use 'resourcecontrol' to disable process affinity and schedular priority\n",
4368 ``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]``
4369 Enable Seccomp mode 2 system call filter. 'on' will enable syscall
4370 filtering and 'off' will disable it. The default is 'off'.
4373 Enable Obsolete system calls
4375 ``elevateprivileges=string``
4376 Disable set\*uid\|gid system calls
4379 Disable \*fork and execve
4381 ``resourcecontrol=string``
4382 Disable process affinity and schedular priority
4385 DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig,
4386 "-readconfig <file>\n", QEMU_ARCH_ALL)
4388 ``-readconfig file``
4389 Read device configuration from file. This approach is useful when
4390 you want to spawn QEMU process with many command line options but
4391 you don't want to exceed the command line character limit.
4393 DEF("writeconfig", HAS_ARG, QEMU_OPTION_writeconfig,
4394 "-writeconfig <file>\n"
4395 " read/write config file (deprecated)\n", QEMU_ARCH_ALL)
4399 DEF("no-user-config", 0, QEMU_OPTION_nouserconfig,
4401 " do not load default user-provided config files at startup\n",
4405 The ``-no-user-config`` option makes QEMU not load any of the
4406 user-provided config files on sysconfdir.
4409 DEF("trace", HAS_ARG, QEMU_OPTION_trace,
4410 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n"
4411 " specify tracing options\n",
4414 ``-trace [[enable=]pattern][,events=file][,file=file]``
4415 .. include:: ../qemu-option-trace.rst.inc
4418 DEF("plugin", HAS_ARG, QEMU_OPTION_plugin,
4419 "-plugin [file=]<file>[,arg=<string>]\n"
4423 ``-plugin file=file[,arg=string]``
4427 Load the given plugin from a shared library file.
4430 Argument string passed to the plugin. (Can be given multiple
4435 DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL)
4436 DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL)
4439 DEF("enable-fips", 0, QEMU_OPTION_enablefips,
4440 "-enable-fips enable FIPS 140-2 compliance\n",
4445 Enable FIPS 140-2 compliance mode.
4448 DEF("msg", HAS_ARG, QEMU_OPTION_msg,
4449 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n"
4450 " control error message format\n"
4451 " timestamp=on enables timestamps (default: off)\n"
4452 " guest-name=on enables guest name prefix but only if\n"
4453 " -name guest option is set (default: off)\n",
4456 ``-msg [timestamp[=on|off]][,guest-name[=on|off]]``
4457 Control error message format.
4459 ``timestamp=on|off``
4460 Prefix messages with a timestamp. Default is off.
4462 ``guest-name=on|off``
4463 Prefix messages with guest name but only if -name guest option is set
4464 otherwise the option is ignored. Default is off.
4467 DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate,
4468 "-dump-vmstate <file>\n"
4469 " Output vmstate information in JSON format to file.\n"
4470 " Use the scripts/vmstate-static-checker.py file to\n"
4471 " check for possible regressions in migration code\n"
4472 " by comparing two such vmstate dumps.\n",
4475 ``-dump-vmstate file``
4476 Dump json-encoded vmstate information for current machine type to
4480 DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile,
4481 "-enable-sync-profile\n"
4482 " enable synchronization profiling\n",
4485 ``-enable-sync-profile``
4486 Enable synchronization profiling.
4491 DEFHEADING(Generic object creation:)
4493 DEF("object", HAS_ARG, QEMU_OPTION_object,
4494 "-object TYPENAME[,PROP1=VALUE1,...]\n"
4495 " create a new object of type TYPENAME setting properties\n"
4496 " in the order they are specified. Note that the 'id'\n"
4497 " property must be set. These objects are placed in the\n"
4498 " '/objects' path.\n",
4501 ``-object typename[,prop1=value1,...]``
4502 Create a new object of type typename setting properties in the order
4503 they are specified. Note that the 'id' property must be set. These
4504 objects are placed in the '/objects' path.
4506 ``-object memory-backend-file,id=id,size=size,mem-path=dir,share=on|off,discard-data=on|off,merge=on|off,dump=on|off,prealloc=on|off,host-nodes=host-nodes,policy=default|preferred|bind|interleave,align=align,readonly=on|off``
4507 Creates a memory file backend object, which can be used to back
4508 the guest RAM with huge pages.
4510 The ``id`` parameter is a unique ID that will be used to
4511 reference this memory region in other parameters, e.g. ``-numa``,
4512 ``-device nvdimm``, etc.
4514 The ``size`` option provides the size of the memory region, and
4515 accepts common suffixes, e.g. ``500M``.
4517 The ``mem-path`` provides the path to either a shared memory or
4518 huge page filesystem mount.
4520 The ``share`` boolean option determines whether the memory
4521 region is marked as private to QEMU, or shared. The latter
4522 allows a co-operating external process to access the QEMU memory
4525 The ``share`` is also required for pvrdma devices due to
4526 limitations in the RDMA API provided by Linux.
4528 Setting share=on might affect the ability to configure NUMA
4529 bindings for the memory backend under some circumstances, see
4530 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel
4531 source tree for additional details.
4533 Setting the ``discard-data`` boolean option to on indicates that
4534 file contents can be destroyed when QEMU exits, to avoid
4535 unnecessarily flushing data to the backing file. Note that
4536 ``discard-data`` is only an optimization, and QEMU might not
4537 discard file contents if it aborts unexpectedly or is terminated
4540 The ``merge`` boolean option enables memory merge, also known as
4541 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider
4542 the pages for memory deduplication.
4544 Setting the ``dump`` boolean option to off excludes the memory
4545 from core dumps. This feature is also known as MADV\_DONTDUMP.
4547 The ``prealloc`` boolean option enables memory preallocation.
4549 The ``host-nodes`` option binds the memory range to a list of
4552 The ``policy`` option sets the NUMA policy to one of the
4559 prefer the given host node list for allocation
4562 restrict memory allocation to the given host node list
4565 interleave memory allocations across the given host node
4568 The ``align`` option specifies the base address alignment when
4569 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg
4570 ``2M``. Some backend store specified by ``mem-path`` requires an
4571 alignment different than the default one used by QEMU, eg the
4572 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
4573 such cases, users can specify the required alignment via this
4576 The ``pmem`` option specifies whether the backing file specified
4577 by ``mem-path`` is in host persistent memory that can be
4578 accessed using the SNIA NVM programming model (e.g. Intel
4579 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary
4580 operations to guarantee the persistence of its own writes to
4581 ``mem-path`` (e.g. in vNVDIMM label emulation and live
4582 migration). Also, we will map the backend-file with MAP\_SYNC
4583 flag, which ensures the file metadata is in sync for
4584 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC
4585 requires support from both the host kernel (since Linux kernel
4586 4.15) and the filesystem of ``mem-path`` mounted with DAX
4589 The ``readonly`` option specifies whether the backing file is opened
4590 read-only or read-write (default).
4592 ``-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``
4593 Creates a memory backend object, which can be used to back the
4594 guest RAM. Memory backend objects offer more control than the
4595 ``-m`` option that is traditionally used to define guest RAM.
4596 Please refer to ``memory-backend-file`` for a description of the
4599 ``-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``
4600 Creates an anonymous memory file backend object, which allows
4601 QEMU to share the memory with an external process (e.g. when
4602 using vhost-user). The memory is allocated with memfd and
4603 optional sealing. (Linux only)
4605 The ``seal`` option creates a sealed-file, that will block
4606 further resizing the memory ('on' by default).
4608 The ``hugetlb`` option specify the file to be created resides in
4609 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction
4610 with the ``hugetlb`` option, the ``hugetlbsize`` option specify
4611 the hugetlb page size on systems that support multiple hugetlb
4612 page sizes (it must be a power of 2 value supported by the
4615 In some versions of Linux, the ``hugetlb`` option is
4616 incompatible with the ``seal`` option (requires at least Linux
4619 Please refer to ``memory-backend-file`` for a description of the
4622 The ``share`` boolean option is on by default with memfd.
4624 ``-object rng-builtin,id=id``
4625 Creates a random number generator backend which obtains entropy
4626 from QEMU builtin functions. The ``id`` parameter is a unique ID
4627 that will be used to reference this entropy backend from the
4628 ``virtio-rng`` device. By default, the ``virtio-rng`` device
4629 uses this RNG backend.
4631 ``-object rng-random,id=id,filename=/dev/random``
4632 Creates a random number generator backend which obtains entropy
4633 from a device on the host. The ``id`` parameter is a unique ID
4634 that will be used to reference this entropy backend from the
4635 ``virtio-rng`` device. The ``filename`` parameter specifies
4636 which file to obtain entropy from and if omitted defaults to
4639 ``-object rng-egd,id=id,chardev=chardevid``
4640 Creates a random number generator backend which obtains entropy
4641 from an external daemon running on the host. The ``id``
4642 parameter is a unique ID that will be used to reference this
4643 entropy backend from the ``virtio-rng`` device. The ``chardev``
4644 parameter is the unique ID of a character device backend that
4645 provides the connection to the RNG daemon.
4647 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off``
4648 Creates a TLS anonymous credentials object, which can be used to
4649 provide TLS support on network backends. The ``id`` parameter is
4650 a unique ID which network backends will use to access the
4651 credentials. The ``endpoint`` is either ``server`` or ``client``
4652 depending on whether the QEMU network backend that uses the
4653 credentials will be acting as a client or as a server. If
4654 ``verify-peer`` is enabled (the default) then once the handshake
4655 is completed, the peer credentials will be verified, though this
4656 is a no-op for anonymous credentials.
4658 The dir parameter tells QEMU where to find the credential files.
4659 For server endpoints, this directory may contain a file
4660 dh-params.pem providing diffie-hellman parameters to use for the
4661 TLS server. If the file is missing, QEMU will generate a set of
4662 DH parameters at startup. This is a computationally expensive
4663 operation that consumes random pool entropy, so it is
4664 recommended that a persistent set of parameters be generated
4667 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]``
4668 Creates a TLS Pre-Shared Keys (PSK) credentials object, which
4669 can be used to provide TLS support on network backends. The
4670 ``id`` parameter is a unique ID which network backends will use
4671 to access the credentials. The ``endpoint`` is either ``server``
4672 or ``client`` depending on whether the QEMU network backend that
4673 uses the credentials will be acting as a client or as a server.
4674 For clients only, ``username`` is the username which will be
4675 sent to the server. If omitted it defaults to "qemu".
4677 The dir parameter tells QEMU where to find the keys file. It is
4678 called "dir/keys.psk" and contains "username:key" pairs. This
4679 file can most easily be created using the GnuTLS ``psktool``
4682 For server endpoints, dir may also contain a file dh-params.pem
4683 providing diffie-hellman parameters to use for the TLS server.
4684 If the file is missing, QEMU will generate a set of DH
4685 parameters at startup. This is a computationally expensive
4686 operation that consumes random pool entropy, so it is
4687 recommended that a persistent set of parameters be generated up
4690 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id``
4691 Creates a TLS anonymous credentials object, which can be used to
4692 provide TLS support on network backends. The ``id`` parameter is
4693 a unique ID which network backends will use to access the
4694 credentials. The ``endpoint`` is either ``server`` or ``client``
4695 depending on whether the QEMU network backend that uses the
4696 credentials will be acting as a client or as a server. If
4697 ``verify-peer`` is enabled (the default) then once the handshake
4698 is completed, the peer credentials will be verified. With x509
4699 certificates, this implies that the clients must be provided
4700 with valid client certificates too.
4702 The dir parameter tells QEMU where to find the credential files.
4703 For server endpoints, this directory may contain a file
4704 dh-params.pem providing diffie-hellman parameters to use for the
4705 TLS server. If the file is missing, QEMU will generate a set of
4706 DH parameters at startup. This is a computationally expensive
4707 operation that consumes random pool entropy, so it is
4708 recommended that a persistent set of parameters be generated
4711 For x509 certificate credentials the directory will contain
4712 further files providing the x509 certificates. The certificates
4713 must be stored in PEM format, in filenames ca-cert.pem,
4714 ca-crl.pem (optional), server-cert.pem (only servers),
4715 server-key.pem (only servers), client-cert.pem (only clients),
4716 and client-key.pem (only clients).
4718 For the server-key.pem and client-key.pem files which contain
4719 sensitive private keys, it is possible to use an encrypted
4720 version by providing the passwordid parameter. This provides the
4721 ID of a previously created ``secret`` object containing the
4722 password for decryption.
4724 The priority parameter allows to override the global default
4725 priority used by gnutls. This can be useful if the system
4726 administrator needs to use a weaker set of crypto priorities for
4727 QEMU without potentially forcing the weakness onto all
4728 applications. Or conversely if one wants wants a stronger
4729 default for QEMU than for all other applications, they can do
4730 this through this parameter. Its format is a gnutls priority
4731 string as described at
4732 https://gnutls.org/manual/html_node/Priority-Strings.html.
4734 ``-object tls-cipher-suites,id=id,priority=priority``
4735 Creates a TLS cipher suites object, which can be used to control
4736 the TLS cipher/protocol algorithms that applications are permitted
4739 The ``id`` parameter is a unique ID which frontends will use to
4740 access the ordered list of permitted TLS cipher suites from the
4743 The ``priority`` parameter allows to override the global default
4744 priority used by gnutls. This can be useful if the system
4745 administrator needs to use a weaker set of crypto priorities for
4746 QEMU without potentially forcing the weakness onto all
4747 applications. Or conversely if one wants wants a stronger
4748 default for QEMU than for all other applications, they can do
4749 this through this parameter. Its format is a gnutls priority
4750 string as described at
4751 https://gnutls.org/manual/html_node/Priority-Strings.html.
4753 An example of use of this object is to control UEFI HTTPS Boot.
4754 The tls-cipher-suites object exposes the ordered list of permitted
4755 TLS cipher suites from the host side to the guest firmware, via
4756 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER
4757 objects. The firmware uses the IANA_TLS_CIPHER array for configuring
4760 In the following example, the priority at which the host-side policy
4761 is retrieved is given by the ``priority`` property.
4762 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to
4763 refer to /etc/crypto-policies/back-ends/gnutls.config.
4768 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\
4769 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0
4771 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]``
4772 Interval t can't be 0, this filter batches the packet delivery:
4773 all packets arriving in a given interval on netdev netdevid are
4774 delayed until the end of the interval. Interval is in
4775 microseconds. ``status`` is optional that indicate whether the
4776 netfilter is on (enabled) or off (disabled), the default status
4777 for netfilter will be 'on'.
4779 queue all\|rx\|tx is an option that can be applied to any
4782 ``all``: the filter is attached both to the receive and the
4783 transmit queue of the netdev (default).
4785 ``rx``: the filter is attached to the receive queue of the
4786 netdev, where it will receive packets sent to the netdev.
4788 ``tx``: the filter is attached to the transmit queue of the
4789 netdev, where it will receive packets sent by the netdev.
4791 position head\|tail\|id=<id> is an option to specify where the
4792 filter should be inserted in the filter list. It can be applied
4795 ``head``: the filter is inserted at the head of the filter list,
4796 before any existing filters.
4798 ``tail``: the filter is inserted at the tail of the filter list,
4799 behind any existing filters (default).
4801 ``id=<id>``: the filter is inserted before or behind the filter
4802 specified by <id>, see the insert option below.
4804 insert behind\|before is an option to specify where to insert
4805 the new filter relative to the one specified with
4806 position=id=<id>. It can be applied to any netfilter.
4808 ``before``: insert before the specified filter.
4810 ``behind``: insert behind the specified filter (default).
4812 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
4813 filter-mirror on netdev netdevid,mirror net packet to
4814 chardevchardevid, if it has the vnet\_hdr\_support flag,
4815 filter-mirror will mirror packet with vnet\_hdr\_len.
4817 ``-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]``
4818 filter-redirector on netdev netdevid,redirect filter's net
4819 packet to chardev chardevid,and redirect indev's packet to
4820 filter.if it has the vnet\_hdr\_support flag, filter-redirector
4821 will redirect packet with vnet\_hdr\_len. Create a
4822 filter-redirector we need to differ outdev id from indev id, id
4823 can not be the same. we can just use indev or outdev, but at
4824 least one of indev or outdev need to be specified.
4826 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
4827 Filter-rewriter is a part of COLO project.It will rewrite tcp
4828 packet to secondary from primary to keep secondary tcp
4829 connection,and rewrite tcp packet to primary from secondary make
4830 tcp packet can be handled by client.if it has the
4831 vnet\_hdr\_support flag, we can parse packet with vnet header.
4833 usage: colo secondary: -object
4834 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object
4835 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object
4836 filter-rewriter,id=rew0,netdev=hn0,queue=all
4838 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]``
4839 Dump the network traffic on netdev dev to the file specified by
4840 filename. At most len bytes (64k by default) per packet are
4841 stored. The file format is libpcap, so it can be analyzed with
4842 tools such as tcpdump or Wireshark.
4844 ``-object colo-compare,id=id,primary_in=chardevid,secondary_in=chardevid,outdev=chardevid,iothread=id[,vnet_hdr_support][,notify_dev=id][,compare_timeout=@var{ms}][,expired_scan_cycle=@var{ms}][,max_queue_size=@var{size}]``
4845 Colo-compare gets packet from primary\_in chardevid and
4846 secondary\_in, then compare whether the payload of primary packet
4847 and secondary packet are the same. If same, it will output
4848 primary packet to out\_dev, else it will notify COLO-framework to do
4849 checkpoint and send primary packet to out\_dev. In order to
4850 improve efficiency, we need to put the task of comparison in
4851 another iothread. If it has the vnet\_hdr\_support flag,
4852 colo compare will send/recv packet with vnet\_hdr\_len.
4853 The compare\_timeout=@var{ms} determines the maximum time of the
4854 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms}
4855 is to set the period of scanning expired primary node network packets.
4856 The max\_queue\_size=@var{size} is to set the max compare queue
4857 size depend on user environment.
4858 If user want to use Xen COLO, need to add the notify\_dev to
4859 notify Xen colo-frame to do checkpoint.
4861 COLO-compare must be used with the help of filter-mirror,
4862 filter-redirector and filter-rewriter.
4869 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
4870 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
4871 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
4872 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
4873 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
4874 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
4875 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
4876 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
4877 -object iothread,id=iothread1
4878 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
4879 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
4880 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
4881 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1
4884 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
4885 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
4886 -chardev socket,id=red0,host=3.3.3.3,port=9003
4887 -chardev socket,id=red1,host=3.3.3.3,port=9004
4888 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
4889 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
4895 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
4896 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
4897 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
4898 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
4899 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
4900 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
4901 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
4902 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
4903 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off
4904 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
4905 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
4906 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
4907 -object iothread,id=iothread1
4908 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1
4911 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
4912 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
4913 -chardev socket,id=red0,host=3.3.3.3,port=9003
4914 -chardev socket,id=red1,host=3.3.3.3,port=9004
4915 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
4916 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
4918 If you want to know the detail of above command line, you can
4919 read the colo-compare git log.
4921 ``-object cryptodev-backend-builtin,id=id[,queues=queues]``
4922 Creates a cryptodev backend which executes crypto opreation from
4923 the QEMU cipher APIS. The id parameter is a unique ID that will
4924 be used to reference this cryptodev backend from the
4925 ``virtio-crypto`` device. The queues parameter is optional,
4926 which specify the queue number of cryptodev backend, the default
4933 -object cryptodev-backend-builtin,id=cryptodev0 \\
4934 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
4937 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]``
4938 Creates a vhost-user cryptodev backend, backed by a chardev
4939 chardevid. The id parameter is a unique ID that will be used to
4940 reference this cryptodev backend from the ``virtio-crypto``
4941 device. The chardev should be a unix domain socket backed one.
4942 The vhost-user uses a specifically defined protocol to pass
4943 vhost ioctl replacement messages to an application on the other
4944 end of the socket. The queues parameter is optional, which
4945 specify the queue number of cryptodev backend for multiqueue
4946 vhost-user, the default of queues is 1.
4952 -chardev socket,id=chardev0,path=/path/to/socket \\
4953 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\
4954 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
4957 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]``
4959 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]``
4960 Defines a secret to store a password, encryption key, or some
4961 other sensitive data. The sensitive data can either be passed
4962 directly via the data parameter, or indirectly via the file
4963 parameter. Using the data parameter is insecure unless the
4964 sensitive data is encrypted.
4966 The sensitive data can be provided in raw format (the default),
4967 or base64. When encoded as JSON, the raw format only supports
4968 valid UTF-8 characters, so base64 is recommended for sending
4969 binary data. QEMU will convert from which ever format is
4970 provided to the format it needs internally. eg, an RBD password
4971 can be provided in raw format, even though it will be base64
4972 encoded when passed onto the RBD sever.
4974 For added protection, it is possible to encrypt the data
4975 associated with a secret using the AES-256-CBC cipher. Use of
4976 encryption is indicated by providing the keyid and iv
4977 parameters. The keyid parameter provides the ID of a previously
4978 defined secret that contains the AES-256 decryption key. This
4979 key should be 32-bytes long and be base64 encoded. The iv
4980 parameter provides the random initialization vector used for
4981 encryption of this particular secret and should be a base64
4982 encrypted string of the 16-byte IV.
4984 The simplest (insecure) usage is to provide the secret inline
4988 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw
4990 The simplest secure usage is to provide the secret via a file
4992 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object
4993 secret,id=sec0,file=mypasswd.txt,format=raw
4995 For greater security, AES-256-CBC should be used. To illustrate
4996 usage, consider the openssl command line tool which can encrypt
4997 the data. Note that when encrypting, the plaintext must be
4998 padded to the cipher block size (32 bytes) using the standard
4999 PKCS#5/6 compatible padding algorithm.
5001 First a master key needs to be created in base64 encoding:
5005 # openssl rand -base64 32 > key.b64
5006 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"')
5008 Each secret to be encrypted needs to have a random
5009 initialization vector generated. These do not need to be kept
5014 # openssl rand -base64 16 > iv.b64
5015 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"')
5017 The secret to be defined can now be encrypted, in this case
5018 we're telling openssl to base64 encode the result, but it could
5019 be left as raw bytes if desired.
5023 # SECRET=$(printf "letmein" |
5024 openssl enc -aes-256-cbc -a -K $KEY -iv $IV)
5026 When launching QEMU, create a master secret pointing to
5027 ``key.b64`` and specify that to be used to decrypt the user
5028 password. Pass the contents of ``iv.b64`` to the second secret
5033 -object secret,id=secmaster0,format=base64,file=key.b64 \\
5034 -object secret,id=sec0,keyid=secmaster0,format=base64,\\
5035 data=$SECRET,iv=$(<iv.b64)
5037 ``-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]``
5038 Create a Secure Encrypted Virtualization (SEV) guest object,
5039 which can be used to provide the guest memory encryption support
5042 When memory encryption is enabled, one of the physical address
5043 bit (aka the C-bit) is utilized to mark if a memory page is
5044 protected. The ``cbitpos`` is used to provide the C-bit
5045 position. The C-bit position is Host family dependent hence user
5046 must provide this value. On EPYC, the value should be 47.
5048 When memory encryption is enabled, we loose certain bits in
5049 physical address space. The ``reduced-phys-bits`` is used to
5050 provide the number of bits we loose in physical address space.
5051 Similar to C-bit, the value is Host family dependent. On EPYC,
5052 the value should be 5.
5054 The ``sev-device`` provides the device file to use for
5055 communicating with the SEV firmware running inside AMD Secure
5056 Processor. The default device is '/dev/sev'. If hardware
5057 supports memory encryption then /dev/sev devices are created by
5060 The ``policy`` provides the guest policy to be enforced by the
5061 SEV firmware and restrict what configuration and operational
5062 commands can be performed on this guest by the hypervisor. The
5063 policy should be provided by the guest owner and is bound to the
5064 guest and cannot be changed throughout the lifetime of the
5065 guest. The default is 0.
5067 If guest ``policy`` allows sharing the key with another SEV
5068 guest then ``handle`` can be use to provide handle of the guest
5069 from which to share the key.
5071 The ``dh-cert-file`` and ``session-file`` provides the guest
5072 owner's Public Diffie-Hillman key defined in SEV spec. The PDH
5073 and session parameters are used for establishing a cryptographic
5074 session with the guest owner to negotiate keys used for
5075 attestation. The file must be encoded in base64.
5077 e.g to launch a SEV guest
5081 # |qemu_system_x86| \\
5083 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \\
5084 -machine ...,memory-encryption=sev0 \\
5087 ``-object authz-simple,id=id,identity=string``
5088 Create an authorization object that will control access to
5091 The ``identity`` parameter is identifies the user and its format
5092 depends on the network service that authorization object is
5093 associated with. For authorizing based on TLS x509 certificates,
5094 the identity must be the x509 distinguished name. Note that care
5095 must be taken to escape any commas in the distinguished name.
5097 An example authorization object to validate a x509 distinguished
5098 name would look like:
5104 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\
5107 Note the use of quotes due to the x509 distinguished name
5108 containing whitespace, and escaping of ','.
5110 ``-object authz-listfile,id=id,filename=path,refresh=on|off``
5111 Create an authorization object that will control access to
5114 The ``filename`` parameter is the fully qualified path to a file
5115 containing the access control list rules in JSON format.
5117 An example set of rules that match against SASL usernames might
5124 { "match": "fred", "policy": "allow", "format": "exact" },
5125 { "match": "bob", "policy": "allow", "format": "exact" },
5126 { "match": "danb", "policy": "deny", "format": "glob" },
5127 { "match": "dan*", "policy": "allow", "format": "exact" },
5132 When checking access the object will iterate over all the rules
5133 and the first rule to match will have its ``policy`` value
5134 returned as the result. If no rules match, then the default
5135 ``policy`` value is returned.
5137 The rules can either be an exact string match, or they can use
5138 the simple UNIX glob pattern matching to allow wildcards to be
5141 If ``refresh`` is set to true the file will be monitored and
5142 automatically reloaded whenever its content changes.
5144 As with the ``authz-simple`` object, the format of the identity
5145 strings being matched depends on the network service, but is
5146 usually a TLS x509 distinguished name, or a SASL username.
5148 An example authorization object to validate a SASL username
5155 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\
5158 ``-object authz-pam,id=id,service=string``
5159 Create an authorization object that will control access to
5162 The ``service`` parameter provides the name of a PAM service to
5163 use for authorization. It requires that a file
5164 ``/etc/pam.d/service`` exist to provide the configuration for
5165 the ``account`` subsystem.
5167 An example authorization object to validate a TLS x509
5168 distinguished name would look like:
5174 -object authz-pam,id=auth0,service=qemu-vnc \\
5177 There would then be a corresponding config file for PAM at
5178 ``/etc/pam.d/qemu-vnc`` that contains:
5182 account requisite pam_listfile.so item=user sense=allow \
5183 file=/etc/qemu/vnc.allow
5185 Finally the ``/etc/qemu/vnc.allow`` file would contain the list
5186 of x509 distingished names that are permitted access
5190 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB
5192 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink``
5193 Creates a dedicated event loop thread that devices can be
5194 assigned to. This is known as an IOThread. By default device
5195 emulation happens in vCPU threads or the main event loop thread.
5196 This can become a scalability bottleneck. IOThreads allow device
5197 emulation and I/O to run on other host CPUs.
5199 The ``id`` parameter is a unique ID that will be used to
5200 reference this IOThread from ``-device ...,iothread=id``.
5201 Multiple devices can be assigned to an IOThread. Note that not
5202 all devices support an ``iothread`` parameter.
5204 The ``query-iothreads`` QMP command lists IOThreads and reports
5205 their thread IDs so that the user can configure host CPU
5208 IOThreads use an adaptive polling algorithm to reduce event loop
5209 latency. Instead of entering a blocking system call to monitor
5210 file descriptors and then pay the cost of being woken up when an
5211 event occurs, the polling algorithm spins waiting for events for
5212 a short time. The algorithm's default parameters are suitable
5213 for many cases but can be adjusted based on knowledge of the
5214 workload and/or host device latency.
5216 The ``poll-max-ns`` parameter is the maximum number of
5217 nanoseconds to busy wait for events. Polling can be disabled by
5218 setting this value to 0.
5220 The ``poll-grow`` parameter is the multiplier used to increase
5221 the polling time when the algorithm detects it is missing events
5222 due to not polling long enough.
5224 The ``poll-shrink`` parameter is the divisor used to decrease
5225 the polling time when the algorithm detects it is spending too
5226 long polling without encountering events.
5228 The polling parameters can be modified at run-time using the
5229 ``qom-set`` command (where ``iothread1`` is the IOThread's
5234 (qemu) qom-set /objects/iothread1 poll-max-ns 100000
5238 HXCOMM This is the last statement. Insert new options before this line!
projects / qemu.git / blob