1 HXCOMM Use DEFHEADING() to define headings in both help text and rST.
2 HXCOMM Text between SRST and ERST is copied to the rST version and
3 HXCOMM discarded from C version.
4 HXCOMM DEF(option, HAS_ARG/0, opt_enum, opt_help, arch_mask) is used to
5 HXCOMM construct option structures, enums and help message for specified
7 HXCOMM HXCOMM can be used for comments, discarded from both rST and C.
9 DEFHEADING(Standard options:)
11 DEF("help", 0, QEMU_OPTION_h,
12 "-h or -help display this help and exit\n", QEMU_ARCH_ALL)
18 DEF("version", 0, QEMU_OPTION_version,
19 "-version display version information and exit\n", QEMU_ARCH_ALL)
22 Display version information and exit
25 DEF("machine", HAS_ARG, QEMU_OPTION_machine, \
26 "-machine [type=]name[,prop[=value][,...]]\n"
27 " selects emulated machine ('-machine help' for list)\n"
28 " property accel=accel1[:accel2[:...]] selects accelerator\n"
29 " supported accelerators are kvm, xen, hax, hvf, nvmm, whpx or tcg (default: tcg)\n"
30 " vmport=on|off|auto controls emulation of vmport (default: auto)\n"
31 " dump-guest-core=on|off include guest memory in a core dump (default=on)\n"
32 " mem-merge=on|off controls memory merge support (default: on)\n"
33 " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n"
34 " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n"
35 " suppress-vmdesc=on|off disables self-describing migration (default=off)\n"
36 " nvdimm=on|off controls NVDIMM support (default=off)\n"
37 " memory-encryption=@var{} memory encryption object to use (default=none)\n"
38 " hmat=on|off controls ACPI HMAT support (default=off)\n"
39 " memory-backend='backend-id' specifies explicitly provided backend for main RAM (default=none)\n"
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",
43 ``-machine [type=]name[,prop=value[,...]]``
44 Select the emulated machine by name. Use ``-machine help`` to list
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
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.
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
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
93 Enables or disables NVDIMM support. The default is off.
95 ``memory-encryption=``
96 Memory encryption object to use. The default is none.
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.
109 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on
110 -machine memory-backend=pc.ram
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.
125 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off
126 -machine memory-backend=pc.ram
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
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.
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
162 DEF("M", HAS_ARG, QEMU_OPTION_M,
163 " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n",
167 ``sgx-epc.0.memdev=@var{memid},sgx-epc.0.node=@var{numaid}``
168 Define an SGX EPC section.
171 DEF("cpu", HAS_ARG, QEMU_OPTION_cpu,
172 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL)
175 Select CPU model (``-cpu help`` for list and additional feature
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)
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
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
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.
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.
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.
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.
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",
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:
347 Note: The cluster topology will only be generated in ACPI and exposed
348 to guest if it's explicitly specified in -smp.
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",
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
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.
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 Legacy '\ ``mem``\ ' assigns a given RAM amount to a node (not supported
409 for 5.1 and newer machine types). '\ ``memdev``\ ' assigns RAM from
410 a given memory backend device to a node. If '\ ``mem``\ ' and
411 '\ ``memdev``\ ' are omitted in all nodes, RAM is split equally between them.
414 '\ ``mem``\ ' and '\ ``memdev``\ ' are mutually exclusive.
415 Furthermore, if one node uses '\ ``memdev``\ ', all of them have to
418 '\ ``initiator``\ ' is an additional option that points to an
419 initiator NUMA node that has best performance (the lowest latency or
420 largest bandwidth) to this NUMA node. Note that this option can be
421 set only when the machine property 'hmat' is set to 'on'.
423 Following example creates a machine with 2 NUMA nodes, node 0 has
424 CPU. node 1 has only memory, and its initiator is node 0. Note that
425 because node 0 has CPU, by default the initiator of node 0 is itself
431 -m 2G,slots=2,maxmem=4G \
432 -object memory-backend-ram,size=1G,id=m0 \
433 -object memory-backend-ram,size=1G,id=m1 \
434 -numa node,nodeid=0,memdev=m0 \
435 -numa node,nodeid=1,memdev=m1,initiator=0 \
436 -smp 2,sockets=2,maxcpus=2 \
437 -numa cpu,node-id=0,socket-id=0 \
438 -numa cpu,node-id=0,socket-id=1
440 source and destination are NUMA node IDs. distance is the NUMA
441 distance from source to destination. The distance from a node to
442 itself is always 10. If any pair of nodes is given a distance, then
443 all pairs must be given distances. Although, when distances are only
444 given in one direction for each pair of nodes, then the distances in
445 the opposite directions are assumed to be the same. If, however, an
446 asymmetrical pair of distances is given for even one node pair, then
447 all node pairs must be provided distance values for both directions,
448 even when they are symmetrical. When a node is unreachable from
449 another node, set the pair's distance to 255.
451 Note that the -``numa`` option doesn't allocate any of the specified
452 resources, it just assigns existing resources to NUMA nodes. This
453 means that one still has to use the ``-m``, ``-smp`` options to
454 allocate RAM and VCPUs respectively.
456 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth
457 Information between initiator and target NUMA nodes in ACPI
458 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can
459 create memory requests, usually it has one or more processors.
460 Target NUMA node contains addressable memory.
462 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is
463 the memory hierarchy of the target NUMA node: if hierarchy is
464 'memory', the structure represents the memory performance; if
465 hierarchy is 'first-level\|second-level\|third-level', this
466 structure represents aggregated performance of memory side caches
467 for each domain. type of 'data-type' is type of data represented by
468 this structure instance: if 'hierarchy' is 'memory', 'data-type' is
469 'access\|read\|write' latency or 'access\|read\|write' bandwidth of
470 the target memory; if 'hierarchy' is
471 'first-level\|second-level\|third-level', 'data-type' is
472 'access\|read\|write' hit latency or 'access\|read\|write' hit
473 bandwidth of the target memory side cache.
475 lat is latency value in nanoseconds. bw is bandwidth value, the
476 possible value and units are NUM[M\|G\|T], mean that the bandwidth
477 value are NUM byte per second (or MB/s, GB/s or TB/s depending on
478 used suffix). Note that if latency or bandwidth value is 0, means
479 the corresponding latency or bandwidth information is not provided.
481 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory
482 belongs. size is the size of memory side cache in bytes. level is
483 the cache level described in this structure, note that the cache
484 level 0 should not be used with '\ ``hmat-cache``\ ' option.
485 associativity is the cache associativity, the possible value is
486 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy
487 is the write policy. line is the cache Line size in bytes.
489 For example, the following options describe 2 NUMA nodes. Node 0 has
490 2 cpus and a ram, node 1 has only a ram. The processors in node 0
491 access memory in node 0 with access-latency 5 nanoseconds,
492 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access
493 memory in NUMA node 1 with access-latency 10 nanoseconds,
494 access-bandwidth is 100 MB/s. And for memory side cache information,
495 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB,
496 policy is write-back, the cache Line size is 8 bytes:
502 -object memory-backend-ram,size=1G,id=m0 \
503 -object memory-backend-ram,size=1G,id=m1 \
504 -smp 2,sockets=2,maxcpus=2 \
505 -numa node,nodeid=0,memdev=m0 \
506 -numa node,nodeid=1,memdev=m1,initiator=0 \
507 -numa cpu,node-id=0,socket-id=0 \
508 -numa cpu,node-id=0,socket-id=1 \
509 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \
510 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \
511 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \
512 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \
513 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \
514 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8
517 DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd,
518 "-add-fd fd=fd,set=set[,opaque=opaque]\n"
519 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL)
521 ``-add-fd fd=fd,set=set[,opaque=opaque]``
522 Add a file descriptor to an fd set. Valid options are:
525 This option defines the file descriptor of which a duplicate is
526 added to fd set. The file descriptor cannot be stdin, stdout, or
530 This option defines the ID of the fd set to add the file
534 This option defines a free-form string that can be used to
537 You can open an image using pre-opened file descriptors from an fd
543 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
544 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
545 -drive file=/dev/fdset/2,index=0,media=disk
548 DEF("set", HAS_ARG, QEMU_OPTION_set,
549 "-set group.id.arg=value\n"
550 " set <arg> parameter for item <id> of type <group>\n"
551 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL)
553 ``-set group.id.arg=value``
554 Set parameter arg for item id of type group
557 DEF("global", HAS_ARG, QEMU_OPTION_global,
558 "-global driver.property=value\n"
559 "-global driver=driver,property=property,value=value\n"
560 " set a global default for a driver property\n",
563 ``-global driver.prop=value``
565 ``-global driver=driver,property=property,value=value``
566 Set default value of driver's property prop to value, e.g.:
570 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img
572 In particular, you can use this to set driver properties for devices
573 which are created automatically by the machine model. To create a
574 device which is not created automatically and set properties on it,
577 -global driver.prop=value is shorthand for -global
578 driver=driver,property=prop,value=value. The longhand syntax works
579 even when driver contains a dot.
582 DEF("boot", HAS_ARG, QEMU_OPTION_boot,
583 "-boot [order=drives][,once=drives][,menu=on|off]\n"
584 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n"
585 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n"
586 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n"
587 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n"
588 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n",
591 ``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]``
592 Specify boot order drives as a string of drive letters. Valid drive
593 letters depend on the target architecture. The x86 PC uses: a, b
594 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p
595 (Etherboot from network adapter 1-4), hard disk boot is the default.
596 To apply a particular boot order only on the first startup, specify
597 it via ``once``. Note that the ``order`` or ``once`` parameter
598 should not be used together with the ``bootindex`` property of
599 devices, since the firmware implementations normally do not support
600 both at the same time.
602 Interactive boot menus/prompts can be enabled via ``menu=on`` as far
603 as firmware/BIOS supports them. The default is non-interactive boot.
605 A splash picture could be passed to bios, enabling user to show it
606 as logo, when option splash=sp\_name is given and menu=on, If
607 firmware/BIOS supports them. Currently Seabios for X86 system
608 support it. limitation: The splash file could be a jpeg file or a
609 BMP file in 24 BPP format(true color). The resolution should be
610 supported by the SVGA mode, so the recommended is 320x240, 640x480,
613 A timeout could be passed to bios, guest will pause for rb\_timeout
614 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will
615 not reboot, qemu passes '-1' to bios by default. Currently Seabios
616 for X86 system support it.
618 Do strict boot via ``strict=on`` as far as firmware/BIOS supports
619 it. This only effects when boot priority is changed by bootindex
620 options. The default is non-strict boot.
624 # try to boot from network first, then from hard disk
625 |qemu_system_x86| -boot order=nc
626 # boot from CD-ROM first, switch back to default order after reboot
627 |qemu_system_x86| -boot once=d
628 # boot with a splash picture for 5 seconds.
629 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000
631 Note: The legacy format '-boot drives' is still supported but its
632 use is discouraged as it may be removed from future versions.
635 DEF("m", HAS_ARG, QEMU_OPTION_m,
636 "-m [size=]megs[,slots=n,maxmem=size]\n"
637 " configure guest RAM\n"
638 " size: initial amount of guest memory\n"
639 " slots: number of hotplug slots (default: none)\n"
640 " maxmem: maximum amount of guest memory (default: none)\n"
641 "NOTE: Some architectures might enforce a specific granularity\n",
644 ``-m [size=]megs[,slots=n,maxmem=size]``
645 Sets guest startup RAM size to megs megabytes. Default is 128 MiB.
646 Optionally, a suffix of "M" or "G" can be used to signify a value in
647 megabytes or gigabytes respectively. Optional pair slots, maxmem
648 could be used to set amount of hotpluggable memory slots and maximum
649 amount of memory. Note that maxmem must be aligned to the page size.
651 For example, the following command-line sets the guest startup RAM
652 size to 1GB, creates 3 slots to hotplug additional memory and sets
653 the maximum memory the guest can reach to 4GB:
657 |qemu_system| -m 1G,slots=3,maxmem=4G
659 If slots and maxmem are not specified, memory hotplug won't be
660 enabled and the guest startup RAM will never increase.
663 DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath,
664 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL)
667 Allocate guest RAM from a temporarily created file in path.
670 DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc,
671 "-mem-prealloc preallocate guest memory (use with -mem-path)\n",
675 Preallocate memory when using -mem-path.
678 DEF("k", HAS_ARG, QEMU_OPTION_k,
679 "-k language use keyboard layout (for example 'fr' for French)\n",
683 Use keyboard layout language (for example ``fr`` for French). This
684 option is only needed where it is not easy to get raw PC keycodes
685 (e.g. on Macs, with some X11 servers or with a VNC or curses
686 display). You don't normally need to use it on PC/Linux or
689 The available layouts are:
693 ar de-ch es fo fr-ca hu ja mk no pt-br sv
694 da en-gb et fr fr-ch is lt nl pl ru th
695 de en-us fi fr-be hr it lv nl-be pt sl tr
697 The default is ``en-us``.
701 HXCOMM Deprecated by -audiodev
702 DEF("audio-help", 0, QEMU_OPTION_audio_help,
703 "-audio-help show -audiodev equivalent of the currently specified audio settings\n",
707 Will show the -audiodev equivalent of the currently specified
708 (deprecated) environment variables.
711 DEF("audio", HAS_ARG, QEMU_OPTION_audio,
712 "-audio [driver=]driver,model=value[,prop[=value][,...]]\n"
713 " specifies the audio backend and device to use;\n"
714 " apart from 'model', options are the same as for -audiodev.\n"
715 " use '-audio model=help' to show possible devices.\n",
718 ``-audio [driver=]driver,model=value[,prop[=value][,...]]``
719 This option is a shortcut for configuring both the guest audio
720 hardware and the host audio backend in one go.
721 The driver option is the same as with the corresponding ``-audiodev`` option below.
722 The guest hardware model can be set with ``model=modelname``.
724 Use ``driver=help`` to list the available drivers,
725 and ``model=help`` to list the available device types.
727 The following two example do exactly the same, to show how ``-audio``
728 can be used to shorten the command line length:
732 |qemu_system| -audiodev pa,id=pa -device sb16,audiodev=pa
733 |qemu_system| -audio pa,model=sb16
736 DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev,
737 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n"
738 " specifies the audio backend to use\n"
739 " Use ``-audiodev help`` to list the available drivers\n"
740 " id= identifier of the backend\n"
741 " timer-period= timer period in microseconds\n"
742 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n"
743 " in|out.fixed-settings= use fixed settings for host audio\n"
744 " in|out.frequency= frequency to use with fixed settings\n"
745 " in|out.channels= number of channels to use with fixed settings\n"
746 " in|out.format= sample format to use with fixed settings\n"
747 " valid values: s8, s16, s32, u8, u16, u32, f32\n"
748 " in|out.voices= number of voices to use\n"
749 " in|out.buffer-length= length of buffer in microseconds\n"
750 "-audiodev none,id=id,[,prop[=value][,...]]\n"
751 " dummy driver that discards all output\n"
752 #ifdef CONFIG_AUDIO_ALSA
753 "-audiodev alsa,id=id[,prop[=value][,...]]\n"
754 " in|out.dev= name of the audio device to use\n"
755 " in|out.period-length= length of period in microseconds\n"
756 " in|out.try-poll= attempt to use poll mode\n"
757 " threshold= threshold (in microseconds) when playback starts\n"
759 #ifdef CONFIG_AUDIO_COREAUDIO
760 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n"
761 " in|out.buffer-count= number of buffers\n"
763 #ifdef CONFIG_AUDIO_DSOUND
764 "-audiodev dsound,id=id[,prop[=value][,...]]\n"
765 " latency= add extra latency to playback in microseconds\n"
767 #ifdef CONFIG_AUDIO_OSS
768 "-audiodev oss,id=id[,prop[=value][,...]]\n"
769 " in|out.dev= path of the audio device to use\n"
770 " in|out.buffer-count= number of buffers\n"
771 " in|out.try-poll= attempt to use poll mode\n"
772 " try-mmap= try using memory mapped access\n"
773 " exclusive= open device in exclusive mode\n"
774 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n"
776 #ifdef CONFIG_AUDIO_PA
777 "-audiodev pa,id=id[,prop[=value][,...]]\n"
778 " server= PulseAudio server address\n"
779 " in|out.name= source/sink device name\n"
780 " in|out.latency= desired latency in microseconds\n"
782 #ifdef CONFIG_AUDIO_SDL
783 "-audiodev sdl,id=id[,prop[=value][,...]]\n"
784 " in|out.buffer-count= number of buffers\n"
786 #ifdef CONFIG_AUDIO_SNDIO
787 "-audiodev sndio,id=id[,prop[=value][,...]]\n"
790 "-audiodev spice,id=id[,prop[=value][,...]]\n"
792 #ifdef CONFIG_DBUS_DISPLAY
793 "-audiodev dbus,id=id[,prop[=value][,...]]\n"
795 "-audiodev wav,id=id[,prop[=value][,...]]\n"
796 " path= path of wav file to record\n",
799 ``-audiodev [driver=]driver,id=id[,prop[=value][,...]]``
800 Adds a new audio backend driver identified by id. There are global
801 and driver specific properties. Some values can be set differently
802 for input and output, they're marked with ``in|out.``. You can set
803 the input's property with ``in.prop`` and the output's property with
804 ``out.prop``. For example:
808 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000
809 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified
811 NOTE: parameter validation is known to be incomplete, in many cases
812 specifying an invalid option causes QEMU to print an error message
813 and continue emulation without sound.
815 Valid global options are:
818 Identifies the audio backend.
820 ``timer-period=period``
821 Sets the timer period used by the audio subsystem in
822 microseconds. Default is 10000 (10 ms).
824 ``in|out.mixing-engine=on|off``
825 Use QEMU's mixing engine to mix all streams inside QEMU and
826 convert audio formats when not supported by the backend. When
827 off, fixed-settings must be off too. Note that disabling this
828 option means that the selected backend must support multiple
829 streams and the audio formats used by the virtual cards,
830 otherwise you'll get no sound. It's not recommended to disable
831 this option unless you want to use 5.1 or 7.1 audio, as mixing
832 engine only supports mono and stereo audio. Default is on.
834 ``in|out.fixed-settings=on|off``
835 Use fixed settings for host audio. When off, it will change
836 based on how the guest opens the sound card. In this case you
837 must not specify frequency, channels or format. Default is on.
839 ``in|out.frequency=frequency``
840 Specify the frequency to use when using fixed-settings. Default
843 ``in|out.channels=channels``
844 Specify the number of channels to use when using fixed-settings.
845 Default is 2 (stereo).
847 ``in|out.format=format``
848 Specify the sample format to use when using fixed-settings.
849 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``,
850 ``u32``, ``f32``. Default is ``s16``.
852 ``in|out.voices=voices``
853 Specify the number of voices to use. Default is 1.
855 ``in|out.buffer-length=usecs``
856 Sets the size of the buffer in microseconds.
858 ``-audiodev none,id=id[,prop[=value][,...]]``
859 Creates a dummy backend that discards all outputs. This backend has
860 no backend specific properties.
862 ``-audiodev alsa,id=id[,prop[=value][,...]]``
863 Creates backend using the ALSA. This backend is only available on
866 ALSA specific options are:
868 ``in|out.dev=device``
869 Specify the ALSA device to use for input and/or output. Default
872 ``in|out.period-length=usecs``
873 Sets the period length in microseconds.
875 ``in|out.try-poll=on|off``
876 Attempt to use poll mode with the device. Default is on.
878 ``threshold=threshold``
879 Threshold (in microseconds) when playback starts. Default is 0.
881 ``-audiodev coreaudio,id=id[,prop[=value][,...]]``
882 Creates a backend using Apple's Core Audio. This backend is only
883 available on Mac OS and only supports playback.
885 Core Audio specific options are:
887 ``in|out.buffer-count=count``
888 Sets the count of the buffers.
890 ``-audiodev dsound,id=id[,prop[=value][,...]]``
891 Creates a backend using Microsoft's DirectSound. This backend is
892 only available on Windows and only supports playback.
894 DirectSound specific options are:
897 Add extra usecs microseconds latency to playback. Default is
900 ``-audiodev oss,id=id[,prop[=value][,...]]``
901 Creates a backend using OSS. This backend is available on most
904 OSS specific options are:
906 ``in|out.dev=device``
907 Specify the file name of the OSS device to use. Default is
910 ``in|out.buffer-count=count``
911 Sets the count of the buffers.
913 ``in|out.try-poll=on|of``
914 Attempt to use poll mode with the device. Default is on.
917 Try using memory mapped device access. Default is off.
920 Open the device in exclusive mode (vmix won't work in this
921 case). Default is off.
923 ``dsp-policy=policy``
924 Sets the timing policy (between 0 and 10, where smaller number
925 means smaller latency but higher CPU usage). Use -1 to use
926 buffer sizes specified by ``buffer`` and ``buffer-count``. This
927 option is ignored if you do not have OSS 4. Default is 5.
929 ``-audiodev pa,id=id[,prop[=value][,...]]``
930 Creates a backend using PulseAudio. This backend is available on
933 PulseAudio specific options are:
936 Sets the PulseAudio server to connect to.
939 Use the specified source/sink for recording/playback.
941 ``in|out.latency=usecs``
942 Desired latency in microseconds. The PulseAudio server will try
943 to honor this value but actual latencies may be lower or higher.
945 ``-audiodev sdl,id=id[,prop[=value][,...]]``
946 Creates a backend using SDL. This backend is available on most
947 systems, but you should use your platform's native backend if
950 SDL specific options are:
952 ``in|out.buffer-count=count``
953 Sets the count of the buffers.
955 ``-audiodev sndio,id=id[,prop[=value][,...]]``
956 Creates a backend using SNDIO. This backend is available on
957 OpenBSD and most other Unix-like systems.
959 Sndio specific options are:
961 ``in|out.dev=device``
962 Specify the sndio device to use for input and/or output. Default
965 ``in|out.latency=usecs``
966 Sets the desired period length in microseconds.
968 ``-audiodev spice,id=id[,prop[=value][,...]]``
969 Creates a backend that sends audio through SPICE. This backend
970 requires ``-spice`` and automatically selected in that case, so
971 usually you can ignore this option. This backend has no backend
974 ``-audiodev wav,id=id[,prop[=value][,...]]``
975 Creates a backend that writes audio to a WAV file.
977 Backend specific options are:
980 Write recorded audio into the specified file. Default is
984 DEF("device", HAS_ARG, QEMU_OPTION_device,
985 "-device driver[,prop[=value][,...]]\n"
986 " add device (based on driver)\n"
987 " prop=value,... sets driver properties\n"
988 " use '-device help' to print all possible drivers\n"
989 " use '-device driver,help' to print all possible properties\n",
992 ``-device driver[,prop[=value][,...]]``
993 Add device driver. prop=value sets driver properties. Valid
994 properties depend on the driver. To get help on possible drivers and
995 properties, use ``-device help`` and ``-device driver,help``.
999 ``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]``
1000 Add an IPMI BMC. This is a simulation of a hardware management
1001 interface processor that normally sits on a system. It provides a
1002 watchdog and the ability to reset and power control the system. You
1003 need to connect this to an IPMI interface to make it useful
1005 The IPMI slave address to use for the BMC. The default is 0x20. This
1006 address is the BMC's address on the I2C network of management
1007 controllers. If you don't know what this means, it is safe to ignore
1011 The BMC id for interfaces to use this device.
1014 Define slave address to use for the BMC. The default is 0x20.
1017 file containing raw Sensor Data Records (SDR) data. The default
1021 size of a Field Replaceable Unit (FRU) area. The default is
1024 ``frudatafile=file``
1025 file containing raw Field Replaceable Unit (FRU) inventory data.
1026 The default is none.
1029 value for the GUID for the BMC, in standard UUID format. If this
1030 is set, get "Get GUID" command to the BMC will return it.
1031 Otherwise "Get GUID" will return an error.
1033 ``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]``
1034 Add a connection to an external IPMI BMC simulator. Instead of
1035 locally emulating the BMC like the above item, instead connect to an
1036 external entity that provides the IPMI services.
1038 A connection is made to an external BMC simulator. If you do this,
1039 it is strongly recommended that you use the "reconnect=" chardev
1040 option to reconnect to the simulator if the connection is lost. Note
1041 that if this is not used carefully, it can be a security issue, as
1042 the interface has the ability to send resets, NMIs, and power off
1043 the VM. It's best if QEMU makes a connection to an external
1044 simulator running on a secure port on localhost, so neither the
1045 simulator nor QEMU is exposed to any outside network.
1047 See the "lanserv/README.vm" file in the OpenIPMI library for more
1048 details on the external interface.
1050 ``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]``
1051 Add a KCS IPMI interface on the ISA bus. This also adds a
1052 corresponding ACPI and SMBIOS entries, if appropriate.
1055 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern
1059 Define the I/O address of the interface. The default is 0xca0
1063 Define the interrupt to use. The default is 5. To disable
1064 interrupts, set this to 0.
1066 ``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]``
1067 Like the KCS interface, but defines a BT interface. The default port
1068 is 0xe4 and the default interrupt is 5.
1070 ``-device pci-ipmi-kcs,bmc=id``
1071 Add a KCS IPMI interface on the PCI bus.
1074 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above.
1076 ``-device pci-ipmi-bt,bmc=id``
1077 Like the KCS interface, but defines a BT interface on the PCI bus.
1079 ``-device intel-iommu[,option=...]``
1080 This is only supported by ``-machine q35``, which will enable Intel VT-d
1081 emulation within the guest. It supports below options:
1083 ``intremap=on|off`` (default: auto)
1084 This enables interrupt remapping feature. It's required to enable
1085 complete x2apic. Currently it only supports kvm kernel-irqchip modes
1086 ``off`` or ``split``, while full kernel-irqchip is not yet supported.
1087 The default value is "auto", which will be decided by the mode of
1090 ``caching-mode=on|off`` (default: off)
1091 This enables caching mode for the VT-d emulated device. When
1092 caching-mode is enabled, each guest DMA buffer mapping will generate an
1093 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in
1094 a synchronous way. It is required for ``-device vfio-pci`` to work
1095 with the VT-d device, because host assigned devices requires to setup
1096 the DMA mapping on the host before guest DMA starts.
1098 ``device-iotlb=on|off`` (default: off)
1099 This enables device-iotlb capability for the emulated VT-d device. So
1100 far virtio/vhost should be the only real user for this parameter,
1101 paired with ats=on configured for the device.
1103 ``aw-bits=39|48`` (default: 39)
1104 This decides the address width of IOVA address space. The address
1105 space has 39 bits width for 3-level IOMMU page tables, and 48 bits for
1106 4-level IOMMU page tables.
1108 Please also refer to the wiki page for general scenarios of VT-d
1109 emulation in QEMU: https://wiki.qemu.org/Features/VT-d.
1113 DEF("name", HAS_ARG, QEMU_OPTION_name,
1114 "-name string1[,process=string2][,debug-threads=on|off]\n"
1115 " set the name of the guest\n"
1116 " string1 sets the window title and string2 the process name\n"
1117 " When debug-threads is enabled, individual threads are given a separate name\n"
1118 " NOTE: The thread names are for debugging and not a stable API.\n",
1122 Sets the name of the guest. This name will be displayed in the SDL
1123 window caption. The name will also be used for the VNC server. Also
1124 optionally set the top visible process name in Linux. Naming of
1125 individual threads can also be enabled on Linux to aid debugging.
1128 DEF("uuid", HAS_ARG, QEMU_OPTION_uuid,
1129 "-uuid %08x-%04x-%04x-%04x-%012x\n"
1130 " specify machine UUID\n", QEMU_ARCH_ALL)
1138 DEFHEADING(Block device options:)
1141 The QEMU block device handling options have a long history and
1142 have gone through several iterations as the feature set and complexity
1143 of the block layer have grown. Many online guides to QEMU often
1144 reference older and deprecated options, which can lead to confusion.
1146 The recommended modern way to describe disks is to use a combination of
1147 ``-device`` to specify the hardware device and ``-blockdev`` to
1148 describe the backend. The device defines what the guest sees and the
1149 backend describes how QEMU handles the data.
1153 DEF("fda", HAS_ARG, QEMU_OPTION_fda,
1154 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
1155 DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
1160 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in
1161 the System Emulation Users Guide).
1164 DEF("hda", HAS_ARG, QEMU_OPTION_hda,
1165 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL)
1166 DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL)
1167 DEF("hdc", HAS_ARG, QEMU_OPTION_hdc,
1168 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL)
1169 DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL)
1178 Use file as hard disk 0, 1, 2 or 3 image (see the :ref:`disk images`
1179 chapter in the System Emulation Users Guide).
1182 DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom,
1183 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n",
1187 Use file as CD-ROM image (you cannot use ``-hdc`` and ``-cdrom`` at
1188 the same time). You can use the host CD-ROM by using ``/dev/cdrom``
1192 DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev,
1193 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n"
1194 " [,cache.direct=on|off][,cache.no-flush=on|off]\n"
1195 " [,read-only=on|off][,auto-read-only=on|off]\n"
1196 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n"
1197 " [,driver specific parameters...]\n"
1198 " configure a block backend\n", QEMU_ARCH_ALL)
1200 ``-blockdev option[,option[,option[,...]]]``
1201 Define a new block driver node. Some of the options apply to all
1202 block drivers, other options are only accepted for a specific block
1203 driver. See below for a list of generic options and options for the
1204 most common block drivers.
1206 Options that expect a reference to another node (e.g. ``file``) can
1207 be given in two ways. Either you specify the node name of an already
1208 existing node (file=node-name), or you define a new node inline,
1209 adding options for the referenced node after a dot
1210 (file.filename=path,file.aio=native).
1212 A block driver node created with ``-blockdev`` can be used for a
1213 guest device by specifying its node name for the ``drive`` property
1214 in a ``-device`` argument that defines a block device.
1216 ``Valid options for any block driver node:``
1218 Specifies the block driver to use for the given node.
1221 This defines the name of the block driver node by which it
1222 will be referenced later. The name must be unique, i.e. it
1223 must not match the name of a different block driver node, or
1224 (if you use ``-drive`` as well) the ID of a drive.
1226 If no node name is specified, it is automatically generated.
1227 The generated node name is not intended to be predictable
1228 and changes between QEMU invocations. For the top level, an
1229 explicit node name must be specified.
1232 Open the node read-only. Guest write attempts will fail.
1234 Note that some block drivers support only read-only access,
1235 either generally or in certain configurations. In this case,
1236 the default value ``read-only=off`` does not work and the
1237 option must be specified explicitly.
1240 If ``auto-read-only=on`` is set, QEMU may fall back to
1241 read-only usage even when ``read-only=off`` is requested, or
1242 even switch between modes as needed, e.g. depending on
1243 whether the image file is writable or whether a writing user
1244 is attached to the node.
1247 Override the image locking system of QEMU by forcing the
1248 node to utilize weaker shared access for permissions where
1249 it would normally request exclusive access. When there is
1250 the potential for multiple instances to have the same file
1251 open (whether this invocation of QEMU is the first or the
1252 second instance), both instances must permit shared access
1253 for the second instance to succeed at opening the file.
1255 Enabling ``force-share=on`` requires ``read-only=on``.
1258 The host page cache can be avoided with ``cache.direct=on``.
1259 This will attempt to do disk IO directly to the guest's
1260 memory. QEMU may still perform an internal copy of the data.
1263 In case you don't care about data integrity over host
1264 failures, you can use ``cache.no-flush=on``. This option
1265 tells QEMU that it never needs to write any data to the disk
1266 but can instead keep things in cache. If anything goes
1267 wrong, like your host losing power, the disk storage getting
1268 disconnected accidentally, etc. your image will most
1269 probably be rendered unusable.
1272 discard is one of "ignore" (or "off") or "unmap" (or "on")
1273 and controls whether ``discard`` (also known as ``trim`` or
1274 ``unmap``) requests are ignored or passed to the filesystem.
1275 Some machine types may not support discard requests.
1277 ``detect-zeroes=detect-zeroes``
1278 detect-zeroes is "off", "on" or "unmap" and enables the
1279 automatic conversion of plain zero writes by the OS to
1280 driver specific optimized zero write commands. You may even
1281 choose "unmap" if discard is set to "unmap" to allow a zero
1282 write to be converted to an ``unmap`` operation.
1284 ``Driver-specific options for file``
1285 This is the protocol-level block driver for accessing regular
1289 The path to the image file in the local filesystem
1292 Specifies the AIO backend (threads/native/io_uring,
1296 Specifies whether the image file is protected with Linux OFD
1297 / POSIX locks. The default is to use the Linux Open File
1298 Descriptor API if available, otherwise no lock is applied.
1299 (auto/on/off, default: auto)
1305 -blockdev driver=file,node-name=disk,filename=disk.img
1307 ``Driver-specific options for raw``
1308 This is the image format block driver for raw images. It is
1309 usually stacked on top of a protocol level block driver such as
1313 Reference to or definition of the data source block driver
1314 node (e.g. a ``file`` driver node)
1320 -blockdev driver=file,node-name=disk_file,filename=disk.img
1321 -blockdev driver=raw,node-name=disk,file=disk_file
1327 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img
1329 ``Driver-specific options for qcow2``
1330 This is the image format block driver for qcow2 images. It is
1331 usually stacked on top of a protocol level block driver such as
1335 Reference to or definition of the data source block driver
1336 node (e.g. a ``file`` driver node)
1339 Reference to or definition of the backing file block device
1340 (default is taken from the image file). It is allowed to
1341 pass ``null`` here in order to disable the default backing
1345 Whether to enable the lazy refcounts feature (on/off;
1346 default is taken from the image file)
1349 The maximum total size of the L2 table and refcount block
1350 caches in bytes (default: the sum of l2-cache-size and
1351 refcount-cache-size)
1354 The maximum size of the L2 table cache in bytes (default: if
1355 cache-size is not specified - 32M on Linux platforms, and 8M
1356 on non-Linux platforms; otherwise, as large as possible
1357 within the cache-size, while permitting the requested or the
1358 minimal refcount cache size)
1360 ``refcount-cache-size``
1361 The maximum size of the refcount block cache in bytes
1362 (default: 4 times the cluster size; or if cache-size is
1363 specified, the part of it which is not used for the L2
1366 ``cache-clean-interval``
1367 Clean unused entries in the L2 and refcount caches. The
1368 interval is in seconds. The default value is 600 on
1369 supporting platforms, and 0 on other platforms. Setting it
1370 to 0 disables this feature.
1372 ``pass-discard-request``
1373 Whether discard requests to the qcow2 device should be
1374 forwarded to the data source (on/off; default: on if
1375 discard=unmap is specified, off otherwise)
1377 ``pass-discard-snapshot``
1378 Whether discard requests for the data source should be
1379 issued when a snapshot operation (e.g. deleting a snapshot)
1380 frees clusters in the qcow2 file (on/off; default: on)
1382 ``pass-discard-other``
1383 Whether discard requests for the data source should be
1384 issued on other occasions where a cluster gets freed
1385 (on/off; default: off)
1388 Which overlap checks to perform for writes to the image
1389 (none/constant/cached/all; default: cached). For details or
1390 finer granularity control refer to the QAPI documentation of
1397 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2
1398 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216
1404 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2
1406 ``Driver-specific options for other drivers``
1407 Please refer to the QAPI documentation of the ``blockdev-add``
1411 DEF("drive", HAS_ARG, QEMU_OPTION_drive,
1412 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n"
1413 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n"
1414 " [,snapshot=on|off][,rerror=ignore|stop|report]\n"
1415 " [,werror=ignore|stop|report|enospc][,id=name]\n"
1416 " [,aio=threads|native|io_uring]\n"
1417 " [,readonly=on|off][,copy-on-read=on|off]\n"
1418 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n"
1419 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n"
1420 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n"
1421 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n"
1422 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n"
1423 " [[,iops_size=is]]\n"
1425 " use 'file' as a drive image\n", QEMU_ARCH_ALL)
1427 ``-drive option[,option[,option[,...]]]``
1428 Define a new drive. This includes creating a block driver node (the
1429 backend) as well as a guest device, and is mostly a shortcut for
1430 defining the corresponding ``-blockdev`` and ``-device`` options.
1432 ``-drive`` accepts all options that are accepted by ``-blockdev``.
1433 In addition, it knows the following options:
1436 This option defines which disk image (see the :ref:`disk images`
1437 chapter in the System Emulation Users Guide) to use with this drive.
1438 If the filename contains comma, you must double it (for instance,
1439 "file=my,,file" to use file "my,file").
1441 Special files such as iSCSI devices can be specified using
1442 protocol specific URLs. See the section for "Device URL Syntax"
1443 for more information.
1446 This option defines on which type on interface the drive is
1447 connected. Available types are: ide, scsi, sd, mtd, floppy,
1448 pflash, virtio, none.
1450 ``bus=bus,unit=unit``
1451 These options define where is connected the drive by defining
1452 the bus number and the unit id.
1455 This option defines where the drive is connected by using an
1456 index in the list of available connectors of a given interface
1460 This option defines the type of the media: disk or cdrom.
1462 ``snapshot=snapshot``
1463 snapshot is "on" or "off" and controls snapshot mode for the
1464 given drive (see ``-snapshot``).
1467 cache is "none", "writeback", "unsafe", "directsync" or
1468 "writethrough" and controls how the host cache is used to access
1469 block data. This is a shortcut that sets the ``cache.direct``
1470 and ``cache.no-flush`` options (as in ``-blockdev``), and
1471 additionally ``cache.writeback``, which provides a default for
1472 the ``write-cache`` option of block guest devices (as in
1473 ``-device``). The modes correspond to the following settings:
1475 ============= =============== ============ ==============
1476 \ cache.writeback cache.direct cache.no-flush
1477 ============= =============== ============ ==============
1478 writeback on off off
1480 writethrough off off off
1481 directsync off on off
1483 ============= =============== ============ ==============
1485 The default mode is ``cache=writeback``.
1488 aio is "threads", "native", or "io_uring" and selects between pthread
1489 based disk I/O, native Linux AIO, or Linux io_uring API.
1492 Specify which disk format will be used rather than detecting the
1493 format. Can be used to specify format=raw to avoid interpreting
1494 an untrusted format header.
1496 ``werror=action,rerror=action``
1497 Specify which action to take on write and read errors. Valid
1498 actions are: "ignore" (ignore the error and try to continue),
1499 "stop" (pause QEMU), "report" (report the error to the guest),
1500 "enospc" (pause QEMU only if the host disk is full; report the
1501 error to the guest otherwise). The default setting is
1502 ``werror=enospc`` and ``rerror=report``.
1504 ``copy-on-read=copy-on-read``
1505 copy-on-read is "on" or "off" and enables whether to copy read
1506 backing file sectors into the image file.
1508 ``bps=b,bps_rd=r,bps_wr=w``
1509 Specify bandwidth throttling limits in bytes per second, either
1510 for all request types or for reads or writes only. Small values
1511 can lead to timeouts or hangs inside the guest. A safe minimum
1512 for disks is 2 MB/s.
1514 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm``
1515 Specify bursts in bytes per second, either for all request types
1516 or for reads or writes only. Bursts allow the guest I/O to spike
1517 above the limit temporarily.
1519 ``iops=i,iops_rd=r,iops_wr=w``
1520 Specify request rate limits in requests per second, either for
1521 all request types or for reads or writes only.
1523 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm``
1524 Specify bursts in requests per second, either for all request
1525 types or for reads or writes only. Bursts allow the guest I/O to
1526 spike above the limit temporarily.
1529 Let every is bytes of a request count as a new request for iops
1530 throttling purposes. Use this option to prevent guests from
1531 circumventing iops limits by sending fewer but larger requests.
1534 Join a throttling quota group with given name g. All drives that
1535 are members of the same group are accounted for together. Use
1536 this option to prevent guests from circumventing throttling
1537 limits by using many small disks instead of a single larger
1540 By default, the ``cache.writeback=on`` mode is used. It will report
1541 data writes as completed as soon as the data is present in the host
1542 page cache. This is safe as long as your guest OS makes sure to
1543 correctly flush disk caches where needed. If your guest OS does not
1544 handle volatile disk write caches correctly and your host crashes or
1545 loses power, then the guest may experience data corruption.
1547 For such guests, you should consider using ``cache.writeback=off``.
1548 This means that the host page cache will be used to read and write
1549 data, but write notification will be sent to the guest only after
1550 QEMU has made sure to flush each write to the disk. Be aware that
1551 this has a major impact on performance.
1553 When using the ``-snapshot`` option, unsafe caching is always used.
1555 Copy-on-read avoids accessing the same backing file sectors
1556 repeatedly and is useful when the backing file is over a slow
1557 network. By default copy-on-read is off.
1559 Instead of ``-cdrom`` you can use:
1563 |qemu_system| -drive file=file,index=2,media=cdrom
1565 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use:
1569 |qemu_system| -drive file=file,index=0,media=disk
1570 |qemu_system| -drive file=file,index=1,media=disk
1571 |qemu_system| -drive file=file,index=2,media=disk
1572 |qemu_system| -drive file=file,index=3,media=disk
1574 You can open an image using pre-opened file descriptors from an fd
1580 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
1581 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
1582 -drive file=/dev/fdset/2,index=0,media=disk
1584 You can connect a CDROM to the slave of ide0:
1588 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom
1590 If you don't specify the "file=" argument, you define an empty
1595 |qemu_system_x86| -drive if=ide,index=1,media=cdrom
1597 Instead of ``-fda``, ``-fdb``, you can use:
1601 |qemu_system_x86| -drive file=file,index=0,if=floppy
1602 |qemu_system_x86| -drive file=file,index=1,if=floppy
1604 By default, interface is "ide" and index is automatically
1609 |qemu_system_x86| -drive file=a -drive file=b"
1611 is interpreted like:
1615 |qemu_system_x86| -hda a -hdb b
1618 DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock,
1619 "-mtdblock file use 'file' as on-board Flash memory image\n",
1623 Use file as on-board Flash memory image.
1626 DEF("sd", HAS_ARG, QEMU_OPTION_sd,
1627 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL)
1630 Use file as SecureDigital card image.
1633 DEF("snapshot", 0, QEMU_OPTION_snapshot,
1634 "-snapshot write to temporary files instead of disk image files\n",
1638 Write to temporary files instead of disk image files. In this case,
1639 the raw disk image you use is not written back. You can however
1640 force the write back by pressing C-a s (see the :ref:`disk images`
1641 chapter in the System Emulation Users Guide).
1644 DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev,
1645 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1646 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n"
1647 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n"
1648 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n"
1649 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n"
1650 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n"
1651 " [[,throttling.iops-size=is]]\n"
1652 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n"
1653 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n"
1654 "-fsdev synth,id=id\n",
1658 ``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]``
1660 ``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]``
1662 ``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]``
1664 ``-fsdev synth,id=id[,readonly=on]``
1665 Define a new file system device. Valid options are:
1668 Accesses to the filesystem are done by QEMU.
1671 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1674 Synthetic filesystem, only used by QTests.
1677 Specifies identifier for this device.
1680 Specifies the export path for the file system device. Files
1681 under this path will be available to the 9p client on the guest.
1683 ``security_model=security_model``
1684 Specifies the security model to be used for this export path.
1685 Supported security models are "passthrough", "mapped-xattr",
1686 "mapped-file" and "none". In "passthrough" security model, files
1687 are stored using the same credentials as they are created on the
1688 guest. This requires QEMU to run as root. In "mapped-xattr"
1689 security model, some of the file attributes like uid, gid, mode
1690 bits and link target are stored as file attributes. For
1691 "mapped-file" these attributes are stored in the hidden
1692 .virtfs\_metadata directory. Directories exported by this
1693 security model cannot interact with other unix tools. "none"
1694 security model is same as passthrough except the sever won't
1695 report failures if it fails to set file attributes like
1696 ownership. Security model is mandatory only for local fsdriver.
1697 Other fsdrivers (like proxy) don't take security model as a
1700 ``writeout=writeout``
1701 This is an optional argument. The only supported value is
1702 "immediate". This means that host page cache will be used to
1703 read and write data but write notification will be sent to the
1704 guest only when the data has been reported as written by the
1708 Enables exporting 9p share as a readonly mount for guests. By
1709 default read-write access is given.
1712 Enables proxy filesystem driver to use passed socket file for
1713 communicating with virtfs-proxy-helper(1).
1716 Enables proxy filesystem driver to use passed socket descriptor
1717 for communicating with virtfs-proxy-helper(1). Usually a helper
1718 like libvirt will create socketpair and pass one of the fds as
1722 Specifies the default mode for newly created files on the host.
1723 Works only with security models "mapped-xattr" and
1727 Specifies the default mode for newly created directories on the
1728 host. Works only with security models "mapped-xattr" and
1731 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w``
1732 Specify bandwidth throttling limits in bytes per second, either
1733 for all request types or for reads or writes only.
1735 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm``
1736 Specify bursts in bytes per second, either for all request types
1737 or for reads or writes only. Bursts allow the guest I/O to spike
1738 above the limit temporarily.
1740 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w``
1741 Specify request rate limits in requests per second, either for
1742 all request types or for reads or writes only.
1744 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm``
1745 Specify bursts in requests per second, either for all request
1746 types or for reads or writes only. Bursts allow the guest I/O to
1747 spike above the limit temporarily.
1749 ``throttling.iops-size=is``
1750 Let every is bytes of a request count as a new request for iops
1751 throttling purposes.
1753 -fsdev option is used along with -device driver "virtio-9p-...".
1755 ``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag``
1756 Options for virtio-9p-... driver are:
1759 Specifies the variant to be used. Supported values are "pci",
1760 "ccw" or "device", depending on the machine type.
1763 Specifies the id value specified along with -fsdev option.
1765 ``mount_tag=mount_tag``
1766 Specifies the tag name to be used by the guest to mount this
1770 DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs,
1771 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1772 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n"
1773 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n"
1774 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n"
1775 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n",
1779 ``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]``
1781 ``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1783 ``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1785 ``-virtfs synth,mount_tag=mount_tag``
1786 Define a new virtual filesystem device and expose it to the guest using
1787 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain
1788 directory on host is made directly accessible by guest as a pass-through
1789 file system by using the 9P network protocol for communication between
1790 host and guests, if desired even accessible, shared by several guests
1793 Note that ``-virtfs`` is actually just a convenience shortcut for its
1794 generalized form ``-fsdev -device virtio-9p-pci``.
1796 The general form of pass-through file system options are:
1799 Accesses to the filesystem are done by QEMU.
1802 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1805 Synthetic filesystem, only used by QTests.
1808 Specifies identifier for the filesystem device
1811 Specifies the export path for the file system device. Files
1812 under this path will be available to the 9p client on the guest.
1814 ``security_model=security_model``
1815 Specifies the security model to be used for this export path.
1816 Supported security models are "passthrough", "mapped-xattr",
1817 "mapped-file" and "none". In "passthrough" security model, files
1818 are stored using the same credentials as they are created on the
1819 guest. This requires QEMU to run as root. In "mapped-xattr"
1820 security model, some of the file attributes like uid, gid, mode
1821 bits and link target are stored as file attributes. For
1822 "mapped-file" these attributes are stored in the hidden
1823 .virtfs\_metadata directory. Directories exported by this
1824 security model cannot interact with other unix tools. "none"
1825 security model is same as passthrough except the sever won't
1826 report failures if it fails to set file attributes like
1827 ownership. Security model is mandatory only for local fsdriver.
1828 Other fsdrivers (like proxy) don't take security model as a
1831 ``writeout=writeout``
1832 This is an optional argument. The only supported value is
1833 "immediate". This means that host page cache will be used to
1834 read and write data but write notification will be sent to the
1835 guest only when the data has been reported as written by the
1839 Enables exporting 9p share as a readonly mount for guests. By
1840 default read-write access is given.
1843 Enables proxy filesystem driver to use passed socket file for
1844 communicating with virtfs-proxy-helper(1). Usually a helper like
1845 libvirt will create socketpair and pass one of the fds as
1849 Enables proxy filesystem driver to use passed 'sock\_fd' as the
1850 socket descriptor for interfacing with virtfs-proxy-helper(1).
1853 Specifies the default mode for newly created files on the host.
1854 Works only with security models "mapped-xattr" and
1858 Specifies the default mode for newly created directories on the
1859 host. Works only with security models "mapped-xattr" and
1862 ``mount_tag=mount_tag``
1863 Specifies the tag name to be used by the guest to mount this
1866 ``multidevs=multidevs``
1867 Specifies how to deal with multiple devices being shared with a
1868 9p export. Supported behaviours are either "remap", "forbid" or
1869 "warn". The latter is the default behaviour on which virtfs 9p
1870 expects only one device to be shared with the same export, and
1871 if more than one device is shared and accessed via the same 9p
1872 export then only a warning message is logged (once) by qemu on
1873 host side. In order to avoid file ID collisions on guest you
1874 should either create a separate virtfs export for each device to
1875 be shared with guests (recommended way) or you might use "remap"
1876 instead which allows you to share multiple devices with only one
1877 export instead, which is achieved by remapping the original
1878 inode numbers from host to guest in a way that would prevent
1879 such collisions. Remapping inodes in such use cases is required
1880 because the original device IDs from host are never passed and
1881 exposed on guest. Instead all files of an export shared with
1882 virtfs always share the same device id on guest. So two files
1883 with identical inode numbers but from actually different devices
1884 on host would otherwise cause a file ID collision and hence
1885 potential misbehaviours on guest. "forbid" on the other hand
1886 assumes like "warn" that only one device is shared by the same
1887 export, however it will not only log a warning message but also
1888 deny access to additional devices on guest. Note though that
1889 "forbid" does currently not block all possible file access
1890 operations (e.g. readdir() would still return entries from other
1894 DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi,
1895 "-iscsi [user=user][,password=password]\n"
1896 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE\n"
1897 " [,initiator-name=initiator-iqn][,id=target-iqn]\n"
1898 " [,timeout=timeout]\n"
1899 " iSCSI session parameters\n", QEMU_ARCH_ALL)
1903 Configure iSCSI session parameters.
1908 DEFHEADING(USB convenience options:)
1910 DEF("usb", 0, QEMU_OPTION_usb,
1911 "-usb enable on-board USB host controller (if not enabled by default)\n",
1915 Enable USB emulation on machine types with an on-board USB host
1916 controller (if not enabled by default). Note that on-board USB host
1917 controllers may not support USB 3.0. In this case
1918 ``-device qemu-xhci`` can be used instead on machines with PCI.
1921 DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice,
1922 "-usbdevice name add the host or guest USB device 'name'\n",
1925 ``-usbdevice devname``
1926 Add the USB device devname, and enable an on-board USB controller
1927 if possible and necessary (just like it can be done via
1928 ``-machine usb=on``). Note that this option is mainly intended for
1929 the user's convenience only. More fine-grained control can be
1930 achieved by selecting a USB host controller (if necessary) and the
1931 desired USB device via the ``-device`` option instead. For example,
1932 instead of using ``-usbdevice mouse`` it is possible to use
1933 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse
1934 to a USB 3.0 controller instead (at least on machines that support
1935 PCI and do not have an USB controller enabled by default yet).
1936 For more details, see the chapter about
1937 :ref:`Connecting USB devices` in the System Emulation Users Guide.
1938 Possible devices for devname are:
1941 Braille device. This will use BrlAPI to display the braille
1942 output on a real or fake device (i.e. it also creates a
1943 corresponding ``braille`` chardev automatically beside the
1944 ``usb-braille`` USB device).
1947 Standard USB keyboard. Will override the PS/2 keyboard (if present).
1950 Virtual Mouse. This will override the PS/2 mouse emulation when
1954 Pointer device that uses absolute coordinates (like a
1955 touchscreen). This means QEMU is able to report the mouse
1956 position without having to grab the mouse. Also overrides the
1957 PS/2 mouse emulation when activated.
1960 Wacom PenPartner USB tablet.
1967 DEFHEADING(Display options:)
1969 DEF("display", HAS_ARG, QEMU_OPTION_display,
1970 #if defined(CONFIG_SPICE)
1971 "-display spice-app[,gl=on|off]\n"
1973 #if defined(CONFIG_SDL)
1974 "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n"
1975 " [,window-close=on|off]\n"
1977 #if defined(CONFIG_GTK)
1978 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n"
1979 " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n"
1980 " [,show-menubar=on|off]\n"
1982 #if defined(CONFIG_VNC)
1983 "-display vnc=<display>[,<optargs>]\n"
1985 #if defined(CONFIG_CURSES)
1986 "-display curses[,charset=<encoding>]\n"
1988 #if defined(CONFIG_COCOA)
1989 "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n"
1991 #if defined(CONFIG_OPENGL)
1992 "-display egl-headless[,rendernode=<file>]\n"
1994 #if defined(CONFIG_DBUS_DISPLAY)
1995 "-display dbus[,addr=<dbusaddr>]\n"
1996 " [,gl=on|core|es|off][,rendernode=<file>]\n"
1998 #if defined(CONFIG_COCOA)
1999 "-display cocoa[,show-cursor=on|off][,left-command-key=on|off]\n"
2002 " select display backend type\n"
2003 " The default display is equivalent to\n "
2004 #if defined(CONFIG_GTK)
2005 "\"-display gtk\"\n"
2006 #elif defined(CONFIG_SDL)
2007 "\"-display sdl\"\n"
2008 #elif defined(CONFIG_COCOA)
2009 "\"-display cocoa\"\n"
2010 #elif defined(CONFIG_VNC)
2011 "\"-vnc localhost:0,to=99,id=default\"\n"
2013 "\"-display none\"\n"
2018 Select type of display to use. Use ``-display help`` to list the available
2019 display types. Valid values for type are
2021 ``spice-app[,gl=on|off]``
2022 Start QEMU as a Spice server and launch the default Spice client
2023 application. The Spice server will redirect the serial consoles
2024 and QEMU monitors. (Since 4.0)
2027 Export the display over D-Bus interfaces. (Since 7.0)
2029 The connection is registered with the "org.qemu" name (and queued when
2032 ``addr=<dbusaddr>`` : D-Bus bus address to connect to.
2034 ``p2p=yes|no`` : Use peer-to-peer connection, accepted via QMP ``add_client``.
2036 ``gl=on|off|core|es`` : Use OpenGL for rendering (the D-Bus interface
2037 will share framebuffers with DMABUF file descriptors).
2040 Display video output via SDL (usually in a separate graphics
2041 window; see the SDL documentation for other possibilities).
2042 Valid parameters are:
2044 ``grab-mod=<mods>`` : Used to select the modifier keys for toggling
2045 the mouse grabbing in conjunction with the "g" key. ``<mods>`` can be
2046 either ``lshift-lctrl-lalt`` or ``rctrl``.
2048 ``gl=on|off|core|es`` : Use OpenGL for displaying
2050 ``show-cursor=on|off`` : Force showing the mouse cursor
2052 ``window-close=on|off`` : Allow to quit qemu with window close button
2055 Display video output in a GTK window. This interface provides
2056 drop-down menus and other UI elements to configure and control
2057 the VM during runtime. Valid parameters are:
2059 ``full-screen=on|off`` : Start in fullscreen mode
2061 ``gl=on|off`` : Use OpenGL for displaying
2063 ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover
2065 ``show-tabs=on|off`` : Display the tab bar for switching between the
2066 various graphical interfaces (e.g. VGA and
2067 virtual console character devices) by default.
2069 ``show-cursor=on|off`` : Force showing the mouse cursor
2071 ``window-close=on|off`` : Allow to quit qemu with window close button
2073 ``show-menubar=on|off`` : Display the main window menubar, defaults to "on"
2075 ``curses[,charset=<encoding>]``
2076 Display video output via curses. For graphics device models
2077 which support a text mode, QEMU can display this output using a
2078 curses/ncurses interface. Nothing is displayed when the graphics
2079 device is in graphical mode or if the graphics device does not
2080 support a text mode. Generally only the VGA device models
2081 support text mode. The font charset used by the guest can be
2082 specified with the ``charset`` option, for example
2083 ``charset=CP850`` for IBM CP850 encoding. The default is
2087 Display video output in a Cocoa window. Mac only. This interface
2088 provides drop-down menus and other UI elements to configure and
2089 control the VM during runtime. Valid parameters are:
2091 ``show-cursor=on|off`` : Force showing the mouse cursor
2093 ``left-command-key=on|off`` : Disable forwarding left command key to host
2095 ``egl-headless[,rendernode=<file>]``
2096 Offload all OpenGL operations to a local DRI device. For any
2097 graphical display, this display needs to be paired with either
2098 VNC or SPICE displays.
2101 Start a VNC server on display <display>
2104 Do not display video output. The guest will still see an
2105 emulated graphics card, but its output will not be displayed to
2106 the QEMU user. This option differs from the -nographic option in
2107 that it only affects what is done with video output; -nographic
2108 also changes the destination of the serial and parallel port
2112 DEF("nographic", 0, QEMU_OPTION_nographic,
2113 "-nographic disable graphical output and redirect serial I/Os to console\n",
2117 Normally, if QEMU is compiled with graphical window support, it
2118 displays output such as guest graphics, guest console, and the QEMU
2119 monitor in a window. With this option, you can totally disable
2120 graphical output so that QEMU is a simple command line application.
2121 The emulated serial port is redirected on the console and muxed with
2122 the monitor (unless redirected elsewhere explicitly). Therefore, you
2123 can still use QEMU to debug a Linux kernel with a serial console.
2124 Use C-a h for help on switching between the console and monitor.
2128 DEF("spice", HAS_ARG, QEMU_OPTION_spice,
2129 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n"
2130 " [,x509-key-file=<file>][,x509-key-password=<file>]\n"
2131 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n"
2132 " [,x509-dh-key-file=<file>][,addr=addr]\n"
2133 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n"
2134 " [,tls-ciphers=<list>]\n"
2135 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n"
2136 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n"
2137 " [,sasl=on|off][,disable-ticketing=on|off]\n"
2138 " [,password=<string>][,password-secret=<secret-id>]\n"
2139 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n"
2140 " [,jpeg-wan-compression=[auto|never|always]]\n"
2141 " [,zlib-glz-wan-compression=[auto|never|always]]\n"
2142 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n"
2143 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n"
2144 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n"
2145 " [,gl=[on|off]][,rendernode=<file>]\n"
2147 " at least one of {port, tls-port} is mandatory\n",
2151 ``-spice option[,option[,...]]``
2152 Enable the spice remote desktop protocol. Valid options are
2155 Set the TCP port spice is listening on for plaintext channels.
2158 Set the IP address spice is listening on. Default is any
2161 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off``
2162 Force using the specified IP version.
2164 ``password=<string>``
2165 Set the password you need to authenticate.
2167 This option is deprecated and insecure because it leaves the
2168 password visible in the process listing. Use ``password-secret``
2171 ``password-secret=<secret-id>``
2172 Set the ID of the ``secret`` object containing the password
2173 you need to authenticate.
2176 Require that the client use SASL to authenticate with the spice.
2177 The exact choice of authentication method used is controlled
2178 from the system / user's SASL configuration file for the 'qemu'
2179 service. This is typically found in /etc/sasl2/qemu.conf. If
2180 running QEMU as an unprivileged user, an environment variable
2181 SASL\_CONF\_PATH can be used to make it search alternate
2182 locations for the service config. While some SASL auth methods
2183 can also provide data encryption (eg GSSAPI), it is recommended
2184 that SASL always be combined with the 'tls' and 'x509' settings
2185 to enable use of SSL and server certificates. This ensures a
2186 data encryption preventing compromise of authentication
2189 ``disable-ticketing=on|off``
2190 Allow client connects without authentication.
2192 ``disable-copy-paste=on|off``
2193 Disable copy paste between the client and the guest.
2195 ``disable-agent-file-xfer=on|off``
2196 Disable spice-vdagent based file-xfer between the client and the
2200 Set the TCP port spice is listening on for encrypted channels.
2203 Set the x509 file directory. Expects same filenames as -vnc
2206 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>``
2207 The x509 file names can also be configured individually.
2209 ``tls-ciphers=<list>``
2210 Specify which ciphers to use.
2212 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]``
2213 Force specific channel to be used with or without TLS
2214 encryption. The options can be specified multiple times to
2215 configure multiple channels. The special name "default" can be
2216 used to set the default mode. For channels which are not
2217 explicitly forced into one mode the spice client is allowed to
2218 pick tls/plaintext as he pleases.
2220 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]``
2221 Configure image compression (lossless). Default is auto\_glz.
2223 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]``
2224 Configure wan image compression (lossy for slow links). Default
2227 ``streaming-video=[off|all|filter]``
2228 Configure video stream detection. Default is off.
2230 ``agent-mouse=[on|off]``
2231 Enable/disable passing mouse events via vdagent. Default is on.
2233 ``playback-compression=[on|off]``
2234 Enable/disable audio stream compression (using celt 0.5.1).
2237 ``seamless-migration=[on|off]``
2238 Enable/disable spice seamless migration. Default is off.
2241 Enable/disable OpenGL context. Default is off.
2243 ``rendernode=<file>``
2244 DRM render node for OpenGL rendering. If not specified, it will
2245 pick the first available. (Since 2.9)
2248 DEF("portrait", 0, QEMU_OPTION_portrait,
2249 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n",
2253 Rotate graphical output 90 deg left (only PXA LCD).
2256 DEF("rotate", HAS_ARG, QEMU_OPTION_rotate,
2257 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n",
2261 Rotate graphical output some deg left (only PXA LCD).
2264 DEF("vga", HAS_ARG, QEMU_OPTION_vga,
2265 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n"
2266 " select video card type\n", QEMU_ARCH_ALL)
2269 Select type of VGA card to emulate. Valid values for type are
2272 Cirrus Logic GD5446 Video card. All Windows versions starting
2273 from Windows 95 should recognize and use this graphic card. For
2274 optimal performances, use 16 bit color depth in the guest and
2275 the host OS. (This card was the default before QEMU 2.2)
2278 Standard VGA card with Bochs VBE extensions. If your guest OS
2279 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if
2280 you want to use high resolution modes (>= 1280x1024x16) then you
2281 should use this option. (This card is the default since QEMU
2285 VMWare SVGA-II compatible adapter. Use it if you have
2286 sufficiently recent XFree86/XOrg server or Windows guest with a
2287 driver for this card.
2290 QXL paravirtual graphic card. It is VGA compatible (including
2291 VESA 2.0 VBE support). Works best with qxl guest drivers
2292 installed though. Recommended choice when using the spice
2296 (sun4m only) Sun TCX framebuffer. This is the default
2297 framebuffer for sun4m machines and offers both 8-bit and 24-bit
2298 colour depths at a fixed resolution of 1024x768.
2301 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit
2302 framebuffer for sun4m machines available in both 1024x768
2303 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people
2304 wishing to run older Solaris versions.
2313 DEF("full-screen", 0, QEMU_OPTION_full_screen,
2314 "-full-screen start in full screen\n", QEMU_ARCH_ALL)
2317 Start in full screen.
2320 DEF("g", HAS_ARG, QEMU_OPTION_g ,
2321 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n",
2322 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K)
2324 ``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]``
2325 Set the initial graphical resolution and depth (PPC, SPARC only).
2327 For PPC the default is 800x600x32.
2329 For SPARC with the TCX graphics device, the default is 1024x768x8
2330 with the option of 1024x768x24. For cgthree, the default is
2331 1024x768x8 with the option of 1152x900x8 for people who wish to use
2335 DEF("vnc", HAS_ARG, QEMU_OPTION_vnc ,
2336 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL)
2338 ``-vnc display[,option[,option[,...]]]``
2339 Normally, if QEMU is compiled with graphical window support, it
2340 displays output such as guest graphics, guest console, and the QEMU
2341 monitor in a window. With this option, you can have QEMU listen on
2342 VNC display display and redirect the VGA display over the VNC
2343 session. It is very useful to enable the usb tablet device when
2344 using this option (option ``-device usb-tablet``). When using the
2345 VNC display, you must use the ``-k`` parameter to set the keyboard
2346 layout if you are not using en-us. Valid syntax for the display is
2349 With this option, QEMU will try next available VNC displays,
2350 until the number L, if the origianlly defined "-vnc display" is
2351 not available, e.g. port 5900+display is already used by another
2352 application. By default, to=0.
2355 TCP connections will only be allowed from host on display d. By
2356 convention the TCP port is 5900+d. Optionally, host can be
2357 omitted in which case the server will accept connections from
2361 Connections will be allowed over UNIX domain sockets where path
2362 is the location of a unix socket to listen for connections on.
2365 VNC is initialized but not started. The monitor ``change``
2366 command can be used to later start the VNC server.
2368 Following the display value there may be one or more option flags
2369 separated by commas. Valid options are
2372 Connect to a listening VNC client via a "reverse" connection.
2373 The client is specified by the display. For reverse network
2374 connections (host:d,``reverse``), the d argument is a TCP port
2375 number, not a display number.
2377 ``websocket=on|off``
2378 Opens an additional TCP listening port dedicated to VNC
2379 Websocket connections. If a bare websocket option is given, the
2380 Websocket port is 5700+display. An alternative port can be
2381 specified with the syntax ``websocket``\ =port.
2383 If host is specified connections will only be allowed from this
2384 host. It is possible to control the websocket listen address
2385 independently, using the syntax ``websocket``\ =host:port.
2387 If no TLS credentials are provided, the websocket connection
2388 runs in unencrypted mode. If TLS credentials are provided, the
2389 websocket connection requires encrypted client connections.
2392 Require that password based authentication is used for client
2395 The password must be set separately using the ``set_password``
2396 command in the :ref:`QEMU monitor`. The
2397 syntax to change your password is:
2398 ``set_password <protocol> <password>`` where <protocol> could be
2399 either "vnc" or "spice".
2401 If you would like to change <protocol> password expiration, you
2402 should use ``expire_password <protocol> <expiration-time>``
2403 where expiration time could be one of the following options:
2404 now, never, +seconds or UNIX time of expiration, e.g. +60 to
2405 make password expire in 60 seconds, or 1335196800 to make
2406 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for
2407 this date and time).
2409 You can also use keywords "now" or "never" for the expiration
2410 time to allow <protocol> password to expire immediately or never
2413 ``password-secret=<secret-id>``
2414 Require that password based authentication is used for client
2415 connections, using the password provided by the ``secret``
2416 object identified by ``secret-id``.
2419 Provides the ID of a set of TLS credentials to use to secure the
2420 VNC server. They will apply to both the normal VNC server socket
2421 and the websocket socket (if enabled). Setting TLS credentials
2422 will cause the VNC server socket to enable the VeNCrypt auth
2423 mechanism. The credentials should have been previously created
2424 using the ``-object tls-creds`` argument.
2427 Provides the ID of the QAuthZ authorization object against which
2428 the client's x509 distinguished name will validated. This object
2429 is only resolved at time of use, so can be deleted and recreated
2430 on the fly while the VNC server is active. If missing, it will
2431 default to denying access.
2434 Require that the client use SASL to authenticate with the VNC
2435 server. The exact choice of authentication method used is
2436 controlled from the system / user's SASL configuration file for
2437 the 'qemu' service. This is typically found in
2438 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user,
2439 an environment variable SASL\_CONF\_PATH can be used to make it
2440 search alternate locations for the service config. While some
2441 SASL auth methods can also provide data encryption (eg GSSAPI),
2442 it is recommended that SASL always be combined with the 'tls'
2443 and 'x509' settings to enable use of SSL and server
2444 certificates. This ensures a data encryption preventing
2445 compromise of authentication credentials. See the
2446 :ref:`VNC security` section in the System Emulation Users Guide
2447 for details on using SASL authentication.
2450 Provides the ID of the QAuthZ authorization object against which
2451 the client's SASL username will validated. This object is only
2452 resolved at time of use, so can be deleted and recreated on the
2453 fly while the VNC server is active. If missing, it will default
2457 Legacy method for enabling authorization of clients against the
2458 x509 distinguished name and SASL username. It results in the
2459 creation of two ``authz-list`` objects with IDs of
2460 ``vnc.username`` and ``vnc.x509dname``. The rules for these
2461 objects must be configured with the HMP ACL commands.
2463 This option is deprecated and should no longer be used. The new
2464 ``sasl-authz`` and ``tls-authz`` options are a replacement.
2467 Enable lossy compression methods (gradient, JPEG, ...). If this
2468 option is set, VNC client may receive lossy framebuffer updates
2469 depending on its encoding settings. Enabling this option can
2470 save a lot of bandwidth at the expense of quality.
2472 ``non-adaptive=on|off``
2473 Disable adaptive encodings. Adaptive encodings are enabled by
2474 default. An adaptive encoding will try to detect frequently
2475 updated screen regions, and send updates in these regions using
2476 a lossy encoding (like JPEG). This can be really helpful to save
2477 bandwidth when playing videos. Disabling adaptive encodings
2478 restores the original static behavior of encodings like Tight.
2480 ``share=[allow-exclusive|force-shared|ignore]``
2481 Set display sharing policy. 'allow-exclusive' allows clients to
2482 ask for exclusive access. As suggested by the rfb spec this is
2483 implemented by dropping other connections. Connecting multiple
2484 clients in parallel requires all clients asking for a shared
2485 session (vncviewer: -shared switch). This is the default.
2486 'force-shared' disables exclusive client access. Useful for
2487 shared desktop sessions, where you don't want someone forgetting
2488 specify -shared disconnect everybody else. 'ignore' completely
2489 ignores the shared flag and allows everybody connect
2490 unconditionally. Doesn't conform to the rfb spec but is
2491 traditional QEMU behavior.
2494 Set keyboard delay, for key down and key up events, in
2495 milliseconds. Default is 10. Keyboards are low-bandwidth
2496 devices, so this slowdown can help the device and guest to keep
2497 up and not lose events in case events are arriving in bulk.
2498 Possible causes for the latter are flaky network connections, or
2499 scripts for automated testing.
2501 ``audiodev=audiodev``
2502 Use the specified audiodev when the VNC client requests audio
2503 transmission. When not using an -audiodev argument, this option
2504 must be omitted, otherwise is must be present and specify a
2507 ``power-control=on|off``
2508 Permit the remote client to issue shutdown, reboot or reset power
2512 ARCHHEADING(, QEMU_ARCH_I386)
2514 ARCHHEADING(i386 target only:, QEMU_ARCH_I386)
2516 DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack,
2517 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n",
2521 Use it when installing Windows 2000 to avoid a disk full bug. After
2522 Windows 2000 is installed, you no longer need this option (this
2523 option slows down the IDE transfers).
2526 DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk,
2527 "-no-fd-bootchk disable boot signature checking for floppy disks\n",
2531 Disable boot signature checking for floppy disks in BIOS. May be
2532 needed to boot from old floppy disks.
2535 DEF("no-acpi", 0, QEMU_OPTION_no_acpi,
2536 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM)
2539 Disable ACPI (Advanced Configuration and Power Interface) support.
2540 Use it if your guest OS complains about ACPI problems (PC target
2544 DEF("no-hpet", 0, QEMU_OPTION_no_hpet,
2545 "-no-hpet disable HPET\n", QEMU_ARCH_I386)
2548 Disable HPET support. Deprecated, use '-machine hpet=off' instead.
2551 DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable,
2552 "-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"
2553 " ACPI table description\n", QEMU_ARCH_I386)
2555 ``-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]...]``
2556 Add ACPI table with specified header fields and context from
2557 specified files. For file=, take whole ACPI table from the specified
2558 files, including all ACPI headers (possible overridden by other
2559 options). For data=, only data portion of the table is used, all
2560 header information is specified in the command line. If a SLIC table
2561 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id
2562 fields will override the same in the RSDT and the FADT (a.k.a.
2563 FACP), in order to ensure the field matches required by the
2564 Microsoft SLIC spec and the ACPI spec.
2567 DEF("smbios", HAS_ARG, QEMU_OPTION_smbios,
2568 "-smbios file=binary\n"
2569 " load SMBIOS entry from binary file\n"
2570 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n"
2572 " specify SMBIOS type 0 fields\n"
2573 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2574 " [,uuid=uuid][,sku=str][,family=str]\n"
2575 " specify SMBIOS type 1 fields\n"
2576 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2577 " [,asset=str][,location=str]\n"
2578 " specify SMBIOS type 2 fields\n"
2579 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n"
2581 " specify SMBIOS type 3 fields\n"
2582 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n"
2583 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n"
2584 " [,processor-id=%d]\n"
2585 " specify SMBIOS type 4 fields\n"
2586 "-smbios type=8[,external_reference=str][,internal_reference=str][,connector_type=%d][,port_type=%d]\n"
2587 " specify SMBIOS type 8 fields\n"
2588 "-smbios type=11[,value=str][,path=filename]\n"
2589 " specify SMBIOS type 11 fields\n"
2590 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n"
2591 " [,asset=str][,part=str][,speed=%d]\n"
2592 " specify SMBIOS type 17 fields\n"
2593 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n"
2594 " specify SMBIOS type 41 fields\n",
2595 QEMU_ARCH_I386 | QEMU_ARCH_ARM)
2597 ``-smbios file=binary``
2598 Load SMBIOS entry from binary file.
2600 ``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]``
2601 Specify SMBIOS type 0 fields
2603 ``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]``
2604 Specify SMBIOS type 1 fields
2606 ``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]``
2607 Specify SMBIOS type 2 fields
2609 ``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]``
2610 Specify SMBIOS type 3 fields
2612 ``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-id=%d]``
2613 Specify SMBIOS type 4 fields
2615 ``-smbios type=11[,value=str][,path=filename]``
2616 Specify SMBIOS type 11 fields
2618 This argument can be repeated multiple times, and values are added in the order they are parsed.
2619 Applications intending to use OEM strings data are encouraged to use their application name as
2620 a prefix for the value string. This facilitates passing information for multiple applications
2623 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax
2624 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes.
2626 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to
2627 the SMBIOS table in the order in which they appear.
2629 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535
2630 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the
2631 guest. Instead it should be used as a indicator to inform the guest where to locate the real
2632 data set, for example, by specifying the serial ID of a block device.
2634 An example passing three strings is
2638 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\
2639 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\
2640 path=/some/file/with/oemstringsdata.txt
2642 In the guest OS this is visible with the ``dmidecode`` command
2647 Handle 0x0E00, DMI type 11, 5 bytes
2649 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/
2650 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os
2651 String 3: myapp:some extra data
2654 ``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]``
2655 Specify SMBIOS type 17 fields
2657 ``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]``
2658 Specify SMBIOS type 41 fields
2660 This argument can be repeated multiple times. Its main use is to allow network interfaces be created
2661 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface
2662 position on the PCI bus.
2664 Here is an example of use:
2668 -netdev user,id=internet \\
2669 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\
2670 -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev
2672 In the guest OS, the device should then appear as ``eno1``:
2677 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP>
2678 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP>
2680 Currently, the PCI device has to be attached to the root bus.
2686 DEFHEADING(Network options:)
2688 DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
2690 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n"
2691 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n"
2692 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n"
2693 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n"
2694 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]"
2696 "[,smb=dir[,smbserver=addr]]\n"
2698 " configure a user mode network backend with ID 'str',\n"
2699 " its DHCP server and optional services\n"
2702 "-netdev tap,id=str,ifname=name\n"
2703 " configure a host TAP network backend with ID 'str'\n"
2705 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n"
2706 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n"
2707 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n"
2709 " configure a host TAP network backend with ID 'str'\n"
2710 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2711 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n"
2712 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n"
2713 " to deconfigure it\n"
2714 " use '[down]script=no' to disable script execution\n"
2715 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n"
2717 " use 'fd=h' to connect to an already opened TAP interface\n"
2718 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n"
2719 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n"
2720 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n"
2721 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n"
2722 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n"
2723 " use vhost=on to enable experimental in kernel accelerator\n"
2724 " (only has effect for virtio guests which use MSIX)\n"
2725 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n"
2726 " use 'vhostfd=h' to connect to an already opened vhost net device\n"
2727 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n"
2728 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n"
2729 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n"
2730 " spent on busy polling for vhost net\n"
2731 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n"
2732 " configure a host TAP network backend with ID 'str' that is\n"
2733 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2734 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n"
2737 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n"
2738 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n"
2739 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n"
2740 " [,rxcookie=rxcookie][,offset=offset]\n"
2741 " configure a network backend with ID 'str' connected to\n"
2742 " an Ethernet over L2TPv3 pseudowire.\n"
2743 " Linux kernel 3.3+ as well as most routers can talk\n"
2744 " L2TPv3. This transport allows connecting a VM to a VM,\n"
2745 " VM to a router and even VM to Host. It is a nearly-universal\n"
2746 " standard (RFC3931). Note - this implementation uses static\n"
2747 " pre-configured tunnels (same as the Linux kernel).\n"
2748 " use 'src=' to specify source address\n"
2749 " use 'dst=' to specify destination address\n"
2750 " use 'udp=on' to specify udp encapsulation\n"
2751 " use 'srcport=' to specify source udp port\n"
2752 " use 'dstport=' to specify destination udp port\n"
2753 " use 'ipv6=on' to force v6\n"
2754 " L2TPv3 uses cookies to prevent misconfiguration as\n"
2755 " well as a weak security measure\n"
2756 " use 'rxcookie=0x012345678' to specify a rxcookie\n"
2757 " use 'txcookie=0x012345678' to specify a txcookie\n"
2758 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n"
2759 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n"
2760 " use 'pincounter=on' to work around broken counter handling in peer\n"
2761 " use 'offset=X' to add an extra offset between header and data\n"
2763 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n"
2764 " configure a network backend to connect to another network\n"
2765 " using a socket connection\n"
2766 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n"
2767 " configure a network backend to connect to a multicast maddr and port\n"
2768 " use 'localaddr=addr' to specify the host address to send packets from\n"
2769 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n"
2770 " configure a network backend to connect to another network\n"
2771 " using an UDP tunnel\n"
2772 "-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]\n"
2773 "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off]\n"
2774 "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor\n"
2775 " configure a network backend to connect to another network\n"
2776 " using a socket connection in stream mode.\n"
2777 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n"
2778 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]\n"
2779 " configure a network backend to connect to a multicast maddr and port\n"
2780 " use ``local.host=addr`` to specify the host address to send packets from\n"
2781 "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n"
2782 "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n"
2783 "-netdev dgram,id=str,local.type=fd,local.str=file-descriptor\n"
2784 " configure a network backend to connect to another network\n"
2785 " using an UDP tunnel\n"
2787 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n"
2788 " configure a network backend to connect to port 'n' of a vde switch\n"
2789 " running on host and listening for incoming connections on 'socketpath'.\n"
2790 " Use group 'groupname' and mode 'octalmode' to change default\n"
2791 " ownership and permissions for communication port.\n"
2793 #ifdef CONFIG_NETMAP
2794 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n"
2795 " attach to the existing netmap-enabled network interface 'name', or to a\n"
2796 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n"
2797 " netmap device, defaults to '/dev/netmap')\n"
2800 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n"
2801 " configure a vhost-user network, backed by a chardev 'dev'\n"
2804 "-netdev vhost-vdpa,id=str[,vhostdev=/path/to/dev][,vhostfd=h]\n"
2805 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n"
2806 " use 'vhostdev=/path/to/dev' to open a vhost vdpa device\n"
2807 " use 'vhostfd=h' to connect to an already opened vhost vdpa device\n"
2810 "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n"
2811 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n"
2812 " configure a vmnet network backend in host mode with ID 'str',\n"
2813 " isolate this interface from others with 'isolated',\n"
2814 " configure the address range and choose a subnet mask,\n"
2815 " specify network UUID 'uuid' to disable DHCP and interact with\n"
2816 " vmnet-host interfaces within this isolated network\n"
2817 "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n"
2818 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n"
2819 " configure a vmnet network backend in shared mode with ID 'str',\n"
2820 " configure the address range and choose a subnet mask,\n"
2821 " set IPv6 ULA prefix (of length 64) to use for internal network,\n"
2822 " isolate this interface from others with 'isolated'\n"
2823 "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n"
2824 " configure a vmnet network backend in bridged mode with ID 'str',\n"
2825 " use 'ifname=name' to select a physical network interface to be bridged,\n"
2826 " isolate this interface from others with 'isolated'\n"
2828 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n"
2829 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL)
2830 DEF("nic", HAS_ARG, QEMU_OPTION_nic,
2841 #ifdef CONFIG_NETMAP
2848 "vmnet-host|vmnet-shared|vmnet-bridged|"
2850 "socket][,option][,...][mac=macaddr]\n"
2851 " initialize an on-board / default host NIC (using MAC address\n"
2852 " macaddr) and connect it to the given host network backend\n"
2853 "-nic none use it alone to have zero network devices (the default is to\n"
2854 " provided a 'user' network connection)\n",
2856 DEF("net", HAS_ARG, QEMU_OPTION_net,
2857 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n"
2858 " configure or create an on-board (or machine default) NIC and\n"
2859 " connect it to hub 0 (please use -nic unless you need a hub)\n"
2869 #ifdef CONFIG_NETMAP
2873 "vmnet-host|vmnet-shared|vmnet-bridged|"
2875 "socket][,option][,option][,...]\n"
2876 " old way to initialize a host network interface\n"
2877 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL)
2879 ``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]``
2880 This option is a shortcut for configuring both the on-board
2881 (default) guest NIC hardware and the host network backend in one go.
2882 The host backend options are the same as with the corresponding
2883 ``-netdev`` options below. The guest NIC model can be set with
2884 ``model=modelname``. Use ``model=help`` to list the available device
2885 types. The hardware MAC address can be set with ``mac=macaddr``.
2887 The following two example do exactly the same, to show how ``-nic``
2888 can be used to shorten the command line length:
2892 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32
2893 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32
2896 Indicate that no network devices should be configured. It is used to
2897 override the default configuration (default NIC with "user" host
2898 network backend) which is activated if no other networking options
2901 ``-netdev user,id=id[,option][,option][,...]``
2902 Configure user mode host network backend which requires no
2903 administrator privilege to run. Valid options are:
2906 Assign symbolic name for use in monitor commands.
2908 ``ipv4=on|off and ipv6=on|off``
2909 Specify that either IPv4 or IPv6 must be enabled. If neither is
2910 specified both protocols are enabled.
2913 Set IP network address the guest will see. Optionally specify
2914 the netmask, either in the form a.b.c.d or as number of valid
2915 top-most bits. Default is 10.0.2.0/24.
2918 Specify the guest-visible address of the host. Default is the
2919 2nd IP in the guest network, i.e. x.x.x.2.
2921 ``ipv6-net=addr[/int]``
2922 Set IPv6 network address the guest will see (default is
2923 fec0::/64). The network prefix is given in the usual hexadecimal
2924 IPv6 address notation. The prefix size is optional, and is given
2925 as the number of valid top-most bits (default is 64).
2928 Specify the guest-visible IPv6 address of the host. Default is
2929 the 2nd IPv6 in the guest network, i.e. xxxx::2.
2932 If this option is enabled, the guest will be isolated, i.e. it
2933 will not be able to contact the host and no guest IP packets
2934 will be routed over the host to the outside. This option does
2935 not affect any explicitly set forwarding rules.
2938 Specifies the client hostname reported by the built-in DHCP
2942 Specify the first of the 16 IPs the built-in DHCP server can
2943 assign. Default is the 15th to 31st IP in the guest network,
2944 i.e. x.x.x.15 to x.x.x.31.
2947 Specify the guest-visible address of the virtual nameserver. The
2948 address must be different from the host address. Default is the
2949 3rd IP in the guest network, i.e. x.x.x.3.
2952 Specify the guest-visible address of the IPv6 virtual
2953 nameserver. The address must be different from the host address.
2954 Default is the 3rd IP in the guest network, i.e. xxxx::3.
2956 ``dnssearch=domain``
2957 Provides an entry for the domain-search list sent by the
2958 built-in DHCP server. More than one domain suffix can be
2959 transmitted by specifying this option multiple times. If
2960 supported, this will cause the guest to automatically try to
2961 append the given domain suffix(es) in case a domain name can not
2968 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org
2970 ``domainname=domain``
2971 Specifies the client domain name reported by the built-in DHCP
2975 When using the user mode network stack, activate a built-in TFTP
2976 server. The files in dir will be exposed as the root of a TFTP
2977 server. The TFTP client on the guest must be configured in
2978 binary mode (use the command ``bin`` of the Unix TFTP client).
2980 ``tftp-server-name=name``
2981 In BOOTP reply, broadcast name as the "TFTP server name"
2982 (RFC2132 option 66). This can be used to advise the guest to
2983 load boot files or configurations from a different server than
2987 When using the user mode network stack, broadcast file as the
2988 BOOTP filename. In conjunction with ``tftp``, this can be used
2989 to network boot a guest from a local directory.
2991 Example (using pxelinux):
2995 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\
2996 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0
2998 ``smb=dir[,smbserver=addr]``
2999 When using the user mode network stack, activate a built-in SMB
3000 server so that Windows OSes can access to the host files in
3001 ``dir`` transparently. The IP address of the SMB server can be
3002 set to addr. By default the 4th IP in the guest network is used,
3005 In the guest Windows OS, the line:
3011 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows
3012 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows
3015 Then ``dir`` can be accessed in ``\\smbserver\qemu``.
3017 Note that a SAMBA server must be installed on the host OS.
3019 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport``
3020 Redirect incoming TCP or UDP connections to the host port
3021 hostport to the guest IP address guestaddr on guest port
3022 guestport. If guestaddr is not specified, its value is x.x.x.15
3023 (default first address given by the built-in DHCP server). By
3024 specifying hostaddr, the rule can be bound to a specific host
3025 interface. If no connection type is set, TCP is used. This
3026 option can be given multiple times.
3028 For example, to redirect host X11 connection from screen 1 to
3029 guest screen 0, use the following:
3034 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000
3035 # this host xterm should open in the guest X11 server
3038 To redirect telnet connections from host port 5555 to telnet
3039 port on the guest, use the following:
3044 |qemu_system| -nic user,hostfwd=tcp::5555-:23
3045 telnet localhost 5555
3047 Then when you use on the host ``telnet localhost 5555``, you
3048 connect to the guest telnet server.
3050 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command``
3051 Forward guest TCP connections to the IP address server on port
3052 port to the character device dev or to a program executed by
3053 cmd:command which gets spawned for each connection. This option
3054 can be given multiple times.
3056 You can either use a chardev directly and have that one used
3057 throughout QEMU's lifetime, like in the following example:
3061 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever
3062 # the guest accesses it
3063 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321
3065 Or you can execute a command on every TCP connection established
3066 by the guest, so that QEMU behaves similar to an inetd process
3067 for that virtual server:
3071 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234
3072 # and connect the TCP stream to its stdin/stdout
3073 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321'
3075 ``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]``
3076 Configure a host TAP network backend with ID id.
3078 Use the network script file to configure it and the network script
3079 dfile to deconfigure it. If name is not provided, the OS
3080 automatically provides one. The default network configure script is
3081 ``/etc/qemu-ifup`` and the default network deconfigure script is
3082 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to
3083 disable script execution.
3085 If running QEMU as an unprivileged user, use the network helper
3086 to configure the TAP interface and attach it to the bridge.
3087 The default network helper executable is
3088 ``/path/to/qemu-bridge-helper`` and the default bridge device is
3091 ``fd``\ =h can be used to specify the handle of an already opened
3098 #launch a QEMU instance with the default network script
3099 |qemu_system| linux.img -nic tap
3103 #launch a QEMU instance with two NICs, each one connected
3105 |qemu_system| linux.img \\
3106 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\
3107 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1
3111 #launch a QEMU instance with the default network helper to
3112 #connect a TAP device to bridge br0
3113 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\
3114 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper"
3116 ``-netdev bridge,id=id[,br=bridge][,helper=helper]``
3117 Connect a host TAP network interface to a host bridge device.
3119 Use the network helper helper to configure the TAP interface and
3120 attach it to the bridge. The default network helper executable is
3121 ``/path/to/qemu-bridge-helper`` and the default bridge device is
3128 #launch a QEMU instance with the default network helper to
3129 #connect a TAP device to bridge br0
3130 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1
3134 #launch a QEMU instance with the default network helper to
3135 #connect a TAP device to bridge qemubr0
3136 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1
3138 ``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]``
3139 This host network backend can be used to connect the guest's network
3140 to another QEMU virtual machine using a TCP socket connection. If
3141 ``listen`` is specified, QEMU waits for incoming connections on port
3142 (host is optional). ``connect`` is used to connect to another QEMU
3143 instance using the ``listen`` option. ``fd``\ =h specifies an
3144 already opened TCP socket.
3150 # launch a first QEMU instance
3151 |qemu_system| linux.img \\
3152 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3153 -netdev socket,id=n1,listen=:1234
3154 # connect the network of this instance to the network of the first instance
3155 |qemu_system| linux.img \\
3156 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
3157 -netdev socket,id=n2,connect=127.0.0.1:1234
3159 ``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]``
3160 Configure a socket host network backend to share the guest's network
3161 traffic with another QEMU virtual machines using a UDP multicast
3162 socket, effectively making a bus for every QEMU with same multicast
3163 address maddr and port. NOTES:
3165 1. Several QEMU can be running on different hosts and share same bus
3166 (assuming correct multicast setup for these hosts).
3168 2. mcast support is compatible with User Mode Linux (argument
3169 ``ethN=mcast``), see http://user-mode-linux.sf.net.
3171 3. Use ``fd=h`` to specify an already opened UDP multicast socket.
3177 # launch one QEMU instance
3178 |qemu_system| linux.img \\
3179 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3180 -netdev socket,id=n1,mcast=230.0.0.1:1234
3181 # launch another QEMU instance on same "bus"
3182 |qemu_system| linux.img \\
3183 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
3184 -netdev socket,id=n2,mcast=230.0.0.1:1234
3185 # launch yet another QEMU instance on same "bus"
3186 |qemu_system| linux.img \\
3187 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\
3188 -netdev socket,id=n3,mcast=230.0.0.1:1234
3190 Example (User Mode Linux compat.):
3194 # launch QEMU instance (note mcast address selected is UML's default)
3195 |qemu_system| linux.img \\
3196 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3197 -netdev socket,id=n1,mcast=239.192.168.1:1102
3199 /path/to/linux ubd0=/path/to/root_fs eth0=mcast
3201 Example (send packets from host's 1.2.3.4):
3205 |qemu_system| linux.img \\
3206 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3207 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4
3209 ``-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]``
3210 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931)
3211 is a popular protocol to transport Ethernet (and other Layer 2) data
3212 frames between two systems. It is present in routers, firewalls and
3213 the Linux kernel (from version 3.3 onwards).
3215 This transport allows a VM to communicate to another VM, router or
3219 source address (mandatory)
3222 destination address (mandatory)
3225 select udp encapsulation (default is ip).
3231 destination udp port.
3234 force v6, otherwise defaults to v4.
3236 ``rxcookie=rxcookie``; \ ``txcookie=txcookie``
3237 Cookies are a weak form of security in the l2tpv3 specification.
3238 Their function is mostly to prevent misconfiguration. By default
3242 Set cookie size to 64 bit instead of the default 32
3245 Force a 'cut-down' L2TPv3 with no counter as in
3246 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00
3249 Work around broken counter handling in peer. This may also help
3250 on networks which have packet reorder.
3253 Add an extra offset between header and data
3255 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to
3256 the bridge br-lan on the remote Linux host 1.2.3.4:
3260 # Setup tunnel on linux host using raw ip as encapsulation
3262 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\
3263 encap udp udp_sport 16384 udp_dport 16384
3264 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\
3265 0xFFFFFFFF peer_session_id 0xFFFFFFFF
3266 ifconfig vmtunnel0 mtu 1500
3267 ifconfig vmtunnel0 up
3268 brctl addif br-lan vmtunnel0
3272 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter
3274 |qemu_system| linux.img -device e1000,netdev=n1 \\
3275 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter
3277 ``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]``
3278 Configure VDE backend to connect to PORT n of a vde switch running
3279 on host and listening for incoming connections on socketpath. Use
3280 GROUP groupname and MODE octalmode to change default ownership and
3281 permissions for communication port. This option is only available if
3282 QEMU has been compiled with vde support enabled.
3289 vde_switch -F -sock /tmp/myswitch
3290 # launch QEMU instance
3291 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch
3293 ``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]``
3294 Establish a vhost-user netdev, backed by a chardev id. The chardev
3295 should be a unix domain socket backed one. The vhost-user uses a
3296 specifically defined protocol to pass vhost ioctl replacement
3297 messages to an application on the other end of the socket. On
3298 non-MSIX guests, the feature can be forced with vhostforce. Use
3299 'queues=n' to specify the number of queues to be created for
3300 multiqueue vhost-user.
3306 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
3307 -numa node,memdev=mem \
3308 -chardev socket,id=chr0,path=/path/to/socket \
3309 -netdev type=vhost-user,id=net0,chardev=chr0 \
3310 -device virtio-net-pci,netdev=net0
3312 ``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]``
3313 Establish a vhost-vdpa netdev.
3315 vDPA device is a device that uses a datapath which complies with
3316 the virtio specifications with a vendor specific control path.
3317 vDPA devices can be both physically located on the hardware or
3318 emulated by software.
3320 ``-netdev hubport,id=id,hubid=hubid[,netdev=nd]``
3321 Create a hub port on the emulated hub with ID hubid.
3323 The hubport netdev lets you connect a NIC to a QEMU emulated hub
3324 instead of a single netdev. Alternatively, you can also connect the
3325 hubport to another netdev with ID nd by using the ``netdev=nd``
3328 ``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]``
3329 Legacy option to configure or create an on-board (or machine
3330 default) Network Interface Card(NIC) and connect it either to the
3331 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd.
3332 If model is omitted, then the default NIC model associated with the
3333 machine type is used. Note that the default NIC model may change in
3334 future QEMU releases, so it is highly recommended to always specify
3335 a model. Optionally, the MAC address can be changed to mac, the
3336 device address set to addr (PCI cards only), and a name can be
3337 assigned for use in monitor commands. Optionally, for PCI cards, you
3338 can specify the number v of MSI-X vectors that the card should have;
3339 this option currently only affects virtio cards; set v = 0 to
3340 disable MSI-X. If no ``-net`` option is specified, a single NIC is
3341 created. QEMU can emulate several different models of network card.
3342 Use ``-net nic,model=help`` for a list of available devices for your
3345 ``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]``
3346 Configure a host network backend (with the options corresponding to
3347 the same ``-netdev`` option) and connect it to the emulated hub 0
3348 (the default hub). Use name to specify the name of the hub port.
3353 DEFHEADING(Character device options:)
3355 DEF("chardev", HAS_ARG, QEMU_OPTION_chardev,
3357 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3358 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n"
3359 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n"
3360 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n"
3361 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n"
3362 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n"
3363 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n"
3364 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n"
3365 " [,logfile=PATH][,logappend=on|off]\n"
3366 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3367 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n"
3368 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3369 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n"
3370 "-chardev file,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3371 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3373 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3374 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3376 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3377 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n"
3379 #ifdef CONFIG_BRLAPI
3380 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3382 #if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \
3383 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__)
3384 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3386 #if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__)
3387 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3389 #if defined(CONFIG_SPICE)
3390 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3391 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3397 The general form of a character device option is:
3399 ``-chardev backend,id=id[,mux=on|off][,options]``
3400 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``,
3401 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``,
3402 ``pty``, ``stdio``, ``braille``, ``parallel``,
3403 ``spicevmc``, ``spiceport``. The specific backend will determine the
3406 Use ``-chardev help`` to print all available chardev backend types.
3408 All devices must have an id, which can be any string up to 127
3409 characters long. It is used to uniquely identify this device in
3410 other command line directives.
3412 A character device may be used in multiplexing mode by multiple
3413 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is
3414 a "1:N" device, and here the "1" end is your specified chardev
3415 backend, and the "N" end is the various parts of QEMU that can talk
3416 to a chardev. If you create a chardev with ``id=myid`` and
3417 ``mux=on``, QEMU will create a multiplexer with your specified ID,
3418 and you can then configure multiple front ends to use that chardev
3419 ID for their input/output. Up to four different front ends can be
3420 connected to a single multiplexed chardev. (Without multiplexing
3421 enabled, a chardev can only be used by a single front end.) For
3422 instance you could use this to allow a single stdio chardev to be
3423 used by two serial ports and the QEMU monitor:
3427 -chardev stdio,mux=on,id=char0 \
3428 -mon chardev=char0,mode=readline \
3429 -serial chardev:char0 \
3430 -serial chardev:char0
3432 You can have more than one multiplexer in a system configuration;
3433 for instance you could have a TCP port multiplexed between UART 0
3434 and UART 1, and stdio multiplexed between the QEMU monitor and a
3439 -chardev stdio,mux=on,id=char0 \
3440 -mon chardev=char0,mode=readline \
3441 -parallel chardev:char0 \
3442 -chardev tcp,...,mux=on,id=char1 \
3443 -serial chardev:char1 \
3444 -serial chardev:char1
3446 When you're using a multiplexed character device, some escape
3447 sequences are interpreted in the input. See the chapter about
3448 :ref:`keys in the character backend multiplexer` in the
3449 System Emulation Users Guide for more details.
3451 Note that some other command line options may implicitly create
3452 multiplexed character backends; for instance ``-serial mon:stdio``
3453 creates a multiplexed stdio backend connected to the serial port and
3454 the QEMU monitor, and ``-nographic`` also multiplexes the console
3455 and the monitor to stdio.
3457 There is currently no support for multiplexing in the other
3458 direction (where a single QEMU front end takes input and output from
3461 Every backend supports the ``logfile`` option, which supplies the
3462 path to a file to record all data transmitted via the backend. The
3463 ``logappend`` option controls whether the log file will be truncated
3464 or appended to when opened.
3466 The available backends are:
3468 ``-chardev null,id=id``
3469 A void device. This device will not emit any data, and will drop any
3470 data it receives. The null backend does not take any options.
3472 ``-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]``
3473 Create a two-way stream socket, which can be either a TCP or a unix
3474 socket. A unix socket will be created if ``path`` is specified.
3475 Behaviour is undefined if TCP options are specified for a unix
3478 ``server=on|off`` specifies that the socket shall be a listening socket.
3480 ``wait=on|off`` specifies that QEMU should not block waiting for a client
3481 to connect to a listening socket.
3483 ``telnet=on|off`` specifies that traffic on the socket should interpret
3484 telnet escape sequences.
3486 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for
3489 ``reconnect`` sets the timeout for reconnecting on non-server
3490 sockets when the remote end goes away. qemu will delay this many
3491 seconds and then attempt to reconnect. Zero disables reconnecting,
3494 ``tls-creds`` requests enablement of the TLS protocol for
3495 encryption, and specifies the id of the TLS credentials to use for
3496 the handshake. The credentials must be previously created with the
3497 ``-object tls-creds`` argument.
3499 ``tls-auth`` provides the ID of the QAuthZ authorization object
3500 against which the client's x509 distinguished name will be
3501 validated. This object is only resolved at time of use, so can be
3502 deleted and recreated on the fly while the chardev server is active.
3503 If missing, it will default to denying access.
3505 TCP and unix socket options are given below:
3507 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]``
3508 ``host`` for a listening socket specifies the local address to
3509 be bound. For a connecting socket species the remote host to
3510 connect to. ``host`` is optional for listening sockets. If not
3511 specified it defaults to ``0.0.0.0``.
3513 ``port`` for a listening socket specifies the local port to be
3514 bound. For a connecting socket specifies the port on the remote
3515 host to connect to. ``port`` can be given as either a port
3516 number or a service name. ``port`` is required.
3518 ``to`` is only relevant to listening sockets. If it is
3519 specified, and ``port`` cannot be bound, QEMU will attempt to
3520 bind to subsequent ports up to and including ``to`` until it
3521 succeeds. ``to`` must be specified as a port number.
3523 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4
3524 or IPv6 must be used. If neither is specified the socket may
3525 use either protocol.
3527 ``nodelay=on|off`` disables the Nagle algorithm.
3529 ``unix options: path=path[,abstract=on|off][,tight=on|off]``
3530 ``path`` specifies the local path of the unix socket. ``path``
3532 ``abstract=on|off`` specifies the use of the abstract socket namespace,
3533 rather than the filesystem. Optional, defaults to false.
3534 ``tight=on|off`` sets the socket length of abstract sockets to their minimum,
3535 rather than the full sun_path length. Optional, defaults to true.
3537 ``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]``
3538 Sends all traffic from the guest to a remote host over UDP.
3540 ``host`` specifies the remote host to connect to. If not specified
3541 it defaults to ``localhost``.
3543 ``port`` specifies the port on the remote host to connect to.
3544 ``port`` is required.
3546 ``localaddr`` specifies the local address to bind to. If not
3547 specified it defaults to ``0.0.0.0``.
3549 ``localport`` specifies the local port to bind to. If not specified
3550 any available local port will be used.
3552 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used.
3553 If neither is specified the device may use either protocol.
3555 ``-chardev msmouse,id=id``
3556 Forward QEMU's emulated msmouse events to the guest. ``msmouse``
3557 does not take any options.
3559 ``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]``
3560 Connect to a QEMU text console. ``vc`` may optionally be given a
3563 ``width`` and ``height`` specify the width and height respectively
3564 of the console, in pixels.
3566 ``cols`` and ``rows`` specify that the console be sized to fit a
3567 text console with the given dimensions.
3569 ``-chardev ringbuf,id=id[,size=size]``
3570 Create a ring buffer with fixed size ``size``. size must be a power
3571 of two and defaults to ``64K``.
3573 ``-chardev file,id=id,path=path``
3574 Log all traffic received from the guest to a file.
3576 ``path`` specifies the path of the file to be opened. This file will
3577 be created if it does not already exist, and overwritten if it does.
3578 ``path`` is required.
3580 ``-chardev pipe,id=id,path=path``
3581 Create a two-way connection to the guest. The behaviour differs
3582 slightly between Windows hosts and other hosts:
3584 On Windows, a single duplex pipe will be created at
3587 On other hosts, 2 pipes will be created called ``path.in`` and
3588 ``path.out``. Data written to ``path.in`` will be received by the
3589 guest. Data written by the guest can be read from ``path.out``. QEMU
3590 will not create these fifos, and requires them to be present.
3592 ``path`` forms part of the pipe path as described above. ``path`` is
3595 ``-chardev console,id=id``
3596 Send traffic from the guest to QEMU's standard output. ``console``
3597 does not take any options.
3599 ``console`` is only available on Windows hosts.
3601 ``-chardev serial,id=id,path=path``
3602 Send traffic from the guest to a serial device on the host.
3604 On Unix hosts serial will actually accept any tty device, not only
3607 ``path`` specifies the name of the serial device to open.
3609 ``-chardev pty,id=id``
3610 Create a new pseudo-terminal on the host and connect to it. ``pty``
3611 does not take any options.
3613 ``pty`` is not available on Windows hosts.
3615 ``-chardev stdio,id=id[,signal=on|off]``
3616 Connect to standard input and standard output of the QEMU process.
3618 ``signal`` controls if signals are enabled on the terminal, that
3619 includes exiting QEMU with the key sequence Control-c. This option
3620 is enabled by default, use ``signal=off`` to disable it.
3622 ``-chardev braille,id=id``
3623 Connect to a local BrlAPI server. ``braille`` does not take any
3626 ``-chardev parallel,id=id,path=path``
3628 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD
3631 Connect to a local parallel port.
3633 ``path`` specifies the path to the parallel port device. ``path`` is
3636 ``-chardev spicevmc,id=id,debug=debug,name=name``
3637 ``spicevmc`` is only available when spice support is built in.
3639 ``debug`` debug level for spicevmc
3641 ``name`` name of spice channel to connect to
3643 Connect to a spice virtual machine channel, such as vdiport.
3645 ``-chardev spiceport,id=id,debug=debug,name=name``
3646 ``spiceport`` is only available when spice support is built in.
3648 ``debug`` debug level for spicevmc
3650 ``name`` name of spice port to connect to
3652 Connect to a spice port, allowing a Spice client to handle the
3653 traffic identified by a name (preferably a fqdn).
3659 DEFHEADING(TPM device options:)
3661 DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \
3662 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n"
3663 " use path to provide path to a character device; default is /dev/tpm0\n"
3664 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n"
3665 " not provided it will be searched for in /sys/class/misc/tpm?/device\n"
3666 "-tpmdev emulator,id=id,chardev=dev\n"
3667 " configure the TPM device using chardev backend\n",
3670 The general form of a TPM device option is:
3672 ``-tpmdev backend,id=id[,options]``
3673 The specific backend type will determine the applicable options. The
3674 ``-tpmdev`` option creates the TPM backend and requires a
3675 ``-device`` option that specifies the TPM frontend interface model.
3677 Use ``-tpmdev help`` to print all available TPM backend types.
3679 The available backends are:
3681 ``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path``
3682 (Linux-host only) Enable access to the host's TPM using the
3685 ``path`` specifies the path to the host's TPM device, i.e., on a
3686 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by
3687 default ``/dev/tpm0`` is used.
3689 ``cancel-path`` specifies the path to the host TPM device's sysfs
3690 entry allowing for cancellation of an ongoing TPM command.
3691 ``cancel-path`` is optional and by default QEMU will search for the
3694 Some notes about using the host's TPM with the passthrough driver:
3696 The TPM device accessed by the passthrough driver must not be used
3697 by any other application on the host.
3699 Since the host's firmware (BIOS/UEFI) has already initialized the
3700 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize
3701 the TPM again and may therefore not show a TPM-specific menu that
3702 would otherwise allow the user to configure the TPM, e.g., allow the
3703 user to enable/disable or activate/deactivate the TPM. Further, if
3704 TPM ownership is released from within a VM then the host's TPM will
3705 get disabled and deactivated. To enable and activate the TPM again
3706 afterwards, the host has to be rebooted and the user is required to
3707 enter the firmware's menu to enable and activate the TPM. If the TPM
3708 is left disabled and/or deactivated most TPM commands will fail.
3710 To create a passthrough TPM use the following two options:
3714 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0
3716 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by
3717 ``tpmdev=tpm0`` in the device option.
3719 ``-tpmdev emulator,id=id,chardev=dev``
3720 (Linux-host only) Enable access to a TPM emulator using Unix domain
3721 socket based chardev backend.
3723 ``chardev`` specifies the unique ID of a character device backend
3724 that provides connection to the software TPM server.
3726 To create a TPM emulator backend device with chardev socket backend:
3730 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0
3737 DEFHEADING(Boot Image or Kernel specific:)
3739 There are broadly 4 ways you can boot a system with QEMU.
3741 - specify a firmware and let it control finding a kernel
3742 - specify a firmware and pass a hint to the kernel to boot
3743 - direct kernel image boot
3744 - manually load files into the guest's address space
3746 The third method is useful for quickly testing kernels but as there is
3747 no firmware to pass configuration information to the kernel the
3748 hardware must either be probeable, the kernel built for the exact
3749 configuration or passed some configuration data (e.g. a DTB blob)
3750 which tells the kernel what drivers it needs. This exact details are
3751 often hardware specific.
3753 The final method is the most generic way of loading images into the
3754 guest address space and used mostly for ``bare metal`` type
3755 development where the reset vectors of the processor are taken into
3762 For x86 machines and some other architectures ``-bios`` will generally
3763 do the right thing with whatever it is given. For other machines the
3764 more strict ``-pflash`` option needs an image that is sized for the
3765 flash device for the given machine type.
3767 Please see the :ref:`system-targets-ref` section of the manual for
3768 more detailed documentation.
3772 DEF("bios", HAS_ARG, QEMU_OPTION_bios, \
3773 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL)
3776 Set the filename for the BIOS.
3779 DEF("pflash", HAS_ARG, QEMU_OPTION_pflash,
3780 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL)
3783 Use file as a parallel flash image.
3788 The kernel options were designed to work with Linux kernels although
3789 other things (like hypervisors) can be packaged up as a kernel
3790 executable image. The exact format of a executable image is usually
3791 architecture specific.
3793 The way in which the kernel is started (what address it is loaded at,
3794 what if any information is passed to it via CPU registers, the state
3795 of the hardware when it is started, and so on) is also architecture
3796 specific. Typically it follows the specification laid down by the
3797 Linux kernel for how kernels for that architecture must be started.
3801 DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \
3802 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL)
3805 Use bzImage as kernel image. The kernel can be either a Linux kernel
3806 or in multiboot format.
3809 DEF("append", HAS_ARG, QEMU_OPTION_append, \
3810 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL)
3813 Use cmdline as kernel command line
3816 DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \
3817 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL)
3820 Use file as initial ram disk.
3822 ``-initrd "file1 arg=foo,file2"``
3823 This syntax is only available with multiboot.
3825 Use file1 and file2 as modules and pass arg=foo as parameter to the
3829 DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \
3830 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL)
3833 Use file as a device tree binary (dtb) image and pass it to the
3839 Finally you can also manually load images directly into the address
3840 space of the guest. This is most useful for developers who already
3841 know the layout of their guest and take care to ensure something sane
3842 will happen when the reset vector executes.
3844 The generic loader can be invoked by using the loader device:
3846 ``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]``
3848 there is also the guest loader which operates in a similar way but
3849 tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where
3852 ``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]``
3858 DEFHEADING(Debug/Expert options:)
3860 DEF("compat", HAS_ARG, QEMU_OPTION_compat,
3861 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n"
3862 " Policy for handling deprecated management interfaces\n"
3863 "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n"
3864 " Policy for handling unstable management interfaces\n",
3867 ``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]``
3868 Set policy for handling deprecated management interfaces (experimental):
3870 ``deprecated-input=accept`` (default)
3871 Accept deprecated commands and arguments
3872 ``deprecated-input=reject``
3873 Reject deprecated commands and arguments
3874 ``deprecated-input=crash``
3875 Crash on deprecated commands and arguments
3876 ``deprecated-output=accept`` (default)
3877 Emit deprecated command results and events
3878 ``deprecated-output=hide``
3879 Suppress deprecated command results and events
3881 Limitation: covers only syntactic aspects of QMP.
3883 ``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]``
3884 Set policy for handling unstable management interfaces (experimental):
3886 ``unstable-input=accept`` (default)
3887 Accept unstable commands and arguments
3888 ``unstable-input=reject``
3889 Reject unstable commands and arguments
3890 ``unstable-input=crash``
3891 Crash on unstable commands and arguments
3892 ``unstable-output=accept`` (default)
3893 Emit unstable command results and events
3894 ``unstable-output=hide``
3895 Suppress unstable command results and events
3897 Limitation: covers only syntactic aspects of QMP.
3900 DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg,
3901 "-fw_cfg [name=]<name>,file=<file>\n"
3902 " add named fw_cfg entry with contents from file\n"
3903 "-fw_cfg [name=]<name>,string=<str>\n"
3904 " add named fw_cfg entry with contents from string\n",
3907 ``-fw_cfg [name=]name,file=file``
3908 Add named fw\_cfg entry with contents from file file.
3910 ``-fw_cfg [name=]name,string=str``
3911 Add named fw\_cfg entry with contents from string str.
3913 The terminating NUL character of the contents of str will not be
3914 included as part of the fw\_cfg item data. To insert contents with
3915 embedded NUL characters, you have to use the file parameter.
3917 The fw\_cfg entries are passed by QEMU through to the guest.
3923 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin
3925 creates an fw\_cfg entry named opt/com.mycompany/blob with contents
3926 from ./my\_blob.bin.
3929 DEF("serial", HAS_ARG, QEMU_OPTION_serial, \
3930 "-serial dev redirect the serial port to char device 'dev'\n",
3934 Redirect the virtual serial port to host character device dev. The
3935 default device is ``vc`` in graphical mode and ``stdio`` in non
3938 This option can be used several times to simulate up to 4 serial
3941 Use ``-serial none`` to disable all serial ports.
3943 Available character devices are:
3946 Virtual console. Optionally, a width and height can be given in
3953 It is also possible to specify width or height in characters:
3960 [Linux only] Pseudo TTY (a new PTY is automatically allocated)
3963 No device is allocated.
3969 Use a named character device defined with the ``-chardev``
3973 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial
3974 port parameters are set according to the emulated ones.
3977 [Linux only, parallel port only] Use host parallel port N.
3978 Currently SPP and EPP parallel port features can be used.
3981 Write output to filename. No character can be read.
3984 [Unix only] standard input/output
3990 [Windows only] Use host serial port n
3992 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]``
3993 This implements UDP Net Console. When remote\_host or src\_ip
3994 are not specified they default to ``0.0.0.0``. When not using a
3995 specified src\_port a random port is automatically chosen.
3997 If you just want a simple readonly console you can use
3998 ``netcat`` or ``nc``, by starting QEMU with:
3999 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time
4000 QEMU writes something to that port it will appear in the
4003 If you plan to send characters back via netconsole or you want
4004 to stop and start QEMU a lot of times, you should have QEMU use
4005 the same source port each time by using something like ``-serial
4006 udp::4555@:4556`` to QEMU. Another approach is to use a patched
4007 version of netcat which can listen to a TCP port and send and
4008 receive characters via udp. If you have a patched version of
4009 netcat which activates telnet remote echo and single char
4010 transfer, then you can use the following options to set up a
4011 netcat redirector to allow telnet on port 5555 to access the
4015 -serial udp::4555@:4556
4018 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T
4023 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]``
4024 The TCP Net Console has two modes of operation. It can send the
4025 serial I/O to a location or wait for a connection from a
4026 location. By default the TCP Net Console is sent to host at the
4027 port. If you use the ``server=on`` option QEMU will wait for a client
4028 socket application to connect to the port before continuing,
4029 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off``
4030 option disables the Nagle buffering algorithm. The ``reconnect=on``
4031 option only applies if ``server=no`` is set, if the connection goes
4032 down it will attempt to reconnect at the given interval. If host
4033 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a
4034 time is accepted. You can use ``telnet=on`` to connect to the
4035 corresponding character device.
4037 ``Example to send tcp console to 192.168.0.2 port 4444``
4038 -serial tcp:192.168.0.2:4444
4040 ``Example to listen and wait on port 4444 for connection``
4041 -serial tcp::4444,server=on
4043 ``Example to not wait and listen on ip 192.168.0.100 port 4444``
4044 -serial tcp:192.168.0.100:4444,server=on,wait=off
4046 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]``
4047 The telnet protocol is used instead of raw tcp sockets. The
4048 options work the same as if you had specified ``-serial tcp``.
4049 The difference is that the port acts like a telnet server or
4050 client using telnet option negotiation. This will also allow you
4051 to send the MAGIC\_SYSRQ sequence if you use a telnet that
4052 supports sending the break sequence. Typically in unix telnet
4053 you do it with Control-] and then type "send break" followed by
4054 pressing the enter key.
4056 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]``
4057 The WebSocket protocol is used instead of raw tcp socket. The
4058 port acts as a WebSocket server. Client mode is not supported.
4060 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]``
4061 A unix domain socket is used instead of a tcp socket. The option
4062 works the same as if you had specified ``-serial tcp`` except
4063 the unix domain socket path is used for connections.
4066 This is a special option to allow the monitor to be multiplexed
4067 onto another serial port. The monitor is accessed with key
4068 sequence of Control-a and then pressing c. dev\_string should be
4069 any one of the serial devices specified above. An example to
4070 multiplex the monitor onto a telnet server listening on port
4073 ``-serial mon:telnet::4444,server=on,wait=off``
4075 When the monitor is multiplexed to stdio in this way, Ctrl+C
4076 will not terminate QEMU any more but will be passed to the guest
4080 Braille device. This will use BrlAPI to display the braille
4081 output on a real or fake device.
4084 Three button serial mouse. Configure the guest to use Microsoft
4088 DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \
4089 "-parallel dev redirect the parallel port to char device 'dev'\n",
4093 Redirect the virtual parallel port to host device dev (same devices
4094 as the serial port). On Linux hosts, ``/dev/parportN`` can be used
4095 to use hardware devices connected on the corresponding host parallel
4098 This option can be used several times to simulate up to 3 parallel
4101 Use ``-parallel none`` to disable all parallel ports.
4104 DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \
4105 "-monitor dev redirect the monitor to char device 'dev'\n",
4109 Redirect the monitor to host device dev (same devices as the serial
4110 port). The default device is ``vc`` in graphical mode and ``stdio``
4111 in non graphical mode. Use ``-monitor none`` to disable the default
4114 DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \
4115 "-qmp dev like -monitor but opens in 'control' mode\n",
4119 Like -monitor but opens in 'control' mode.
4121 DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \
4122 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n",
4126 Like -qmp but uses pretty JSON formatting.
4129 DEF("mon", HAS_ARG, QEMU_OPTION_mon, \
4130 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL)
4132 ``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]``
4133 Setup monitor on chardev name. ``mode=control`` configures
4134 a QMP monitor (a JSON RPC-style protocol) and it is not the
4135 same as HMP, the human monitor that has a "(qemu)" prompt.
4136 ``pretty`` is only valid when ``mode=control``,
4137 turning on JSON pretty printing to ease
4138 human reading and debugging.
4141 DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \
4142 "-debugcon dev redirect the debug console to char device 'dev'\n",
4146 Redirect the debug console to host device dev (same devices as the
4147 serial port). The debug console is an I/O port which is typically
4148 port 0xe9; writing to that I/O port sends output to this device. The
4149 default device is ``vc`` in graphical mode and ``stdio`` in non
4153 DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \
4154 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL)
4157 Store the QEMU process PID in file. It is useful if you launch QEMU
4161 DEF("singlestep", 0, QEMU_OPTION_singlestep, \
4162 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL)
4165 Run the emulation in single step mode.
4168 DEF("preconfig", 0, QEMU_OPTION_preconfig, \
4169 "--preconfig pause QEMU before machine is initialized (experimental)\n",
4173 Pause QEMU for interactive configuration before the machine is
4174 created, which allows querying and configuring properties that will
4175 affect machine initialization. Use QMP command 'x-exit-preconfig' to
4176 exit the preconfig state and move to the next state (i.e. run guest
4177 if -S isn't used or pause the second time if -S is used). This
4178 option is experimental.
4181 DEF("S", 0, QEMU_OPTION_S, \
4182 "-S freeze CPU at startup (use 'c' to start execution)\n",
4186 Do not start CPU at startup (you must type 'c' in the monitor).
4189 DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit,
4190 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n"
4191 " run qemu with overcommit hints\n"
4192 " mem-lock=on|off controls memory lock support (default: off)\n"
4193 " cpu-pm=on|off controls cpu power management (default: off)\n",
4196 ``-overcommit mem-lock=on|off``
4198 ``-overcommit cpu-pm=on|off``
4199 Run qemu with hints about host resource overcommit. The default is
4200 to assume that host overcommits all resources.
4202 Locking qemu and guest memory can be enabled via ``mem-lock=on``
4203 (disabled by default). This works when host memory is not
4204 overcommitted and reduces the worst-case latency for guest.
4206 Guest ability to manage power state of host cpus (increasing latency
4207 for other processes on the same host cpu, but decreasing latency for
4208 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This
4209 works best when host CPU is not overcommitted. When used, host
4210 estimates of CPU cycle and power utilization will be incorrect, not
4211 taking into account guest idle time.
4214 DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \
4215 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n"
4216 " the guest without waiting for gdb to connect; use -S too\n"
4217 " if you want it to not start execution.)\n",
4221 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter
4222 in the System Emulation Users Guide). Note that this option does not pause QEMU
4223 execution -- if you want QEMU to not start the guest until you
4224 connect with gdb and issue a ``continue`` command, you will need to
4225 also pass the ``-S`` option to QEMU.
4227 The most usual configuration is to listen on a local TCP socket::
4231 but you can specify other backends; UDP, pseudo TTY, or even stdio
4232 are all reasonable use cases. For example, a stdio connection
4233 allows you to start QEMU from within gdb and establish the
4234 connection via a pipe:
4238 (gdb) target remote | exec |qemu_system| -gdb stdio ...
4241 DEF("s", 0, QEMU_OPTION_s, \
4242 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n",
4246 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234
4247 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide).
4250 DEF("d", HAS_ARG, QEMU_OPTION_d, \
4251 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n",
4255 Enable logging of specified items. Use '-d help' for a list of log
4259 DEF("D", HAS_ARG, QEMU_OPTION_D, \
4260 "-D logfile output log to logfile (default stderr)\n",
4264 Output log in logfile instead of to stderr
4267 DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \
4268 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n",
4271 ``-dfilter range1[,...]``
4272 Filter debug output to that relevant to a range of target addresses.
4273 The filter spec can be either start+size, start-size or start..end
4274 where start end and size are the addresses and sizes required. For
4279 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000
4281 Will dump output for any code in the 0x1000 sized block starting at
4282 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and
4283 another 0x1000 sized block starting at 0xffffffc00005f000.
4286 DEF("seed", HAS_ARG, QEMU_OPTION_seed, \
4287 "-seed number seed the pseudo-random number generator\n",
4291 Force the guest to use a deterministic pseudo-random number
4292 generator, seeded with number. This does not affect crypto routines
4296 DEF("L", HAS_ARG, QEMU_OPTION_L, \
4297 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n",
4301 Set the directory for the BIOS, VGA BIOS and keymaps.
4303 To list all the data directories, use ``-L help``.
4306 DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \
4307 "-enable-kvm enable KVM full virtualization support\n",
4308 QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC |
4309 QEMU_ARCH_RISCV | QEMU_ARCH_S390X)
4312 Enable KVM full virtualization support. This option is only
4313 available if KVM support is enabled when compiling.
4316 DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid,
4317 "-xen-domid id specify xen guest domain id\n",
4318 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4319 DEF("xen-attach", 0, QEMU_OPTION_xen_attach,
4320 "-xen-attach attach to existing xen domain\n"
4321 " libxl will use this when starting QEMU\n",
4322 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4323 DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict,
4324 "-xen-domid-restrict restrict set of available xen operations\n"
4325 " to specified domain id. (Does not affect\n"
4326 " xenpv machine type).\n",
4327 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4330 Specify xen guest domain id (XEN only).
4333 Attach to existing xen domain. libxl will use this when starting
4334 QEMU (XEN only). Restrict set of available xen operations to
4335 specified domain id (XEN only).
4338 DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \
4339 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL)
4342 Exit instead of rebooting.
4345 DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \
4346 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL)
4349 Don't exit QEMU on guest shutdown, but instead only stop the
4350 emulation. This allows for instance switching to monitor to commit
4351 changes to the disk image.
4354 DEF("action", HAS_ARG, QEMU_OPTION_action,
4355 "-action reboot=reset|shutdown\n"
4356 " action when guest reboots [default=reset]\n"
4357 "-action shutdown=poweroff|pause\n"
4358 " action when guest shuts down [default=poweroff]\n"
4359 "-action panic=pause|shutdown|exit-failure|none\n"
4360 " action when guest panics [default=shutdown]\n"
4361 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n"
4362 " action when watchdog fires [default=reset]\n",
4365 ``-action event=action``
4366 The action parameter serves to modify QEMU's default behavior when
4367 certain guest events occur. It provides a generic method for specifying the
4368 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown``
4373 ``-action panic=none``
4374 ``-action reboot=shutdown,shutdown=pause``
4375 ``-device i6300esb -action watchdog=pause``
4379 DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \
4380 "-loadvm [tag|id]\n" \
4381 " start right away with a saved state (loadvm in monitor)\n",
4385 Start right away with a saved state (``loadvm`` in monitor)
4389 DEF("daemonize", 0, QEMU_OPTION_daemonize, \
4390 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL)
4394 Daemonize the QEMU process after initialization. QEMU will not
4395 detach from standard IO until it is ready to receive connections on
4396 any of its devices. This option is a useful way for external
4397 programs to launch QEMU without having to cope with initialization
4401 DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \
4402 "-option-rom rom load a file, rom, into the option ROM space\n",
4405 ``-option-rom file``
4406 Load the contents of file as an option ROM. This option is useful to
4407 load things like EtherBoot.
4410 DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \
4411 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \
4412 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n",
4416 ``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]``
4417 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at
4418 the current UTC or local time, respectively. ``localtime`` is
4419 required for correct date in MS-DOS or Windows. To start at a
4420 specific point in time, provide datetime in the format
4421 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC.
4423 By default the RTC is driven by the host system time. This allows
4424 using of the RTC as accurate reference clock inside the guest,
4425 specifically if the host time is smoothly following an accurate
4426 external reference clock, e.g. via NTP. If you want to isolate the
4427 guest time from the host, you can set ``clock`` to ``rt`` instead,
4428 which provides a host monotonic clock if host support it. To even
4429 prevent the RTC from progressing during suspension, you can set
4430 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is
4431 recommended especially in icount mode in order to preserve
4432 determinism; however, note that in icount mode the speed of the
4433 virtual clock is variable and can in general differ from the host
4436 Enable ``driftfix`` (i386 targets only) if you experience time drift
4437 problems, specifically with Windows' ACPI HAL. This option will try
4438 to figure out how many timer interrupts were not processed by the
4439 Windows guest and will re-inject them.
4442 DEF("icount", HAS_ARG, QEMU_OPTION_icount, \
4443 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \
4444 " enable virtual instruction counter with 2^N clock ticks per\n" \
4445 " instruction, enable aligning the host and virtual clocks\n" \
4446 " or disable real time cpu sleeping, and optionally enable\n" \
4447 " record-and-replay mode\n", QEMU_ARCH_ALL)
4449 ``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]``
4450 Enable virtual instruction counter. The virtual cpu will execute one
4451 instruction every 2^N ns of virtual time. If ``auto`` is specified
4452 then the virtual cpu speed will be automatically adjusted to keep
4453 virtual time within a few seconds of real time.
4455 Note that while this option can give deterministic behavior, it does
4456 not provide cycle accurate emulation. Modern CPUs contain
4457 superscalar out of order cores with complex cache hierarchies. The
4458 number of instructions executed often has little or no correlation
4459 with actual performance.
4461 When the virtual cpu is sleeping, the virtual time will advance at
4462 default speed unless ``sleep=on`` is specified. With
4463 ``sleep=on``, the virtual time will jump to the next timer
4464 deadline instantly whenever the virtual cpu goes to sleep mode and
4465 will not advance if no timer is enabled. This behavior gives
4466 deterministic execution times from the guest point of view.
4467 The default if icount is enabled is ``sleep=off``.
4468 ``sleep=on`` cannot be used together with either ``shift=auto``
4471 ``align=on`` will activate the delay algorithm which will try to
4472 synchronise the host clock and the virtual clock. The goal is to
4473 have a guest running at the real frequency imposed by the shift
4474 option. Whenever the guest clock is behind the host clock and if
4475 ``align=on`` is specified then we print a message to the user to
4476 inform about the delay. Currently this option does not work when
4477 ``shift`` is ``auto``. Note: The sync algorithm will work for those
4478 shift values for which the guest clock runs ahead of the host clock.
4479 Typically this happens when the shift value is high (how high
4480 depends on the host machine). The default if icount is enabled
4483 When the ``rr`` option is specified deterministic record/replay is
4484 enabled. The ``rrfile=`` option must also be provided to
4485 specify the path to the replay log. In record mode data is written
4486 to this file, and in replay mode it is read back.
4487 If the ``rrsnapshot`` option is given then it specifies a VM snapshot
4488 name. In record mode, a new VM snapshot with the given name is created
4489 at the start of execution recording. In replay mode this option
4490 specifies the snapshot name used to load the initial VM state.
4493 DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \
4494 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \
4495 " action when watchdog fires [default=reset]\n",
4498 ``-watchdog-action action``
4499 The action controls what QEMU will do when the watchdog timer
4500 expires. The default is ``reset`` (forcefully reset the guest).
4501 Other possible actions are: ``shutdown`` (attempt to gracefully
4502 shutdown the guest), ``poweroff`` (forcefully poweroff the guest),
4503 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the
4504 guest), ``debug`` (print a debug message and continue), or ``none``
4507 Note that the ``shutdown`` action requires that the guest responds
4508 to ACPI signals, which it may not be able to do in the sort of
4509 situations where the watchdog would have expired, and thus
4510 ``-watchdog-action shutdown`` is not recommended for production use.
4514 ``-device i6300esb -watchdog-action pause``
4518 DEF("echr", HAS_ARG, QEMU_OPTION_echr, \
4519 "-echr chr set terminal escape character instead of ctrl-a\n",
4522 ``-echr numeric_ascii_value``
4523 Change the escape character used for switching to the monitor when
4524 using monitor and serial sharing. The default is ``0x01`` when using
4525 the ``-nographic`` option. ``0x01`` is equal to pressing
4526 ``Control-a``. You can select a different character from the ascii
4527 control keys where 1 through 26 map to Control-a through Control-z.
4528 For instance you could use the either of the following to change the
4529 escape character to Control-t.
4531 ``-echr 0x14``; \ ``-echr 20``
4535 DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \
4536 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \
4537 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \
4538 "-incoming unix:socketpath\n" \
4539 " prepare for incoming migration, listen on\n" \
4540 " specified protocol and socket address\n" \
4541 "-incoming fd:fd\n" \
4542 "-incoming exec:cmdline\n" \
4543 " accept incoming migration on given file descriptor\n" \
4544 " or from given external command\n" \
4545 "-incoming defer\n" \
4546 " wait for the URI to be specified via migrate_incoming\n",
4549 ``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]``
4551 ``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]``
4552 Prepare for incoming migration, listen on a given tcp port.
4554 ``-incoming unix:socketpath``
4555 Prepare for incoming migration, listen on a given unix socket.
4558 Accept incoming migration from a given filedescriptor.
4560 ``-incoming exec:cmdline``
4561 Accept incoming migration as an output from specified external
4565 Wait for the URI to be specified via migrate\_incoming. The monitor
4566 can be used to change settings (such as migration parameters) prior
4567 to issuing the migrate\_incoming to allow the migration to begin.
4570 DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \
4571 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL)
4573 ``-only-migratable``
4574 Only allow migratable devices. Devices will not be allowed to enter
4575 an unmigratable state.
4578 DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \
4579 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL)
4582 Don't create default devices. Normally, QEMU sets the default
4583 devices like serial port, parallel port, virtual console, monitor
4584 device, VGA adapter, floppy and CD-ROM drive and others. The
4585 ``-nodefaults`` option will disable all those default devices.
4589 DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \
4590 "-chroot dir chroot to dir just before starting the VM\n",
4595 Immediately before starting guest execution, chroot to the specified
4596 directory. Especially useful in combination with -runas.
4600 DEF("runas", HAS_ARG, QEMU_OPTION_runas, \
4601 "-runas user change to user id user just before starting the VM\n" \
4602 " user can be numeric uid:gid instead\n",
4607 Immediately before starting guest execution, drop root privileges,
4608 switching to the specified user.
4611 DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env,
4612 "-prom-env variable=value\n"
4613 " set OpenBIOS nvram variables\n",
4614 QEMU_ARCH_PPC | QEMU_ARCH_SPARC)
4616 ``-prom-env variable=value``
4617 Set OpenBIOS nvram variable to given value (PPC, SPARC only).
4621 qemu-system-sparc -prom-env 'auto-boot?=false' \
4622 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single'
4626 qemu-system-ppc -prom-env 'auto-boot?=false' \
4627 -prom-env 'boot-device=hd:2,\yaboot' \
4628 -prom-env 'boot-args=conf=hd:2,\yaboot.conf'
4630 DEF("semihosting", 0, QEMU_OPTION_semihosting,
4631 "-semihosting semihosting mode\n",
4632 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA |
4633 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4636 Enable :ref:`Semihosting` mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only).
4639 Note that this allows guest direct access to the host filesystem, so
4640 should only be used with a trusted guest OS.
4642 See the -semihosting-config option documentation for further
4643 information about the facilities this enables.
4645 DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config,
4646 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]\n" \
4647 " semihosting configuration\n",
4648 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA |
4649 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4651 ``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]``
4652 Enable and configure :ref:`Semihosting` (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V
4656 Note that this allows guest direct access to the host filesystem, so
4657 should only be used with a trusted guest OS.
4659 ``target=native|gdb|auto``
4660 Defines where the semihosting calls will be addressed, to QEMU
4661 (``native``) or to GDB (``gdb``). The default is ``auto``, which
4662 means ``gdb`` during debug sessions and ``native`` otherwise.
4665 Send the output to a chardev backend output for native or auto
4666 output when not in gdb
4668 ``userspace=on|off``
4669 Allows code running in guest userspace to access the semihosting
4670 interface. The default is that only privileged guest code can
4671 make semihosting calls. Note that setting ``userspace=on`` should
4672 only be used if all guest code is trusted (for example, in
4673 bare-metal test case code).
4675 ``arg=str1,arg=str2,...``
4676 Allows the user to pass input arguments, and can be used
4677 multiple times to build up a list. The old-style
4678 ``-kernel``/``-append`` method of passing a command line is
4679 still supported for backward compatibility. If both the
4680 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are
4681 specified, the former is passed to semihosting as it always
4684 DEF("old-param", 0, QEMU_OPTION_old_param,
4685 "-old-param old param mode\n", QEMU_ARCH_ARM)
4688 Old param mode (ARM only).
4691 DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \
4692 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \
4693 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \
4694 " Enable seccomp mode 2 system call filter (default 'off').\n" \
4695 " use 'obsolete' to allow obsolete system calls that are provided\n" \
4696 " by the kernel, but typically no longer used by modern\n" \
4697 " C library implementations.\n" \
4698 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \
4699 " to elevate privileges using set*uid|gid system calls.\n" \
4700 " The value 'children' will deny set*uid|gid system calls for\n" \
4701 " main QEMU process but will allow forks and execves to run unprivileged\n" \
4702 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \
4703 " blocking *fork and execve\n" \
4704 " use 'resourcecontrol' to disable process affinity and schedular priority\n",
4707 ``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]``
4708 Enable Seccomp mode 2 system call filter. 'on' will enable syscall
4709 filtering and 'off' will disable it. The default is 'off'.
4712 Enable Obsolete system calls
4714 ``elevateprivileges=string``
4715 Disable set\*uid\|gid system calls
4718 Disable \*fork and execve
4720 ``resourcecontrol=string``
4721 Disable process affinity and schedular priority
4724 DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig,
4725 "-readconfig <file>\n"
4726 " read config file\n", QEMU_ARCH_ALL)
4728 ``-readconfig file``
4729 Read device configuration from file. This approach is useful when
4730 you want to spawn QEMU process with many command line options but
4731 you don't want to exceed the command line character limit.
4734 DEF("no-user-config", 0, QEMU_OPTION_nouserconfig,
4736 " do not load default user-provided config files at startup\n",
4740 The ``-no-user-config`` option makes QEMU not load any of the
4741 user-provided config files on sysconfdir.
4744 DEF("trace", HAS_ARG, QEMU_OPTION_trace,
4745 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n"
4746 " specify tracing options\n",
4749 ``-trace [[enable=]pattern][,events=file][,file=file]``
4750 .. include:: ../qemu-option-trace.rst.inc
4753 DEF("plugin", HAS_ARG, QEMU_OPTION_plugin,
4754 "-plugin [file=]<file>[,<argname>=<argvalue>]\n"
4758 ``-plugin file=file[,argname=argvalue]``
4762 Load the given plugin from a shared library file.
4764 ``argname=argvalue``
4765 Argument passed to the plugin. (Can be given multiple times.)
4769 DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL)
4770 DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL)
4773 DEF("async-teardown", 0, QEMU_OPTION_asyncteardown,
4774 "-async-teardown enable asynchronous teardown\n",
4779 Enable asynchronous teardown. A new process called "cleanup/<QEMU_PID>"
4780 will be created at startup sharing the address space with the main qemu
4781 process, using clone. It will wait for the main qemu process to
4782 terminate completely, and then exit.
4783 This allows qemu to terminate very quickly even if the guest was
4784 huge, leaving the teardown of the address space to the cleanup
4785 process. Since the cleanup process shares the same cgroups as the
4786 main qemu process, accounting is performed correctly. This only
4787 works if the cleanup process is not forcefully killed with SIGKILL
4788 before the main qemu process has terminated completely.
4791 DEF("msg", HAS_ARG, QEMU_OPTION_msg,
4792 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n"
4793 " control error message format\n"
4794 " timestamp=on enables timestamps (default: off)\n"
4795 " guest-name=on enables guest name prefix but only if\n"
4796 " -name guest option is set (default: off)\n",
4799 ``-msg [timestamp[=on|off]][,guest-name[=on|off]]``
4800 Control error message format.
4802 ``timestamp=on|off``
4803 Prefix messages with a timestamp. Default is off.
4805 ``guest-name=on|off``
4806 Prefix messages with guest name but only if -name guest option is set
4807 otherwise the option is ignored. Default is off.
4810 DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate,
4811 "-dump-vmstate <file>\n"
4812 " Output vmstate information in JSON format to file.\n"
4813 " Use the scripts/vmstate-static-checker.py file to\n"
4814 " check for possible regressions in migration code\n"
4815 " by comparing two such vmstate dumps.\n",
4818 ``-dump-vmstate file``
4819 Dump json-encoded vmstate information for current machine type to
4823 DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile,
4824 "-enable-sync-profile\n"
4825 " enable synchronization profiling\n",
4828 ``-enable-sync-profile``
4829 Enable synchronization profiling.
4832 #if defined(CONFIG_TCG) && defined(CONFIG_LINUX)
4833 DEF("perfmap", 0, QEMU_OPTION_perfmap,
4834 "-perfmap generate a /tmp/perf-${pid}.map file for perf\n",
4838 Generate a map file for Linux perf tools that will allow basic profiling
4839 information to be broken down into basic blocks.
4842 DEF("jitdump", 0, QEMU_OPTION_jitdump,
4843 "-jitdump generate a jit-${pid}.dump file for perf\n",
4847 Generate a dump file for Linux perf tools that maps basic blocks to symbol
4848 names, line numbers and JITted code.
4854 DEFHEADING(Generic object creation:)
4856 DEF("object", HAS_ARG, QEMU_OPTION_object,
4857 "-object TYPENAME[,PROP1=VALUE1,...]\n"
4858 " create a new object of type TYPENAME setting properties\n"
4859 " in the order they are specified. Note that the 'id'\n"
4860 " property must be set. These objects are placed in the\n"
4861 " '/objects' path.\n",
4864 ``-object typename[,prop1=value1,...]``
4865 Create a new object of type typename setting properties in the order
4866 they are specified. Note that the 'id' property must be set. These
4867 objects are placed in the '/objects' path.
4869 ``-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``
4870 Creates a memory file backend object, which can be used to back
4871 the guest RAM with huge pages.
4873 The ``id`` parameter is a unique ID that will be used to
4874 reference this memory region in other parameters, e.g. ``-numa``,
4875 ``-device nvdimm``, etc.
4877 The ``size`` option provides the size of the memory region, and
4878 accepts common suffixes, e.g. ``500M``.
4880 The ``mem-path`` provides the path to either a shared memory or
4881 huge page filesystem mount.
4883 The ``share`` boolean option determines whether the memory
4884 region is marked as private to QEMU, or shared. The latter
4885 allows a co-operating external process to access the QEMU memory
4888 The ``share`` is also required for pvrdma devices due to
4889 limitations in the RDMA API provided by Linux.
4891 Setting share=on might affect the ability to configure NUMA
4892 bindings for the memory backend under some circumstances, see
4893 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel
4894 source tree for additional details.
4896 Setting the ``discard-data`` boolean option to on indicates that
4897 file contents can be destroyed when QEMU exits, to avoid
4898 unnecessarily flushing data to the backing file. Note that
4899 ``discard-data`` is only an optimization, and QEMU might not
4900 discard file contents if it aborts unexpectedly or is terminated
4903 The ``merge`` boolean option enables memory merge, also known as
4904 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider
4905 the pages for memory deduplication.
4907 Setting the ``dump`` boolean option to off excludes the memory
4908 from core dumps. This feature is also known as MADV\_DONTDUMP.
4910 The ``prealloc`` boolean option enables memory preallocation.
4912 The ``host-nodes`` option binds the memory range to a list of
4915 The ``policy`` option sets the NUMA policy to one of the
4922 prefer the given host node list for allocation
4925 restrict memory allocation to the given host node list
4928 interleave memory allocations across the given host node
4931 The ``align`` option specifies the base address alignment when
4932 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg
4933 ``2M``. Some backend store specified by ``mem-path`` requires an
4934 alignment different than the default one used by QEMU, eg the
4935 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
4936 such cases, users can specify the required alignment via this
4939 The ``pmem`` option specifies whether the backing file specified
4940 by ``mem-path`` is in host persistent memory that can be
4941 accessed using the SNIA NVM programming model (e.g. Intel
4942 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary
4943 operations to guarantee the persistence of its own writes to
4944 ``mem-path`` (e.g. in vNVDIMM label emulation and live
4945 migration). Also, we will map the backend-file with MAP\_SYNC
4946 flag, which ensures the file metadata is in sync for
4947 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC
4948 requires support from both the host kernel (since Linux kernel
4949 4.15) and the filesystem of ``mem-path`` mounted with DAX
4952 The ``readonly`` option specifies whether the backing file is opened
4953 read-only or read-write (default).
4955 ``-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``
4956 Creates a memory backend object, which can be used to back the
4957 guest RAM. Memory backend objects offer more control than the
4958 ``-m`` option that is traditionally used to define guest RAM.
4959 Please refer to ``memory-backend-file`` for a description of the
4962 ``-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``
4963 Creates an anonymous memory file backend object, which allows
4964 QEMU to share the memory with an external process (e.g. when
4965 using vhost-user). The memory is allocated with memfd and
4966 optional sealing. (Linux only)
4968 The ``seal`` option creates a sealed-file, that will block
4969 further resizing the memory ('on' by default).
4971 The ``hugetlb`` option specify the file to be created resides in
4972 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction
4973 with the ``hugetlb`` option, the ``hugetlbsize`` option specify
4974 the hugetlb page size on systems that support multiple hugetlb
4975 page sizes (it must be a power of 2 value supported by the
4978 In some versions of Linux, the ``hugetlb`` option is
4979 incompatible with the ``seal`` option (requires at least Linux
4982 Please refer to ``memory-backend-file`` for a description of the
4985 The ``share`` boolean option is on by default with memfd.
4987 ``-object rng-builtin,id=id``
4988 Creates a random number generator backend which obtains entropy
4989 from QEMU builtin functions. The ``id`` parameter is a unique ID
4990 that will be used to reference this entropy backend from the
4991 ``virtio-rng`` device. By default, the ``virtio-rng`` device
4992 uses this RNG backend.
4994 ``-object rng-random,id=id,filename=/dev/random``
4995 Creates a random number generator backend which obtains entropy
4996 from a device on the host. The ``id`` parameter is a unique ID
4997 that will be used to reference this entropy backend from the
4998 ``virtio-rng`` device. The ``filename`` parameter specifies
4999 which file to obtain entropy from and if omitted defaults to
5002 ``-object rng-egd,id=id,chardev=chardevid``
5003 Creates a random number generator backend which obtains entropy
5004 from an external daemon running on the host. The ``id``
5005 parameter is a unique ID that will be used to reference this
5006 entropy backend from the ``virtio-rng`` device. The ``chardev``
5007 parameter is the unique ID of a character device backend that
5008 provides the connection to the RNG daemon.
5010 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off``
5011 Creates a TLS anonymous credentials object, which can be used to
5012 provide TLS support on network backends. The ``id`` parameter is
5013 a unique ID which network backends will use to access the
5014 credentials. The ``endpoint`` is either ``server`` or ``client``
5015 depending on whether the QEMU network backend that uses the
5016 credentials will be acting as a client or as a server. If
5017 ``verify-peer`` is enabled (the default) then once the handshake
5018 is completed, the peer credentials will be verified, though this
5019 is a no-op for anonymous credentials.
5021 The dir parameter tells QEMU where to find the credential files.
5022 For server endpoints, this directory may contain a file
5023 dh-params.pem providing diffie-hellman parameters to use for the
5024 TLS server. If the file is missing, QEMU will generate a set of
5025 DH parameters at startup. This is a computationally expensive
5026 operation that consumes random pool entropy, so it is
5027 recommended that a persistent set of parameters be generated
5030 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]``
5031 Creates a TLS Pre-Shared Keys (PSK) credentials object, which
5032 can be used to provide TLS support on network backends. The
5033 ``id`` parameter is a unique ID which network backends will use
5034 to access the credentials. The ``endpoint`` is either ``server``
5035 or ``client`` depending on whether the QEMU network backend that
5036 uses the credentials will be acting as a client or as a server.
5037 For clients only, ``username`` is the username which will be
5038 sent to the server. If omitted it defaults to "qemu".
5040 The dir parameter tells QEMU where to find the keys file. It is
5041 called "dir/keys.psk" and contains "username:key" pairs. This
5042 file can most easily be created using the GnuTLS ``psktool``
5045 For server endpoints, dir may also contain a file dh-params.pem
5046 providing diffie-hellman parameters to use for the TLS server.
5047 If the file is missing, QEMU will generate a set of DH
5048 parameters at startup. This is a computationally expensive
5049 operation that consumes random pool entropy, so it is
5050 recommended that a persistent set of parameters be generated up
5053 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id``
5054 Creates a TLS anonymous credentials object, which can be used to
5055 provide TLS support on network backends. The ``id`` parameter is
5056 a unique ID which network backends will use to access the
5057 credentials. The ``endpoint`` is either ``server`` or ``client``
5058 depending on whether the QEMU network backend that uses the
5059 credentials will be acting as a client or as a server. If
5060 ``verify-peer`` is enabled (the default) then once the handshake
5061 is completed, the peer credentials will be verified. With x509
5062 certificates, this implies that the clients must be provided
5063 with valid client certificates too.
5065 The dir parameter tells QEMU where to find the credential files.
5066 For server endpoints, this directory may contain a file
5067 dh-params.pem providing diffie-hellman parameters to use for the
5068 TLS server. If the file is missing, QEMU will generate a set of
5069 DH parameters at startup. This is a computationally expensive
5070 operation that consumes random pool entropy, so it is
5071 recommended that a persistent set of parameters be generated
5074 For x509 certificate credentials the directory will contain
5075 further files providing the x509 certificates. The certificates
5076 must be stored in PEM format, in filenames ca-cert.pem,
5077 ca-crl.pem (optional), server-cert.pem (only servers),
5078 server-key.pem (only servers), client-cert.pem (only clients),
5079 and client-key.pem (only clients).
5081 For the server-key.pem and client-key.pem files which contain
5082 sensitive private keys, it is possible to use an encrypted
5083 version by providing the passwordid parameter. This provides the
5084 ID of a previously created ``secret`` object containing the
5085 password for decryption.
5087 The priority parameter allows to override the global default
5088 priority used by gnutls. This can be useful if the system
5089 administrator needs to use a weaker set of crypto priorities for
5090 QEMU without potentially forcing the weakness onto all
5091 applications. Or conversely if one wants wants a stronger
5092 default for QEMU than for all other applications, they can do
5093 this through this parameter. Its format is a gnutls priority
5094 string as described at
5095 https://gnutls.org/manual/html_node/Priority-Strings.html.
5097 ``-object tls-cipher-suites,id=id,priority=priority``
5098 Creates a TLS cipher suites object, which can be used to control
5099 the TLS cipher/protocol algorithms that applications are permitted
5102 The ``id`` parameter is a unique ID which frontends will use to
5103 access the ordered list of permitted TLS cipher suites from the
5106 The ``priority`` parameter allows to override the global default
5107 priority used by gnutls. This can be useful if the system
5108 administrator needs to use a weaker set of crypto priorities for
5109 QEMU without potentially forcing the weakness onto all
5110 applications. Or conversely if one wants wants a stronger
5111 default for QEMU than for all other applications, they can do
5112 this through this parameter. Its format is a gnutls priority
5113 string as described at
5114 https://gnutls.org/manual/html_node/Priority-Strings.html.
5116 An example of use of this object is to control UEFI HTTPS Boot.
5117 The tls-cipher-suites object exposes the ordered list of permitted
5118 TLS cipher suites from the host side to the guest firmware, via
5119 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER
5120 objects. The firmware uses the IANA_TLS_CIPHER array for configuring
5123 In the following example, the priority at which the host-side policy
5124 is retrieved is given by the ``priority`` property.
5125 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to
5126 refer to /etc/crypto-policies/back-ends/gnutls.config.
5131 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\
5132 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0
5134 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]``
5135 Interval t can't be 0, this filter batches the packet delivery:
5136 all packets arriving in a given interval on netdev netdevid are
5137 delayed until the end of the interval. Interval is in
5138 microseconds. ``status`` is optional that indicate whether the
5139 netfilter is on (enabled) or off (disabled), the default status
5140 for netfilter will be 'on'.
5142 queue all\|rx\|tx is an option that can be applied to any
5145 ``all``: the filter is attached both to the receive and the
5146 transmit queue of the netdev (default).
5148 ``rx``: the filter is attached to the receive queue of the
5149 netdev, where it will receive packets sent to the netdev.
5151 ``tx``: the filter is attached to the transmit queue of the
5152 netdev, where it will receive packets sent by the netdev.
5154 position head\|tail\|id=<id> is an option to specify where the
5155 filter should be inserted in the filter list. It can be applied
5158 ``head``: the filter is inserted at the head of the filter list,
5159 before any existing filters.
5161 ``tail``: the filter is inserted at the tail of the filter list,
5162 behind any existing filters (default).
5164 ``id=<id>``: the filter is inserted before or behind the filter
5165 specified by <id>, see the insert option below.
5167 insert behind\|before is an option to specify where to insert
5168 the new filter relative to the one specified with
5169 position=id=<id>. It can be applied to any netfilter.
5171 ``before``: insert before the specified filter.
5173 ``behind``: insert behind the specified filter (default).
5175 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
5176 filter-mirror on netdev netdevid,mirror net packet to
5177 chardevchardevid, if it has the vnet\_hdr\_support flag,
5178 filter-mirror will mirror packet with vnet\_hdr\_len.
5180 ``-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]``
5181 filter-redirector on netdev netdevid,redirect filter's net
5182 packet to chardev chardevid,and redirect indev's packet to
5183 filter.if it has the vnet\_hdr\_support flag, filter-redirector
5184 will redirect packet with vnet\_hdr\_len. Create a
5185 filter-redirector we need to differ outdev id from indev id, id
5186 can not be the same. we can just use indev or outdev, but at
5187 least one of indev or outdev need to be specified.
5189 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
5190 Filter-rewriter is a part of COLO project.It will rewrite tcp
5191 packet to secondary from primary to keep secondary tcp
5192 connection,and rewrite tcp packet to primary from secondary make
5193 tcp packet can be handled by client.if it has the
5194 vnet\_hdr\_support flag, we can parse packet with vnet header.
5196 usage: colo secondary: -object
5197 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object
5198 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object
5199 filter-rewriter,id=rew0,netdev=hn0,queue=all
5201 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]``
5202 Dump the network traffic on netdev dev to the file specified by
5203 filename. At most len bytes (64k by default) per packet are
5204 stored. The file format is libpcap, so it can be analyzed with
5205 tools such as tcpdump or Wireshark.
5207 ``-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}]``
5208 Colo-compare gets packet from primary\_in chardevid and
5209 secondary\_in, then compare whether the payload of primary packet
5210 and secondary packet are the same. If same, it will output
5211 primary packet to out\_dev, else it will notify COLO-framework to do
5212 checkpoint and send primary packet to out\_dev. In order to
5213 improve efficiency, we need to put the task of comparison in
5214 another iothread. If it has the vnet\_hdr\_support flag,
5215 colo compare will send/recv packet with vnet\_hdr\_len.
5216 The compare\_timeout=@var{ms} determines the maximum time of the
5217 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms}
5218 is to set the period of scanning expired primary node network packets.
5219 The max\_queue\_size=@var{size} is to set the max compare queue
5220 size depend on user environment.
5221 If user want to use Xen COLO, need to add the notify\_dev to
5222 notify Xen colo-frame to do checkpoint.
5224 COLO-compare must be used with the help of filter-mirror,
5225 filter-redirector and filter-rewriter.
5232 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
5233 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
5234 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
5235 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
5236 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
5237 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
5238 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
5239 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
5240 -object iothread,id=iothread1
5241 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
5242 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
5243 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
5244 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1
5247 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
5248 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
5249 -chardev socket,id=red0,host=3.3.3.3,port=9003
5250 -chardev socket,id=red1,host=3.3.3.3,port=9004
5251 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
5252 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
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 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off
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 iothread,id=iothread1
5271 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1
5274 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
5275 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
5276 -chardev socket,id=red0,host=3.3.3.3,port=9003
5277 -chardev socket,id=red1,host=3.3.3.3,port=9004
5278 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
5279 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
5281 If you want to know the detail of above command line, you can
5282 read the colo-compare git log.
5284 ``-object cryptodev-backend-builtin,id=id[,queues=queues]``
5285 Creates a cryptodev backend which executes crypto operations from
5286 the QEMU cipher APIs. The id parameter is a unique ID that will
5287 be used to reference this cryptodev backend from the
5288 ``virtio-crypto`` device. The queues parameter is optional,
5289 which specify the queue number of cryptodev backend, the default
5296 -object cryptodev-backend-builtin,id=cryptodev0 \\
5297 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
5300 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]``
5301 Creates a vhost-user cryptodev backend, backed by a chardev
5302 chardevid. The id parameter is a unique ID that will be used to
5303 reference this cryptodev backend from the ``virtio-crypto``
5304 device. The chardev should be a unix domain socket backed one.
5305 The vhost-user uses a specifically defined protocol to pass
5306 vhost ioctl replacement messages to an application on the other
5307 end of the socket. The queues parameter is optional, which
5308 specify the queue number of cryptodev backend for multiqueue
5309 vhost-user, the default of queues is 1.
5315 -chardev socket,id=chardev0,path=/path/to/socket \\
5316 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\
5317 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
5320 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]``
5322 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]``
5323 Defines a secret to store a password, encryption key, or some
5324 other sensitive data. The sensitive data can either be passed
5325 directly via the data parameter, or indirectly via the file
5326 parameter. Using the data parameter is insecure unless the
5327 sensitive data is encrypted.
5329 The sensitive data can be provided in raw format (the default),
5330 or base64. When encoded as JSON, the raw format only supports
5331 valid UTF-8 characters, so base64 is recommended for sending
5332 binary data. QEMU will convert from which ever format is
5333 provided to the format it needs internally. eg, an RBD password
5334 can be provided in raw format, even though it will be base64
5335 encoded when passed onto the RBD sever.
5337 For added protection, it is possible to encrypt the data
5338 associated with a secret using the AES-256-CBC cipher. Use of
5339 encryption is indicated by providing the keyid and iv
5340 parameters. The keyid parameter provides the ID of a previously
5341 defined secret that contains the AES-256 decryption key. This
5342 key should be 32-bytes long and be base64 encoded. The iv
5343 parameter provides the random initialization vector used for
5344 encryption of this particular secret and should be a base64
5345 encrypted string of the 16-byte IV.
5347 The simplest (insecure) usage is to provide the secret inline
5351 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw
5353 The simplest secure usage is to provide the secret via a file
5355 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object
5356 secret,id=sec0,file=mypasswd.txt,format=raw
5358 For greater security, AES-256-CBC should be used. To illustrate
5359 usage, consider the openssl command line tool which can encrypt
5360 the data. Note that when encrypting, the plaintext must be
5361 padded to the cipher block size (32 bytes) using the standard
5362 PKCS#5/6 compatible padding algorithm.
5364 First a master key needs to be created in base64 encoding:
5368 # openssl rand -base64 32 > key.b64
5369 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"')
5371 Each secret to be encrypted needs to have a random
5372 initialization vector generated. These do not need to be kept
5377 # openssl rand -base64 16 > iv.b64
5378 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"')
5380 The secret to be defined can now be encrypted, in this case
5381 we're telling openssl to base64 encode the result, but it could
5382 be left as raw bytes if desired.
5386 # SECRET=$(printf "letmein" |
5387 openssl enc -aes-256-cbc -a -K $KEY -iv $IV)
5389 When launching QEMU, create a master secret pointing to
5390 ``key.b64`` and specify that to be used to decrypt the user
5391 password. Pass the contents of ``iv.b64`` to the second secret
5396 -object secret,id=secmaster0,format=base64,file=key.b64 \\
5397 -object secret,id=sec0,keyid=secmaster0,format=base64,\\
5398 data=$SECRET,iv=$(<iv.b64)
5400 ``-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]``
5401 Create a Secure Encrypted Virtualization (SEV) guest object,
5402 which can be used to provide the guest memory encryption support
5405 When memory encryption is enabled, one of the physical address
5406 bit (aka the C-bit) is utilized to mark if a memory page is
5407 protected. The ``cbitpos`` is used to provide the C-bit
5408 position. The C-bit position is Host family dependent hence user
5409 must provide this value. On EPYC, the value should be 47.
5411 When memory encryption is enabled, we loose certain bits in
5412 physical address space. The ``reduced-phys-bits`` is used to
5413 provide the number of bits we loose in physical address space.
5414 Similar to C-bit, the value is Host family dependent. On EPYC,
5415 the value should be 5.
5417 The ``sev-device`` provides the device file to use for
5418 communicating with the SEV firmware running inside AMD Secure
5419 Processor. The default device is '/dev/sev'. If hardware
5420 supports memory encryption then /dev/sev devices are created by
5423 The ``policy`` provides the guest policy to be enforced by the
5424 SEV firmware and restrict what configuration and operational
5425 commands can be performed on this guest by the hypervisor. The
5426 policy should be provided by the guest owner and is bound to the
5427 guest and cannot be changed throughout the lifetime of the
5428 guest. The default is 0.
5430 If guest ``policy`` allows sharing the key with another SEV
5431 guest then ``handle`` can be use to provide handle of the guest
5432 from which to share the key.
5434 The ``dh-cert-file`` and ``session-file`` provides the guest
5435 owner's Public Diffie-Hillman key defined in SEV spec. The PDH
5436 and session parameters are used for establishing a cryptographic
5437 session with the guest owner to negotiate keys used for
5438 attestation. The file must be encoded in base64.
5440 The ``kernel-hashes`` adds the hashes of given kernel/initrd/
5441 cmdline to a designated guest firmware page for measured Linux
5442 boot with -kernel. The default is off. (Since 6.2)
5444 e.g to launch a SEV guest
5448 # |qemu_system_x86| \\
5450 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \\
5451 -machine ...,memory-encryption=sev0 \\
5454 ``-object authz-simple,id=id,identity=string``
5455 Create an authorization object that will control access to
5458 The ``identity`` parameter is identifies the user and its format
5459 depends on the network service that authorization object is
5460 associated with. For authorizing based on TLS x509 certificates,
5461 the identity must be the x509 distinguished name. Note that care
5462 must be taken to escape any commas in the distinguished name.
5464 An example authorization object to validate a x509 distinguished
5465 name would look like:
5471 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\
5474 Note the use of quotes due to the x509 distinguished name
5475 containing whitespace, and escaping of ','.
5477 ``-object authz-listfile,id=id,filename=path,refresh=on|off``
5478 Create an authorization object that will control access to
5481 The ``filename`` parameter is the fully qualified path to a file
5482 containing the access control list rules in JSON format.
5484 An example set of rules that match against SASL usernames might
5491 { "match": "fred", "policy": "allow", "format": "exact" },
5492 { "match": "bob", "policy": "allow", "format": "exact" },
5493 { "match": "danb", "policy": "deny", "format": "glob" },
5494 { "match": "dan*", "policy": "allow", "format": "exact" },
5499 When checking access the object will iterate over all the rules
5500 and the first rule to match will have its ``policy`` value
5501 returned as the result. If no rules match, then the default
5502 ``policy`` value is returned.
5504 The rules can either be an exact string match, or they can use
5505 the simple UNIX glob pattern matching to allow wildcards to be
5508 If ``refresh`` is set to true the file will be monitored and
5509 automatically reloaded whenever its content changes.
5511 As with the ``authz-simple`` object, the format of the identity
5512 strings being matched depends on the network service, but is
5513 usually a TLS x509 distinguished name, or a SASL username.
5515 An example authorization object to validate a SASL username
5522 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\
5525 ``-object authz-pam,id=id,service=string``
5526 Create an authorization object that will control access to
5529 The ``service`` parameter provides the name of a PAM service to
5530 use for authorization. It requires that a file
5531 ``/etc/pam.d/service`` exist to provide the configuration for
5532 the ``account`` subsystem.
5534 An example authorization object to validate a TLS x509
5535 distinguished name would look like:
5541 -object authz-pam,id=auth0,service=qemu-vnc \\
5544 There would then be a corresponding config file for PAM at
5545 ``/etc/pam.d/qemu-vnc`` that contains:
5549 account requisite pam_listfile.so item=user sense=allow \
5550 file=/etc/qemu/vnc.allow
5552 Finally the ``/etc/qemu/vnc.allow`` file would contain the list
5553 of x509 distinguished names that are permitted access
5557 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB
5559 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch``
5560 Creates a dedicated event loop thread that devices can be
5561 assigned to. This is known as an IOThread. By default device
5562 emulation happens in vCPU threads or the main event loop thread.
5563 This can become a scalability bottleneck. IOThreads allow device
5564 emulation and I/O to run on other host CPUs.
5566 The ``id`` parameter is a unique ID that will be used to
5567 reference this IOThread from ``-device ...,iothread=id``.
5568 Multiple devices can be assigned to an IOThread. Note that not
5569 all devices support an ``iothread`` parameter.
5571 The ``query-iothreads`` QMP command lists IOThreads and reports
5572 their thread IDs so that the user can configure host CPU
5575 IOThreads use an adaptive polling algorithm to reduce event loop
5576 latency. Instead of entering a blocking system call to monitor
5577 file descriptors and then pay the cost of being woken up when an
5578 event occurs, the polling algorithm spins waiting for events for
5579 a short time. The algorithm's default parameters are suitable
5580 for many cases but can be adjusted based on knowledge of the
5581 workload and/or host device latency.
5583 The ``poll-max-ns`` parameter is the maximum number of
5584 nanoseconds to busy wait for events. Polling can be disabled by
5585 setting this value to 0.
5587 The ``poll-grow`` parameter is the multiplier used to increase
5588 the polling time when the algorithm detects it is missing events
5589 due to not polling long enough.
5591 The ``poll-shrink`` parameter is the divisor used to decrease
5592 the polling time when the algorithm detects it is spending too
5593 long polling without encountering events.
5595 The ``aio-max-batch`` parameter is the maximum number of requests
5596 in a batch for the AIO engine, 0 means that the engine will use
5599 The IOThread parameters can be modified at run-time using the
5600 ``qom-set`` command (where ``iothread1`` is the IOThread's
5605 (qemu) qom-set /objects/iothread1 poll-max-ns 100000
5609 HXCOMM This is the last statement. Insert new options before this line!
OSDN Git Service
