qapi: Replace ad hoc "since" documentation by member documentation
[qemu/armbru.git] / qemu-options.hx
blobb5efa648bad1f34446d049a59687679f176dea0b
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
6 HXCOMM architectures.
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)
13 SRST
14 ``-h``
15 Display help and exit
16 ERST
18 DEF("version", 0, QEMU_OPTION_version,
19 "-version display version information and exit\n", QEMU_ARCH_ALL)
20 SRST
21 ``-version``
22 Display version information and exit
23 ERST
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"
40 " cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]\n",
41 QEMU_ARCH_ALL)
42 SRST
43 ``-machine [type=]name[,prop=value[,...]]``
44 Select the emulated machine by name. Use ``-machine help`` to list
45 available machines.
47 For architectures which aim to support live migration compatibility
48 across releases, each release will introduce a new versioned machine
49 type. For example, the 2.8.0 release introduced machine types
50 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures.
52 To allow live migration of guests from QEMU version 2.8.0, to QEMU
53 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8"
54 and "pc-q35-2.8" machines too. To allow users live migrating VMs to
55 skip multiple intermediate releases when upgrading, new releases of
56 QEMU will support machine types from many previous versions.
58 Supported machine properties are:
60 ``accel=accels1[:accels2[:...]]``
61 This is used to enable an accelerator. Depending on the target
62 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available.
63 By default, tcg is used. If there is more than one accelerator
64 specified, the next one is used if the previous one fails to
65 initialize.
67 ``vmport=on|off|auto``
68 Enables emulation of VMWare IO port, for vmmouse etc. auto says
69 to select the value based on accel. For accel=xen the default is
70 off otherwise the default is on.
72 ``dump-guest-core=on|off``
73 Include guest memory in a core dump. The default is on.
75 ``mem-merge=on|off``
76 Enables or disables memory merge support. This feature, when
77 supported by the host, de-duplicates identical memory pages
78 among VMs instances (enabled by default).
80 ``aes-key-wrap=on|off``
81 Enables or disables AES key wrapping support on s390-ccw hosts.
82 This feature controls whether AES wrapping keys will be created
83 to allow execution of AES cryptographic functions. The default
84 is on.
86 ``dea-key-wrap=on|off``
87 Enables or disables DEA key wrapping support on s390-ccw hosts.
88 This feature controls whether DEA wrapping keys will be created
89 to allow execution of DEA cryptographic functions. The default
90 is on.
92 ``nvdimm=on|off``
93 Enables or disables NVDIMM support. The default is off.
95 ``memory-encryption=``
96 Memory encryption object to use. The default is none.
98 ``hmat=on|off``
99 Enables or disables ACPI Heterogeneous Memory Attribute Table
100 (HMAT) support. The default is off.
102 ``memory-backend='id'``
103 An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options.
104 Allows to use a memory backend as main RAM.
106 For example:
109 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on
110 -machine memory-backend=pc.ram
111 -m 512M
113 Migration compatibility note:
115 * as backend id one shall use value of 'default-ram-id', advertised by
116 machine type (available via ``query-machines`` QMP command), if migration
117 to/from old QEMU (<5.0) is expected.
118 * for machine types 4.0 and older, user shall
119 use ``x-use-canonical-path-for-ramblock-id=off`` backend option
120 if migration to/from old QEMU (<5.0) is expected.
122 For example:
125 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off
126 -machine memory-backend=pc.ram
127 -m 512M
129 ``cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]``
130 Define a CXL Fixed Memory Window (CFMW).
132 Described in the CXL 2.0 ECN: CEDT CFMWS & QTG _DSM.
134 They are regions of Host Physical Addresses (HPA) on a system which
135 may be interleaved across one or more CXL host bridges. The system
136 software will assign particular devices into these windows and
137 configure the downstream Host-managed Device Memory (HDM) decoders
138 in root ports, switch ports and devices appropriately to meet the
139 interleave requirements before enabling the memory devices.
141 ``targets.X=target`` provides the mapping to CXL host bridges
142 which may be identified by the id provided in the -device entry.
143 Multiple entries are needed to specify all the targets when
144 the fixed memory window represents interleaved memory. X is the
145 target index from 0.
147 ``size=size`` sets the size of the CFMW. This must be a multiple of
148 256MiB. The region will be aligned to 256MiB but the location is
149 platform and configuration dependent.
151 ``interleave-granularity=granularity`` sets the granularity of
152 interleave. Default 256KiB. Only 256KiB, 512KiB, 1024KiB, 2048KiB
153 4096KiB, 8192KiB and 16384KiB granularities supported.
155 Example:
159 -machine cxl-fmw.0.targets.0=cxl.0,cxl-fmw.0.targets.1=cxl.1,cxl-fmw.0.size=128G,cxl-fmw.0.interleave-granularity=512k
160 ERST
162 DEF("M", HAS_ARG, QEMU_OPTION_M,
163 " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n",
164 QEMU_ARCH_ALL)
166 SRST
167 ``sgx-epc.0.memdev=@var{memid},sgx-epc.0.node=@var{numaid}``
168 Define an SGX EPC section.
169 ERST
171 DEF("cpu", HAS_ARG, QEMU_OPTION_cpu,
172 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL)
173 SRST
174 ``-cpu model``
175 Select CPU model (``-cpu help`` for list and additional feature
176 selection)
177 ERST
179 DEF("accel", HAS_ARG, QEMU_OPTION_accel,
180 "-accel [accel=]accelerator[,prop[=value][,...]]\n"
181 " select accelerator (kvm, xen, hax, hvf, nvmm, whpx or tcg; use 'help' for a list)\n"
182 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n"
183 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n"
184 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n"
185 " split-wx=on|off (enable TCG split w^x mapping)\n"
186 " tb-size=n (TCG translation block cache size)\n"
187 " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n"
188 " notify-vmexit=run|internal-error|disable,notify-window=n (enable notify VM exit and set notify window, x86 only)\n"
189 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL)
190 SRST
191 ``-accel name[,prop=value[,...]]``
192 This is used to enable an accelerator. Depending on the target
193 architecture, kvm, xen, hax, hvf, nvmm, whpx or tcg can be available. By
194 default, tcg is used. If there is more than one accelerator
195 specified, the next one is used if the previous one fails to
196 initialize.
198 ``igd-passthru=on|off``
199 When Xen is in use, this option controls whether Intel
200 integrated graphics devices can be passed through to the guest
201 (default=off)
203 ``kernel-irqchip=on|off|split``
204 Controls KVM in-kernel irqchip support. The default is full
205 acceleration of the interrupt controllers. On x86, split irqchip
206 reduces the kernel attack surface, at a performance cost for
207 non-MSI interrupts. Disabling the in-kernel irqchip completely
208 is not recommended except for debugging purposes.
210 ``kvm-shadow-mem=size``
211 Defines the size of the KVM shadow MMU.
213 ``split-wx=on|off``
214 Controls the use of split w^x mapping for the TCG code generation
215 buffer. Some operating systems require this to be enabled, and in
216 such a case this will default on. On other operating systems, this
217 will default off, but one may enable this for testing or debugging.
219 ``tb-size=n``
220 Controls the size (in MiB) of the TCG translation block cache.
222 ``thread=single|multi``
223 Controls number of TCG threads. When the TCG is multi-threaded
224 there will be one thread per vCPU therefore taking advantage of
225 additional host cores. The default is to enable multi-threading
226 where both the back-end and front-ends support it and no
227 incompatible TCG features have been enabled (e.g.
228 icount/replay).
230 ``dirty-ring-size=n``
231 When the KVM accelerator is used, it controls the size of the per-vCPU
232 dirty page ring buffer (number of entries for each vCPU). It should
233 be a value that is power of two, and it should be 1024 or bigger (but
234 still less than the maximum value that the kernel supports). 4096
235 could be a good initial value if you have no idea which is the best.
236 Set this value to 0 to disable the feature. By default, this feature
237 is disabled (dirty-ring-size=0). When enabled, KVM will instead
238 record dirty pages in a bitmap.
240 ``notify-vmexit=run|internal-error|disable,notify-window=n``
241 Enables or disables notify VM exit support on x86 host and specify
242 the corresponding notify window to trigger the VM exit if enabled.
243 ``run`` option enables the feature. It does nothing and continue
244 if the exit happens. ``internal-error`` option enables the feature.
245 It raises a internal error. ``disable`` option doesn't enable the feature.
246 This feature can mitigate the CPU stuck issue due to event windows don't
247 open up for a specified of time (i.e. notify-window).
248 Default: notify-vmexit=run,notify-window=0.
250 ERST
252 DEF("smp", HAS_ARG, QEMU_OPTION_smp,
253 "-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]\n"
254 " set the number of initial CPUs to 'n' [default=1]\n"
255 " maxcpus= maximum number of total CPUs, including\n"
256 " offline CPUs for hotplug, etc\n"
257 " sockets= number of sockets on the machine board\n"
258 " dies= number of dies in one socket\n"
259 " clusters= number of clusters in one die\n"
260 " cores= number of cores in one cluster\n"
261 " threads= number of threads in one core\n"
262 "Note: Different machines may have different subsets of the CPU topology\n"
263 " parameters supported, so the actual meaning of the supported parameters\n"
264 " will vary accordingly. For example, for a machine type that supports a\n"
265 " three-level CPU hierarchy of sockets/cores/threads, the parameters will\n"
266 " sequentially mean as below:\n"
267 " sockets means the number of sockets on the machine board\n"
268 " cores means the number of cores in one socket\n"
269 " threads means the number of threads in one core\n"
270 " For a particular machine type board, an expected CPU topology hierarchy\n"
271 " can be defined through the supported sub-option. Unsupported parameters\n"
272 " can also be provided in addition to the sub-option, but their values\n"
273 " must be set as 1 in the purpose of correct parsing.\n",
274 QEMU_ARCH_ALL)
275 SRST
276 ``-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]``
277 Simulate a SMP system with '\ ``n``\ ' CPUs initially present on
278 the machine type board. On boards supporting CPU hotplug, the optional
279 '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be
280 added at runtime. When both parameters are omitted, the maximum number
281 of CPUs will be calculated from the provided topology members and the
282 initial CPU count will match the maximum number. When only one of them
283 is given then the omitted one will be set to its counterpart's value.
284 Both parameters may be specified, but the maximum number of CPUs must
285 be equal to or greater than the initial CPU count. Product of the
286 CPU topology hierarchy must be equal to the maximum number of CPUs.
287 Both parameters are subject to an upper limit that is determined by
288 the specific machine type chosen.
290 To control reporting of CPU topology information, values of the topology
291 parameters can be specified. Machines may only support a subset of the
292 parameters and different machines may have different subsets supported
293 which vary depending on capacity of the corresponding CPU targets. So
294 for a particular machine type board, an expected topology hierarchy can
295 be defined through the supported sub-option. Unsupported parameters can
296 also be provided in addition to the sub-option, but their values must be
297 set as 1 in the purpose of correct parsing.
299 Either the initial CPU count, or at least one of the topology parameters
300 must be specified. The specified parameters must be greater than zero,
301 explicit configuration like "cpus=0" is not allowed. Values for any
302 omitted parameters will be computed from those which are given.
304 For example, the following sub-option defines a CPU topology hierarchy
305 (2 sockets totally on the machine, 2 cores per socket, 2 threads per
306 core) for a machine that only supports sockets/cores/threads.
307 Some members of the option can be omitted but their values will be
308 automatically computed:
312 -smp 8,sockets=2,cores=2,threads=2,maxcpus=8
314 The following sub-option defines a CPU topology hierarchy (2 sockets
315 totally on the machine, 2 dies per socket, 2 cores per die, 2 threads
316 per core) for PC machines which support sockets/dies/cores/threads.
317 Some members of the option can be omitted but their values will be
318 automatically computed:
322 -smp 16,sockets=2,dies=2,cores=2,threads=2,maxcpus=16
324 The following sub-option defines a CPU topology hierarchy (2 sockets
325 totally on the machine, 2 clusters per socket, 2 cores per cluster,
326 2 threads per core) for ARM virt machines which support sockets/clusters
327 /cores/threads. Some members of the option can be omitted but their values
328 will be automatically computed:
332 -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16
334 Historically preference was given to the coarsest topology parameters
335 when computing missing values (ie sockets preferred over cores, which
336 were preferred over threads), however, this behaviour is considered
337 liable to change. Prior to 6.2 the preference was sockets over cores
338 over threads. Since 6.2 the preference is cores over sockets over threads.
340 For example, the following option defines a machine board with 2 sockets
341 of 1 core before 6.2 and 1 socket of 2 cores after 6.2:
345 -smp 2
347 Note: The cluster topology will only be generated in ACPI and exposed
348 to guest if it's explicitly specified in -smp.
349 ERST
351 DEF("numa", HAS_ARG, QEMU_OPTION_numa,
352 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
353 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
354 "-numa dist,src=source,dst=destination,val=distance\n"
355 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n"
356 "-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"
357 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n",
358 QEMU_ARCH_ALL)
359 SRST
360 ``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
362 ``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
364 ``-numa dist,src=source,dst=destination,val=distance``
366 ``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]``
368 ``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=type[,latency=lat][,bandwidth=bw]``
370 ``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]``
371 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA
372 distance from a source node to a destination node. Set the ACPI
373 Heterogeneous Memory Attributes for the given nodes.
375 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and
376 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a
377 contiguous range of CPU indexes (or a single VCPU if lastcpu is
378 omitted). A non-contiguous set of VCPUs can be represented by
379 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is
380 omitted on all nodes, VCPUs are automatically split between them.
382 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a
383 NUMA node:
387 -numa node,cpus=0-2,cpus=5
389 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option
390 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to
391 assign CPU objects to a node using topology layout properties of
392 CPU. The set of properties is machine specific, and depends on used
393 machine type/'\ ``smp``\ ' options. It could be queried with
394 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ '
395 property specifies node to which CPU object will be assigned, it's
396 required for node to be declared with '\ ``node``\ ' option before
397 it's used with '\ ``cpu``\ ' option.
399 For example:
403 -M pc \
404 -smp 1,sockets=2,maxcpus=2 \
405 -numa node,nodeid=0 -numa node,nodeid=1 \
406 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1
408 '\ ``memdev``\ ' option assigns RAM from a given memory backend
409 device to a node. It is recommended to use '\ ``memdev``\ ' option
410 over legacy '\ ``mem``\ ' option. This is because '\ ``memdev``\ '
411 option provides better performance and more control over the
412 backend's RAM (e.g. '\ ``prealloc``\ ' parameter of
413 '\ ``-memory-backend-ram``\ ' allows memory preallocation).
415 For compatibility reasons, legacy '\ ``mem``\ ' option is
416 supported in 5.0 and older machine types. Note that '\ ``mem``\ '
417 and '\ ``memdev``\ ' are mutually exclusive. If one node uses
418 '\ ``memdev``\ ', the rest nodes have to use '\ ``memdev``\ '
419 option, and vice versa.
421 Users must specify memory for all NUMA nodes by '\ ``memdev``\ '
422 (or legacy '\ ``mem``\ ' if available). In QEMU 5.2, the support
423 for '\ ``-numa node``\ ' without memory specified was removed.
425 '\ ``initiator``\ ' is an additional option that points to an
426 initiator NUMA node that has best performance (the lowest latency or
427 largest bandwidth) to this NUMA node. Note that this option can be
428 set only when the machine property 'hmat' is set to 'on'.
430 Following example creates a machine with 2 NUMA nodes, node 0 has
431 CPU. node 1 has only memory, and its initiator is node 0. Note that
432 because node 0 has CPU, by default the initiator of node 0 is itself
433 and must be itself.
437 -machine hmat=on \
438 -m 2G,slots=2,maxmem=4G \
439 -object memory-backend-ram,size=1G,id=m0 \
440 -object memory-backend-ram,size=1G,id=m1 \
441 -numa node,nodeid=0,memdev=m0 \
442 -numa node,nodeid=1,memdev=m1,initiator=0 \
443 -smp 2,sockets=2,maxcpus=2 \
444 -numa cpu,node-id=0,socket-id=0 \
445 -numa cpu,node-id=0,socket-id=1
447 source and destination are NUMA node IDs. distance is the NUMA
448 distance from source to destination. The distance from a node to
449 itself is always 10. If any pair of nodes is given a distance, then
450 all pairs must be given distances. Although, when distances are only
451 given in one direction for each pair of nodes, then the distances in
452 the opposite directions are assumed to be the same. If, however, an
453 asymmetrical pair of distances is given for even one node pair, then
454 all node pairs must be provided distance values for both directions,
455 even when they are symmetrical. When a node is unreachable from
456 another node, set the pair's distance to 255.
458 Note that the -``numa`` option doesn't allocate any of the specified
459 resources, it just assigns existing resources to NUMA nodes. This
460 means that one still has to use the ``-m``, ``-smp`` options to
461 allocate RAM and VCPUs respectively.
463 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth
464 Information between initiator and target NUMA nodes in ACPI
465 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can
466 create memory requests, usually it has one or more processors.
467 Target NUMA node contains addressable memory.
469 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is
470 the memory hierarchy of the target NUMA node: if hierarchy is
471 'memory', the structure represents the memory performance; if
472 hierarchy is 'first-level\|second-level\|third-level', this
473 structure represents aggregated performance of memory side caches
474 for each domain. type of 'data-type' is type of data represented by
475 this structure instance: if 'hierarchy' is 'memory', 'data-type' is
476 'access\|read\|write' latency or 'access\|read\|write' bandwidth of
477 the target memory; if 'hierarchy' is
478 'first-level\|second-level\|third-level', 'data-type' is
479 'access\|read\|write' hit latency or 'access\|read\|write' hit
480 bandwidth of the target memory side cache.
482 lat is latency value in nanoseconds. bw is bandwidth value, the
483 possible value and units are NUM[M\|G\|T], mean that the bandwidth
484 value are NUM byte per second (or MB/s, GB/s or TB/s depending on
485 used suffix). Note that if latency or bandwidth value is 0, means
486 the corresponding latency or bandwidth information is not provided.
488 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory
489 belongs. size is the size of memory side cache in bytes. level is
490 the cache level described in this structure, note that the cache
491 level 0 should not be used with '\ ``hmat-cache``\ ' option.
492 associativity is the cache associativity, the possible value is
493 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy
494 is the write policy. line is the cache Line size in bytes.
496 For example, the following options describe 2 NUMA nodes. Node 0 has
497 2 cpus and a ram, node 1 has only a ram. The processors in node 0
498 access memory in node 0 with access-latency 5 nanoseconds,
499 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access
500 memory in NUMA node 1 with access-latency 10 nanoseconds,
501 access-bandwidth is 100 MB/s. And for memory side cache information,
502 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB,
503 policy is write-back, the cache Line size is 8 bytes:
507 -machine hmat=on \
508 -m 2G \
509 -object memory-backend-ram,size=1G,id=m0 \
510 -object memory-backend-ram,size=1G,id=m1 \
511 -smp 2,sockets=2,maxcpus=2 \
512 -numa node,nodeid=0,memdev=m0 \
513 -numa node,nodeid=1,memdev=m1,initiator=0 \
514 -numa cpu,node-id=0,socket-id=0 \
515 -numa cpu,node-id=0,socket-id=1 \
516 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \
517 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \
518 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \
519 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \
520 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \
521 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8
522 ERST
524 DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd,
525 "-add-fd fd=fd,set=set[,opaque=opaque]\n"
526 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL)
527 SRST
528 ``-add-fd fd=fd,set=set[,opaque=opaque]``
529 Add a file descriptor to an fd set. Valid options are:
531 ``fd=fd``
532 This option defines the file descriptor of which a duplicate is
533 added to fd set. The file descriptor cannot be stdin, stdout, or
534 stderr.
536 ``set=set``
537 This option defines the ID of the fd set to add the file
538 descriptor to.
540 ``opaque=opaque``
541 This option defines a free-form string that can be used to
542 describe fd.
544 You can open an image using pre-opened file descriptors from an fd
545 set:
547 .. parsed-literal::
549 |qemu_system| \\
550 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
551 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
552 -drive file=/dev/fdset/2,index=0,media=disk
553 ERST
555 DEF("set", HAS_ARG, QEMU_OPTION_set,
556 "-set group.id.arg=value\n"
557 " set <arg> parameter for item <id> of type <group>\n"
558 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL)
559 SRST
560 ``-set group.id.arg=value``
561 Set parameter arg for item id of type group
562 ERST
564 DEF("global", HAS_ARG, QEMU_OPTION_global,
565 "-global driver.property=value\n"
566 "-global driver=driver,property=property,value=value\n"
567 " set a global default for a driver property\n",
568 QEMU_ARCH_ALL)
569 SRST
570 ``-global driver.prop=value``
572 ``-global driver=driver,property=property,value=value``
573 Set default value of driver's property prop to value, e.g.:
575 .. parsed-literal::
577 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img
579 In particular, you can use this to set driver properties for devices
580 which are created automatically by the machine model. To create a
581 device which is not created automatically and set properties on it,
582 use -``device``.
584 -global driver.prop=value is shorthand for -global
585 driver=driver,property=prop,value=value. The longhand syntax works
586 even when driver contains a dot.
587 ERST
589 DEF("boot", HAS_ARG, QEMU_OPTION_boot,
590 "-boot [order=drives][,once=drives][,menu=on|off]\n"
591 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n"
592 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n"
593 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n"
594 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n"
595 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n",
596 QEMU_ARCH_ALL)
597 SRST
598 ``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]``
599 Specify boot order drives as a string of drive letters. Valid drive
600 letters depend on the target architecture. The x86 PC uses: a, b
601 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p
602 (Etherboot from network adapter 1-4), hard disk boot is the default.
603 To apply a particular boot order only on the first startup, specify
604 it via ``once``. Note that the ``order`` or ``once`` parameter
605 should not be used together with the ``bootindex`` property of
606 devices, since the firmware implementations normally do not support
607 both at the same time.
609 Interactive boot menus/prompts can be enabled via ``menu=on`` as far
610 as firmware/BIOS supports them. The default is non-interactive boot.
612 A splash picture could be passed to bios, enabling user to show it
613 as logo, when option splash=sp\_name is given and menu=on, If
614 firmware/BIOS supports them. Currently Seabios for X86 system
615 support it. limitation: The splash file could be a jpeg file or a
616 BMP file in 24 BPP format(true color). The resolution should be
617 supported by the SVGA mode, so the recommended is 320x240, 640x480,
618 800x640.
620 A timeout could be passed to bios, guest will pause for rb\_timeout
621 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will
622 not reboot, qemu passes '-1' to bios by default. Currently Seabios
623 for X86 system support it.
625 Do strict boot via ``strict=on`` as far as firmware/BIOS supports
626 it. This only effects when boot priority is changed by bootindex
627 options. The default is non-strict boot.
629 .. parsed-literal::
631 # try to boot from network first, then from hard disk
632 |qemu_system_x86| -boot order=nc
633 # boot from CD-ROM first, switch back to default order after reboot
634 |qemu_system_x86| -boot once=d
635 # boot with a splash picture for 5 seconds.
636 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000
638 Note: The legacy format '-boot drives' is still supported but its
639 use is discouraged as it may be removed from future versions.
640 ERST
642 DEF("m", HAS_ARG, QEMU_OPTION_m,
643 "-m [size=]megs[,slots=n,maxmem=size]\n"
644 " configure guest RAM\n"
645 " size: initial amount of guest memory\n"
646 " slots: number of hotplug slots (default: none)\n"
647 " maxmem: maximum amount of guest memory (default: none)\n"
648 "NOTE: Some architectures might enforce a specific granularity\n",
649 QEMU_ARCH_ALL)
650 SRST
651 ``-m [size=]megs[,slots=n,maxmem=size]``
652 Sets guest startup RAM size to megs megabytes. Default is 128 MiB.
653 Optionally, a suffix of "M" or "G" can be used to signify a value in
654 megabytes or gigabytes respectively. Optional pair slots, maxmem
655 could be used to set amount of hotpluggable memory slots and maximum
656 amount of memory. Note that maxmem must be aligned to the page size.
658 For example, the following command-line sets the guest startup RAM
659 size to 1GB, creates 3 slots to hotplug additional memory and sets
660 the maximum memory the guest can reach to 4GB:
662 .. parsed-literal::
664 |qemu_system| -m 1G,slots=3,maxmem=4G
666 If slots and maxmem are not specified, memory hotplug won't be
667 enabled and the guest startup RAM will never increase.
668 ERST
670 DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath,
671 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL)
672 SRST
673 ``-mem-path path``
674 Allocate guest RAM from a temporarily created file in path.
675 ERST
677 DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc,
678 "-mem-prealloc preallocate guest memory (use with -mem-path)\n",
679 QEMU_ARCH_ALL)
680 SRST
681 ``-mem-prealloc``
682 Preallocate memory when using -mem-path.
683 ERST
685 DEF("k", HAS_ARG, QEMU_OPTION_k,
686 "-k language use keyboard layout (for example 'fr' for French)\n",
687 QEMU_ARCH_ALL)
688 SRST
689 ``-k language``
690 Use keyboard layout language (for example ``fr`` for French). This
691 option is only needed where it is not easy to get raw PC keycodes
692 (e.g. on Macs, with some X11 servers or with a VNC or curses
693 display). You don't normally need to use it on PC/Linux or
694 PC/Windows hosts.
696 The available layouts are:
700 ar de-ch es fo fr-ca hu ja mk no pt-br sv
701 da en-gb et fr fr-ch is lt nl pl ru th
702 de en-us fi fr-be hr it lv nl-be pt sl tr
704 The default is ``en-us``.
705 ERST
708 HXCOMM Deprecated by -audiodev
709 DEF("audio-help", 0, QEMU_OPTION_audio_help,
710 "-audio-help show -audiodev equivalent of the currently specified audio settings\n",
711 QEMU_ARCH_ALL)
712 SRST
713 ``-audio-help``
714 Will show the -audiodev equivalent of the currently specified
715 (deprecated) environment variables.
716 ERST
718 DEF("audio", HAS_ARG, QEMU_OPTION_audio,
719 "-audio [driver=]driver,model=value[,prop[=value][,...]]\n"
720 " specifies the audio backend and device to use;\n"
721 " apart from 'model', options are the same as for -audiodev.\n"
722 " use '-audio model=help' to show possible devices.\n",
723 QEMU_ARCH_ALL)
724 SRST
725 ``-audio [driver=]driver,model=value[,prop[=value][,...]]``
726 This option is a shortcut for configuring both the guest audio
727 hardware and the host audio backend in one go.
728 The driver option is the same as with the corresponding ``-audiodev`` option below.
729 The guest hardware model can be set with ``model=modelname``.
731 Use ``driver=help`` to list the available drivers,
732 and ``model=help`` to list the available device types.
734 The following two example do exactly the same, to show how ``-audio``
735 can be used to shorten the command line length:
737 .. parsed-literal::
739 |qemu_system| -audiodev pa,id=pa -device sb16,audiodev=pa
740 |qemu_system| -audio pa,model=sb16
741 ERST
743 DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev,
744 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n"
745 " specifies the audio backend to use\n"
746 " Use ``-audiodev help`` to list the available drivers\n"
747 " id= identifier of the backend\n"
748 " timer-period= timer period in microseconds\n"
749 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n"
750 " in|out.fixed-settings= use fixed settings for host audio\n"
751 " in|out.frequency= frequency to use with fixed settings\n"
752 " in|out.channels= number of channels to use with fixed settings\n"
753 " in|out.format= sample format to use with fixed settings\n"
754 " valid values: s8, s16, s32, u8, u16, u32, f32\n"
755 " in|out.voices= number of voices to use\n"
756 " in|out.buffer-length= length of buffer in microseconds\n"
757 "-audiodev none,id=id,[,prop[=value][,...]]\n"
758 " dummy driver that discards all output\n"
759 #ifdef CONFIG_AUDIO_ALSA
760 "-audiodev alsa,id=id[,prop[=value][,...]]\n"
761 " in|out.dev= name of the audio device to use\n"
762 " in|out.period-length= length of period in microseconds\n"
763 " in|out.try-poll= attempt to use poll mode\n"
764 " threshold= threshold (in microseconds) when playback starts\n"
765 #endif
766 #ifdef CONFIG_AUDIO_COREAUDIO
767 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n"
768 " in|out.buffer-count= number of buffers\n"
769 #endif
770 #ifdef CONFIG_AUDIO_DSOUND
771 "-audiodev dsound,id=id[,prop[=value][,...]]\n"
772 " latency= add extra latency to playback in microseconds\n"
773 #endif
774 #ifdef CONFIG_AUDIO_OSS
775 "-audiodev oss,id=id[,prop[=value][,...]]\n"
776 " in|out.dev= path of the audio device to use\n"
777 " in|out.buffer-count= number of buffers\n"
778 " in|out.try-poll= attempt to use poll mode\n"
779 " try-mmap= try using memory mapped access\n"
780 " exclusive= open device in exclusive mode\n"
781 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n"
782 #endif
783 #ifdef CONFIG_AUDIO_PA
784 "-audiodev pa,id=id[,prop[=value][,...]]\n"
785 " server= PulseAudio server address\n"
786 " in|out.name= source/sink device name\n"
787 " in|out.latency= desired latency in microseconds\n"
788 #endif
789 #ifdef CONFIG_AUDIO_SDL
790 "-audiodev sdl,id=id[,prop[=value][,...]]\n"
791 " in|out.buffer-count= number of buffers\n"
792 #endif
793 #ifdef CONFIG_AUDIO_SNDIO
794 "-audiodev sndio,id=id[,prop[=value][,...]]\n"
795 #endif
796 #ifdef CONFIG_SPICE
797 "-audiodev spice,id=id[,prop[=value][,...]]\n"
798 #endif
799 #ifdef CONFIG_DBUS_DISPLAY
800 "-audiodev dbus,id=id[,prop[=value][,...]]\n"
801 #endif
802 "-audiodev wav,id=id[,prop[=value][,...]]\n"
803 " path= path of wav file to record\n",
804 QEMU_ARCH_ALL)
805 SRST
806 ``-audiodev [driver=]driver,id=id[,prop[=value][,...]]``
807 Adds a new audio backend driver identified by id. There are global
808 and driver specific properties. Some values can be set differently
809 for input and output, they're marked with ``in|out.``. You can set
810 the input's property with ``in.prop`` and the output's property with
811 ``out.prop``. For example:
815 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000
816 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified
818 NOTE: parameter validation is known to be incomplete, in many cases
819 specifying an invalid option causes QEMU to print an error message
820 and continue emulation without sound.
822 Valid global options are:
824 ``id=identifier``
825 Identifies the audio backend.
827 ``timer-period=period``
828 Sets the timer period used by the audio subsystem in
829 microseconds. Default is 10000 (10 ms).
831 ``in|out.mixing-engine=on|off``
832 Use QEMU's mixing engine to mix all streams inside QEMU and
833 convert audio formats when not supported by the backend. When
834 off, fixed-settings must be off too. Note that disabling this
835 option means that the selected backend must support multiple
836 streams and the audio formats used by the virtual cards,
837 otherwise you'll get no sound. It's not recommended to disable
838 this option unless you want to use 5.1 or 7.1 audio, as mixing
839 engine only supports mono and stereo audio. Default is on.
841 ``in|out.fixed-settings=on|off``
842 Use fixed settings for host audio. When off, it will change
843 based on how the guest opens the sound card. In this case you
844 must not specify frequency, channels or format. Default is on.
846 ``in|out.frequency=frequency``
847 Specify the frequency to use when using fixed-settings. Default
848 is 44100Hz.
850 ``in|out.channels=channels``
851 Specify the number of channels to use when using fixed-settings.
852 Default is 2 (stereo).
854 ``in|out.format=format``
855 Specify the sample format to use when using fixed-settings.
856 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``,
857 ``u32``, ``f32``. Default is ``s16``.
859 ``in|out.voices=voices``
860 Specify the number of voices to use. Default is 1.
862 ``in|out.buffer-length=usecs``
863 Sets the size of the buffer in microseconds.
865 ``-audiodev none,id=id[,prop[=value][,...]]``
866 Creates a dummy backend that discards all outputs. This backend has
867 no backend specific properties.
869 ``-audiodev alsa,id=id[,prop[=value][,...]]``
870 Creates backend using the ALSA. This backend is only available on
871 Linux.
873 ALSA specific options are:
875 ``in|out.dev=device``
876 Specify the ALSA device to use for input and/or output. Default
877 is ``default``.
879 ``in|out.period-length=usecs``
880 Sets the period length in microseconds.
882 ``in|out.try-poll=on|off``
883 Attempt to use poll mode with the device. Default is on.
885 ``threshold=threshold``
886 Threshold (in microseconds) when playback starts. Default is 0.
888 ``-audiodev coreaudio,id=id[,prop[=value][,...]]``
889 Creates a backend using Apple's Core Audio. This backend is only
890 available on Mac OS and only supports playback.
892 Core Audio specific options are:
894 ``in|out.buffer-count=count``
895 Sets the count of the buffers.
897 ``-audiodev dsound,id=id[,prop[=value][,...]]``
898 Creates a backend using Microsoft's DirectSound. This backend is
899 only available on Windows and only supports playback.
901 DirectSound specific options are:
903 ``latency=usecs``
904 Add extra usecs microseconds latency to playback. Default is
905 10000 (10 ms).
907 ``-audiodev oss,id=id[,prop[=value][,...]]``
908 Creates a backend using OSS. This backend is available on most
909 Unix-like systems.
911 OSS specific options are:
913 ``in|out.dev=device``
914 Specify the file name of the OSS device to use. Default is
915 ``/dev/dsp``.
917 ``in|out.buffer-count=count``
918 Sets the count of the buffers.
920 ``in|out.try-poll=on|of``
921 Attempt to use poll mode with the device. Default is on.
923 ``try-mmap=on|off``
924 Try using memory mapped device access. Default is off.
926 ``exclusive=on|off``
927 Open the device in exclusive mode (vmix won't work in this
928 case). Default is off.
930 ``dsp-policy=policy``
931 Sets the timing policy (between 0 and 10, where smaller number
932 means smaller latency but higher CPU usage). Use -1 to use
933 buffer sizes specified by ``buffer`` and ``buffer-count``. This
934 option is ignored if you do not have OSS 4. Default is 5.
936 ``-audiodev pa,id=id[,prop[=value][,...]]``
937 Creates a backend using PulseAudio. This backend is available on
938 most systems.
940 PulseAudio specific options are:
942 ``server=server``
943 Sets the PulseAudio server to connect to.
945 ``in|out.name=sink``
946 Use the specified source/sink for recording/playback.
948 ``in|out.latency=usecs``
949 Desired latency in microseconds. The PulseAudio server will try
950 to honor this value but actual latencies may be lower or higher.
952 ``-audiodev sdl,id=id[,prop[=value][,...]]``
953 Creates a backend using SDL. This backend is available on most
954 systems, but you should use your platform's native backend if
955 possible.
957 SDL specific options are:
959 ``in|out.buffer-count=count``
960 Sets the count of the buffers.
962 ``-audiodev sndio,id=id[,prop[=value][,...]]``
963 Creates a backend using SNDIO. This backend is available on
964 OpenBSD and most other Unix-like systems.
966 Sndio specific options are:
968 ``in|out.dev=device``
969 Specify the sndio device to use for input and/or output. Default
970 is ``default``.
972 ``in|out.latency=usecs``
973 Sets the desired period length in microseconds.
975 ``-audiodev spice,id=id[,prop[=value][,...]]``
976 Creates a backend that sends audio through SPICE. This backend
977 requires ``-spice`` and automatically selected in that case, so
978 usually you can ignore this option. This backend has no backend
979 specific properties.
981 ``-audiodev wav,id=id[,prop[=value][,...]]``
982 Creates a backend that writes audio to a WAV file.
984 Backend specific options are:
986 ``path=path``
987 Write recorded audio into the specified file. Default is
988 ``qemu.wav``.
989 ERST
991 DEF("device", HAS_ARG, QEMU_OPTION_device,
992 "-device driver[,prop[=value][,...]]\n"
993 " add device (based on driver)\n"
994 " prop=value,... sets driver properties\n"
995 " use '-device help' to print all possible drivers\n"
996 " use '-device driver,help' to print all possible properties\n",
997 QEMU_ARCH_ALL)
998 SRST
999 ``-device driver[,prop[=value][,...]]``
1000 Add device driver. prop=value sets driver properties. Valid
1001 properties depend on the driver. To get help on possible drivers and
1002 properties, use ``-device help`` and ``-device driver,help``.
1004 Some drivers are:
1006 ``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]``
1007 Add an IPMI BMC. This is a simulation of a hardware management
1008 interface processor that normally sits on a system. It provides a
1009 watchdog and the ability to reset and power control the system. You
1010 need to connect this to an IPMI interface to make it useful
1012 The IPMI slave address to use for the BMC. The default is 0x20. This
1013 address is the BMC's address on the I2C network of management
1014 controllers. If you don't know what this means, it is safe to ignore
1017 ``id=id``
1018 The BMC id for interfaces to use this device.
1020 ``slave_addr=val``
1021 Define slave address to use for the BMC. The default is 0x20.
1023 ``sdrfile=file``
1024 file containing raw Sensor Data Records (SDR) data. The default
1025 is none.
1027 ``fruareasize=val``
1028 size of a Field Replaceable Unit (FRU) area. The default is
1029 1024.
1031 ``frudatafile=file``
1032 file containing raw Field Replaceable Unit (FRU) inventory data.
1033 The default is none.
1035 ``guid=uuid``
1036 value for the GUID for the BMC, in standard UUID format. If this
1037 is set, get "Get GUID" command to the BMC will return it.
1038 Otherwise "Get GUID" will return an error.
1040 ``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]``
1041 Add a connection to an external IPMI BMC simulator. Instead of
1042 locally emulating the BMC like the above item, instead connect to an
1043 external entity that provides the IPMI services.
1045 A connection is made to an external BMC simulator. If you do this,
1046 it is strongly recommended that you use the "reconnect=" chardev
1047 option to reconnect to the simulator if the connection is lost. Note
1048 that if this is not used carefully, it can be a security issue, as
1049 the interface has the ability to send resets, NMIs, and power off
1050 the VM. It's best if QEMU makes a connection to an external
1051 simulator running on a secure port on localhost, so neither the
1052 simulator nor QEMU is exposed to any outside network.
1054 See the "lanserv/README.vm" file in the OpenIPMI library for more
1055 details on the external interface.
1057 ``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]``
1058 Add a KCS IPMI interface on the ISA bus. This also adds a
1059 corresponding ACPI and SMBIOS entries, if appropriate.
1061 ``bmc=id``
1062 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern
1063 above.
1065 ``ioport=val``
1066 Define the I/O address of the interface. The default is 0xca0
1067 for KCS.
1069 ``irq=val``
1070 Define the interrupt to use. The default is 5. To disable
1071 interrupts, set this to 0.
1073 ``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]``
1074 Like the KCS interface, but defines a BT interface. The default port
1075 is 0xe4 and the default interrupt is 5.
1077 ``-device pci-ipmi-kcs,bmc=id``
1078 Add a KCS IPMI interface on the PCI bus.
1080 ``bmc=id``
1081 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above.
1083 ``-device pci-ipmi-bt,bmc=id``
1084 Like the KCS interface, but defines a BT interface on the PCI bus.
1086 ``-device intel-iommu[,option=...]``
1087 This is only supported by ``-machine q35``, which will enable Intel VT-d
1088 emulation within the guest. It supports below options:
1090 ``intremap=on|off`` (default: auto)
1091 This enables interrupt remapping feature. It's required to enable
1092 complete x2apic. Currently it only supports kvm kernel-irqchip modes
1093 ``off`` or ``split``, while full kernel-irqchip is not yet supported.
1094 The default value is "auto", which will be decided by the mode of
1095 kernel-irqchip.
1097 ``caching-mode=on|off`` (default: off)
1098 This enables caching mode for the VT-d emulated device. When
1099 caching-mode is enabled, each guest DMA buffer mapping will generate an
1100 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in
1101 a synchronous way. It is required for ``-device vfio-pci`` to work
1102 with the VT-d device, because host assigned devices requires to setup
1103 the DMA mapping on the host before guest DMA starts.
1105 ``device-iotlb=on|off`` (default: off)
1106 This enables device-iotlb capability for the emulated VT-d device. So
1107 far virtio/vhost should be the only real user for this parameter,
1108 paired with ats=on configured for the device.
1110 ``aw-bits=39|48`` (default: 39)
1111 This decides the address width of IOVA address space. The address
1112 space has 39 bits width for 3-level IOMMU page tables, and 48 bits for
1113 4-level IOMMU page tables.
1115 Please also refer to the wiki page for general scenarios of VT-d
1116 emulation in QEMU: https://wiki.qemu.org/Features/VT-d.
1118 ERST
1120 DEF("name", HAS_ARG, QEMU_OPTION_name,
1121 "-name string1[,process=string2][,debug-threads=on|off]\n"
1122 " set the name of the guest\n"
1123 " string1 sets the window title and string2 the process name\n"
1124 " When debug-threads is enabled, individual threads are given a separate name\n"
1125 " NOTE: The thread names are for debugging and not a stable API.\n",
1126 QEMU_ARCH_ALL)
1127 SRST
1128 ``-name name``
1129 Sets the name of the guest. This name will be displayed in the SDL
1130 window caption. The name will also be used for the VNC server. Also
1131 optionally set the top visible process name in Linux. Naming of
1132 individual threads can also be enabled on Linux to aid debugging.
1133 ERST
1135 DEF("uuid", HAS_ARG, QEMU_OPTION_uuid,
1136 "-uuid %08x-%04x-%04x-%04x-%012x\n"
1137 " specify machine UUID\n", QEMU_ARCH_ALL)
1138 SRST
1139 ``-uuid uuid``
1140 Set system UUID.
1141 ERST
1143 DEFHEADING()
1145 DEFHEADING(Block device options:)
1147 SRST
1148 The QEMU block device handling options have a long history and
1149 have gone through several iterations as the feature set and complexity
1150 of the block layer have grown. Many online guides to QEMU often
1151 reference older and deprecated options, which can lead to confusion.
1153 The most explicit way to describe disks is to use a combination of
1154 ``-device`` to specify the hardware device and ``-blockdev`` to
1155 describe the backend. The device defines what the guest sees and the
1156 backend describes how QEMU handles the data. It is the only guaranteed
1157 stable interface for describing block devices and as such is
1158 recommended for management tools and scripting.
1160 The ``-drive`` option combines the device and backend into a single
1161 command line option which is a more human friendly. There is however no
1162 interface stability guarantee although some older board models still
1163 need updating to work with the modern blockdev forms.
1165 Older options like ``-hda`` are essentially macros which expand into
1166 ``-drive`` options for various drive interfaces. The original forms
1167 bake in a lot of assumptions from the days when QEMU was emulating a
1168 legacy PC, they are not recommended for modern configurations.
1170 ERST
1172 DEF("fda", HAS_ARG, QEMU_OPTION_fda,
1173 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
1174 DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
1175 SRST
1176 ``-fda file``
1178 ``-fdb file``
1179 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in
1180 the System Emulation Users Guide).
1181 ERST
1183 DEF("hda", HAS_ARG, QEMU_OPTION_hda,
1184 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL)
1185 DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL)
1186 DEF("hdc", HAS_ARG, QEMU_OPTION_hdc,
1187 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL)
1188 DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL)
1189 SRST
1190 ``-hda file``
1192 ``-hdb file``
1194 ``-hdc file``
1196 ``-hdd file``
1197 Use file as hard disk 0, 1, 2 or 3 image (see the :ref:`disk images`
1198 chapter in the System Emulation Users Guide).
1199 ERST
1201 DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom,
1202 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n",
1203 QEMU_ARCH_ALL)
1204 SRST
1205 ``-cdrom file``
1206 Use file as CD-ROM image (you cannot use ``-hdc`` and ``-cdrom`` at
1207 the same time). You can use the host CD-ROM by using ``/dev/cdrom``
1208 as filename.
1209 ERST
1211 DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev,
1212 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n"
1213 " [,cache.direct=on|off][,cache.no-flush=on|off]\n"
1214 " [,read-only=on|off][,auto-read-only=on|off]\n"
1215 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n"
1216 " [,driver specific parameters...]\n"
1217 " configure a block backend\n", QEMU_ARCH_ALL)
1218 SRST
1219 ``-blockdev option[,option[,option[,...]]]``
1220 Define a new block driver node. Some of the options apply to all
1221 block drivers, other options are only accepted for a specific block
1222 driver. See below for a list of generic options and options for the
1223 most common block drivers.
1225 Options that expect a reference to another node (e.g. ``file``) can
1226 be given in two ways. Either you specify the node name of an already
1227 existing node (file=node-name), or you define a new node inline,
1228 adding options for the referenced node after a dot
1229 (file.filename=path,file.aio=native).
1231 A block driver node created with ``-blockdev`` can be used for a
1232 guest device by specifying its node name for the ``drive`` property
1233 in a ``-device`` argument that defines a block device.
1235 ``Valid options for any block driver node:``
1236 ``driver``
1237 Specifies the block driver to use for the given node.
1239 ``node-name``
1240 This defines the name of the block driver node by which it
1241 will be referenced later. The name must be unique, i.e. it
1242 must not match the name of a different block driver node, or
1243 (if you use ``-drive`` as well) the ID of a drive.
1245 If no node name is specified, it is automatically generated.
1246 The generated node name is not intended to be predictable
1247 and changes between QEMU invocations. For the top level, an
1248 explicit node name must be specified.
1250 ``read-only``
1251 Open the node read-only. Guest write attempts will fail.
1253 Note that some block drivers support only read-only access,
1254 either generally or in certain configurations. In this case,
1255 the default value ``read-only=off`` does not work and the
1256 option must be specified explicitly.
1258 ``auto-read-only``
1259 If ``auto-read-only=on`` is set, QEMU may fall back to
1260 read-only usage even when ``read-only=off`` is requested, or
1261 even switch between modes as needed, e.g. depending on
1262 whether the image file is writable or whether a writing user
1263 is attached to the node.
1265 ``force-share``
1266 Override the image locking system of QEMU by forcing the
1267 node to utilize weaker shared access for permissions where
1268 it would normally request exclusive access. When there is
1269 the potential for multiple instances to have the same file
1270 open (whether this invocation of QEMU is the first or the
1271 second instance), both instances must permit shared access
1272 for the second instance to succeed at opening the file.
1274 Enabling ``force-share=on`` requires ``read-only=on``.
1276 ``cache.direct``
1277 The host page cache can be avoided with ``cache.direct=on``.
1278 This will attempt to do disk IO directly to the guest's
1279 memory. QEMU may still perform an internal copy of the data.
1281 ``cache.no-flush``
1282 In case you don't care about data integrity over host
1283 failures, you can use ``cache.no-flush=on``. This option
1284 tells QEMU that it never needs to write any data to the disk
1285 but can instead keep things in cache. If anything goes
1286 wrong, like your host losing power, the disk storage getting
1287 disconnected accidentally, etc. your image will most
1288 probably be rendered unusable.
1290 ``discard=discard``
1291 discard is one of "ignore" (or "off") or "unmap" (or "on")
1292 and controls whether ``discard`` (also known as ``trim`` or
1293 ``unmap``) requests are ignored or passed to the filesystem.
1294 Some machine types may not support discard requests.
1296 ``detect-zeroes=detect-zeroes``
1297 detect-zeroes is "off", "on" or "unmap" and enables the
1298 automatic conversion of plain zero writes by the OS to
1299 driver specific optimized zero write commands. You may even
1300 choose "unmap" if discard is set to "unmap" to allow a zero
1301 write to be converted to an ``unmap`` operation.
1303 ``Driver-specific options for file``
1304 This is the protocol-level block driver for accessing regular
1305 files.
1307 ``filename``
1308 The path to the image file in the local filesystem
1310 ``aio``
1311 Specifies the AIO backend (threads/native/io_uring,
1312 default: threads)
1314 ``locking``
1315 Specifies whether the image file is protected with Linux OFD
1316 / POSIX locks. The default is to use the Linux Open File
1317 Descriptor API if available, otherwise no lock is applied.
1318 (auto/on/off, default: auto)
1320 Example:
1324 -blockdev driver=file,node-name=disk,filename=disk.img
1326 ``Driver-specific options for raw``
1327 This is the image format block driver for raw images. It is
1328 usually stacked on top of a protocol level block driver such as
1329 ``file``.
1331 ``file``
1332 Reference to or definition of the data source block driver
1333 node (e.g. a ``file`` driver node)
1335 Example 1:
1339 -blockdev driver=file,node-name=disk_file,filename=disk.img
1340 -blockdev driver=raw,node-name=disk,file=disk_file
1342 Example 2:
1346 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img
1348 ``Driver-specific options for qcow2``
1349 This is the image format block driver for qcow2 images. It is
1350 usually stacked on top of a protocol level block driver such as
1351 ``file``.
1353 ``file``
1354 Reference to or definition of the data source block driver
1355 node (e.g. a ``file`` driver node)
1357 ``backing``
1358 Reference to or definition of the backing file block device
1359 (default is taken from the image file). It is allowed to
1360 pass ``null`` here in order to disable the default backing
1361 file.
1363 ``lazy-refcounts``
1364 Whether to enable the lazy refcounts feature (on/off;
1365 default is taken from the image file)
1367 ``cache-size``
1368 The maximum total size of the L2 table and refcount block
1369 caches in bytes (default: the sum of l2-cache-size and
1370 refcount-cache-size)
1372 ``l2-cache-size``
1373 The maximum size of the L2 table cache in bytes (default: if
1374 cache-size is not specified - 32M on Linux platforms, and 8M
1375 on non-Linux platforms; otherwise, as large as possible
1376 within the cache-size, while permitting the requested or the
1377 minimal refcount cache size)
1379 ``refcount-cache-size``
1380 The maximum size of the refcount block cache in bytes
1381 (default: 4 times the cluster size; or if cache-size is
1382 specified, the part of it which is not used for the L2
1383 cache)
1385 ``cache-clean-interval``
1386 Clean unused entries in the L2 and refcount caches. The
1387 interval is in seconds. The default value is 600 on
1388 supporting platforms, and 0 on other platforms. Setting it
1389 to 0 disables this feature.
1391 ``pass-discard-request``
1392 Whether discard requests to the qcow2 device should be
1393 forwarded to the data source (on/off; default: on if
1394 discard=unmap is specified, off otherwise)
1396 ``pass-discard-snapshot``
1397 Whether discard requests for the data source should be
1398 issued when a snapshot operation (e.g. deleting a snapshot)
1399 frees clusters in the qcow2 file (on/off; default: on)
1401 ``pass-discard-other``
1402 Whether discard requests for the data source should be
1403 issued on other occasions where a cluster gets freed
1404 (on/off; default: off)
1406 ``overlap-check``
1407 Which overlap checks to perform for writes to the image
1408 (none/constant/cached/all; default: cached). For details or
1409 finer granularity control refer to the QAPI documentation of
1410 ``blockdev-add``.
1412 Example 1:
1416 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2
1417 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216
1419 Example 2:
1423 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2
1425 ``Driver-specific options for other drivers``
1426 Please refer to the QAPI documentation of the ``blockdev-add``
1427 QMP command.
1428 ERST
1430 DEF("drive", HAS_ARG, QEMU_OPTION_drive,
1431 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n"
1432 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n"
1433 " [,snapshot=on|off][,rerror=ignore|stop|report]\n"
1434 " [,werror=ignore|stop|report|enospc][,id=name]\n"
1435 " [,aio=threads|native|io_uring]\n"
1436 " [,readonly=on|off][,copy-on-read=on|off]\n"
1437 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n"
1438 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n"
1439 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n"
1440 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n"
1441 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n"
1442 " [[,iops_size=is]]\n"
1443 " [[,group=g]]\n"
1444 " use 'file' as a drive image\n", QEMU_ARCH_ALL)
1445 SRST
1446 ``-drive option[,option[,option[,...]]]``
1447 Define a new drive. This includes creating a block driver node (the
1448 backend) as well as a guest device, and is mostly a shortcut for
1449 defining the corresponding ``-blockdev`` and ``-device`` options.
1451 ``-drive`` accepts all options that are accepted by ``-blockdev``.
1452 In addition, it knows the following options:
1454 ``file=file``
1455 This option defines which disk image (see the :ref:`disk images`
1456 chapter in the System Emulation Users Guide) to use with this drive.
1457 If the filename contains comma, you must double it (for instance,
1458 "file=my,,file" to use file "my,file").
1460 Special files such as iSCSI devices can be specified using
1461 protocol specific URLs. See the section for "Device URL Syntax"
1462 for more information.
1464 ``if=interface``
1465 This option defines on which type on interface the drive is
1466 connected. Available types are: ide, scsi, sd, mtd, floppy,
1467 pflash, virtio, none.
1469 ``bus=bus,unit=unit``
1470 These options define where is connected the drive by defining
1471 the bus number and the unit id.
1473 ``index=index``
1474 This option defines where the drive is connected by using an
1475 index in the list of available connectors of a given interface
1476 type.
1478 ``media=media``
1479 This option defines the type of the media: disk or cdrom.
1481 ``snapshot=snapshot``
1482 snapshot is "on" or "off" and controls snapshot mode for the
1483 given drive (see ``-snapshot``).
1485 ``cache=cache``
1486 cache is "none", "writeback", "unsafe", "directsync" or
1487 "writethrough" and controls how the host cache is used to access
1488 block data. This is a shortcut that sets the ``cache.direct``
1489 and ``cache.no-flush`` options (as in ``-blockdev``), and
1490 additionally ``cache.writeback``, which provides a default for
1491 the ``write-cache`` option of block guest devices (as in
1492 ``-device``). The modes correspond to the following settings:
1494 ============= =============== ============ ==============
1495 \ cache.writeback cache.direct cache.no-flush
1496 ============= =============== ============ ==============
1497 writeback on off off
1498 none on on off
1499 writethrough off off off
1500 directsync off on off
1501 unsafe on off on
1502 ============= =============== ============ ==============
1504 The default mode is ``cache=writeback``.
1506 ``aio=aio``
1507 aio is "threads", "native", or "io_uring" and selects between pthread
1508 based disk I/O, native Linux AIO, or Linux io_uring API.
1510 ``format=format``
1511 Specify which disk format will be used rather than detecting the
1512 format. Can be used to specify format=raw to avoid interpreting
1513 an untrusted format header.
1515 ``werror=action,rerror=action``
1516 Specify which action to take on write and read errors. Valid
1517 actions are: "ignore" (ignore the error and try to continue),
1518 "stop" (pause QEMU), "report" (report the error to the guest),
1519 "enospc" (pause QEMU only if the host disk is full; report the
1520 error to the guest otherwise). The default setting is
1521 ``werror=enospc`` and ``rerror=report``.
1523 ``copy-on-read=copy-on-read``
1524 copy-on-read is "on" or "off" and enables whether to copy read
1525 backing file sectors into the image file.
1527 ``bps=b,bps_rd=r,bps_wr=w``
1528 Specify bandwidth throttling limits in bytes per second, either
1529 for all request types or for reads or writes only. Small values
1530 can lead to timeouts or hangs inside the guest. A safe minimum
1531 for disks is 2 MB/s.
1533 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm``
1534 Specify bursts in bytes per second, either for all request types
1535 or for reads or writes only. Bursts allow the guest I/O to spike
1536 above the limit temporarily.
1538 ``iops=i,iops_rd=r,iops_wr=w``
1539 Specify request rate limits in requests per second, either for
1540 all request types or for reads or writes only.
1542 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm``
1543 Specify bursts in requests per second, either for all request
1544 types or for reads or writes only. Bursts allow the guest I/O to
1545 spike above the limit temporarily.
1547 ``iops_size=is``
1548 Let every is bytes of a request count as a new request for iops
1549 throttling purposes. Use this option to prevent guests from
1550 circumventing iops limits by sending fewer but larger requests.
1552 ``group=g``
1553 Join a throttling quota group with given name g. All drives that
1554 are members of the same group are accounted for together. Use
1555 this option to prevent guests from circumventing throttling
1556 limits by using many small disks instead of a single larger
1557 disk.
1559 By default, the ``cache.writeback=on`` mode is used. It will report
1560 data writes as completed as soon as the data is present in the host
1561 page cache. This is safe as long as your guest OS makes sure to
1562 correctly flush disk caches where needed. If your guest OS does not
1563 handle volatile disk write caches correctly and your host crashes or
1564 loses power, then the guest may experience data corruption.
1566 For such guests, you should consider using ``cache.writeback=off``.
1567 This means that the host page cache will be used to read and write
1568 data, but write notification will be sent to the guest only after
1569 QEMU has made sure to flush each write to the disk. Be aware that
1570 this has a major impact on performance.
1572 When using the ``-snapshot`` option, unsafe caching is always used.
1574 Copy-on-read avoids accessing the same backing file sectors
1575 repeatedly and is useful when the backing file is over a slow
1576 network. By default copy-on-read is off.
1578 Instead of ``-cdrom`` you can use:
1580 .. parsed-literal::
1582 |qemu_system| -drive file=file,index=2,media=cdrom
1584 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use:
1586 .. parsed-literal::
1588 |qemu_system| -drive file=file,index=0,media=disk
1589 |qemu_system| -drive file=file,index=1,media=disk
1590 |qemu_system| -drive file=file,index=2,media=disk
1591 |qemu_system| -drive file=file,index=3,media=disk
1593 You can open an image using pre-opened file descriptors from an fd
1594 set:
1596 .. parsed-literal::
1598 |qemu_system| \\
1599 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
1600 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
1601 -drive file=/dev/fdset/2,index=0,media=disk
1603 You can connect a CDROM to the slave of ide0:
1605 .. parsed-literal::
1607 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom
1609 If you don't specify the "file=" argument, you define an empty
1610 drive:
1612 .. parsed-literal::
1614 |qemu_system_x86| -drive if=ide,index=1,media=cdrom
1616 Instead of ``-fda``, ``-fdb``, you can use:
1618 .. parsed-literal::
1620 |qemu_system_x86| -drive file=file,index=0,if=floppy
1621 |qemu_system_x86| -drive file=file,index=1,if=floppy
1623 By default, interface is "ide" and index is automatically
1624 incremented:
1626 .. parsed-literal::
1628 |qemu_system_x86| -drive file=a -drive file=b
1630 is interpreted like:
1632 .. parsed-literal::
1634 |qemu_system_x86| -hda a -hdb b
1635 ERST
1637 DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock,
1638 "-mtdblock file use 'file' as on-board Flash memory image\n",
1639 QEMU_ARCH_ALL)
1640 SRST
1641 ``-mtdblock file``
1642 Use file as on-board Flash memory image.
1643 ERST
1645 DEF("sd", HAS_ARG, QEMU_OPTION_sd,
1646 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL)
1647 SRST
1648 ``-sd file``
1649 Use file as SecureDigital card image.
1650 ERST
1652 DEF("snapshot", 0, QEMU_OPTION_snapshot,
1653 "-snapshot write to temporary files instead of disk image files\n",
1654 QEMU_ARCH_ALL)
1655 SRST
1656 ``-snapshot``
1657 Write to temporary files instead of disk image files. In this case,
1658 the raw disk image you use is not written back. You can however
1659 force the write back by pressing C-a s (see the :ref:`disk images`
1660 chapter in the System Emulation Users Guide).
1662 .. warning::
1663 snapshot is incompatible with ``-blockdev`` (instead use qemu-img
1664 to manually create snapshot images to attach to your blockdev).
1665 If you have mixed ``-blockdev`` and ``-drive`` declarations you
1666 can use the 'snapshot' property on your drive declarations
1667 instead of this global option.
1669 ERST
1671 DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev,
1672 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1673 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n"
1674 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n"
1675 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n"
1676 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n"
1677 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n"
1678 " [[,throttling.iops-size=is]]\n"
1679 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n"
1680 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n"
1681 "-fsdev synth,id=id\n",
1682 QEMU_ARCH_ALL)
1684 SRST
1685 ``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]``
1687 ``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]``
1689 ``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]``
1691 ``-fsdev synth,id=id[,readonly=on]``
1692 Define a new file system device. Valid options are:
1694 ``local``
1695 Accesses to the filesystem are done by QEMU.
1697 ``proxy``
1698 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1700 ``synth``
1701 Synthetic filesystem, only used by QTests.
1703 ``id=id``
1704 Specifies identifier for this device.
1706 ``path=path``
1707 Specifies the export path for the file system device. Files
1708 under this path will be available to the 9p client on the guest.
1710 ``security_model=security_model``
1711 Specifies the security model to be used for this export path.
1712 Supported security models are "passthrough", "mapped-xattr",
1713 "mapped-file" and "none". In "passthrough" security model, files
1714 are stored using the same credentials as they are created on the
1715 guest. This requires QEMU to run as root. In "mapped-xattr"
1716 security model, some of the file attributes like uid, gid, mode
1717 bits and link target are stored as file attributes. For
1718 "mapped-file" these attributes are stored in the hidden
1719 .virtfs\_metadata directory. Directories exported by this
1720 security model cannot interact with other unix tools. "none"
1721 security model is same as passthrough except the sever won't
1722 report failures if it fails to set file attributes like
1723 ownership. Security model is mandatory only for local fsdriver.
1724 Other fsdrivers (like proxy) don't take security model as a
1725 parameter.
1727 ``writeout=writeout``
1728 This is an optional argument. The only supported value is
1729 "immediate". This means that host page cache will be used to
1730 read and write data but write notification will be sent to the
1731 guest only when the data has been reported as written by the
1732 storage subsystem.
1734 ``readonly=on``
1735 Enables exporting 9p share as a readonly mount for guests. By
1736 default read-write access is given.
1738 ``socket=socket``
1739 Enables proxy filesystem driver to use passed socket file for
1740 communicating with virtfs-proxy-helper(1).
1742 ``sock_fd=sock_fd``
1743 Enables proxy filesystem driver to use passed socket descriptor
1744 for communicating with virtfs-proxy-helper(1). Usually a helper
1745 like libvirt will create socketpair and pass one of the fds as
1746 sock\_fd.
1748 ``fmode=fmode``
1749 Specifies the default mode for newly created files on the host.
1750 Works only with security models "mapped-xattr" and
1751 "mapped-file".
1753 ``dmode=dmode``
1754 Specifies the default mode for newly created directories on the
1755 host. Works only with security models "mapped-xattr" and
1756 "mapped-file".
1758 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w``
1759 Specify bandwidth throttling limits in bytes per second, either
1760 for all request types or for reads or writes only.
1762 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm``
1763 Specify bursts in bytes per second, either for all request types
1764 or for reads or writes only. Bursts allow the guest I/O to spike
1765 above the limit temporarily.
1767 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w``
1768 Specify request rate limits in requests per second, either for
1769 all request types or for reads or writes only.
1771 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm``
1772 Specify bursts in requests per second, either for all request
1773 types or for reads or writes only. Bursts allow the guest I/O to
1774 spike above the limit temporarily.
1776 ``throttling.iops-size=is``
1777 Let every is bytes of a request count as a new request for iops
1778 throttling purposes.
1780 -fsdev option is used along with -device driver "virtio-9p-...".
1782 ``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag``
1783 Options for virtio-9p-... driver are:
1785 ``type``
1786 Specifies the variant to be used. Supported values are "pci",
1787 "ccw" or "device", depending on the machine type.
1789 ``fsdev=id``
1790 Specifies the id value specified along with -fsdev option.
1792 ``mount_tag=mount_tag``
1793 Specifies the tag name to be used by the guest to mount this
1794 export point.
1795 ERST
1797 DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs,
1798 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1799 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n"
1800 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n"
1801 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n"
1802 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n",
1803 QEMU_ARCH_ALL)
1805 SRST
1806 ``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]``
1808 ``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1810 ``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1812 ``-virtfs synth,mount_tag=mount_tag``
1813 Define a new virtual filesystem device and expose it to the guest using
1814 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain
1815 directory on host is made directly accessible by guest as a pass-through
1816 file system by using the 9P network protocol for communication between
1817 host and guests, if desired even accessible, shared by several guests
1818 simultaneously.
1820 Note that ``-virtfs`` is actually just a convenience shortcut for its
1821 generalized form ``-fsdev -device virtio-9p-pci``.
1823 The general form of pass-through file system options are:
1825 ``local``
1826 Accesses to the filesystem are done by QEMU.
1828 ``proxy``
1829 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1831 ``synth``
1832 Synthetic filesystem, only used by QTests.
1834 ``id=id``
1835 Specifies identifier for the filesystem device
1837 ``path=path``
1838 Specifies the export path for the file system device. Files
1839 under this path will be available to the 9p client on the guest.
1841 ``security_model=security_model``
1842 Specifies the security model to be used for this export path.
1843 Supported security models are "passthrough", "mapped-xattr",
1844 "mapped-file" and "none". In "passthrough" security model, files
1845 are stored using the same credentials as they are created on the
1846 guest. This requires QEMU to run as root. In "mapped-xattr"
1847 security model, some of the file attributes like uid, gid, mode
1848 bits and link target are stored as file attributes. For
1849 "mapped-file" these attributes are stored in the hidden
1850 .virtfs\_metadata directory. Directories exported by this
1851 security model cannot interact with other unix tools. "none"
1852 security model is same as passthrough except the sever won't
1853 report failures if it fails to set file attributes like
1854 ownership. Security model is mandatory only for local fsdriver.
1855 Other fsdrivers (like proxy) don't take security model as a
1856 parameter.
1858 ``writeout=writeout``
1859 This is an optional argument. The only supported value is
1860 "immediate". This means that host page cache will be used to
1861 read and write data but write notification will be sent to the
1862 guest only when the data has been reported as written by the
1863 storage subsystem.
1865 ``readonly=on``
1866 Enables exporting 9p share as a readonly mount for guests. By
1867 default read-write access is given.
1869 ``socket=socket``
1870 Enables proxy filesystem driver to use passed socket file for
1871 communicating with virtfs-proxy-helper(1). Usually a helper like
1872 libvirt will create socketpair and pass one of the fds as
1873 sock\_fd.
1875 ``sock_fd``
1876 Enables proxy filesystem driver to use passed 'sock\_fd' as the
1877 socket descriptor for interfacing with virtfs-proxy-helper(1).
1879 ``fmode=fmode``
1880 Specifies the default mode for newly created files on the host.
1881 Works only with security models "mapped-xattr" and
1882 "mapped-file".
1884 ``dmode=dmode``
1885 Specifies the default mode for newly created directories on the
1886 host. Works only with security models "mapped-xattr" and
1887 "mapped-file".
1889 ``mount_tag=mount_tag``
1890 Specifies the tag name to be used by the guest to mount this
1891 export point.
1893 ``multidevs=multidevs``
1894 Specifies how to deal with multiple devices being shared with a
1895 9p export. Supported behaviours are either "remap", "forbid" or
1896 "warn". The latter is the default behaviour on which virtfs 9p
1897 expects only one device to be shared with the same export, and
1898 if more than one device is shared and accessed via the same 9p
1899 export then only a warning message is logged (once) by qemu on
1900 host side. In order to avoid file ID collisions on guest you
1901 should either create a separate virtfs export for each device to
1902 be shared with guests (recommended way) or you might use "remap"
1903 instead which allows you to share multiple devices with only one
1904 export instead, which is achieved by remapping the original
1905 inode numbers from host to guest in a way that would prevent
1906 such collisions. Remapping inodes in such use cases is required
1907 because the original device IDs from host are never passed and
1908 exposed on guest. Instead all files of an export shared with
1909 virtfs always share the same device id on guest. So two files
1910 with identical inode numbers but from actually different devices
1911 on host would otherwise cause a file ID collision and hence
1912 potential misbehaviours on guest. "forbid" on the other hand
1913 assumes like "warn" that only one device is shared by the same
1914 export, however it will not only log a warning message but also
1915 deny access to additional devices on guest. Note though that
1916 "forbid" does currently not block all possible file access
1917 operations (e.g. readdir() would still return entries from other
1918 devices).
1919 ERST
1921 DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi,
1922 "-iscsi [user=user][,password=password][,password-secret=secret-id]\n"
1923 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE]\n"
1924 " [,initiator-name=initiator-iqn][,id=target-iqn]\n"
1925 " [,timeout=timeout]\n"
1926 " iSCSI session parameters\n", QEMU_ARCH_ALL)
1928 SRST
1929 ``-iscsi``
1930 Configure iSCSI session parameters.
1931 ERST
1933 DEFHEADING()
1935 DEFHEADING(USB convenience options:)
1937 DEF("usb", 0, QEMU_OPTION_usb,
1938 "-usb enable on-board USB host controller (if not enabled by default)\n",
1939 QEMU_ARCH_ALL)
1940 SRST
1941 ``-usb``
1942 Enable USB emulation on machine types with an on-board USB host
1943 controller (if not enabled by default). Note that on-board USB host
1944 controllers may not support USB 3.0. In this case
1945 ``-device qemu-xhci`` can be used instead on machines with PCI.
1946 ERST
1948 DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice,
1949 "-usbdevice name add the host or guest USB device 'name'\n",
1950 QEMU_ARCH_ALL)
1951 SRST
1952 ``-usbdevice devname``
1953 Add the USB device devname, and enable an on-board USB controller
1954 if possible and necessary (just like it can be done via
1955 ``-machine usb=on``). Note that this option is mainly intended for
1956 the user's convenience only. More fine-grained control can be
1957 achieved by selecting a USB host controller (if necessary) and the
1958 desired USB device via the ``-device`` option instead. For example,
1959 instead of using ``-usbdevice mouse`` it is possible to use
1960 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse
1961 to a USB 3.0 controller instead (at least on machines that support
1962 PCI and do not have an USB controller enabled by default yet).
1963 For more details, see the chapter about
1964 :ref:`Connecting USB devices` in the System Emulation Users Guide.
1965 Possible devices for devname are:
1967 ``braille``
1968 Braille device. This will use BrlAPI to display the braille
1969 output on a real or fake device (i.e. it also creates a
1970 corresponding ``braille`` chardev automatically beside the
1971 ``usb-braille`` USB device).
1973 ``keyboard``
1974 Standard USB keyboard. Will override the PS/2 keyboard (if present).
1976 ``mouse``
1977 Virtual Mouse. This will override the PS/2 mouse emulation when
1978 activated.
1980 ``tablet``
1981 Pointer device that uses absolute coordinates (like a
1982 touchscreen). This means QEMU is able to report the mouse
1983 position without having to grab the mouse. Also overrides the
1984 PS/2 mouse emulation when activated.
1986 ``wacom-tablet``
1987 Wacom PenPartner USB tablet.
1990 ERST
1992 DEFHEADING()
1994 DEFHEADING(Display options:)
1996 DEF("display", HAS_ARG, QEMU_OPTION_display,
1997 #if defined(CONFIG_SPICE)
1998 "-display spice-app[,gl=on|off]\n"
1999 #endif
2000 #if defined(CONFIG_SDL)
2001 "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n"
2002 " [,window-close=on|off]\n"
2003 #endif
2004 #if defined(CONFIG_GTK)
2005 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n"
2006 " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n"
2007 " [,show-menubar=on|off]\n"
2008 #endif
2009 #if defined(CONFIG_VNC)
2010 "-display vnc=<display>[,<optargs>]\n"
2011 #endif
2012 #if defined(CONFIG_CURSES)
2013 "-display curses[,charset=<encoding>]\n"
2014 #endif
2015 #if defined(CONFIG_COCOA)
2016 "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n"
2017 #endif
2018 #if defined(CONFIG_OPENGL)
2019 "-display egl-headless[,rendernode=<file>]\n"
2020 #endif
2021 #if defined(CONFIG_DBUS_DISPLAY)
2022 "-display dbus[,addr=<dbusaddr>]\n"
2023 " [,gl=on|core|es|off][,rendernode=<file>]\n"
2024 #endif
2025 #if defined(CONFIG_COCOA)
2026 "-display cocoa[,show-cursor=on|off][,left-command-key=on|off]\n"
2027 #endif
2028 "-display none\n"
2029 " select display backend type\n"
2030 " The default display is equivalent to\n "
2031 #if defined(CONFIG_GTK)
2032 "\"-display gtk\"\n"
2033 #elif defined(CONFIG_SDL)
2034 "\"-display sdl\"\n"
2035 #elif defined(CONFIG_COCOA)
2036 "\"-display cocoa\"\n"
2037 #elif defined(CONFIG_VNC)
2038 "\"-vnc localhost:0,to=99,id=default\"\n"
2039 #else
2040 "\"-display none\"\n"
2041 #endif
2042 , QEMU_ARCH_ALL)
2043 SRST
2044 ``-display type``
2045 Select type of display to use. Use ``-display help`` to list the available
2046 display types. Valid values for type are
2048 ``spice-app[,gl=on|off]``
2049 Start QEMU as a Spice server and launch the default Spice client
2050 application. The Spice server will redirect the serial consoles
2051 and QEMU monitors. (Since 4.0)
2053 ``dbus``
2054 Export the display over D-Bus interfaces. (Since 7.0)
2056 The connection is registered with the "org.qemu" name (and queued when
2057 already owned).
2059 ``addr=<dbusaddr>`` : D-Bus bus address to connect to.
2061 ``p2p=yes|no`` : Use peer-to-peer connection, accepted via QMP ``add_client``.
2063 ``gl=on|off|core|es`` : Use OpenGL for rendering (the D-Bus interface
2064 will share framebuffers with DMABUF file descriptors).
2066 ``sdl``
2067 Display video output via SDL (usually in a separate graphics
2068 window; see the SDL documentation for other possibilities).
2069 Valid parameters are:
2071 ``grab-mod=<mods>`` : Used to select the modifier keys for toggling
2072 the mouse grabbing in conjunction with the "g" key. ``<mods>`` can be
2073 either ``lshift-lctrl-lalt`` or ``rctrl``.
2075 ``gl=on|off|core|es`` : Use OpenGL for displaying
2077 ``show-cursor=on|off`` : Force showing the mouse cursor
2079 ``window-close=on|off`` : Allow to quit qemu with window close button
2081 ``gtk``
2082 Display video output in a GTK window. This interface provides
2083 drop-down menus and other UI elements to configure and control
2084 the VM during runtime. Valid parameters are:
2086 ``full-screen=on|off`` : Start in fullscreen mode
2088 ``gl=on|off`` : Use OpenGL for displaying
2090 ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover
2092 ``show-tabs=on|off`` : Display the tab bar for switching between the
2093 various graphical interfaces (e.g. VGA and
2094 virtual console character devices) by default.
2096 ``show-cursor=on|off`` : Force showing the mouse cursor
2098 ``window-close=on|off`` : Allow to quit qemu with window close button
2100 ``show-menubar=on|off`` : Display the main window menubar, defaults to "on"
2102 ``curses[,charset=<encoding>]``
2103 Display video output via curses. For graphics device models
2104 which support a text mode, QEMU can display this output using a
2105 curses/ncurses interface. Nothing is displayed when the graphics
2106 device is in graphical mode or if the graphics device does not
2107 support a text mode. Generally only the VGA device models
2108 support text mode. The font charset used by the guest can be
2109 specified with the ``charset`` option, for example
2110 ``charset=CP850`` for IBM CP850 encoding. The default is
2111 ``CP437``.
2113 ``cocoa``
2114 Display video output in a Cocoa window. Mac only. This interface
2115 provides drop-down menus and other UI elements to configure and
2116 control the VM during runtime. Valid parameters are:
2118 ``show-cursor=on|off`` : Force showing the mouse cursor
2120 ``left-command-key=on|off`` : Disable forwarding left command key to host
2122 ``egl-headless[,rendernode=<file>]``
2123 Offload all OpenGL operations to a local DRI device. For any
2124 graphical display, this display needs to be paired with either
2125 VNC or SPICE displays.
2127 ``vnc=<display>``
2128 Start a VNC server on display <display>
2130 ``none``
2131 Do not display video output. The guest will still see an
2132 emulated graphics card, but its output will not be displayed to
2133 the QEMU user. This option differs from the -nographic option in
2134 that it only affects what is done with video output; -nographic
2135 also changes the destination of the serial and parallel port
2136 data.
2137 ERST
2139 DEF("nographic", 0, QEMU_OPTION_nographic,
2140 "-nographic disable graphical output and redirect serial I/Os to console\n",
2141 QEMU_ARCH_ALL)
2142 SRST
2143 ``-nographic``
2144 Normally, if QEMU is compiled with graphical window support, it
2145 displays output such as guest graphics, guest console, and the QEMU
2146 monitor in a window. With this option, you can totally disable
2147 graphical output so that QEMU is a simple command line application.
2148 The emulated serial port is redirected on the console and muxed with
2149 the monitor (unless redirected elsewhere explicitly). Therefore, you
2150 can still use QEMU to debug a Linux kernel with a serial console.
2151 Use C-a h for help on switching between the console and monitor.
2152 ERST
2154 #ifdef CONFIG_SPICE
2155 DEF("spice", HAS_ARG, QEMU_OPTION_spice,
2156 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n"
2157 " [,x509-key-file=<file>][,x509-key-password=<file>]\n"
2158 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n"
2159 " [,x509-dh-key-file=<file>][,addr=addr]\n"
2160 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n"
2161 " [,tls-ciphers=<list>]\n"
2162 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n"
2163 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n"
2164 " [,sasl=on|off][,disable-ticketing=on|off]\n"
2165 " [,password-secret=<secret-id>]\n"
2166 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n"
2167 " [,jpeg-wan-compression=[auto|never|always]]\n"
2168 " [,zlib-glz-wan-compression=[auto|never|always]]\n"
2169 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n"
2170 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n"
2171 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n"
2172 " [,gl=[on|off]][,rendernode=<file>]\n"
2173 " enable spice\n"
2174 " at least one of {port, tls-port} is mandatory\n",
2175 QEMU_ARCH_ALL)
2176 #endif
2177 SRST
2178 ``-spice option[,option[,...]]``
2179 Enable the spice remote desktop protocol. Valid options are
2181 ``port=<nr>``
2182 Set the TCP port spice is listening on for plaintext channels.
2184 ``addr=<addr>``
2185 Set the IP address spice is listening on. Default is any
2186 address.
2188 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off``
2189 Force using the specified IP version.
2191 ``password-secret=<secret-id>``
2192 Set the ID of the ``secret`` object containing the password
2193 you need to authenticate.
2195 ``sasl=on|off``
2196 Require that the client use SASL to authenticate with the spice.
2197 The exact choice of authentication method used is controlled
2198 from the system / user's SASL configuration file for the 'qemu'
2199 service. This is typically found in /etc/sasl2/qemu.conf. If
2200 running QEMU as an unprivileged user, an environment variable
2201 SASL\_CONF\_PATH can be used to make it search alternate
2202 locations for the service config. While some SASL auth methods
2203 can also provide data encryption (eg GSSAPI), it is recommended
2204 that SASL always be combined with the 'tls' and 'x509' settings
2205 to enable use of SSL and server certificates. This ensures a
2206 data encryption preventing compromise of authentication
2207 credentials.
2209 ``disable-ticketing=on|off``
2210 Allow client connects without authentication.
2212 ``disable-copy-paste=on|off``
2213 Disable copy paste between the client and the guest.
2215 ``disable-agent-file-xfer=on|off``
2216 Disable spice-vdagent based file-xfer between the client and the
2217 guest.
2219 ``tls-port=<nr>``
2220 Set the TCP port spice is listening on for encrypted channels.
2222 ``x509-dir=<dir>``
2223 Set the x509 file directory. Expects same filenames as -vnc
2224 $display,x509=$dir
2226 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>``
2227 The x509 file names can also be configured individually.
2229 ``tls-ciphers=<list>``
2230 Specify which ciphers to use.
2232 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]``
2233 Force specific channel to be used with or without TLS
2234 encryption. The options can be specified multiple times to
2235 configure multiple channels. The special name "default" can be
2236 used to set the default mode. For channels which are not
2237 explicitly forced into one mode the spice client is allowed to
2238 pick tls/plaintext as he pleases.
2240 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]``
2241 Configure image compression (lossless). Default is auto\_glz.
2243 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]``
2244 Configure wan image compression (lossy for slow links). Default
2245 is auto.
2247 ``streaming-video=[off|all|filter]``
2248 Configure video stream detection. Default is off.
2250 ``agent-mouse=[on|off]``
2251 Enable/disable passing mouse events via vdagent. Default is on.
2253 ``playback-compression=[on|off]``
2254 Enable/disable audio stream compression (using celt 0.5.1).
2255 Default is on.
2257 ``seamless-migration=[on|off]``
2258 Enable/disable spice seamless migration. Default is off.
2260 ``gl=[on|off]``
2261 Enable/disable OpenGL context. Default is off.
2263 ``rendernode=<file>``
2264 DRM render node for OpenGL rendering. If not specified, it will
2265 pick the first available. (Since 2.9)
2266 ERST
2268 DEF("portrait", 0, QEMU_OPTION_portrait,
2269 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n",
2270 QEMU_ARCH_ALL)
2271 SRST
2272 ``-portrait``
2273 Rotate graphical output 90 deg left (only PXA LCD).
2274 ERST
2276 DEF("rotate", HAS_ARG, QEMU_OPTION_rotate,
2277 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n",
2278 QEMU_ARCH_ALL)
2279 SRST
2280 ``-rotate deg``
2281 Rotate graphical output some deg left (only PXA LCD).
2282 ERST
2284 DEF("vga", HAS_ARG, QEMU_OPTION_vga,
2285 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n"
2286 " select video card type\n", QEMU_ARCH_ALL)
2287 SRST
2288 ``-vga type``
2289 Select type of VGA card to emulate. Valid values for type are
2291 ``cirrus``
2292 Cirrus Logic GD5446 Video card. All Windows versions starting
2293 from Windows 95 should recognize and use this graphic card. For
2294 optimal performances, use 16 bit color depth in the guest and
2295 the host OS. (This card was the default before QEMU 2.2)
2297 ``std``
2298 Standard VGA card with Bochs VBE extensions. If your guest OS
2299 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if
2300 you want to use high resolution modes (>= 1280x1024x16) then you
2301 should use this option. (This card is the default since QEMU
2302 2.2)
2304 ``vmware``
2305 VMWare SVGA-II compatible adapter. Use it if you have
2306 sufficiently recent XFree86/XOrg server or Windows guest with a
2307 driver for this card.
2309 ``qxl``
2310 QXL paravirtual graphic card. It is VGA compatible (including
2311 VESA 2.0 VBE support). Works best with qxl guest drivers
2312 installed though. Recommended choice when using the spice
2313 protocol.
2315 ``tcx``
2316 (sun4m only) Sun TCX framebuffer. This is the default
2317 framebuffer for sun4m machines and offers both 8-bit and 24-bit
2318 colour depths at a fixed resolution of 1024x768.
2320 ``cg3``
2321 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit
2322 framebuffer for sun4m machines available in both 1024x768
2323 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people
2324 wishing to run older Solaris versions.
2326 ``virtio``
2327 Virtio VGA card.
2329 ``none``
2330 Disable VGA card.
2331 ERST
2333 DEF("full-screen", 0, QEMU_OPTION_full_screen,
2334 "-full-screen start in full screen\n", QEMU_ARCH_ALL)
2335 SRST
2336 ``-full-screen``
2337 Start in full screen.
2338 ERST
2340 DEF("g", HAS_ARG, QEMU_OPTION_g ,
2341 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n",
2342 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K)
2343 SRST
2344 ``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]``
2345 Set the initial graphical resolution and depth (PPC, SPARC only).
2347 For PPC the default is 800x600x32.
2349 For SPARC with the TCX graphics device, the default is 1024x768x8
2350 with the option of 1024x768x24. For cgthree, the default is
2351 1024x768x8 with the option of 1152x900x8 for people who wish to use
2352 OBP.
2353 ERST
2355 DEF("vnc", HAS_ARG, QEMU_OPTION_vnc ,
2356 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL)
2357 SRST
2358 ``-vnc display[,option[,option[,...]]]``
2359 Normally, if QEMU is compiled with graphical window support, it
2360 displays output such as guest graphics, guest console, and the QEMU
2361 monitor in a window. With this option, you can have QEMU listen on
2362 VNC display display and redirect the VGA display over the VNC
2363 session. It is very useful to enable the usb tablet device when
2364 using this option (option ``-device usb-tablet``). When using the
2365 VNC display, you must use the ``-k`` parameter to set the keyboard
2366 layout if you are not using en-us. Valid syntax for the display is
2368 ``to=L``
2369 With this option, QEMU will try next available VNC displays,
2370 until the number L, if the origianlly defined "-vnc display" is
2371 not available, e.g. port 5900+display is already used by another
2372 application. By default, to=0.
2374 ``host:d``
2375 TCP connections will only be allowed from host on display d. By
2376 convention the TCP port is 5900+d. Optionally, host can be
2377 omitted in which case the server will accept connections from
2378 any host.
2380 ``unix:path``
2381 Connections will be allowed over UNIX domain sockets where path
2382 is the location of a unix socket to listen for connections on.
2384 ``none``
2385 VNC is initialized but not started. The monitor ``change``
2386 command can be used to later start the VNC server.
2388 Following the display value there may be one or more option flags
2389 separated by commas. Valid options are
2391 ``reverse=on|off``
2392 Connect to a listening VNC client via a "reverse" connection.
2393 The client is specified by the display. For reverse network
2394 connections (host:d,``reverse``), the d argument is a TCP port
2395 number, not a display number.
2397 ``websocket=on|off``
2398 Opens an additional TCP listening port dedicated to VNC
2399 Websocket connections. If a bare websocket option is given, the
2400 Websocket port is 5700+display. An alternative port can be
2401 specified with the syntax ``websocket``\ =port.
2403 If host is specified connections will only be allowed from this
2404 host. It is possible to control the websocket listen address
2405 independently, using the syntax ``websocket``\ =host:port.
2407 If no TLS credentials are provided, the websocket connection
2408 runs in unencrypted mode. If TLS credentials are provided, the
2409 websocket connection requires encrypted client connections.
2411 ``password=on|off``
2412 Require that password based authentication is used for client
2413 connections.
2415 The password must be set separately using the ``set_password``
2416 command in the :ref:`QEMU monitor`. The
2417 syntax to change your password is:
2418 ``set_password <protocol> <password>`` where <protocol> could be
2419 either "vnc" or "spice".
2421 If you would like to change <protocol> password expiration, you
2422 should use ``expire_password <protocol> <expiration-time>``
2423 where expiration time could be one of the following options:
2424 now, never, +seconds or UNIX time of expiration, e.g. +60 to
2425 make password expire in 60 seconds, or 1335196800 to make
2426 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for
2427 this date and time).
2429 You can also use keywords "now" or "never" for the expiration
2430 time to allow <protocol> password to expire immediately or never
2431 expire.
2433 ``password-secret=<secret-id>``
2434 Require that password based authentication is used for client
2435 connections, using the password provided by the ``secret``
2436 object identified by ``secret-id``.
2438 ``tls-creds=ID``
2439 Provides the ID of a set of TLS credentials to use to secure the
2440 VNC server. They will apply to both the normal VNC server socket
2441 and the websocket socket (if enabled). Setting TLS credentials
2442 will cause the VNC server socket to enable the VeNCrypt auth
2443 mechanism. The credentials should have been previously created
2444 using the ``-object tls-creds`` argument.
2446 ``tls-authz=ID``
2447 Provides the ID of the QAuthZ authorization object against which
2448 the client's x509 distinguished name will validated. This object
2449 is only resolved at time of use, so can be deleted and recreated
2450 on the fly while the VNC server is active. If missing, it will
2451 default to denying access.
2453 ``sasl=on|off``
2454 Require that the client use SASL to authenticate with the VNC
2455 server. The exact choice of authentication method used is
2456 controlled from the system / user's SASL configuration file for
2457 the 'qemu' service. This is typically found in
2458 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user,
2459 an environment variable SASL\_CONF\_PATH can be used to make it
2460 search alternate locations for the service config. While some
2461 SASL auth methods can also provide data encryption (eg GSSAPI),
2462 it is recommended that SASL always be combined with the 'tls'
2463 and 'x509' settings to enable use of SSL and server
2464 certificates. This ensures a data encryption preventing
2465 compromise of authentication credentials. See the
2466 :ref:`VNC security` section in the System Emulation Users Guide
2467 for details on using SASL authentication.
2469 ``sasl-authz=ID``
2470 Provides the ID of the QAuthZ authorization object against which
2471 the client's SASL username will validated. This object is only
2472 resolved at time of use, so can be deleted and recreated on the
2473 fly while the VNC server is active. If missing, it will default
2474 to denying access.
2476 ``acl=on|off``
2477 Legacy method for enabling authorization of clients against the
2478 x509 distinguished name and SASL username. It results in the
2479 creation of two ``authz-list`` objects with IDs of
2480 ``vnc.username`` and ``vnc.x509dname``. The rules for these
2481 objects must be configured with the HMP ACL commands.
2483 This option is deprecated and should no longer be used. The new
2484 ``sasl-authz`` and ``tls-authz`` options are a replacement.
2486 ``lossy=on|off``
2487 Enable lossy compression methods (gradient, JPEG, ...). If this
2488 option is set, VNC client may receive lossy framebuffer updates
2489 depending on its encoding settings. Enabling this option can
2490 save a lot of bandwidth at the expense of quality.
2492 ``non-adaptive=on|off``
2493 Disable adaptive encodings. Adaptive encodings are enabled by
2494 default. An adaptive encoding will try to detect frequently
2495 updated screen regions, and send updates in these regions using
2496 a lossy encoding (like JPEG). This can be really helpful to save
2497 bandwidth when playing videos. Disabling adaptive encodings
2498 restores the original static behavior of encodings like Tight.
2500 ``share=[allow-exclusive|force-shared|ignore]``
2501 Set display sharing policy. 'allow-exclusive' allows clients to
2502 ask for exclusive access. As suggested by the rfb spec this is
2503 implemented by dropping other connections. Connecting multiple
2504 clients in parallel requires all clients asking for a shared
2505 session (vncviewer: -shared switch). This is the default.
2506 'force-shared' disables exclusive client access. Useful for
2507 shared desktop sessions, where you don't want someone forgetting
2508 specify -shared disconnect everybody else. 'ignore' completely
2509 ignores the shared flag and allows everybody connect
2510 unconditionally. Doesn't conform to the rfb spec but is
2511 traditional QEMU behavior.
2513 ``key-delay-ms``
2514 Set keyboard delay, for key down and key up events, in
2515 milliseconds. Default is 10. Keyboards are low-bandwidth
2516 devices, so this slowdown can help the device and guest to keep
2517 up and not lose events in case events are arriving in bulk.
2518 Possible causes for the latter are flaky network connections, or
2519 scripts for automated testing.
2521 ``audiodev=audiodev``
2522 Use the specified audiodev when the VNC client requests audio
2523 transmission. When not using an -audiodev argument, this option
2524 must be omitted, otherwise is must be present and specify a
2525 valid audiodev.
2527 ``power-control=on|off``
2528 Permit the remote client to issue shutdown, reboot or reset power
2529 control requests.
2530 ERST
2532 ARCHHEADING(, QEMU_ARCH_I386)
2534 ARCHHEADING(i386 target only:, QEMU_ARCH_I386)
2536 DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack,
2537 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n",
2538 QEMU_ARCH_I386)
2539 SRST
2540 ``-win2k-hack``
2541 Use it when installing Windows 2000 to avoid a disk full bug. After
2542 Windows 2000 is installed, you no longer need this option (this
2543 option slows down the IDE transfers).
2544 ERST
2546 DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk,
2547 "-no-fd-bootchk disable boot signature checking for floppy disks\n",
2548 QEMU_ARCH_I386)
2549 SRST
2550 ``-no-fd-bootchk``
2551 Disable boot signature checking for floppy disks in BIOS. May be
2552 needed to boot from old floppy disks.
2553 ERST
2555 DEF("no-acpi", 0, QEMU_OPTION_no_acpi,
2556 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM)
2557 SRST
2558 ``-no-acpi``
2559 Disable ACPI (Advanced Configuration and Power Interface) support.
2560 Use it if your guest OS complains about ACPI problems (PC target
2561 machine only).
2562 ERST
2564 DEF("no-hpet", 0, QEMU_OPTION_no_hpet,
2565 "-no-hpet disable HPET\n", QEMU_ARCH_I386)
2566 SRST
2567 ``-no-hpet``
2568 Disable HPET support. Deprecated, use '-machine hpet=off' instead.
2569 ERST
2571 DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable,
2572 "-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"
2573 " ACPI table description\n", QEMU_ARCH_I386)
2574 SRST
2575 ``-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]...]``
2576 Add ACPI table with specified header fields and context from
2577 specified files. For file=, take whole ACPI table from the specified
2578 files, including all ACPI headers (possible overridden by other
2579 options). For data=, only data portion of the table is used, all
2580 header information is specified in the command line. If a SLIC table
2581 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id
2582 fields will override the same in the RSDT and the FADT (a.k.a.
2583 FACP), in order to ensure the field matches required by the
2584 Microsoft SLIC spec and the ACPI spec.
2585 ERST
2587 DEF("smbios", HAS_ARG, QEMU_OPTION_smbios,
2588 "-smbios file=binary\n"
2589 " load SMBIOS entry from binary file\n"
2590 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n"
2591 " [,uefi=on|off]\n"
2592 " specify SMBIOS type 0 fields\n"
2593 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2594 " [,uuid=uuid][,sku=str][,family=str]\n"
2595 " specify SMBIOS type 1 fields\n"
2596 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2597 " [,asset=str][,location=str]\n"
2598 " specify SMBIOS type 2 fields\n"
2599 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n"
2600 " [,sku=str]\n"
2601 " specify SMBIOS type 3 fields\n"
2602 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n"
2603 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n"
2604 " [,processor-id=%d]\n"
2605 " specify SMBIOS type 4 fields\n"
2606 "-smbios type=8[,external_reference=str][,internal_reference=str][,connector_type=%d][,port_type=%d]\n"
2607 " specify SMBIOS type 8 fields\n"
2608 "-smbios type=11[,value=str][,path=filename]\n"
2609 " specify SMBIOS type 11 fields\n"
2610 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n"
2611 " [,asset=str][,part=str][,speed=%d]\n"
2612 " specify SMBIOS type 17 fields\n"
2613 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n"
2614 " specify SMBIOS type 41 fields\n",
2615 QEMU_ARCH_I386 | QEMU_ARCH_ARM | QEMU_ARCH_LOONGARCH)
2616 SRST
2617 ``-smbios file=binary``
2618 Load SMBIOS entry from binary file.
2620 ``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]``
2621 Specify SMBIOS type 0 fields
2623 ``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]``
2624 Specify SMBIOS type 1 fields
2626 ``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]``
2627 Specify SMBIOS type 2 fields
2629 ``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]``
2630 Specify SMBIOS type 3 fields
2632 ``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-id=%d]``
2633 Specify SMBIOS type 4 fields
2635 ``-smbios type=11[,value=str][,path=filename]``
2636 Specify SMBIOS type 11 fields
2638 This argument can be repeated multiple times, and values are added in the order they are parsed.
2639 Applications intending to use OEM strings data are encouraged to use their application name as
2640 a prefix for the value string. This facilitates passing information for multiple applications
2641 concurrently.
2643 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax
2644 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes.
2646 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to
2647 the SMBIOS table in the order in which they appear.
2649 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535
2650 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the
2651 guest. Instead it should be used as a indicator to inform the guest where to locate the real
2652 data set, for example, by specifying the serial ID of a block device.
2654 An example passing three strings is
2656 .. parsed-literal::
2658 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\
2659 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\
2660 path=/some/file/with/oemstringsdata.txt
2662 In the guest OS this is visible with the ``dmidecode`` command
2664 .. parsed-literal::
2666 $ dmidecode -t 11
2667 Handle 0x0E00, DMI type 11, 5 bytes
2668 OEM Strings
2669 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/
2670 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os
2671 String 3: myapp:some extra data
2674 ``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]``
2675 Specify SMBIOS type 17 fields
2677 ``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]``
2678 Specify SMBIOS type 41 fields
2680 This argument can be repeated multiple times. Its main use is to allow network interfaces be created
2681 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface
2682 position on the PCI bus.
2684 Here is an example of use:
2686 .. parsed-literal::
2688 -netdev user,id=internet \\
2689 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\
2690 -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev
2692 In the guest OS, the device should then appear as ``eno1``:
2694 ..parsed-literal::
2696 $ ip -brief l
2697 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP>
2698 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP>
2700 Currently, the PCI device has to be attached to the root bus.
2702 ERST
2704 DEFHEADING()
2706 DEFHEADING(Network options:)
2708 DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
2709 #ifdef CONFIG_SLIRP
2710 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n"
2711 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n"
2712 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n"
2713 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n"
2714 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]"
2715 #ifndef _WIN32
2716 "[,smb=dir[,smbserver=addr]]\n"
2717 #endif
2718 " configure a user mode network backend with ID 'str',\n"
2719 " its DHCP server and optional services\n"
2720 #endif
2721 #ifdef _WIN32
2722 "-netdev tap,id=str,ifname=name\n"
2723 " configure a host TAP network backend with ID 'str'\n"
2724 #else
2725 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n"
2726 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n"
2727 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n"
2728 " [,poll-us=n]\n"
2729 " configure a host TAP network backend with ID 'str'\n"
2730 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2731 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n"
2732 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n"
2733 " to deconfigure it\n"
2734 " use '[down]script=no' to disable script execution\n"
2735 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n"
2736 " configure it\n"
2737 " use 'fd=h' to connect to an already opened TAP interface\n"
2738 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n"
2739 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n"
2740 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n"
2741 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n"
2742 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n"
2743 " use vhost=on to enable experimental in kernel accelerator\n"
2744 " (only has effect for virtio guests which use MSIX)\n"
2745 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n"
2746 " use 'vhostfd=h' to connect to an already opened vhost net device\n"
2747 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n"
2748 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n"
2749 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n"
2750 " spent on busy polling for vhost net\n"
2751 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n"
2752 " configure a host TAP network backend with ID 'str' that is\n"
2753 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2754 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n"
2755 #endif
2756 #ifdef __linux__
2757 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n"
2758 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n"
2759 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n"
2760 " [,rxcookie=rxcookie][,offset=offset]\n"
2761 " configure a network backend with ID 'str' connected to\n"
2762 " an Ethernet over L2TPv3 pseudowire.\n"
2763 " Linux kernel 3.3+ as well as most routers can talk\n"
2764 " L2TPv3. This transport allows connecting a VM to a VM,\n"
2765 " VM to a router and even VM to Host. It is a nearly-universal\n"
2766 " standard (RFC3931). Note - this implementation uses static\n"
2767 " pre-configured tunnels (same as the Linux kernel).\n"
2768 " use 'src=' to specify source address\n"
2769 " use 'dst=' to specify destination address\n"
2770 " use 'udp=on' to specify udp encapsulation\n"
2771 " use 'srcport=' to specify source udp port\n"
2772 " use 'dstport=' to specify destination udp port\n"
2773 " use 'ipv6=on' to force v6\n"
2774 " L2TPv3 uses cookies to prevent misconfiguration as\n"
2775 " well as a weak security measure\n"
2776 " use 'rxcookie=0x012345678' to specify a rxcookie\n"
2777 " use 'txcookie=0x012345678' to specify a txcookie\n"
2778 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n"
2779 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n"
2780 " use 'pincounter=on' to work around broken counter handling in peer\n"
2781 " use 'offset=X' to add an extra offset between header and data\n"
2782 #endif
2783 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n"
2784 " configure a network backend to connect to another network\n"
2785 " using a socket connection\n"
2786 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n"
2787 " configure a network backend to connect to a multicast maddr and port\n"
2788 " use 'localaddr=addr' to specify the host address to send packets from\n"
2789 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n"
2790 " configure a network backend to connect to another network\n"
2791 " using an UDP tunnel\n"
2792 "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port[,to=maxport][,numeric=on|off][,keep-alive=on|off][,mptcp=on|off][,addr.ipv4=on|off][,addr.ipv6=on|off][,reconnect=seconds]\n"
2793 "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off][,reconnect=seconds]\n"
2794 "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor[,reconnect=seconds]\n"
2795 " configure a network backend to connect to another network\n"
2796 " using a socket connection in stream mode.\n"
2797 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n"
2798 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]\n"
2799 " configure a network backend to connect to a multicast maddr and port\n"
2800 " use ``local.host=addr`` to specify the host address to send packets from\n"
2801 "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n"
2802 "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n"
2803 "-netdev dgram,id=str,local.type=fd,local.str=file-descriptor\n"
2804 " configure a network backend to connect to another network\n"
2805 " using an UDP tunnel\n"
2806 #ifdef CONFIG_VDE
2807 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n"
2808 " configure a network backend to connect to port 'n' of a vde switch\n"
2809 " running on host and listening for incoming connections on 'socketpath'.\n"
2810 " Use group 'groupname' and mode 'octalmode' to change default\n"
2811 " ownership and permissions for communication port.\n"
2812 #endif
2813 #ifdef CONFIG_NETMAP
2814 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n"
2815 " attach to the existing netmap-enabled network interface 'name', or to a\n"
2816 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n"
2817 " netmap device, defaults to '/dev/netmap')\n"
2818 #endif
2819 #ifdef CONFIG_POSIX
2820 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n"
2821 " configure a vhost-user network, backed by a chardev 'dev'\n"
2822 #endif
2823 #ifdef __linux__
2824 "-netdev vhost-vdpa,id=str[,vhostdev=/path/to/dev][,vhostfd=h]\n"
2825 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n"
2826 " use 'vhostdev=/path/to/dev' to open a vhost vdpa device\n"
2827 " use 'vhostfd=h' to connect to an already opened vhost vdpa device\n"
2828 #endif
2829 #ifdef CONFIG_VMNET
2830 "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n"
2831 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n"
2832 " configure a vmnet network backend in host mode with ID 'str',\n"
2833 " isolate this interface from others with 'isolated',\n"
2834 " configure the address range and choose a subnet mask,\n"
2835 " specify network UUID 'uuid' to disable DHCP and interact with\n"
2836 " vmnet-host interfaces within this isolated network\n"
2837 "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n"
2838 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n"
2839 " configure a vmnet network backend in shared mode with ID 'str',\n"
2840 " configure the address range and choose a subnet mask,\n"
2841 " set IPv6 ULA prefix (of length 64) to use for internal network,\n"
2842 " isolate this interface from others with 'isolated'\n"
2843 "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n"
2844 " configure a vmnet network backend in bridged mode with ID 'str',\n"
2845 " use 'ifname=name' to select a physical network interface to be bridged,\n"
2846 " isolate this interface from others with 'isolated'\n"
2847 #endif
2848 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n"
2849 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL)
2850 DEF("nic", HAS_ARG, QEMU_OPTION_nic,
2851 "-nic [tap|bridge|"
2852 #ifdef CONFIG_SLIRP
2853 "user|"
2854 #endif
2855 #ifdef __linux__
2856 "l2tpv3|"
2857 #endif
2858 #ifdef CONFIG_VDE
2859 "vde|"
2860 #endif
2861 #ifdef CONFIG_NETMAP
2862 "netmap|"
2863 #endif
2864 #ifdef CONFIG_POSIX
2865 "vhost-user|"
2866 #endif
2867 #ifdef CONFIG_VMNET
2868 "vmnet-host|vmnet-shared|vmnet-bridged|"
2869 #endif
2870 "socket][,option][,...][mac=macaddr]\n"
2871 " initialize an on-board / default host NIC (using MAC address\n"
2872 " macaddr) and connect it to the given host network backend\n"
2873 "-nic none use it alone to have zero network devices (the default is to\n"
2874 " provided a 'user' network connection)\n",
2875 QEMU_ARCH_ALL)
2876 DEF("net", HAS_ARG, QEMU_OPTION_net,
2877 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n"
2878 " configure or create an on-board (or machine default) NIC and\n"
2879 " connect it to hub 0 (please use -nic unless you need a hub)\n"
2880 "-net ["
2881 #ifdef CONFIG_SLIRP
2882 "user|"
2883 #endif
2884 "tap|"
2885 "bridge|"
2886 #ifdef CONFIG_VDE
2887 "vde|"
2888 #endif
2889 #ifdef CONFIG_NETMAP
2890 "netmap|"
2891 #endif
2892 #ifdef CONFIG_VMNET
2893 "vmnet-host|vmnet-shared|vmnet-bridged|"
2894 #endif
2895 "socket][,option][,option][,...]\n"
2896 " old way to initialize a host network interface\n"
2897 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL)
2898 SRST
2899 ``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]``
2900 This option is a shortcut for configuring both the on-board
2901 (default) guest NIC hardware and the host network backend in one go.
2902 The host backend options are the same as with the corresponding
2903 ``-netdev`` options below. The guest NIC model can be set with
2904 ``model=modelname``. Use ``model=help`` to list the available device
2905 types. The hardware MAC address can be set with ``mac=macaddr``.
2907 The following two example do exactly the same, to show how ``-nic``
2908 can be used to shorten the command line length:
2910 .. parsed-literal::
2912 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32
2913 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32
2915 ``-nic none``
2916 Indicate that no network devices should be configured. It is used to
2917 override the default configuration (default NIC with "user" host
2918 network backend) which is activated if no other networking options
2919 are provided.
2921 ``-netdev user,id=id[,option][,option][,...]``
2922 Configure user mode host network backend which requires no
2923 administrator privilege to run. Valid options are:
2925 ``id=id``
2926 Assign symbolic name for use in monitor commands.
2928 ``ipv4=on|off and ipv6=on|off``
2929 Specify that either IPv4 or IPv6 must be enabled. If neither is
2930 specified both protocols are enabled.
2932 ``net=addr[/mask]``
2933 Set IP network address the guest will see. Optionally specify
2934 the netmask, either in the form a.b.c.d or as number of valid
2935 top-most bits. Default is 10.0.2.0/24.
2937 ``host=addr``
2938 Specify the guest-visible address of the host. Default is the
2939 2nd IP in the guest network, i.e. x.x.x.2.
2941 ``ipv6-net=addr[/int]``
2942 Set IPv6 network address the guest will see (default is
2943 fec0::/64). The network prefix is given in the usual hexadecimal
2944 IPv6 address notation. The prefix size is optional, and is given
2945 as the number of valid top-most bits (default is 64).
2947 ``ipv6-host=addr``
2948 Specify the guest-visible IPv6 address of the host. Default is
2949 the 2nd IPv6 in the guest network, i.e. xxxx::2.
2951 ``restrict=on|off``
2952 If this option is enabled, the guest will be isolated, i.e. it
2953 will not be able to contact the host and no guest IP packets
2954 will be routed over the host to the outside. This option does
2955 not affect any explicitly set forwarding rules.
2957 ``hostname=name``
2958 Specifies the client hostname reported by the built-in DHCP
2959 server.
2961 ``dhcpstart=addr``
2962 Specify the first of the 16 IPs the built-in DHCP server can
2963 assign. Default is the 15th to 31st IP in the guest network,
2964 i.e. x.x.x.15 to x.x.x.31.
2966 ``dns=addr``
2967 Specify the guest-visible address of the virtual nameserver. The
2968 address must be different from the host address. Default is the
2969 3rd IP in the guest network, i.e. x.x.x.3.
2971 ``ipv6-dns=addr``
2972 Specify the guest-visible address of the IPv6 virtual
2973 nameserver. The address must be different from the host address.
2974 Default is the 3rd IP in the guest network, i.e. xxxx::3.
2976 ``dnssearch=domain``
2977 Provides an entry for the domain-search list sent by the
2978 built-in DHCP server. More than one domain suffix can be
2979 transmitted by specifying this option multiple times. If
2980 supported, this will cause the guest to automatically try to
2981 append the given domain suffix(es) in case a domain name can not
2982 be resolved.
2984 Example:
2986 .. parsed-literal::
2988 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org
2990 ``domainname=domain``
2991 Specifies the client domain name reported by the built-in DHCP
2992 server.
2994 ``tftp=dir``
2995 When using the user mode network stack, activate a built-in TFTP
2996 server. The files in dir will be exposed as the root of a TFTP
2997 server. The TFTP client on the guest must be configured in
2998 binary mode (use the command ``bin`` of the Unix TFTP client).
3000 ``tftp-server-name=name``
3001 In BOOTP reply, broadcast name as the "TFTP server name"
3002 (RFC2132 option 66). This can be used to advise the guest to
3003 load boot files or configurations from a different server than
3004 the host address.
3006 ``bootfile=file``
3007 When using the user mode network stack, broadcast file as the
3008 BOOTP filename. In conjunction with ``tftp``, this can be used
3009 to network boot a guest from a local directory.
3011 Example (using pxelinux):
3013 .. parsed-literal::
3015 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\
3016 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0
3018 ``smb=dir[,smbserver=addr]``
3019 When using the user mode network stack, activate a built-in SMB
3020 server so that Windows OSes can access to the host files in
3021 ``dir`` transparently. The IP address of the SMB server can be
3022 set to addr. By default the 4th IP in the guest network is used,
3023 i.e. x.x.x.4.
3025 In the guest Windows OS, the line:
3029 10.0.2.4 smbserver
3031 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows
3032 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows
3033 NT/2000).
3035 Then ``dir`` can be accessed in ``\\smbserver\qemu``.
3037 Note that a SAMBA server must be installed on the host OS.
3039 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport``
3040 Redirect incoming TCP or UDP connections to the host port
3041 hostport to the guest IP address guestaddr on guest port
3042 guestport. If guestaddr is not specified, its value is x.x.x.15
3043 (default first address given by the built-in DHCP server). By
3044 specifying hostaddr, the rule can be bound to a specific host
3045 interface. If no connection type is set, TCP is used. This
3046 option can be given multiple times.
3048 For example, to redirect host X11 connection from screen 1 to
3049 guest screen 0, use the following:
3051 .. parsed-literal::
3053 # on the host
3054 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000
3055 # this host xterm should open in the guest X11 server
3056 xterm -display :1
3058 To redirect telnet connections from host port 5555 to telnet
3059 port on the guest, use the following:
3061 .. parsed-literal::
3063 # on the host
3064 |qemu_system| -nic user,hostfwd=tcp::5555-:23
3065 telnet localhost 5555
3067 Then when you use on the host ``telnet localhost 5555``, you
3068 connect to the guest telnet server.
3070 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command``
3071 Forward guest TCP connections to the IP address server on port
3072 port to the character device dev or to a program executed by
3073 cmd:command which gets spawned for each connection. This option
3074 can be given multiple times.
3076 You can either use a chardev directly and have that one used
3077 throughout QEMU's lifetime, like in the following example:
3079 .. parsed-literal::
3081 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever
3082 # the guest accesses it
3083 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321
3085 Or you can execute a command on every TCP connection established
3086 by the guest, so that QEMU behaves similar to an inetd process
3087 for that virtual server:
3089 .. parsed-literal::
3091 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234
3092 # and connect the TCP stream to its stdin/stdout
3093 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321'
3095 ``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]``
3096 Configure a host TAP network backend with ID id.
3098 Use the network script file to configure it and the network script
3099 dfile to deconfigure it. If name is not provided, the OS
3100 automatically provides one. The default network configure script is
3101 ``/etc/qemu-ifup`` and the default network deconfigure script is
3102 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to
3103 disable script execution.
3105 If running QEMU as an unprivileged user, use the network helper
3106 to configure the TAP interface and attach it to the bridge.
3107 The default network helper executable is
3108 ``/path/to/qemu-bridge-helper`` and the default bridge device is
3109 ``br0``.
3111 ``fd``\ =h can be used to specify the handle of an already opened
3112 host TAP interface.
3114 Examples:
3116 .. parsed-literal::
3118 #launch a QEMU instance with the default network script
3119 |qemu_system| linux.img -nic tap
3121 .. parsed-literal::
3123 #launch a QEMU instance with two NICs, each one connected
3124 #to a TAP device
3125 |qemu_system| linux.img \\
3126 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\
3127 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1
3129 .. parsed-literal::
3131 #launch a QEMU instance with the default network helper to
3132 #connect a TAP device to bridge br0
3133 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\
3134 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper"
3136 ``-netdev bridge,id=id[,br=bridge][,helper=helper]``
3137 Connect a host TAP network interface to a host bridge device.
3139 Use the network helper helper to configure the TAP interface and
3140 attach it to the bridge. The default network helper executable is
3141 ``/path/to/qemu-bridge-helper`` and the default bridge device is
3142 ``br0``.
3144 Examples:
3146 .. parsed-literal::
3148 #launch a QEMU instance with the default network helper to
3149 #connect a TAP device to bridge br0
3150 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1
3152 .. parsed-literal::
3154 #launch a QEMU instance with the default network helper to
3155 #connect a TAP device to bridge qemubr0
3156 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1
3158 ``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]``
3159 This host network backend can be used to connect the guest's network
3160 to another QEMU virtual machine using a TCP socket connection. If
3161 ``listen`` is specified, QEMU waits for incoming connections on port
3162 (host is optional). ``connect`` is used to connect to another QEMU
3163 instance using the ``listen`` option. ``fd``\ =h specifies an
3164 already opened TCP socket.
3166 Example:
3168 .. parsed-literal::
3170 # launch a first QEMU instance
3171 |qemu_system| linux.img \\
3172 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3173 -netdev socket,id=n1,listen=:1234
3174 # connect the network of this instance to the network of the first instance
3175 |qemu_system| linux.img \\
3176 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
3177 -netdev socket,id=n2,connect=127.0.0.1:1234
3179 ``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]``
3180 Configure a socket host network backend to share the guest's network
3181 traffic with another QEMU virtual machines using a UDP multicast
3182 socket, effectively making a bus for every QEMU with same multicast
3183 address maddr and port. NOTES:
3185 1. Several QEMU can be running on different hosts and share same bus
3186 (assuming correct multicast setup for these hosts).
3188 2. mcast support is compatible with User Mode Linux (argument
3189 ``ethN=mcast``), see http://user-mode-linux.sf.net.
3191 3. Use ``fd=h`` to specify an already opened UDP multicast socket.
3193 Example:
3195 .. parsed-literal::
3197 # launch one QEMU instance
3198 |qemu_system| linux.img \\
3199 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3200 -netdev socket,id=n1,mcast=230.0.0.1:1234
3201 # launch another QEMU instance on same "bus"
3202 |qemu_system| linux.img \\
3203 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
3204 -netdev socket,id=n2,mcast=230.0.0.1:1234
3205 # launch yet another QEMU instance on same "bus"
3206 |qemu_system| linux.img \\
3207 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\
3208 -netdev socket,id=n3,mcast=230.0.0.1:1234
3210 Example (User Mode Linux compat.):
3212 .. parsed-literal::
3214 # launch QEMU instance (note mcast address selected is UML's default)
3215 |qemu_system| linux.img \\
3216 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3217 -netdev socket,id=n1,mcast=239.192.168.1:1102
3218 # launch UML
3219 /path/to/linux ubd0=/path/to/root_fs eth0=mcast
3221 Example (send packets from host's 1.2.3.4):
3223 .. parsed-literal::
3225 |qemu_system| linux.img \\
3226 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3227 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4
3229 ``-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]``
3230 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931)
3231 is a popular protocol to transport Ethernet (and other Layer 2) data
3232 frames between two systems. It is present in routers, firewalls and
3233 the Linux kernel (from version 3.3 onwards).
3235 This transport allows a VM to communicate to another VM, router or
3236 firewall directly.
3238 ``src=srcaddr``
3239 source address (mandatory)
3241 ``dst=dstaddr``
3242 destination address (mandatory)
3244 ``udp``
3245 select udp encapsulation (default is ip).
3247 ``srcport=srcport``
3248 source udp port.
3250 ``dstport=dstport``
3251 destination udp port.
3253 ``ipv6``
3254 force v6, otherwise defaults to v4.
3256 ``rxcookie=rxcookie``; \ ``txcookie=txcookie``
3257 Cookies are a weak form of security in the l2tpv3 specification.
3258 Their function is mostly to prevent misconfiguration. By default
3259 they are 32 bit.
3261 ``cookie64``
3262 Set cookie size to 64 bit instead of the default 32
3264 ``counter=off``
3265 Force a 'cut-down' L2TPv3 with no counter as in
3266 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00
3268 ``pincounter=on``
3269 Work around broken counter handling in peer. This may also help
3270 on networks which have packet reorder.
3272 ``offset=offset``
3273 Add an extra offset between header and data
3275 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to
3276 the bridge br-lan on the remote Linux host 1.2.3.4:
3278 .. parsed-literal::
3280 # Setup tunnel on linux host using raw ip as encapsulation
3281 # on 1.2.3.4
3282 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\
3283 encap udp udp_sport 16384 udp_dport 16384
3284 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\
3285 0xFFFFFFFF peer_session_id 0xFFFFFFFF
3286 ifconfig vmtunnel0 mtu 1500
3287 ifconfig vmtunnel0 up
3288 brctl addif br-lan vmtunnel0
3291 # on 4.3.2.1
3292 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter
3294 |qemu_system| linux.img -device e1000,netdev=n1 \\
3295 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter
3297 ``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]``
3298 Configure VDE backend to connect to PORT n of a vde switch running
3299 on host and listening for incoming connections on socketpath. Use
3300 GROUP groupname and MODE octalmode to change default ownership and
3301 permissions for communication port. This option is only available if
3302 QEMU has been compiled with vde support enabled.
3304 Example:
3306 .. parsed-literal::
3308 # launch vde switch
3309 vde_switch -F -sock /tmp/myswitch
3310 # launch QEMU instance
3311 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch
3313 ``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]``
3314 Establish a vhost-user netdev, backed by a chardev id. The chardev
3315 should be a unix domain socket backed one. The vhost-user uses a
3316 specifically defined protocol to pass vhost ioctl replacement
3317 messages to an application on the other end of the socket. On
3318 non-MSIX guests, the feature can be forced with vhostforce. Use
3319 'queues=n' to specify the number of queues to be created for
3320 multiqueue vhost-user.
3322 Example:
3326 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
3327 -numa node,memdev=mem \
3328 -chardev socket,id=chr0,path=/path/to/socket \
3329 -netdev type=vhost-user,id=net0,chardev=chr0 \
3330 -device virtio-net-pci,netdev=net0
3332 ``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]``
3333 Establish a vhost-vdpa netdev.
3335 vDPA device is a device that uses a datapath which complies with
3336 the virtio specifications with a vendor specific control path.
3337 vDPA devices can be both physically located on the hardware or
3338 emulated by software.
3340 ``-netdev hubport,id=id,hubid=hubid[,netdev=nd]``
3341 Create a hub port on the emulated hub with ID hubid.
3343 The hubport netdev lets you connect a NIC to a QEMU emulated hub
3344 instead of a single netdev. Alternatively, you can also connect the
3345 hubport to another netdev with ID nd by using the ``netdev=nd``
3346 option.
3348 ``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]``
3349 Legacy option to configure or create an on-board (or machine
3350 default) Network Interface Card(NIC) and connect it either to the
3351 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd.
3352 If model is omitted, then the default NIC model associated with the
3353 machine type is used. Note that the default NIC model may change in
3354 future QEMU releases, so it is highly recommended to always specify
3355 a model. Optionally, the MAC address can be changed to mac, the
3356 device address set to addr (PCI cards only), and a name can be
3357 assigned for use in monitor commands. Optionally, for PCI cards, you
3358 can specify the number v of MSI-X vectors that the card should have;
3359 this option currently only affects virtio cards; set v = 0 to
3360 disable MSI-X. If no ``-net`` option is specified, a single NIC is
3361 created. QEMU can emulate several different models of network card.
3362 Use ``-net nic,model=help`` for a list of available devices for your
3363 target.
3365 ``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]``
3366 Configure a host network backend (with the options corresponding to
3367 the same ``-netdev`` option) and connect it to the emulated hub 0
3368 (the default hub). Use name to specify the name of the hub port.
3369 ERST
3371 DEFHEADING()
3373 DEFHEADING(Character device options:)
3375 DEF("chardev", HAS_ARG, QEMU_OPTION_chardev,
3376 "-chardev help\n"
3377 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3378 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n"
3379 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n"
3380 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n"
3381 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n"
3382 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n"
3383 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n"
3384 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n"
3385 " [,logfile=PATH][,logappend=on|off]\n"
3386 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3387 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n"
3388 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3389 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n"
3390 "-chardev file,id=id,path=path[,input-path=input-file][,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3391 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3392 #ifdef _WIN32
3393 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3394 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3395 #else
3396 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3397 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n"
3398 #endif
3399 #ifdef CONFIG_BRLAPI
3400 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3401 #endif
3402 #if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \
3403 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__)
3404 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3405 #endif
3406 #if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__)
3407 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3408 #endif
3409 #if defined(CONFIG_SPICE)
3410 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3411 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3412 #endif
3413 , QEMU_ARCH_ALL
3416 SRST
3417 The general form of a character device option is:
3419 ``-chardev backend,id=id[,mux=on|off][,options]``
3420 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``,
3421 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``,
3422 ``pty``, ``stdio``, ``braille``, ``parallel``,
3423 ``spicevmc``, ``spiceport``. The specific backend will determine the
3424 applicable options.
3426 Use ``-chardev help`` to print all available chardev backend types.
3428 All devices must have an id, which can be any string up to 127
3429 characters long. It is used to uniquely identify this device in
3430 other command line directives.
3432 A character device may be used in multiplexing mode by multiple
3433 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is
3434 a "1:N" device, and here the "1" end is your specified chardev
3435 backend, and the "N" end is the various parts of QEMU that can talk
3436 to a chardev. If you create a chardev with ``id=myid`` and
3437 ``mux=on``, QEMU will create a multiplexer with your specified ID,
3438 and you can then configure multiple front ends to use that chardev
3439 ID for their input/output. Up to four different front ends can be
3440 connected to a single multiplexed chardev. (Without multiplexing
3441 enabled, a chardev can only be used by a single front end.) For
3442 instance you could use this to allow a single stdio chardev to be
3443 used by two serial ports and the QEMU monitor:
3447 -chardev stdio,mux=on,id=char0 \
3448 -mon chardev=char0,mode=readline \
3449 -serial chardev:char0 \
3450 -serial chardev:char0
3452 You can have more than one multiplexer in a system configuration;
3453 for instance you could have a TCP port multiplexed between UART 0
3454 and UART 1, and stdio multiplexed between the QEMU monitor and a
3455 parallel port:
3459 -chardev stdio,mux=on,id=char0 \
3460 -mon chardev=char0,mode=readline \
3461 -parallel chardev:char0 \
3462 -chardev tcp,...,mux=on,id=char1 \
3463 -serial chardev:char1 \
3464 -serial chardev:char1
3466 When you're using a multiplexed character device, some escape
3467 sequences are interpreted in the input. See the chapter about
3468 :ref:`keys in the character backend multiplexer` in the
3469 System Emulation Users Guide for more details.
3471 Note that some other command line options may implicitly create
3472 multiplexed character backends; for instance ``-serial mon:stdio``
3473 creates a multiplexed stdio backend connected to the serial port and
3474 the QEMU monitor, and ``-nographic`` also multiplexes the console
3475 and the monitor to stdio.
3477 There is currently no support for multiplexing in the other
3478 direction (where a single QEMU front end takes input and output from
3479 multiple chardevs).
3481 Every backend supports the ``logfile`` option, which supplies the
3482 path to a file to record all data transmitted via the backend. The
3483 ``logappend`` option controls whether the log file will be truncated
3484 or appended to when opened.
3486 The available backends are:
3488 ``-chardev null,id=id``
3489 A void device. This device will not emit any data, and will drop any
3490 data it receives. The null backend does not take any options.
3492 ``-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]``
3493 Create a two-way stream socket, which can be either a TCP or a unix
3494 socket. A unix socket will be created if ``path`` is specified.
3495 Behaviour is undefined if TCP options are specified for a unix
3496 socket.
3498 ``server=on|off`` specifies that the socket shall be a listening socket.
3500 ``wait=on|off`` specifies that QEMU should not block waiting for a client
3501 to connect to a listening socket.
3503 ``telnet=on|off`` specifies that traffic on the socket should interpret
3504 telnet escape sequences.
3506 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for
3507 communication.
3509 ``reconnect`` sets the timeout for reconnecting on non-server
3510 sockets when the remote end goes away. qemu will delay this many
3511 seconds and then attempt to reconnect. Zero disables reconnecting,
3512 and is the default.
3514 ``tls-creds`` requests enablement of the TLS protocol for
3515 encryption, and specifies the id of the TLS credentials to use for
3516 the handshake. The credentials must be previously created with the
3517 ``-object tls-creds`` argument.
3519 ``tls-auth`` provides the ID of the QAuthZ authorization object
3520 against which the client's x509 distinguished name will be
3521 validated. This object is only resolved at time of use, so can be
3522 deleted and recreated on the fly while the chardev server is active.
3523 If missing, it will default to denying access.
3525 TCP and unix socket options are given below:
3527 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]``
3528 ``host`` for a listening socket specifies the local address to
3529 be bound. For a connecting socket species the remote host to
3530 connect to. ``host`` is optional for listening sockets. If not
3531 specified it defaults to ``0.0.0.0``.
3533 ``port`` for a listening socket specifies the local port to be
3534 bound. For a connecting socket specifies the port on the remote
3535 host to connect to. ``port`` can be given as either a port
3536 number or a service name. ``port`` is required.
3538 ``to`` is only relevant to listening sockets. If it is
3539 specified, and ``port`` cannot be bound, QEMU will attempt to
3540 bind to subsequent ports up to and including ``to`` until it
3541 succeeds. ``to`` must be specified as a port number.
3543 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4
3544 or IPv6 must be used. If neither is specified the socket may
3545 use either protocol.
3547 ``nodelay=on|off`` disables the Nagle algorithm.
3549 ``unix options: path=path[,abstract=on|off][,tight=on|off]``
3550 ``path`` specifies the local path of the unix socket. ``path``
3551 is required.
3552 ``abstract=on|off`` specifies the use of the abstract socket namespace,
3553 rather than the filesystem. Optional, defaults to false.
3554 ``tight=on|off`` sets the socket length of abstract sockets to their minimum,
3555 rather than the full sun_path length. Optional, defaults to true.
3557 ``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]``
3558 Sends all traffic from the guest to a remote host over UDP.
3560 ``host`` specifies the remote host to connect to. If not specified
3561 it defaults to ``localhost``.
3563 ``port`` specifies the port on the remote host to connect to.
3564 ``port`` is required.
3566 ``localaddr`` specifies the local address to bind to. If not
3567 specified it defaults to ``0.0.0.0``.
3569 ``localport`` specifies the local port to bind to. If not specified
3570 any available local port will be used.
3572 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used.
3573 If neither is specified the device may use either protocol.
3575 ``-chardev msmouse,id=id``
3576 Forward QEMU's emulated msmouse events to the guest. ``msmouse``
3577 does not take any options.
3579 ``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]``
3580 Connect to a QEMU text console. ``vc`` may optionally be given a
3581 specific size.
3583 ``width`` and ``height`` specify the width and height respectively
3584 of the console, in pixels.
3586 ``cols`` and ``rows`` specify that the console be sized to fit a
3587 text console with the given dimensions.
3589 ``-chardev ringbuf,id=id[,size=size]``
3590 Create a ring buffer with fixed size ``size``. size must be a power
3591 of two and defaults to ``64K``.
3593 ``-chardev file,id=id,path=path[,input-path=input-path]``
3594 Log all traffic received from the guest to a file.
3596 ``path`` specifies the path of the file to be opened. This file will
3597 be created if it does not already exist, and overwritten if it does.
3598 ``path`` is required.
3600 If ``input-path`` is specified, this is the path of a second file
3601 which will be used for input. If ``input-path`` is not specified,
3602 no input will be available from the chardev.
3604 Note that ``input-path`` is not supported on Windows hosts.
3606 ``-chardev pipe,id=id,path=path``
3607 Create a two-way connection to the guest. The behaviour differs
3608 slightly between Windows hosts and other hosts:
3610 On Windows, a single duplex pipe will be created at
3611 ``\\.pipe\path``.
3613 On other hosts, 2 pipes will be created called ``path.in`` and
3614 ``path.out``. Data written to ``path.in`` will be received by the
3615 guest. Data written by the guest can be read from ``path.out``. QEMU
3616 will not create these fifos, and requires them to be present.
3618 ``path`` forms part of the pipe path as described above. ``path`` is
3619 required.
3621 ``-chardev console,id=id``
3622 Send traffic from the guest to QEMU's standard output. ``console``
3623 does not take any options.
3625 ``console`` is only available on Windows hosts.
3627 ``-chardev serial,id=id,path=path``
3628 Send traffic from the guest to a serial device on the host.
3630 On Unix hosts serial will actually accept any tty device, not only
3631 serial lines.
3633 ``path`` specifies the name of the serial device to open.
3635 ``-chardev pty,id=id``
3636 Create a new pseudo-terminal on the host and connect to it. ``pty``
3637 does not take any options.
3639 ``pty`` is not available on Windows hosts.
3641 ``-chardev stdio,id=id[,signal=on|off]``
3642 Connect to standard input and standard output of the QEMU process.
3644 ``signal`` controls if signals are enabled on the terminal, that
3645 includes exiting QEMU with the key sequence Control-c. This option
3646 is enabled by default, use ``signal=off`` to disable it.
3648 ``-chardev braille,id=id``
3649 Connect to a local BrlAPI server. ``braille`` does not take any
3650 options.
3652 ``-chardev parallel,id=id,path=path``
3654 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD
3655 hosts.
3657 Connect to a local parallel port.
3659 ``path`` specifies the path to the parallel port device. ``path`` is
3660 required.
3662 ``-chardev spicevmc,id=id,debug=debug,name=name``
3663 ``spicevmc`` is only available when spice support is built in.
3665 ``debug`` debug level for spicevmc
3667 ``name`` name of spice channel to connect to
3669 Connect to a spice virtual machine channel, such as vdiport.
3671 ``-chardev spiceport,id=id,debug=debug,name=name``
3672 ``spiceport`` is only available when spice support is built in.
3674 ``debug`` debug level for spicevmc
3676 ``name`` name of spice port to connect to
3678 Connect to a spice port, allowing a Spice client to handle the
3679 traffic identified by a name (preferably a fqdn).
3680 ERST
3682 DEFHEADING()
3684 #ifdef CONFIG_TPM
3685 DEFHEADING(TPM device options:)
3687 DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \
3688 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n"
3689 " use path to provide path to a character device; default is /dev/tpm0\n"
3690 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n"
3691 " not provided it will be searched for in /sys/class/misc/tpm?/device\n"
3692 "-tpmdev emulator,id=id,chardev=dev\n"
3693 " configure the TPM device using chardev backend\n",
3694 QEMU_ARCH_ALL)
3695 SRST
3696 The general form of a TPM device option is:
3698 ``-tpmdev backend,id=id[,options]``
3699 The specific backend type will determine the applicable options. The
3700 ``-tpmdev`` option creates the TPM backend and requires a
3701 ``-device`` option that specifies the TPM frontend interface model.
3703 Use ``-tpmdev help`` to print all available TPM backend types.
3705 The available backends are:
3707 ``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path``
3708 (Linux-host only) Enable access to the host's TPM using the
3709 passthrough driver.
3711 ``path`` specifies the path to the host's TPM device, i.e., on a
3712 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by
3713 default ``/dev/tpm0`` is used.
3715 ``cancel-path`` specifies the path to the host TPM device's sysfs
3716 entry allowing for cancellation of an ongoing TPM command.
3717 ``cancel-path`` is optional and by default QEMU will search for the
3718 sysfs entry to use.
3720 Some notes about using the host's TPM with the passthrough driver:
3722 The TPM device accessed by the passthrough driver must not be used
3723 by any other application on the host.
3725 Since the host's firmware (BIOS/UEFI) has already initialized the
3726 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize
3727 the TPM again and may therefore not show a TPM-specific menu that
3728 would otherwise allow the user to configure the TPM, e.g., allow the
3729 user to enable/disable or activate/deactivate the TPM. Further, if
3730 TPM ownership is released from within a VM then the host's TPM will
3731 get disabled and deactivated. To enable and activate the TPM again
3732 afterwards, the host has to be rebooted and the user is required to
3733 enter the firmware's menu to enable and activate the TPM. If the TPM
3734 is left disabled and/or deactivated most TPM commands will fail.
3736 To create a passthrough TPM use the following two options:
3740 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0
3742 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by
3743 ``tpmdev=tpm0`` in the device option.
3745 ``-tpmdev emulator,id=id,chardev=dev``
3746 (Linux-host only) Enable access to a TPM emulator using Unix domain
3747 socket based chardev backend.
3749 ``chardev`` specifies the unique ID of a character device backend
3750 that provides connection to the software TPM server.
3752 To create a TPM emulator backend device with chardev socket backend:
3756 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0
3757 ERST
3759 DEFHEADING()
3761 #endif
3763 DEFHEADING(Boot Image or Kernel specific:)
3764 SRST
3765 There are broadly 4 ways you can boot a system with QEMU.
3767 - specify a firmware and let it control finding a kernel
3768 - specify a firmware and pass a hint to the kernel to boot
3769 - direct kernel image boot
3770 - manually load files into the guest's address space
3772 The third method is useful for quickly testing kernels but as there is
3773 no firmware to pass configuration information to the kernel the
3774 hardware must either be probeable, the kernel built for the exact
3775 configuration or passed some configuration data (e.g. a DTB blob)
3776 which tells the kernel what drivers it needs. This exact details are
3777 often hardware specific.
3779 The final method is the most generic way of loading images into the
3780 guest address space and used mostly for ``bare metal`` type
3781 development where the reset vectors of the processor are taken into
3782 account.
3784 ERST
3786 SRST
3788 For x86 machines and some other architectures ``-bios`` will generally
3789 do the right thing with whatever it is given. For other machines the
3790 more strict ``-pflash`` option needs an image that is sized for the
3791 flash device for the given machine type.
3793 Please see the :ref:`system-targets-ref` section of the manual for
3794 more detailed documentation.
3796 ERST
3798 DEF("bios", HAS_ARG, QEMU_OPTION_bios, \
3799 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL)
3800 SRST
3801 ``-bios file``
3802 Set the filename for the BIOS.
3803 ERST
3805 DEF("pflash", HAS_ARG, QEMU_OPTION_pflash,
3806 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL)
3807 SRST
3808 ``-pflash file``
3809 Use file as a parallel flash image.
3810 ERST
3812 SRST
3814 The kernel options were designed to work with Linux kernels although
3815 other things (like hypervisors) can be packaged up as a kernel
3816 executable image. The exact format of a executable image is usually
3817 architecture specific.
3819 The way in which the kernel is started (what address it is loaded at,
3820 what if any information is passed to it via CPU registers, the state
3821 of the hardware when it is started, and so on) is also architecture
3822 specific. Typically it follows the specification laid down by the
3823 Linux kernel for how kernels for that architecture must be started.
3825 ERST
3827 DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \
3828 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL)
3829 SRST
3830 ``-kernel bzImage``
3831 Use bzImage as kernel image. The kernel can be either a Linux kernel
3832 or in multiboot format.
3833 ERST
3835 DEF("append", HAS_ARG, QEMU_OPTION_append, \
3836 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL)
3837 SRST
3838 ``-append cmdline``
3839 Use cmdline as kernel command line
3840 ERST
3842 DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \
3843 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL)
3844 SRST
3845 ``-initrd file``
3846 Use file as initial ram disk.
3848 ``-initrd "file1 arg=foo,file2"``
3849 This syntax is only available with multiboot.
3851 Use file1 and file2 as modules and pass arg=foo as parameter to the
3852 first module.
3853 ERST
3855 DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \
3856 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL)
3857 SRST
3858 ``-dtb file``
3859 Use file as a device tree binary (dtb) image and pass it to the
3860 kernel on boot.
3861 ERST
3863 SRST
3865 Finally you can also manually load images directly into the address
3866 space of the guest. This is most useful for developers who already
3867 know the layout of their guest and take care to ensure something sane
3868 will happen when the reset vector executes.
3870 The generic loader can be invoked by using the loader device:
3872 ``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]``
3874 there is also the guest loader which operates in a similar way but
3875 tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where
3876 the guest image is:
3878 ``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]``
3880 ERST
3882 DEFHEADING()
3884 DEFHEADING(Debug/Expert options:)
3886 DEF("compat", HAS_ARG, QEMU_OPTION_compat,
3887 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n"
3888 " Policy for handling deprecated management interfaces\n"
3889 "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n"
3890 " Policy for handling unstable management interfaces\n",
3891 QEMU_ARCH_ALL)
3892 SRST
3893 ``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]``
3894 Set policy for handling deprecated management interfaces (experimental):
3896 ``deprecated-input=accept`` (default)
3897 Accept deprecated commands and arguments
3898 ``deprecated-input=reject``
3899 Reject deprecated commands and arguments
3900 ``deprecated-input=crash``
3901 Crash on deprecated commands and arguments
3902 ``deprecated-output=accept`` (default)
3903 Emit deprecated command results and events
3904 ``deprecated-output=hide``
3905 Suppress deprecated command results and events
3907 Limitation: covers only syntactic aspects of QMP.
3909 ``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]``
3910 Set policy for handling unstable management interfaces (experimental):
3912 ``unstable-input=accept`` (default)
3913 Accept unstable commands and arguments
3914 ``unstable-input=reject``
3915 Reject unstable commands and arguments
3916 ``unstable-input=crash``
3917 Crash on unstable commands and arguments
3918 ``unstable-output=accept`` (default)
3919 Emit unstable command results and events
3920 ``unstable-output=hide``
3921 Suppress unstable command results and events
3923 Limitation: covers only syntactic aspects of QMP.
3924 ERST
3926 DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg,
3927 "-fw_cfg [name=]<name>,file=<file>\n"
3928 " add named fw_cfg entry with contents from file\n"
3929 "-fw_cfg [name=]<name>,string=<str>\n"
3930 " add named fw_cfg entry with contents from string\n",
3931 QEMU_ARCH_ALL)
3932 SRST
3933 ``-fw_cfg [name=]name,file=file``
3934 Add named fw\_cfg entry with contents from file file.
3936 ``-fw_cfg [name=]name,string=str``
3937 Add named fw\_cfg entry with contents from string str.
3939 The terminating NUL character of the contents of str will not be
3940 included as part of the fw\_cfg item data. To insert contents with
3941 embedded NUL characters, you have to use the file parameter.
3943 The fw\_cfg entries are passed by QEMU through to the guest.
3945 Example:
3949 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin
3951 creates an fw\_cfg entry named opt/com.mycompany/blob with contents
3952 from ./my\_blob.bin.
3953 ERST
3955 DEF("serial", HAS_ARG, QEMU_OPTION_serial, \
3956 "-serial dev redirect the serial port to char device 'dev'\n",
3957 QEMU_ARCH_ALL)
3958 SRST
3959 ``-serial dev``
3960 Redirect the virtual serial port to host character device dev. The
3961 default device is ``vc`` in graphical mode and ``stdio`` in non
3962 graphical mode.
3964 This option can be used several times to simulate up to 4 serial
3965 ports.
3967 Use ``-serial none`` to disable all serial ports.
3969 Available character devices are:
3971 ``vc[:WxH]``
3972 Virtual console. Optionally, a width and height can be given in
3973 pixel with
3977 vc:800x600
3979 It is also possible to specify width or height in characters:
3983 vc:80Cx24C
3985 ``pty``
3986 [Linux only] Pseudo TTY (a new PTY is automatically allocated)
3988 ``none``
3989 No device is allocated.
3991 ``null``
3992 void device
3994 ``chardev:id``
3995 Use a named character device defined with the ``-chardev``
3996 option.
3998 ``/dev/XXX``
3999 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial
4000 port parameters are set according to the emulated ones.
4002 ``/dev/parportN``
4003 [Linux only, parallel port only] Use host parallel port N.
4004 Currently SPP and EPP parallel port features can be used.
4006 ``file:filename``
4007 Write output to filename. No character can be read.
4009 ``stdio``
4010 [Unix only] standard input/output
4012 ``pipe:filename``
4013 name pipe filename
4015 ``COMn``
4016 [Windows only] Use host serial port n
4018 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]``
4019 This implements UDP Net Console. When remote\_host or src\_ip
4020 are not specified they default to ``0.0.0.0``. When not using a
4021 specified src\_port a random port is automatically chosen.
4023 If you just want a simple readonly console you can use
4024 ``netcat`` or ``nc``, by starting QEMU with:
4025 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time
4026 QEMU writes something to that port it will appear in the
4027 netconsole session.
4029 If you plan to send characters back via netconsole or you want
4030 to stop and start QEMU a lot of times, you should have QEMU use
4031 the same source port each time by using something like ``-serial
4032 udp::4555@:4556`` to QEMU. Another approach is to use a patched
4033 version of netcat which can listen to a TCP port and send and
4034 receive characters via udp. If you have a patched version of
4035 netcat which activates telnet remote echo and single char
4036 transfer, then you can use the following options to set up a
4037 netcat redirector to allow telnet on port 5555 to access the
4038 QEMU port.
4040 ``QEMU Options:``
4041 -serial udp::4555@:4556
4043 ``netcat options:``
4044 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T
4046 ``telnet options:``
4047 localhost 5555
4049 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]``
4050 The TCP Net Console has two modes of operation. It can send the
4051 serial I/O to a location or wait for a connection from a
4052 location. By default the TCP Net Console is sent to host at the
4053 port. If you use the ``server=on`` option QEMU will wait for a client
4054 socket application to connect to the port before continuing,
4055 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off``
4056 option disables the Nagle buffering algorithm. The ``reconnect=on``
4057 option only applies if ``server=no`` is set, if the connection goes
4058 down it will attempt to reconnect at the given interval. If host
4059 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a
4060 time is accepted. You can use ``telnet=on`` to connect to the
4061 corresponding character device.
4063 ``Example to send tcp console to 192.168.0.2 port 4444``
4064 -serial tcp:192.168.0.2:4444
4066 ``Example to listen and wait on port 4444 for connection``
4067 -serial tcp::4444,server=on
4069 ``Example to not wait and listen on ip 192.168.0.100 port 4444``
4070 -serial tcp:192.168.0.100:4444,server=on,wait=off
4072 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]``
4073 The telnet protocol is used instead of raw tcp sockets. The
4074 options work the same as if you had specified ``-serial tcp``.
4075 The difference is that the port acts like a telnet server or
4076 client using telnet option negotiation. This will also allow you
4077 to send the MAGIC\_SYSRQ sequence if you use a telnet that
4078 supports sending the break sequence. Typically in unix telnet
4079 you do it with Control-] and then type "send break" followed by
4080 pressing the enter key.
4082 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]``
4083 The WebSocket protocol is used instead of raw tcp socket. The
4084 port acts as a WebSocket server. Client mode is not supported.
4086 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]``
4087 A unix domain socket is used instead of a tcp socket. The option
4088 works the same as if you had specified ``-serial tcp`` except
4089 the unix domain socket path is used for connections.
4091 ``mon:dev_string``
4092 This is a special option to allow the monitor to be multiplexed
4093 onto another serial port. The monitor is accessed with key
4094 sequence of Control-a and then pressing c. dev\_string should be
4095 any one of the serial devices specified above. An example to
4096 multiplex the monitor onto a telnet server listening on port
4097 4444 would be:
4099 ``-serial mon:telnet::4444,server=on,wait=off``
4101 When the monitor is multiplexed to stdio in this way, Ctrl+C
4102 will not terminate QEMU any more but will be passed to the guest
4103 instead.
4105 ``braille``
4106 Braille device. This will use BrlAPI to display the braille
4107 output on a real or fake device.
4109 ``msmouse``
4110 Three button serial mouse. Configure the guest to use Microsoft
4111 protocol.
4112 ERST
4114 DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \
4115 "-parallel dev redirect the parallel port to char device 'dev'\n",
4116 QEMU_ARCH_ALL)
4117 SRST
4118 ``-parallel dev``
4119 Redirect the virtual parallel port to host device dev (same devices
4120 as the serial port). On Linux hosts, ``/dev/parportN`` can be used
4121 to use hardware devices connected on the corresponding host parallel
4122 port.
4124 This option can be used several times to simulate up to 3 parallel
4125 ports.
4127 Use ``-parallel none`` to disable all parallel ports.
4128 ERST
4130 DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \
4131 "-monitor dev redirect the monitor to char device 'dev'\n",
4132 QEMU_ARCH_ALL)
4133 SRST
4134 ``-monitor dev``
4135 Redirect the monitor to host device dev (same devices as the serial
4136 port). The default device is ``vc`` in graphical mode and ``stdio``
4137 in non graphical mode. Use ``-monitor none`` to disable the default
4138 monitor.
4139 ERST
4140 DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \
4141 "-qmp dev like -monitor but opens in 'control' mode\n",
4142 QEMU_ARCH_ALL)
4143 SRST
4144 ``-qmp dev``
4145 Like -monitor but opens in 'control' mode.
4146 ERST
4147 DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \
4148 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n",
4149 QEMU_ARCH_ALL)
4150 SRST
4151 ``-qmp-pretty dev``
4152 Like -qmp but uses pretty JSON formatting.
4153 ERST
4155 DEF("mon", HAS_ARG, QEMU_OPTION_mon, \
4156 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL)
4157 SRST
4158 ``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]``
4159 Setup monitor on chardev name. ``mode=control`` configures
4160 a QMP monitor (a JSON RPC-style protocol) and it is not the
4161 same as HMP, the human monitor that has a "(qemu)" prompt.
4162 ``pretty`` is only valid when ``mode=control``,
4163 turning on JSON pretty printing to ease
4164 human reading and debugging.
4165 ERST
4167 DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \
4168 "-debugcon dev redirect the debug console to char device 'dev'\n",
4169 QEMU_ARCH_ALL)
4170 SRST
4171 ``-debugcon dev``
4172 Redirect the debug console to host device dev (same devices as the
4173 serial port). The debug console is an I/O port which is typically
4174 port 0xe9; writing to that I/O port sends output to this device. The
4175 default device is ``vc`` in graphical mode and ``stdio`` in non
4176 graphical mode.
4177 ERST
4179 DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \
4180 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL)
4181 SRST
4182 ``-pidfile file``
4183 Store the QEMU process PID in file. It is useful if you launch QEMU
4184 from a script.
4185 ERST
4187 DEF("singlestep", 0, QEMU_OPTION_singlestep, \
4188 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL)
4189 SRST
4190 ``-singlestep``
4191 Run the emulation in single step mode.
4192 ERST
4194 DEF("preconfig", 0, QEMU_OPTION_preconfig, \
4195 "--preconfig pause QEMU before machine is initialized (experimental)\n",
4196 QEMU_ARCH_ALL)
4197 SRST
4198 ``--preconfig``
4199 Pause QEMU for interactive configuration before the machine is
4200 created, which allows querying and configuring properties that will
4201 affect machine initialization. Use QMP command 'x-exit-preconfig' to
4202 exit the preconfig state and move to the next state (i.e. run guest
4203 if -S isn't used or pause the second time if -S is used). This
4204 option is experimental.
4205 ERST
4207 DEF("S", 0, QEMU_OPTION_S, \
4208 "-S freeze CPU at startup (use 'c' to start execution)\n",
4209 QEMU_ARCH_ALL)
4210 SRST
4211 ``-S``
4212 Do not start CPU at startup (you must type 'c' in the monitor).
4213 ERST
4215 DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit,
4216 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n"
4217 " run qemu with overcommit hints\n"
4218 " mem-lock=on|off controls memory lock support (default: off)\n"
4219 " cpu-pm=on|off controls cpu power management (default: off)\n",
4220 QEMU_ARCH_ALL)
4221 SRST
4222 ``-overcommit mem-lock=on|off``
4224 ``-overcommit cpu-pm=on|off``
4225 Run qemu with hints about host resource overcommit. The default is
4226 to assume that host overcommits all resources.
4228 Locking qemu and guest memory can be enabled via ``mem-lock=on``
4229 (disabled by default). This works when host memory is not
4230 overcommitted and reduces the worst-case latency for guest.
4232 Guest ability to manage power state of host cpus (increasing latency
4233 for other processes on the same host cpu, but decreasing latency for
4234 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This
4235 works best when host CPU is not overcommitted. When used, host
4236 estimates of CPU cycle and power utilization will be incorrect, not
4237 taking into account guest idle time.
4238 ERST
4240 DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \
4241 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n"
4242 " the guest without waiting for gdb to connect; use -S too\n"
4243 " if you want it to not start execution.)\n",
4244 QEMU_ARCH_ALL)
4245 SRST
4246 ``-gdb dev``
4247 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter
4248 in the System Emulation Users Guide). Note that this option does not pause QEMU
4249 execution -- if you want QEMU to not start the guest until you
4250 connect with gdb and issue a ``continue`` command, you will need to
4251 also pass the ``-S`` option to QEMU.
4253 The most usual configuration is to listen on a local TCP socket::
4255 -gdb tcp::3117
4257 but you can specify other backends; UDP, pseudo TTY, or even stdio
4258 are all reasonable use cases. For example, a stdio connection
4259 allows you to start QEMU from within gdb and establish the
4260 connection via a pipe:
4262 .. parsed-literal::
4264 (gdb) target remote | exec |qemu_system| -gdb stdio ...
4265 ERST
4267 DEF("s", 0, QEMU_OPTION_s, \
4268 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n",
4269 QEMU_ARCH_ALL)
4270 SRST
4271 ``-s``
4272 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234
4273 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide).
4274 ERST
4276 DEF("d", HAS_ARG, QEMU_OPTION_d, \
4277 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n",
4278 QEMU_ARCH_ALL)
4279 SRST
4280 ``-d item1[,...]``
4281 Enable logging of specified items. Use '-d help' for a list of log
4282 items.
4283 ERST
4285 DEF("D", HAS_ARG, QEMU_OPTION_D, \
4286 "-D logfile output log to logfile (default stderr)\n",
4287 QEMU_ARCH_ALL)
4288 SRST
4289 ``-D logfile``
4290 Output log in logfile instead of to stderr
4291 ERST
4293 DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \
4294 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n",
4295 QEMU_ARCH_ALL)
4296 SRST
4297 ``-dfilter range1[,...]``
4298 Filter debug output to that relevant to a range of target addresses.
4299 The filter spec can be either start+size, start-size or start..end
4300 where start end and size are the addresses and sizes required. For
4301 example:
4305 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000
4307 Will dump output for any code in the 0x1000 sized block starting at
4308 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and
4309 another 0x1000 sized block starting at 0xffffffc00005f000.
4310 ERST
4312 DEF("seed", HAS_ARG, QEMU_OPTION_seed, \
4313 "-seed number seed the pseudo-random number generator\n",
4314 QEMU_ARCH_ALL)
4315 SRST
4316 ``-seed number``
4317 Force the guest to use a deterministic pseudo-random number
4318 generator, seeded with number. This does not affect crypto routines
4319 within the host.
4320 ERST
4322 DEF("L", HAS_ARG, QEMU_OPTION_L, \
4323 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n",
4324 QEMU_ARCH_ALL)
4325 SRST
4326 ``-L path``
4327 Set the directory for the BIOS, VGA BIOS and keymaps.
4329 To list all the data directories, use ``-L help``.
4330 ERST
4332 DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \
4333 "-enable-kvm enable KVM full virtualization support\n",
4334 QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC |
4335 QEMU_ARCH_RISCV | QEMU_ARCH_S390X)
4336 SRST
4337 ``-enable-kvm``
4338 Enable KVM full virtualization support. This option is only
4339 available if KVM support is enabled when compiling.
4340 ERST
4342 DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid,
4343 "-xen-domid id specify xen guest domain id\n",
4344 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4345 DEF("xen-attach", 0, QEMU_OPTION_xen_attach,
4346 "-xen-attach attach to existing xen domain\n"
4347 " libxl will use this when starting QEMU\n",
4348 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4349 DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict,
4350 "-xen-domid-restrict restrict set of available xen operations\n"
4351 " to specified domain id. (Does not affect\n"
4352 " xenpv machine type).\n",
4353 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4354 SRST
4355 ``-xen-domid id``
4356 Specify xen guest domain id (XEN only).
4358 ``-xen-attach``
4359 Attach to existing xen domain. libxl will use this when starting
4360 QEMU (XEN only). Restrict set of available xen operations to
4361 specified domain id (XEN only).
4362 ERST
4364 DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \
4365 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL)
4366 SRST
4367 ``-no-reboot``
4368 Exit instead of rebooting.
4369 ERST
4371 DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \
4372 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL)
4373 SRST
4374 ``-no-shutdown``
4375 Don't exit QEMU on guest shutdown, but instead only stop the
4376 emulation. This allows for instance switching to monitor to commit
4377 changes to the disk image.
4378 ERST
4380 DEF("action", HAS_ARG, QEMU_OPTION_action,
4381 "-action reboot=reset|shutdown\n"
4382 " action when guest reboots [default=reset]\n"
4383 "-action shutdown=poweroff|pause\n"
4384 " action when guest shuts down [default=poweroff]\n"
4385 "-action panic=pause|shutdown|exit-failure|none\n"
4386 " action when guest panics [default=shutdown]\n"
4387 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n"
4388 " action when watchdog fires [default=reset]\n",
4389 QEMU_ARCH_ALL)
4390 SRST
4391 ``-action event=action``
4392 The action parameter serves to modify QEMU's default behavior when
4393 certain guest events occur. It provides a generic method for specifying the
4394 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown``
4395 parameters.
4397 Examples:
4399 ``-action panic=none``
4400 ``-action reboot=shutdown,shutdown=pause``
4401 ``-device i6300esb -action watchdog=pause``
4403 ERST
4405 DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \
4406 "-loadvm [tag|id]\n" \
4407 " start right away with a saved state (loadvm in monitor)\n",
4408 QEMU_ARCH_ALL)
4409 SRST
4410 ``-loadvm file``
4411 Start right away with a saved state (``loadvm`` in monitor)
4412 ERST
4414 #ifndef _WIN32
4415 DEF("daemonize", 0, QEMU_OPTION_daemonize, \
4416 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL)
4417 #endif
4418 SRST
4419 ``-daemonize``
4420 Daemonize the QEMU process after initialization. QEMU will not
4421 detach from standard IO until it is ready to receive connections on
4422 any of its devices. This option is a useful way for external
4423 programs to launch QEMU without having to cope with initialization
4424 race conditions.
4425 ERST
4427 DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \
4428 "-option-rom rom load a file, rom, into the option ROM space\n",
4429 QEMU_ARCH_ALL)
4430 SRST
4431 ``-option-rom file``
4432 Load the contents of file as an option ROM. This option is useful to
4433 load things like EtherBoot.
4434 ERST
4436 DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \
4437 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \
4438 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n",
4439 QEMU_ARCH_ALL)
4441 SRST
4442 ``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]``
4443 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at
4444 the current UTC or local time, respectively. ``localtime`` is
4445 required for correct date in MS-DOS or Windows. To start at a
4446 specific point in time, provide datetime in the format
4447 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC.
4449 By default the RTC is driven by the host system time. This allows
4450 using of the RTC as accurate reference clock inside the guest,
4451 specifically if the host time is smoothly following an accurate
4452 external reference clock, e.g. via NTP. If you want to isolate the
4453 guest time from the host, you can set ``clock`` to ``rt`` instead,
4454 which provides a host monotonic clock if host support it. To even
4455 prevent the RTC from progressing during suspension, you can set
4456 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is
4457 recommended especially in icount mode in order to preserve
4458 determinism; however, note that in icount mode the speed of the
4459 virtual clock is variable and can in general differ from the host
4460 clock.
4462 Enable ``driftfix`` (i386 targets only) if you experience time drift
4463 problems, specifically with Windows' ACPI HAL. This option will try
4464 to figure out how many timer interrupts were not processed by the
4465 Windows guest and will re-inject them.
4466 ERST
4468 DEF("icount", HAS_ARG, QEMU_OPTION_icount, \
4469 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \
4470 " enable virtual instruction counter with 2^N clock ticks per\n" \
4471 " instruction, enable aligning the host and virtual clocks\n" \
4472 " or disable real time cpu sleeping, and optionally enable\n" \
4473 " record-and-replay mode\n", QEMU_ARCH_ALL)
4474 SRST
4475 ``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]``
4476 Enable virtual instruction counter. The virtual cpu will execute one
4477 instruction every 2^N ns of virtual time. If ``auto`` is specified
4478 then the virtual cpu speed will be automatically adjusted to keep
4479 virtual time within a few seconds of real time.
4481 Note that while this option can give deterministic behavior, it does
4482 not provide cycle accurate emulation. Modern CPUs contain
4483 superscalar out of order cores with complex cache hierarchies. The
4484 number of instructions executed often has little or no correlation
4485 with actual performance.
4487 When the virtual cpu is sleeping, the virtual time will advance at
4488 default speed unless ``sleep=on`` is specified. With
4489 ``sleep=on``, the virtual time will jump to the next timer
4490 deadline instantly whenever the virtual cpu goes to sleep mode and
4491 will not advance if no timer is enabled. This behavior gives
4492 deterministic execution times from the guest point of view.
4493 The default if icount is enabled is ``sleep=off``.
4494 ``sleep=on`` cannot be used together with either ``shift=auto``
4495 or ``align=on``.
4497 ``align=on`` will activate the delay algorithm which will try to
4498 synchronise the host clock and the virtual clock. The goal is to
4499 have a guest running at the real frequency imposed by the shift
4500 option. Whenever the guest clock is behind the host clock and if
4501 ``align=on`` is specified then we print a message to the user to
4502 inform about the delay. Currently this option does not work when
4503 ``shift`` is ``auto``. Note: The sync algorithm will work for those
4504 shift values for which the guest clock runs ahead of the host clock.
4505 Typically this happens when the shift value is high (how high
4506 depends on the host machine). The default if icount is enabled
4507 is ``align=off``.
4509 When the ``rr`` option is specified deterministic record/replay is
4510 enabled. The ``rrfile=`` option must also be provided to
4511 specify the path to the replay log. In record mode data is written
4512 to this file, and in replay mode it is read back.
4513 If the ``rrsnapshot`` option is given then it specifies a VM snapshot
4514 name. In record mode, a new VM snapshot with the given name is created
4515 at the start of execution recording. In replay mode this option
4516 specifies the snapshot name used to load the initial VM state.
4517 ERST
4519 DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \
4520 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \
4521 " action when watchdog fires [default=reset]\n",
4522 QEMU_ARCH_ALL)
4523 SRST
4524 ``-watchdog-action action``
4525 The action controls what QEMU will do when the watchdog timer
4526 expires. The default is ``reset`` (forcefully reset the guest).
4527 Other possible actions are: ``shutdown`` (attempt to gracefully
4528 shutdown the guest), ``poweroff`` (forcefully poweroff the guest),
4529 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the
4530 guest), ``debug`` (print a debug message and continue), or ``none``
4531 (do nothing).
4533 Note that the ``shutdown`` action requires that the guest responds
4534 to ACPI signals, which it may not be able to do in the sort of
4535 situations where the watchdog would have expired, and thus
4536 ``-watchdog-action shutdown`` is not recommended for production use.
4538 Examples:
4540 ``-device i6300esb -watchdog-action pause``
4542 ERST
4544 DEF("echr", HAS_ARG, QEMU_OPTION_echr, \
4545 "-echr chr set terminal escape character instead of ctrl-a\n",
4546 QEMU_ARCH_ALL)
4547 SRST
4548 ``-echr numeric_ascii_value``
4549 Change the escape character used for switching to the monitor when
4550 using monitor and serial sharing. The default is ``0x01`` when using
4551 the ``-nographic`` option. ``0x01`` is equal to pressing
4552 ``Control-a``. You can select a different character from the ascii
4553 control keys where 1 through 26 map to Control-a through Control-z.
4554 For instance you could use the either of the following to change the
4555 escape character to Control-t.
4557 ``-echr 0x14``; \ ``-echr 20``
4559 ERST
4561 DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \
4562 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \
4563 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \
4564 "-incoming unix:socketpath\n" \
4565 " prepare for incoming migration, listen on\n" \
4566 " specified protocol and socket address\n" \
4567 "-incoming fd:fd\n" \
4568 "-incoming exec:cmdline\n" \
4569 " accept incoming migration on given file descriptor\n" \
4570 " or from given external command\n" \
4571 "-incoming defer\n" \
4572 " wait for the URI to be specified via migrate_incoming\n",
4573 QEMU_ARCH_ALL)
4574 SRST
4575 ``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]``
4577 ``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]``
4578 Prepare for incoming migration, listen on a given tcp port.
4580 ``-incoming unix:socketpath``
4581 Prepare for incoming migration, listen on a given unix socket.
4583 ``-incoming fd:fd``
4584 Accept incoming migration from a given filedescriptor.
4586 ``-incoming exec:cmdline``
4587 Accept incoming migration as an output from specified external
4588 command.
4590 ``-incoming defer``
4591 Wait for the URI to be specified via migrate\_incoming. The monitor
4592 can be used to change settings (such as migration parameters) prior
4593 to issuing the migrate\_incoming to allow the migration to begin.
4594 ERST
4596 DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \
4597 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL)
4598 SRST
4599 ``-only-migratable``
4600 Only allow migratable devices. Devices will not be allowed to enter
4601 an unmigratable state.
4602 ERST
4604 DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \
4605 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL)
4606 SRST
4607 ``-nodefaults``
4608 Don't create default devices. Normally, QEMU sets the default
4609 devices like serial port, parallel port, virtual console, monitor
4610 device, VGA adapter, floppy and CD-ROM drive and others. The
4611 ``-nodefaults`` option will disable all those default devices.
4612 ERST
4614 #ifndef _WIN32
4615 DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \
4616 "-chroot dir chroot to dir just before starting the VM\n",
4617 QEMU_ARCH_ALL)
4618 #endif
4619 SRST
4620 ``-chroot dir``
4621 Immediately before starting guest execution, chroot to the specified
4622 directory. Especially useful in combination with -runas.
4623 ERST
4625 #ifndef _WIN32
4626 DEF("runas", HAS_ARG, QEMU_OPTION_runas, \
4627 "-runas user change to user id user just before starting the VM\n" \
4628 " user can be numeric uid:gid instead\n",
4629 QEMU_ARCH_ALL)
4630 #endif
4631 SRST
4632 ``-runas user``
4633 Immediately before starting guest execution, drop root privileges,
4634 switching to the specified user.
4635 ERST
4637 DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env,
4638 "-prom-env variable=value\n"
4639 " set OpenBIOS nvram variables\n",
4640 QEMU_ARCH_PPC | QEMU_ARCH_SPARC)
4641 SRST
4642 ``-prom-env variable=value``
4643 Set OpenBIOS nvram variable to given value (PPC, SPARC only).
4647 qemu-system-sparc -prom-env 'auto-boot?=false' \
4648 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single'
4652 qemu-system-ppc -prom-env 'auto-boot?=false' \
4653 -prom-env 'boot-device=hd:2,\yaboot' \
4654 -prom-env 'boot-args=conf=hd:2,\yaboot.conf'
4655 ERST
4656 DEF("semihosting", 0, QEMU_OPTION_semihosting,
4657 "-semihosting semihosting mode\n",
4658 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA |
4659 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4660 SRST
4661 ``-semihosting``
4662 Enable :ref:`Semihosting` mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only).
4664 .. warning::
4665 Note that this allows guest direct access to the host filesystem, so
4666 should only be used with a trusted guest OS.
4668 See the -semihosting-config option documentation for further
4669 information about the facilities this enables.
4670 ERST
4671 DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config,
4672 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]\n" \
4673 " semihosting configuration\n",
4674 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA |
4675 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4676 SRST
4677 ``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]``
4678 Enable and configure :ref:`Semihosting` (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V
4679 only).
4681 .. warning::
4682 Note that this allows guest direct access to the host filesystem, so
4683 should only be used with a trusted guest OS.
4685 ``target=native|gdb|auto``
4686 Defines where the semihosting calls will be addressed, to QEMU
4687 (``native``) or to GDB (``gdb``). The default is ``auto``, which
4688 means ``gdb`` during debug sessions and ``native`` otherwise.
4690 ``chardev=str1``
4691 Send the output to a chardev backend output for native or auto
4692 output when not in gdb
4694 ``userspace=on|off``
4695 Allows code running in guest userspace to access the semihosting
4696 interface. The default is that only privileged guest code can
4697 make semihosting calls. Note that setting ``userspace=on`` should
4698 only be used if all guest code is trusted (for example, in
4699 bare-metal test case code).
4701 ``arg=str1,arg=str2,...``
4702 Allows the user to pass input arguments, and can be used
4703 multiple times to build up a list. The old-style
4704 ``-kernel``/``-append`` method of passing a command line is
4705 still supported for backward compatibility. If both the
4706 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are
4707 specified, the former is passed to semihosting as it always
4708 takes precedence.
4709 ERST
4710 DEF("old-param", 0, QEMU_OPTION_old_param,
4711 "-old-param old param mode\n", QEMU_ARCH_ARM)
4712 SRST
4713 ``-old-param``
4714 Old param mode (ARM only).
4715 ERST
4717 DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \
4718 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \
4719 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \
4720 " Enable seccomp mode 2 system call filter (default 'off').\n" \
4721 " use 'obsolete' to allow obsolete system calls that are provided\n" \
4722 " by the kernel, but typically no longer used by modern\n" \
4723 " C library implementations.\n" \
4724 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \
4725 " to elevate privileges using set*uid|gid system calls.\n" \
4726 " The value 'children' will deny set*uid|gid system calls for\n" \
4727 " main QEMU process but will allow forks and execves to run unprivileged\n" \
4728 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \
4729 " blocking *fork and execve\n" \
4730 " use 'resourcecontrol' to disable process affinity and schedular priority\n",
4731 QEMU_ARCH_ALL)
4732 SRST
4733 ``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]``
4734 Enable Seccomp mode 2 system call filter. 'on' will enable syscall
4735 filtering and 'off' will disable it. The default is 'off'.
4737 ``obsolete=string``
4738 Enable Obsolete system calls
4740 ``elevateprivileges=string``
4741 Disable set\*uid\|gid system calls
4743 ``spawn=string``
4744 Disable \*fork and execve
4746 ``resourcecontrol=string``
4747 Disable process affinity and schedular priority
4748 ERST
4750 DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig,
4751 "-readconfig <file>\n"
4752 " read config file\n", QEMU_ARCH_ALL)
4753 SRST
4754 ``-readconfig file``
4755 Read device configuration from file. This approach is useful when
4756 you want to spawn QEMU process with many command line options but
4757 you don't want to exceed the command line character limit.
4758 ERST
4760 DEF("no-user-config", 0, QEMU_OPTION_nouserconfig,
4761 "-no-user-config\n"
4762 " do not load default user-provided config files at startup\n",
4763 QEMU_ARCH_ALL)
4764 SRST
4765 ``-no-user-config``
4766 The ``-no-user-config`` option makes QEMU not load any of the
4767 user-provided config files on sysconfdir.
4768 ERST
4770 DEF("trace", HAS_ARG, QEMU_OPTION_trace,
4771 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n"
4772 " specify tracing options\n",
4773 QEMU_ARCH_ALL)
4774 SRST
4775 ``-trace [[enable=]pattern][,events=file][,file=file]``
4776 .. include:: ../qemu-option-trace.rst.inc
4778 ERST
4779 DEF("plugin", HAS_ARG, QEMU_OPTION_plugin,
4780 "-plugin [file=]<file>[,<argname>=<argvalue>]\n"
4781 " load a plugin\n",
4782 QEMU_ARCH_ALL)
4783 SRST
4784 ``-plugin file=file[,argname=argvalue]``
4785 Load a plugin.
4787 ``file=file``
4788 Load the given plugin from a shared library file.
4790 ``argname=argvalue``
4791 Argument passed to the plugin. (Can be given multiple times.)
4792 ERST
4794 HXCOMM Internal use
4795 DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL)
4796 DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL)
4798 #ifdef __linux__
4799 DEF("async-teardown", 0, QEMU_OPTION_asyncteardown,
4800 "-async-teardown enable asynchronous teardown\n",
4801 QEMU_ARCH_ALL)
4802 #endif
4803 SRST
4804 ``-async-teardown``
4805 Enable asynchronous teardown. A new process called "cleanup/<QEMU_PID>"
4806 will be created at startup sharing the address space with the main qemu
4807 process, using clone. It will wait for the main qemu process to
4808 terminate completely, and then exit.
4809 This allows qemu to terminate very quickly even if the guest was
4810 huge, leaving the teardown of the address space to the cleanup
4811 process. Since the cleanup process shares the same cgroups as the
4812 main qemu process, accounting is performed correctly. This only
4813 works if the cleanup process is not forcefully killed with SIGKILL
4814 before the main qemu process has terminated completely.
4815 ERST
4817 DEF("msg", HAS_ARG, QEMU_OPTION_msg,
4818 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n"
4819 " control error message format\n"
4820 " timestamp=on enables timestamps (default: off)\n"
4821 " guest-name=on enables guest name prefix but only if\n"
4822 " -name guest option is set (default: off)\n",
4823 QEMU_ARCH_ALL)
4824 SRST
4825 ``-msg [timestamp[=on|off]][,guest-name[=on|off]]``
4826 Control error message format.
4828 ``timestamp=on|off``
4829 Prefix messages with a timestamp. Default is off.
4831 ``guest-name=on|off``
4832 Prefix messages with guest name but only if -name guest option is set
4833 otherwise the option is ignored. Default is off.
4834 ERST
4836 DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate,
4837 "-dump-vmstate <file>\n"
4838 " Output vmstate information in JSON format to file.\n"
4839 " Use the scripts/vmstate-static-checker.py file to\n"
4840 " check for possible regressions in migration code\n"
4841 " by comparing two such vmstate dumps.\n",
4842 QEMU_ARCH_ALL)
4843 SRST
4844 ``-dump-vmstate file``
4845 Dump json-encoded vmstate information for current machine type to
4846 file in file
4847 ERST
4849 DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile,
4850 "-enable-sync-profile\n"
4851 " enable synchronization profiling\n",
4852 QEMU_ARCH_ALL)
4853 SRST
4854 ``-enable-sync-profile``
4855 Enable synchronization profiling.
4856 ERST
4858 #if defined(CONFIG_TCG) && defined(CONFIG_LINUX)
4859 DEF("perfmap", 0, QEMU_OPTION_perfmap,
4860 "-perfmap generate a /tmp/perf-${pid}.map file for perf\n",
4861 QEMU_ARCH_ALL)
4862 SRST
4863 ``-perfmap``
4864 Generate a map file for Linux perf tools that will allow basic profiling
4865 information to be broken down into basic blocks.
4866 ERST
4868 DEF("jitdump", 0, QEMU_OPTION_jitdump,
4869 "-jitdump generate a jit-${pid}.dump file for perf\n",
4870 QEMU_ARCH_ALL)
4871 SRST
4872 ``-jitdump``
4873 Generate a dump file for Linux perf tools that maps basic blocks to symbol
4874 names, line numbers and JITted code.
4875 ERST
4876 #endif
4878 DEFHEADING()
4880 DEFHEADING(Generic object creation:)
4882 DEF("object", HAS_ARG, QEMU_OPTION_object,
4883 "-object TYPENAME[,PROP1=VALUE1,...]\n"
4884 " create a new object of type TYPENAME setting properties\n"
4885 " in the order they are specified. Note that the 'id'\n"
4886 " property must be set. These objects are placed in the\n"
4887 " '/objects' path.\n",
4888 QEMU_ARCH_ALL)
4889 SRST
4890 ``-object typename[,prop1=value1,...]``
4891 Create a new object of type typename setting properties in the order
4892 they are specified. Note that the 'id' property must be set. These
4893 objects are placed in the '/objects' path.
4895 ``-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``
4896 Creates a memory file backend object, which can be used to back
4897 the guest RAM with huge pages.
4899 The ``id`` parameter is a unique ID that will be used to
4900 reference this memory region in other parameters, e.g. ``-numa``,
4901 ``-device nvdimm``, etc.
4903 The ``size`` option provides the size of the memory region, and
4904 accepts common suffixes, e.g. ``500M``.
4906 The ``mem-path`` provides the path to either a shared memory or
4907 huge page filesystem mount.
4909 The ``share`` boolean option determines whether the memory
4910 region is marked as private to QEMU, or shared. The latter
4911 allows a co-operating external process to access the QEMU memory
4912 region.
4914 The ``share`` is also required for pvrdma devices due to
4915 limitations in the RDMA API provided by Linux.
4917 Setting share=on might affect the ability to configure NUMA
4918 bindings for the memory backend under some circumstances, see
4919 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel
4920 source tree for additional details.
4922 Setting the ``discard-data`` boolean option to on indicates that
4923 file contents can be destroyed when QEMU exits, to avoid
4924 unnecessarily flushing data to the backing file. Note that
4925 ``discard-data`` is only an optimization, and QEMU might not
4926 discard file contents if it aborts unexpectedly or is terminated
4927 using SIGKILL.
4929 The ``merge`` boolean option enables memory merge, also known as
4930 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider
4931 the pages for memory deduplication.
4933 Setting the ``dump`` boolean option to off excludes the memory
4934 from core dumps. This feature is also known as MADV\_DONTDUMP.
4936 The ``prealloc`` boolean option enables memory preallocation.
4938 The ``host-nodes`` option binds the memory range to a list of
4939 NUMA host nodes.
4941 The ``policy`` option sets the NUMA policy to one of the
4942 following values:
4944 ``default``
4945 default host policy
4947 ``preferred``
4948 prefer the given host node list for allocation
4950 ``bind``
4951 restrict memory allocation to the given host node list
4953 ``interleave``
4954 interleave memory allocations across the given host node
4955 list
4957 The ``align`` option specifies the base address alignment when
4958 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg
4959 ``2M``. Some backend store specified by ``mem-path`` requires an
4960 alignment different than the default one used by QEMU, eg the
4961 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
4962 such cases, users can specify the required alignment via this
4963 option.
4965 The ``pmem`` option specifies whether the backing file specified
4966 by ``mem-path`` is in host persistent memory that can be
4967 accessed using the SNIA NVM programming model (e.g. Intel
4968 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary
4969 operations to guarantee the persistence of its own writes to
4970 ``mem-path`` (e.g. in vNVDIMM label emulation and live
4971 migration). Also, we will map the backend-file with MAP\_SYNC
4972 flag, which ensures the file metadata is in sync for
4973 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC
4974 requires support from both the host kernel (since Linux kernel
4975 4.15) and the filesystem of ``mem-path`` mounted with DAX
4976 option.
4978 The ``readonly`` option specifies whether the backing file is opened
4979 read-only or read-write (default).
4981 ``-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``
4982 Creates a memory backend object, which can be used to back the
4983 guest RAM. Memory backend objects offer more control than the
4984 ``-m`` option that is traditionally used to define guest RAM.
4985 Please refer to ``memory-backend-file`` for a description of the
4986 options.
4988 ``-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``
4989 Creates an anonymous memory file backend object, which allows
4990 QEMU to share the memory with an external process (e.g. when
4991 using vhost-user). The memory is allocated with memfd and
4992 optional sealing. (Linux only)
4994 The ``seal`` option creates a sealed-file, that will block
4995 further resizing the memory ('on' by default).
4997 The ``hugetlb`` option specify the file to be created resides in
4998 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction
4999 with the ``hugetlb`` option, the ``hugetlbsize`` option specify
5000 the hugetlb page size on systems that support multiple hugetlb
5001 page sizes (it must be a power of 2 value supported by the
5002 system).
5004 In some versions of Linux, the ``hugetlb`` option is
5005 incompatible with the ``seal`` option (requires at least Linux
5006 4.16).
5008 Please refer to ``memory-backend-file`` for a description of the
5009 other options.
5011 The ``share`` boolean option is on by default with memfd.
5013 ``-object rng-builtin,id=id``
5014 Creates a random number generator backend which obtains entropy
5015 from QEMU builtin functions. The ``id`` parameter is a unique ID
5016 that will be used to reference this entropy backend from the
5017 ``virtio-rng`` device. By default, the ``virtio-rng`` device
5018 uses this RNG backend.
5020 ``-object rng-random,id=id,filename=/dev/random``
5021 Creates a random number generator backend which obtains entropy
5022 from a device on the host. The ``id`` parameter is a unique ID
5023 that will be used to reference this entropy backend from the
5024 ``virtio-rng`` device. The ``filename`` parameter specifies
5025 which file to obtain entropy from and if omitted defaults to
5026 ``/dev/urandom``.
5028 ``-object rng-egd,id=id,chardev=chardevid``
5029 Creates a random number generator backend which obtains entropy
5030 from an external daemon running on the host. The ``id``
5031 parameter is a unique ID that will be used to reference this
5032 entropy backend from the ``virtio-rng`` device. The ``chardev``
5033 parameter is the unique ID of a character device backend that
5034 provides the connection to the RNG daemon.
5036 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off``
5037 Creates a TLS anonymous credentials object, which can be used to
5038 provide TLS support on network backends. The ``id`` parameter is
5039 a unique ID which network backends will use to access the
5040 credentials. The ``endpoint`` is either ``server`` or ``client``
5041 depending on whether the QEMU network backend that uses the
5042 credentials will be acting as a client or as a server. If
5043 ``verify-peer`` is enabled (the default) then once the handshake
5044 is completed, the peer credentials will be verified, though this
5045 is a no-op for anonymous credentials.
5047 The dir parameter tells QEMU where to find the credential files.
5048 For server endpoints, this directory may contain a file
5049 dh-params.pem providing diffie-hellman parameters to use for the
5050 TLS server. If the file is missing, QEMU will generate a set of
5051 DH parameters at startup. This is a computationally expensive
5052 operation that consumes random pool entropy, so it is
5053 recommended that a persistent set of parameters be generated
5054 upfront and saved.
5056 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]``
5057 Creates a TLS Pre-Shared Keys (PSK) credentials object, which
5058 can be used to provide TLS support on network backends. The
5059 ``id`` parameter is a unique ID which network backends will use
5060 to access the credentials. The ``endpoint`` is either ``server``
5061 or ``client`` depending on whether the QEMU network backend that
5062 uses the credentials will be acting as a client or as a server.
5063 For clients only, ``username`` is the username which will be
5064 sent to the server. If omitted it defaults to "qemu".
5066 The dir parameter tells QEMU where to find the keys file. It is
5067 called "dir/keys.psk" and contains "username:key" pairs. This
5068 file can most easily be created using the GnuTLS ``psktool``
5069 program.
5071 For server endpoints, dir may also contain a file dh-params.pem
5072 providing diffie-hellman parameters to use for the TLS server.
5073 If the file is missing, QEMU will generate a set of DH
5074 parameters at startup. This is a computationally expensive
5075 operation that consumes random pool entropy, so it is
5076 recommended that a persistent set of parameters be generated up
5077 front and saved.
5079 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id``
5080 Creates a TLS anonymous credentials object, which can be used to
5081 provide TLS support on network backends. The ``id`` parameter is
5082 a unique ID which network backends will use to access the
5083 credentials. The ``endpoint`` is either ``server`` or ``client``
5084 depending on whether the QEMU network backend that uses the
5085 credentials will be acting as a client or as a server. If
5086 ``verify-peer`` is enabled (the default) then once the handshake
5087 is completed, the peer credentials will be verified. With x509
5088 certificates, this implies that the clients must be provided
5089 with valid client certificates too.
5091 The dir parameter tells QEMU where to find the credential files.
5092 For server endpoints, this directory may contain a file
5093 dh-params.pem providing diffie-hellman parameters to use for the
5094 TLS server. If the file is missing, QEMU will generate a set of
5095 DH parameters at startup. This is a computationally expensive
5096 operation that consumes random pool entropy, so it is
5097 recommended that a persistent set of parameters be generated
5098 upfront and saved.
5100 For x509 certificate credentials the directory will contain
5101 further files providing the x509 certificates. The certificates
5102 must be stored in PEM format, in filenames ca-cert.pem,
5103 ca-crl.pem (optional), server-cert.pem (only servers),
5104 server-key.pem (only servers), client-cert.pem (only clients),
5105 and client-key.pem (only clients).
5107 For the server-key.pem and client-key.pem files which contain
5108 sensitive private keys, it is possible to use an encrypted
5109 version by providing the passwordid parameter. This provides the
5110 ID of a previously created ``secret`` object containing the
5111 password for decryption.
5113 The priority parameter allows to override the global default
5114 priority used by gnutls. This can be useful if the system
5115 administrator needs to use a weaker set of crypto priorities for
5116 QEMU without potentially forcing the weakness onto all
5117 applications. Or conversely if one wants wants a stronger
5118 default for QEMU than for all other applications, they can do
5119 this through this parameter. Its format is a gnutls priority
5120 string as described at
5121 https://gnutls.org/manual/html_node/Priority-Strings.html.
5123 ``-object tls-cipher-suites,id=id,priority=priority``
5124 Creates a TLS cipher suites object, which can be used to control
5125 the TLS cipher/protocol algorithms that applications are permitted
5126 to use.
5128 The ``id`` parameter is a unique ID which frontends will use to
5129 access the ordered list of permitted TLS cipher suites from the
5130 host.
5132 The ``priority`` parameter allows to override the global default
5133 priority used by gnutls. This can be useful if the system
5134 administrator needs to use a weaker set of crypto priorities for
5135 QEMU without potentially forcing the weakness onto all
5136 applications. Or conversely if one wants wants a stronger
5137 default for QEMU than for all other applications, they can do
5138 this through this parameter. Its format is a gnutls priority
5139 string as described at
5140 https://gnutls.org/manual/html_node/Priority-Strings.html.
5142 An example of use of this object is to control UEFI HTTPS Boot.
5143 The tls-cipher-suites object exposes the ordered list of permitted
5144 TLS cipher suites from the host side to the guest firmware, via
5145 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER
5146 objects. The firmware uses the IANA_TLS_CIPHER array for configuring
5147 guest-side TLS.
5149 In the following example, the priority at which the host-side policy
5150 is retrieved is given by the ``priority`` property.
5151 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to
5152 refer to /etc/crypto-policies/back-ends/gnutls.config.
5154 .. parsed-literal::
5156 # |qemu_system| \\
5157 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\
5158 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0
5160 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]``
5161 Interval t can't be 0, this filter batches the packet delivery:
5162 all packets arriving in a given interval on netdev netdevid are
5163 delayed until the end of the interval. Interval is in
5164 microseconds. ``status`` is optional that indicate whether the
5165 netfilter is on (enabled) or off (disabled), the default status
5166 for netfilter will be 'on'.
5168 queue all\|rx\|tx is an option that can be applied to any
5169 netfilter.
5171 ``all``: the filter is attached both to the receive and the
5172 transmit queue of the netdev (default).
5174 ``rx``: the filter is attached to the receive queue of the
5175 netdev, where it will receive packets sent to the netdev.
5177 ``tx``: the filter is attached to the transmit queue of the
5178 netdev, where it will receive packets sent by the netdev.
5180 position head\|tail\|id=<id> is an option to specify where the
5181 filter should be inserted in the filter list. It can be applied
5182 to any netfilter.
5184 ``head``: the filter is inserted at the head of the filter list,
5185 before any existing filters.
5187 ``tail``: the filter is inserted at the tail of the filter list,
5188 behind any existing filters (default).
5190 ``id=<id>``: the filter is inserted before or behind the filter
5191 specified by <id>, see the insert option below.
5193 insert behind\|before is an option to specify where to insert
5194 the new filter relative to the one specified with
5195 position=id=<id>. It can be applied to any netfilter.
5197 ``before``: insert before the specified filter.
5199 ``behind``: insert behind the specified filter (default).
5201 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
5202 filter-mirror on netdev netdevid,mirror net packet to
5203 chardevchardevid, if it has the vnet\_hdr\_support flag,
5204 filter-mirror will mirror packet with vnet\_hdr\_len.
5206 ``-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]``
5207 filter-redirector on netdev netdevid,redirect filter's net
5208 packet to chardev chardevid,and redirect indev's packet to
5209 filter.if it has the vnet\_hdr\_support flag, filter-redirector
5210 will redirect packet with vnet\_hdr\_len. Create a
5211 filter-redirector we need to differ outdev id from indev id, id
5212 can not be the same. we can just use indev or outdev, but at
5213 least one of indev or outdev need to be specified.
5215 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
5216 Filter-rewriter is a part of COLO project.It will rewrite tcp
5217 packet to secondary from primary to keep secondary tcp
5218 connection,and rewrite tcp packet to primary from secondary make
5219 tcp packet can be handled by client.if it has the
5220 vnet\_hdr\_support flag, we can parse packet with vnet header.
5222 usage: colo secondary: -object
5223 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object
5224 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object
5225 filter-rewriter,id=rew0,netdev=hn0,queue=all
5227 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]``
5228 Dump the network traffic on netdev dev to the file specified by
5229 filename. At most len bytes (64k by default) per packet are
5230 stored. The file format is libpcap, so it can be analyzed with
5231 tools such as tcpdump or Wireshark.
5233 ``-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}]``
5234 Colo-compare gets packet from primary\_in chardevid and
5235 secondary\_in, then compare whether the payload of primary packet
5236 and secondary packet are the same. If same, it will output
5237 primary packet to out\_dev, else it will notify COLO-framework to do
5238 checkpoint and send primary packet to out\_dev. In order to
5239 improve efficiency, we need to put the task of comparison in
5240 another iothread. If it has the vnet\_hdr\_support flag,
5241 colo compare will send/recv packet with vnet\_hdr\_len.
5242 The compare\_timeout=@var{ms} determines the maximum time of the
5243 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms}
5244 is to set the period of scanning expired primary node network packets.
5245 The max\_queue\_size=@var{size} is to set the max compare queue
5246 size depend on user environment.
5247 If user want to use Xen COLO, need to add the notify\_dev to
5248 notify Xen colo-frame to do checkpoint.
5250 COLO-compare must be used with the help of filter-mirror,
5251 filter-redirector and filter-rewriter.
5255 KVM COLO
5257 primary:
5258 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
5259 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
5260 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
5261 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
5262 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
5263 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
5264 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
5265 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
5266 -object iothread,id=iothread1
5267 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
5268 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
5269 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
5270 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1
5272 secondary:
5273 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
5274 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
5275 -chardev socket,id=red0,host=3.3.3.3,port=9003
5276 -chardev socket,id=red1,host=3.3.3.3,port=9004
5277 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
5278 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
5281 Xen COLO
5283 primary:
5284 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
5285 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
5286 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
5287 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
5288 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
5289 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
5290 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
5291 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
5292 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off
5293 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
5294 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
5295 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
5296 -object iothread,id=iothread1
5297 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1
5299 secondary:
5300 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
5301 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
5302 -chardev socket,id=red0,host=3.3.3.3,port=9003
5303 -chardev socket,id=red1,host=3.3.3.3,port=9004
5304 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
5305 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
5307 If you want to know the detail of above command line, you can
5308 read the colo-compare git log.
5310 ``-object cryptodev-backend-builtin,id=id[,queues=queues]``
5311 Creates a cryptodev backend which executes crypto operations from
5312 the QEMU cipher APIs. The id parameter is a unique ID that will
5313 be used to reference this cryptodev backend from the
5314 ``virtio-crypto`` device. The queues parameter is optional,
5315 which specify the queue number of cryptodev backend, the default
5316 of queues is 1.
5318 .. parsed-literal::
5320 # |qemu_system| \\
5321 [...] \\
5322 -object cryptodev-backend-builtin,id=cryptodev0 \\
5323 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
5324 [...]
5326 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]``
5327 Creates a vhost-user cryptodev backend, backed by a chardev
5328 chardevid. The id parameter is a unique ID that will be used to
5329 reference this cryptodev backend from the ``virtio-crypto``
5330 device. The chardev should be a unix domain socket backed one.
5331 The vhost-user uses a specifically defined protocol to pass
5332 vhost ioctl replacement messages to an application on the other
5333 end of the socket. The queues parameter is optional, which
5334 specify the queue number of cryptodev backend for multiqueue
5335 vhost-user, the default of queues is 1.
5337 .. parsed-literal::
5339 # |qemu_system| \\
5340 [...] \\
5341 -chardev socket,id=chardev0,path=/path/to/socket \\
5342 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\
5343 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
5344 [...]
5346 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]``
5348 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]``
5349 Defines a secret to store a password, encryption key, or some
5350 other sensitive data. The sensitive data can either be passed
5351 directly via the data parameter, or indirectly via the file
5352 parameter. Using the data parameter is insecure unless the
5353 sensitive data is encrypted.
5355 The sensitive data can be provided in raw format (the default),
5356 or base64. When encoded as JSON, the raw format only supports
5357 valid UTF-8 characters, so base64 is recommended for sending
5358 binary data. QEMU will convert from which ever format is
5359 provided to the format it needs internally. eg, an RBD password
5360 can be provided in raw format, even though it will be base64
5361 encoded when passed onto the RBD sever.
5363 For added protection, it is possible to encrypt the data
5364 associated with a secret using the AES-256-CBC cipher. Use of
5365 encryption is indicated by providing the keyid and iv
5366 parameters. The keyid parameter provides the ID of a previously
5367 defined secret that contains the AES-256 decryption key. This
5368 key should be 32-bytes long and be base64 encoded. The iv
5369 parameter provides the random initialization vector used for
5370 encryption of this particular secret and should be a base64
5371 encrypted string of the 16-byte IV.
5373 The simplest (insecure) usage is to provide the secret inline
5375 .. parsed-literal::
5377 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw
5379 The simplest secure usage is to provide the secret via a file
5381 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object
5382 secret,id=sec0,file=mypasswd.txt,format=raw
5384 For greater security, AES-256-CBC should be used. To illustrate
5385 usage, consider the openssl command line tool which can encrypt
5386 the data. Note that when encrypting, the plaintext must be
5387 padded to the cipher block size (32 bytes) using the standard
5388 PKCS#5/6 compatible padding algorithm.
5390 First a master key needs to be created in base64 encoding:
5394 # openssl rand -base64 32 > key.b64
5395 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"')
5397 Each secret to be encrypted needs to have a random
5398 initialization vector generated. These do not need to be kept
5399 secret
5403 # openssl rand -base64 16 > iv.b64
5404 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"')
5406 The secret to be defined can now be encrypted, in this case
5407 we're telling openssl to base64 encode the result, but it could
5408 be left as raw bytes if desired.
5412 # SECRET=$(printf "letmein" |
5413 openssl enc -aes-256-cbc -a -K $KEY -iv $IV)
5415 When launching QEMU, create a master secret pointing to
5416 ``key.b64`` and specify that to be used to decrypt the user
5417 password. Pass the contents of ``iv.b64`` to the second secret
5419 .. parsed-literal::
5421 # |qemu_system| \\
5422 -object secret,id=secmaster0,format=base64,file=key.b64 \\
5423 -object secret,id=sec0,keyid=secmaster0,format=base64,\\
5424 data=$SECRET,iv=$(<iv.b64)
5426 ``-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,kernel-hashes=on|off]``
5427 Create a Secure Encrypted Virtualization (SEV) guest object,
5428 which can be used to provide the guest memory encryption support
5429 on AMD processors.
5431 When memory encryption is enabled, one of the physical address
5432 bit (aka the C-bit) is utilized to mark if a memory page is
5433 protected. The ``cbitpos`` is used to provide the C-bit
5434 position. The C-bit position is Host family dependent hence user
5435 must provide this value. On EPYC, the value should be 47.
5437 When memory encryption is enabled, we loose certain bits in
5438 physical address space. The ``reduced-phys-bits`` is used to
5439 provide the number of bits we loose in physical address space.
5440 Similar to C-bit, the value is Host family dependent. On EPYC,
5441 the value should be 5.
5443 The ``sev-device`` provides the device file to use for
5444 communicating with the SEV firmware running inside AMD Secure
5445 Processor. The default device is '/dev/sev'. If hardware
5446 supports memory encryption then /dev/sev devices are created by
5447 CCP driver.
5449 The ``policy`` provides the guest policy to be enforced by the
5450 SEV firmware and restrict what configuration and operational
5451 commands can be performed on this guest by the hypervisor. The
5452 policy should be provided by the guest owner and is bound to the
5453 guest and cannot be changed throughout the lifetime of the
5454 guest. The default is 0.
5456 If guest ``policy`` allows sharing the key with another SEV
5457 guest then ``handle`` can be use to provide handle of the guest
5458 from which to share the key.
5460 The ``dh-cert-file`` and ``session-file`` provides the guest
5461 owner's Public Diffie-Hillman key defined in SEV spec. The PDH
5462 and session parameters are used for establishing a cryptographic
5463 session with the guest owner to negotiate keys used for
5464 attestation. The file must be encoded in base64.
5466 The ``kernel-hashes`` adds the hashes of given kernel/initrd/
5467 cmdline to a designated guest firmware page for measured Linux
5468 boot with -kernel. The default is off. (Since 6.2)
5470 e.g to launch a SEV guest
5472 .. parsed-literal::
5474 # |qemu_system_x86| \\
5475 ...... \\
5476 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \\
5477 -machine ...,memory-encryption=sev0 \\
5478 .....
5480 ``-object authz-simple,id=id,identity=string``
5481 Create an authorization object that will control access to
5482 network services.
5484 The ``identity`` parameter is identifies the user and its format
5485 depends on the network service that authorization object is
5486 associated with. For authorizing based on TLS x509 certificates,
5487 the identity must be the x509 distinguished name. Note that care
5488 must be taken to escape any commas in the distinguished name.
5490 An example authorization object to validate a x509 distinguished
5491 name would look like:
5493 .. parsed-literal::
5495 # |qemu_system| \\
5496 ... \\
5497 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\
5500 Note the use of quotes due to the x509 distinguished name
5501 containing whitespace, and escaping of ','.
5503 ``-object authz-listfile,id=id,filename=path,refresh=on|off``
5504 Create an authorization object that will control access to
5505 network services.
5507 The ``filename`` parameter is the fully qualified path to a file
5508 containing the access control list rules in JSON format.
5510 An example set of rules that match against SASL usernames might
5511 look like:
5516 "rules": [
5517 { "match": "fred", "policy": "allow", "format": "exact" },
5518 { "match": "bob", "policy": "allow", "format": "exact" },
5519 { "match": "danb", "policy": "deny", "format": "glob" },
5520 { "match": "dan*", "policy": "allow", "format": "exact" },
5522 "policy": "deny"
5525 When checking access the object will iterate over all the rules
5526 and the first rule to match will have its ``policy`` value
5527 returned as the result. If no rules match, then the default
5528 ``policy`` value is returned.
5530 The rules can either be an exact string match, or they can use
5531 the simple UNIX glob pattern matching to allow wildcards to be
5532 used.
5534 If ``refresh`` is set to true the file will be monitored and
5535 automatically reloaded whenever its content changes.
5537 As with the ``authz-simple`` object, the format of the identity
5538 strings being matched depends on the network service, but is
5539 usually a TLS x509 distinguished name, or a SASL username.
5541 An example authorization object to validate a SASL username
5542 would look like:
5544 .. parsed-literal::
5546 # |qemu_system| \\
5547 ... \\
5548 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\
5551 ``-object authz-pam,id=id,service=string``
5552 Create an authorization object that will control access to
5553 network services.
5555 The ``service`` parameter provides the name of a PAM service to
5556 use for authorization. It requires that a file
5557 ``/etc/pam.d/service`` exist to provide the configuration for
5558 the ``account`` subsystem.
5560 An example authorization object to validate a TLS x509
5561 distinguished name would look like:
5563 .. parsed-literal::
5565 # |qemu_system| \\
5566 ... \\
5567 -object authz-pam,id=auth0,service=qemu-vnc \\
5570 There would then be a corresponding config file for PAM at
5571 ``/etc/pam.d/qemu-vnc`` that contains:
5575 account requisite pam_listfile.so item=user sense=allow \
5576 file=/etc/qemu/vnc.allow
5578 Finally the ``/etc/qemu/vnc.allow`` file would contain the list
5579 of x509 distinguished names that are permitted access
5583 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB
5585 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch``
5586 Creates a dedicated event loop thread that devices can be
5587 assigned to. This is known as an IOThread. By default device
5588 emulation happens in vCPU threads or the main event loop thread.
5589 This can become a scalability bottleneck. IOThreads allow device
5590 emulation and I/O to run on other host CPUs.
5592 The ``id`` parameter is a unique ID that will be used to
5593 reference this IOThread from ``-device ...,iothread=id``.
5594 Multiple devices can be assigned to an IOThread. Note that not
5595 all devices support an ``iothread`` parameter.
5597 The ``query-iothreads`` QMP command lists IOThreads and reports
5598 their thread IDs so that the user can configure host CPU
5599 pinning/affinity.
5601 IOThreads use an adaptive polling algorithm to reduce event loop
5602 latency. Instead of entering a blocking system call to monitor
5603 file descriptors and then pay the cost of being woken up when an
5604 event occurs, the polling algorithm spins waiting for events for
5605 a short time. The algorithm's default parameters are suitable
5606 for many cases but can be adjusted based on knowledge of the
5607 workload and/or host device latency.
5609 The ``poll-max-ns`` parameter is the maximum number of
5610 nanoseconds to busy wait for events. Polling can be disabled by
5611 setting this value to 0.
5613 The ``poll-grow`` parameter is the multiplier used to increase
5614 the polling time when the algorithm detects it is missing events
5615 due to not polling long enough.
5617 The ``poll-shrink`` parameter is the divisor used to decrease
5618 the polling time when the algorithm detects it is spending too
5619 long polling without encountering events.
5621 The ``aio-max-batch`` parameter is the maximum number of requests
5622 in a batch for the AIO engine, 0 means that the engine will use
5623 its default.
5625 The IOThread parameters can be modified at run-time using the
5626 ``qom-set`` command (where ``iothread1`` is the IOThread's
5627 ``id``):
5631 (qemu) qom-set /objects/iothread1 poll-max-ns 100000
5632 ERST
5635 HXCOMM This is the last statement. Insert new options before this line!
5637 #undef DEF
5638 #undef DEFHEADING
5639 #undef ARCHHEADING