qemu-options: Remove the deprecated -no-acpi option
[qemu/kevin.git] / qemu-options.hx
blob844a189fb06a848e45643fea1d12178bf1727708
1 HXCOMM See docs/devel/docs.rst for the format of this file.
2 HXCOMM
3 HXCOMM Use DEFHEADING() to define headings in both help text and rST.
4 HXCOMM Text between SRST and ERST is copied to the rST version and
5 HXCOMM discarded from C version.
6 HXCOMM DEF(option, HAS_ARG/0, opt_enum, opt_help, arch_mask) is used to
7 HXCOMM construct option structures, enums and help message for specified
8 HXCOMM architectures.
9 HXCOMM HXCOMM can be used for comments, discarded from both rST and C.
11 DEFHEADING(Standard options:)
13 DEF("help", 0, QEMU_OPTION_h,
14 "-h or -help display this help and exit\n", QEMU_ARCH_ALL)
15 SRST
16 ``-h``
17 Display help and exit
18 ERST
20 DEF("version", 0, QEMU_OPTION_version,
21 "-version display version information and exit\n", QEMU_ARCH_ALL)
22 SRST
23 ``-version``
24 Display version information and exit
25 ERST
27 DEF("machine", HAS_ARG, QEMU_OPTION_machine, \
28 "-machine [type=]name[,prop[=value][,...]]\n"
29 " selects emulated machine ('-machine help' for list)\n"
30 " property accel=accel1[:accel2[:...]] selects accelerator\n"
31 " supported accelerators are kvm, xen, hvf, nvmm, whpx or tcg (default: tcg)\n"
32 " vmport=on|off|auto controls emulation of vmport (default: auto)\n"
33 " dump-guest-core=on|off include guest memory in a core dump (default=on)\n"
34 " mem-merge=on|off controls memory merge support (default: on)\n"
35 " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n"
36 " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n"
37 " suppress-vmdesc=on|off disables self-describing migration (default=off)\n"
38 " nvdimm=on|off controls NVDIMM support (default=off)\n"
39 " memory-encryption=@var{} memory encryption object to use (default=none)\n"
40 " hmat=on|off controls ACPI HMAT support (default=off)\n"
41 " memory-backend='backend-id' specifies explicitly provided backend for main RAM (default=none)\n"
42 " 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 QEMU_ARCH_ALL)
44 SRST
45 ``-machine [type=]name[,prop=value[,...]]``
46 Select the emulated machine by name. Use ``-machine help`` to list
47 available machines.
49 For architectures which aim to support live migration compatibility
50 across releases, each release will introduce a new versioned machine
51 type. For example, the 2.8.0 release introduced machine types
52 "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures.
54 To allow live migration of guests from QEMU version 2.8.0, to QEMU
55 version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8"
56 and "pc-q35-2.8" machines too. To allow users live migrating VMs to
57 skip multiple intermediate releases when upgrading, new releases of
58 QEMU will support machine types from many previous versions.
60 Supported machine properties are:
62 ``accel=accels1[:accels2[:...]]``
63 This is used to enable an accelerator. Depending on the target
64 architecture, kvm, xen, hvf, nvmm, whpx or tcg can be available.
65 By default, tcg is used. If there is more than one accelerator
66 specified, the next one is used if the previous one fails to
67 initialize.
69 ``vmport=on|off|auto``
70 Enables emulation of VMWare IO port, for vmmouse etc. auto says
71 to select the value based on accel. For accel=xen the default is
72 off otherwise the default is on.
74 ``dump-guest-core=on|off``
75 Include guest memory in a core dump. The default is on.
77 ``mem-merge=on|off``
78 Enables or disables memory merge support. This feature, when
79 supported by the host, de-duplicates identical memory pages
80 among VMs instances (enabled by default).
82 ``aes-key-wrap=on|off``
83 Enables or disables AES key wrapping support on s390-ccw hosts.
84 This feature controls whether AES wrapping keys will be created
85 to allow execution of AES cryptographic functions. The default
86 is on.
88 ``dea-key-wrap=on|off``
89 Enables or disables DEA key wrapping support on s390-ccw hosts.
90 This feature controls whether DEA wrapping keys will be created
91 to allow execution of DEA cryptographic functions. The default
92 is on.
94 ``nvdimm=on|off``
95 Enables or disables NVDIMM support. The default is off.
97 ``memory-encryption=``
98 Memory encryption object to use. The default is none.
100 ``hmat=on|off``
101 Enables or disables ACPI Heterogeneous Memory Attribute Table
102 (HMAT) support. The default is off.
104 ``memory-backend='id'``
105 An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options.
106 Allows to use a memory backend as main RAM.
108 For example:
111 -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on
112 -machine memory-backend=pc.ram
113 -m 512M
115 Migration compatibility note:
117 * as backend id one shall use value of 'default-ram-id', advertised by
118 machine type (available via ``query-machines`` QMP command), if migration
119 to/from old QEMU (<5.0) is expected.
120 * for machine types 4.0 and older, user shall
121 use ``x-use-canonical-path-for-ramblock-id=off`` backend option
122 if migration to/from old QEMU (<5.0) is expected.
124 For example:
127 -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off
128 -machine memory-backend=pc.ram
129 -m 512M
131 ``cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]``
132 Define a CXL Fixed Memory Window (CFMW).
134 Described in the CXL 2.0 ECN: CEDT CFMWS & QTG _DSM.
136 They are regions of Host Physical Addresses (HPA) on a system which
137 may be interleaved across one or more CXL host bridges. The system
138 software will assign particular devices into these windows and
139 configure the downstream Host-managed Device Memory (HDM) decoders
140 in root ports, switch ports and devices appropriately to meet the
141 interleave requirements before enabling the memory devices.
143 ``targets.X=target`` provides the mapping to CXL host bridges
144 which may be identified by the id provided in the -device entry.
145 Multiple entries are needed to specify all the targets when
146 the fixed memory window represents interleaved memory. X is the
147 target index from 0.
149 ``size=size`` sets the size of the CFMW. This must be a multiple of
150 256MiB. The region will be aligned to 256MiB but the location is
151 platform and configuration dependent.
153 ``interleave-granularity=granularity`` sets the granularity of
154 interleave. Default 256KiB. Only 256KiB, 512KiB, 1024KiB, 2048KiB
155 4096KiB, 8192KiB and 16384KiB granularities supported.
157 Example:
161 -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 ERST
164 DEF("M", HAS_ARG, QEMU_OPTION_M,
165 " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n",
166 QEMU_ARCH_ALL)
168 SRST
169 ``sgx-epc.0.memdev=@var{memid},sgx-epc.0.node=@var{numaid}``
170 Define an SGX EPC section.
171 ERST
173 DEF("cpu", HAS_ARG, QEMU_OPTION_cpu,
174 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL)
175 SRST
176 ``-cpu model``
177 Select CPU model (``-cpu help`` for list and additional feature
178 selection)
179 ERST
181 DEF("accel", HAS_ARG, QEMU_OPTION_accel,
182 "-accel [accel=]accelerator[,prop[=value][,...]]\n"
183 " select accelerator (kvm, xen, hvf, nvmm, whpx or tcg; use 'help' for a list)\n"
184 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n"
185 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n"
186 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n"
187 " one-insn-per-tb=on|off (one guest instruction per TCG translation block)\n"
188 " split-wx=on|off (enable TCG split w^x mapping)\n"
189 " tb-size=n (TCG translation block cache size)\n"
190 " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n"
191 " eager-split-size=n (KVM Eager Page Split chunk size, default 0, disabled. ARM only)\n"
192 " notify-vmexit=run|internal-error|disable,notify-window=n (enable notify VM exit and set notify window, x86 only)\n"
193 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL)
194 SRST
195 ``-accel name[,prop=value[,...]]``
196 This is used to enable an accelerator. Depending on the target
197 architecture, kvm, xen, hvf, nvmm, whpx or tcg can be available. By
198 default, tcg is used. If there is more than one accelerator
199 specified, the next one is used if the previous one fails to
200 initialize.
202 ``igd-passthru=on|off``
203 When Xen is in use, this option controls whether Intel
204 integrated graphics devices can be passed through to the guest
205 (default=off)
207 ``kernel-irqchip=on|off|split``
208 Controls KVM in-kernel irqchip support. The default is full
209 acceleration of the interrupt controllers. On x86, split irqchip
210 reduces the kernel attack surface, at a performance cost for
211 non-MSI interrupts. Disabling the in-kernel irqchip completely
212 is not recommended except for debugging purposes.
214 ``kvm-shadow-mem=size``
215 Defines the size of the KVM shadow MMU.
217 ``one-insn-per-tb=on|off``
218 Makes the TCG accelerator put only one guest instruction into
219 each translation block. This slows down emulation a lot, but
220 can be useful in some situations, such as when trying to analyse
221 the logs produced by the ``-d`` option.
223 ``split-wx=on|off``
224 Controls the use of split w^x mapping for the TCG code generation
225 buffer. Some operating systems require this to be enabled, and in
226 such a case this will default on. On other operating systems, this
227 will default off, but one may enable this for testing or debugging.
229 ``tb-size=n``
230 Controls the size (in MiB) of the TCG translation block cache.
232 ``thread=single|multi``
233 Controls number of TCG threads. When the TCG is multi-threaded
234 there will be one thread per vCPU therefore taking advantage of
235 additional host cores. The default is to enable multi-threading
236 where both the back-end and front-ends support it and no
237 incompatible TCG features have been enabled (e.g.
238 icount/replay).
240 ``dirty-ring-size=n``
241 When the KVM accelerator is used, it controls the size of the per-vCPU
242 dirty page ring buffer (number of entries for each vCPU). It should
243 be a value that is power of two, and it should be 1024 or bigger (but
244 still less than the maximum value that the kernel supports). 4096
245 could be a good initial value if you have no idea which is the best.
246 Set this value to 0 to disable the feature. By default, this feature
247 is disabled (dirty-ring-size=0). When enabled, KVM will instead
248 record dirty pages in a bitmap.
250 ``eager-split-size=n``
251 KVM implements dirty page logging at the PAGE_SIZE granularity and
252 enabling dirty-logging on a huge-page requires breaking it into
253 PAGE_SIZE pages in the first place. KVM on ARM does this splitting
254 lazily by default. There are performance benefits in doing huge-page
255 split eagerly, especially in situations where TLBI costs associated
256 with break-before-make sequences are considerable and also if guest
257 workloads are read intensive. The size here specifies how many pages
258 to break at a time and needs to be a valid block size which is
259 1GB/2MB/4KB, 32MB/16KB and 512MB/64KB for 4KB/16KB/64KB PAGE_SIZE
260 respectively. Be wary of specifying a higher size as it will have an
261 impact on the memory. By default, this feature is disabled
262 (eager-split-size=0).
264 ``notify-vmexit=run|internal-error|disable,notify-window=n``
265 Enables or disables notify VM exit support on x86 host and specify
266 the corresponding notify window to trigger the VM exit if enabled.
267 ``run`` option enables the feature. It does nothing and continue
268 if the exit happens. ``internal-error`` option enables the feature.
269 It raises a internal error. ``disable`` option doesn't enable the feature.
270 This feature can mitigate the CPU stuck issue due to event windows don't
271 open up for a specified of time (i.e. notify-window).
272 Default: notify-vmexit=run,notify-window=0.
274 ERST
276 DEF("smp", HAS_ARG, QEMU_OPTION_smp,
277 "-smp [[cpus=]n][,maxcpus=maxcpus][,drawers=drawers][,books=books][,sockets=sockets]\n"
278 " [,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]\n"
279 " set the number of initial CPUs to 'n' [default=1]\n"
280 " maxcpus= maximum number of total CPUs, including\n"
281 " offline CPUs for hotplug, etc\n"
282 " drawers= number of drawers on the machine board\n"
283 " books= number of books in one drawer\n"
284 " sockets= number of sockets in one book\n"
285 " dies= number of dies in one socket\n"
286 " clusters= number of clusters in one die\n"
287 " cores= number of cores in one cluster\n"
288 " threads= number of threads in one core\n"
289 "Note: Different machines may have different subsets of the CPU topology\n"
290 " parameters supported, so the actual meaning of the supported parameters\n"
291 " will vary accordingly. For example, for a machine type that supports a\n"
292 " three-level CPU hierarchy of sockets/cores/threads, the parameters will\n"
293 " sequentially mean as below:\n"
294 " sockets means the number of sockets on the machine board\n"
295 " cores means the number of cores in one socket\n"
296 " threads means the number of threads in one core\n"
297 " For a particular machine type board, an expected CPU topology hierarchy\n"
298 " can be defined through the supported sub-option. Unsupported parameters\n"
299 " can also be provided in addition to the sub-option, but their values\n"
300 " must be set as 1 in the purpose of correct parsing.\n",
301 QEMU_ARCH_ALL)
302 SRST
303 ``-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]``
304 Simulate a SMP system with '\ ``n``\ ' CPUs initially present on
305 the machine type board. On boards supporting CPU hotplug, the optional
306 '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be
307 added at runtime. When both parameters are omitted, the maximum number
308 of CPUs will be calculated from the provided topology members and the
309 initial CPU count will match the maximum number. When only one of them
310 is given then the omitted one will be set to its counterpart's value.
311 Both parameters may be specified, but the maximum number of CPUs must
312 be equal to or greater than the initial CPU count. Product of the
313 CPU topology hierarchy must be equal to the maximum number of CPUs.
314 Both parameters are subject to an upper limit that is determined by
315 the specific machine type chosen.
317 To control reporting of CPU topology information, values of the topology
318 parameters can be specified. Machines may only support a subset of the
319 parameters and different machines may have different subsets supported
320 which vary depending on capacity of the corresponding CPU targets. So
321 for a particular machine type board, an expected topology hierarchy can
322 be defined through the supported sub-option. Unsupported parameters can
323 also be provided in addition to the sub-option, but their values must be
324 set as 1 in the purpose of correct parsing.
326 Either the initial CPU count, or at least one of the topology parameters
327 must be specified. The specified parameters must be greater than zero,
328 explicit configuration like "cpus=0" is not allowed. Values for any
329 omitted parameters will be computed from those which are given.
331 For example, the following sub-option defines a CPU topology hierarchy
332 (2 sockets totally on the machine, 2 cores per socket, 2 threads per
333 core) for a machine that only supports sockets/cores/threads.
334 Some members of the option can be omitted but their values will be
335 automatically computed:
339 -smp 8,sockets=2,cores=2,threads=2,maxcpus=8
341 The following sub-option defines a CPU topology hierarchy (2 sockets
342 totally on the machine, 2 dies per socket, 2 cores per die, 2 threads
343 per core) for PC machines which support sockets/dies/cores/threads.
344 Some members of the option can be omitted but their values will be
345 automatically computed:
349 -smp 16,sockets=2,dies=2,cores=2,threads=2,maxcpus=16
351 The following sub-option defines a CPU topology hierarchy (2 sockets
352 totally on the machine, 2 clusters per socket, 2 cores per cluster,
353 2 threads per core) for ARM virt machines which support sockets/clusters
354 /cores/threads. Some members of the option can be omitted but their values
355 will be automatically computed:
359 -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16
361 Historically preference was given to the coarsest topology parameters
362 when computing missing values (ie sockets preferred over cores, which
363 were preferred over threads), however, this behaviour is considered
364 liable to change. Prior to 6.2 the preference was sockets over cores
365 over threads. Since 6.2 the preference is cores over sockets over threads.
367 For example, the following option defines a machine board with 2 sockets
368 of 1 core before 6.2 and 1 socket of 2 cores after 6.2:
372 -smp 2
374 Note: The cluster topology will only be generated in ACPI and exposed
375 to guest if it's explicitly specified in -smp.
376 ERST
378 DEF("numa", HAS_ARG, QEMU_OPTION_numa,
379 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
380 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
381 "-numa dist,src=source,dst=destination,val=distance\n"
382 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n"
383 "-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"
384 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n",
385 QEMU_ARCH_ALL)
386 SRST
387 ``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
389 ``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
391 ``-numa dist,src=source,dst=destination,val=distance``
393 ``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]``
395 ``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=type[,latency=lat][,bandwidth=bw]``
397 ``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]``
398 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA
399 distance from a source node to a destination node. Set the ACPI
400 Heterogeneous Memory Attributes for the given nodes.
402 Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and
403 lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a
404 contiguous range of CPU indexes (or a single VCPU if lastcpu is
405 omitted). A non-contiguous set of VCPUs can be represented by
406 providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is
407 omitted on all nodes, VCPUs are automatically split between them.
409 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a
410 NUMA node:
414 -numa node,cpus=0-2,cpus=5
416 '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option
417 which uses '\ ``socket-id|core-id|thread-id``\ ' properties to
418 assign CPU objects to a node using topology layout properties of
419 CPU. The set of properties is machine specific, and depends on used
420 machine type/'\ ``smp``\ ' options. It could be queried with
421 '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ '
422 property specifies node to which CPU object will be assigned, it's
423 required for node to be declared with '\ ``node``\ ' option before
424 it's used with '\ ``cpu``\ ' option.
426 For example:
430 -M pc \
431 -smp 1,sockets=2,maxcpus=2 \
432 -numa node,nodeid=0 -numa node,nodeid=1 \
433 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1
435 '\ ``memdev``\ ' option assigns RAM from a given memory backend
436 device to a node. It is recommended to use '\ ``memdev``\ ' option
437 over legacy '\ ``mem``\ ' option. This is because '\ ``memdev``\ '
438 option provides better performance and more control over the
439 backend's RAM (e.g. '\ ``prealloc``\ ' parameter of
440 '\ ``-memory-backend-ram``\ ' allows memory preallocation).
442 For compatibility reasons, legacy '\ ``mem``\ ' option is
443 supported in 5.0 and older machine types. Note that '\ ``mem``\ '
444 and '\ ``memdev``\ ' are mutually exclusive. If one node uses
445 '\ ``memdev``\ ', the rest nodes have to use '\ ``memdev``\ '
446 option, and vice versa.
448 Users must specify memory for all NUMA nodes by '\ ``memdev``\ '
449 (or legacy '\ ``mem``\ ' if available). In QEMU 5.2, the support
450 for '\ ``-numa node``\ ' without memory specified was removed.
452 '\ ``initiator``\ ' is an additional option that points to an
453 initiator NUMA node that has best performance (the lowest latency or
454 largest bandwidth) to this NUMA node. Note that this option can be
455 set only when the machine property 'hmat' is set to 'on'.
457 Following example creates a machine with 2 NUMA nodes, node 0 has
458 CPU. node 1 has only memory, and its initiator is node 0. Note that
459 because node 0 has CPU, by default the initiator of node 0 is itself
460 and must be itself.
464 -machine hmat=on \
465 -m 2G,slots=2,maxmem=4G \
466 -object memory-backend-ram,size=1G,id=m0 \
467 -object memory-backend-ram,size=1G,id=m1 \
468 -numa node,nodeid=0,memdev=m0 \
469 -numa node,nodeid=1,memdev=m1,initiator=0 \
470 -smp 2,sockets=2,maxcpus=2 \
471 -numa cpu,node-id=0,socket-id=0 \
472 -numa cpu,node-id=0,socket-id=1
474 source and destination are NUMA node IDs. distance is the NUMA
475 distance from source to destination. The distance from a node to
476 itself is always 10. If any pair of nodes is given a distance, then
477 all pairs must be given distances. Although, when distances are only
478 given in one direction for each pair of nodes, then the distances in
479 the opposite directions are assumed to be the same. If, however, an
480 asymmetrical pair of distances is given for even one node pair, then
481 all node pairs must be provided distance values for both directions,
482 even when they are symmetrical. When a node is unreachable from
483 another node, set the pair's distance to 255.
485 Note that the -``numa`` option doesn't allocate any of the specified
486 resources, it just assigns existing resources to NUMA nodes. This
487 means that one still has to use the ``-m``, ``-smp`` options to
488 allocate RAM and VCPUs respectively.
490 Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth
491 Information between initiator and target NUMA nodes in ACPI
492 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can
493 create memory requests, usually it has one or more processors.
494 Target NUMA node contains addressable memory.
496 In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is
497 the memory hierarchy of the target NUMA node: if hierarchy is
498 'memory', the structure represents the memory performance; if
499 hierarchy is 'first-level\|second-level\|third-level', this
500 structure represents aggregated performance of memory side caches
501 for each domain. type of 'data-type' is type of data represented by
502 this structure instance: if 'hierarchy' is 'memory', 'data-type' is
503 'access\|read\|write' latency or 'access\|read\|write' bandwidth of
504 the target memory; if 'hierarchy' is
505 'first-level\|second-level\|third-level', 'data-type' is
506 'access\|read\|write' hit latency or 'access\|read\|write' hit
507 bandwidth of the target memory side cache.
509 lat is latency value in nanoseconds. bw is bandwidth value, the
510 possible value and units are NUM[M\|G\|T], mean that the bandwidth
511 value are NUM byte per second (or MB/s, GB/s or TB/s depending on
512 used suffix). Note that if latency or bandwidth value is 0, means
513 the corresponding latency or bandwidth information is not provided.
515 In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory
516 belongs. size is the size of memory side cache in bytes. level is
517 the cache level described in this structure, note that the cache
518 level 0 should not be used with '\ ``hmat-cache``\ ' option.
519 associativity is the cache associativity, the possible value is
520 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy
521 is the write policy. line is the cache Line size in bytes.
523 For example, the following options describe 2 NUMA nodes. Node 0 has
524 2 cpus and a ram, node 1 has only a ram. The processors in node 0
525 access memory in node 0 with access-latency 5 nanoseconds,
526 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access
527 memory in NUMA node 1 with access-latency 10 nanoseconds,
528 access-bandwidth is 100 MB/s. And for memory side cache information,
529 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB,
530 policy is write-back, the cache Line size is 8 bytes:
534 -machine hmat=on \
535 -m 2G \
536 -object memory-backend-ram,size=1G,id=m0 \
537 -object memory-backend-ram,size=1G,id=m1 \
538 -smp 2,sockets=2,maxcpus=2 \
539 -numa node,nodeid=0,memdev=m0 \
540 -numa node,nodeid=1,memdev=m1,initiator=0 \
541 -numa cpu,node-id=0,socket-id=0 \
542 -numa cpu,node-id=0,socket-id=1 \
543 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \
544 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \
545 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \
546 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \
547 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \
548 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8
549 ERST
551 DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd,
552 "-add-fd fd=fd,set=set[,opaque=opaque]\n"
553 " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL)
554 SRST
555 ``-add-fd fd=fd,set=set[,opaque=opaque]``
556 Add a file descriptor to an fd set. Valid options are:
558 ``fd=fd``
559 This option defines the file descriptor of which a duplicate is
560 added to fd set. The file descriptor cannot be stdin, stdout, or
561 stderr.
563 ``set=set``
564 This option defines the ID of the fd set to add the file
565 descriptor to.
567 ``opaque=opaque``
568 This option defines a free-form string that can be used to
569 describe fd.
571 You can open an image using pre-opened file descriptors from an fd
572 set:
574 .. parsed-literal::
576 |qemu_system| \\
577 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
578 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
579 -drive file=/dev/fdset/2,index=0,media=disk
580 ERST
582 DEF("set", HAS_ARG, QEMU_OPTION_set,
583 "-set group.id.arg=value\n"
584 " set <arg> parameter for item <id> of type <group>\n"
585 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL)
586 SRST
587 ``-set group.id.arg=value``
588 Set parameter arg for item id of type group
589 ERST
591 DEF("global", HAS_ARG, QEMU_OPTION_global,
592 "-global driver.property=value\n"
593 "-global driver=driver,property=property,value=value\n"
594 " set a global default for a driver property\n",
595 QEMU_ARCH_ALL)
596 SRST
597 ``-global driver.prop=value``
599 ``-global driver=driver,property=property,value=value``
600 Set default value of driver's property prop to value, e.g.:
602 .. parsed-literal::
604 |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img
606 In particular, you can use this to set driver properties for devices
607 which are created automatically by the machine model. To create a
608 device which is not created automatically and set properties on it,
609 use -``device``.
611 -global driver.prop=value is shorthand for -global
612 driver=driver,property=prop,value=value. The longhand syntax works
613 even when driver contains a dot.
614 ERST
616 DEF("boot", HAS_ARG, QEMU_OPTION_boot,
617 "-boot [order=drives][,once=drives][,menu=on|off]\n"
618 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n"
619 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n"
620 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n"
621 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n"
622 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n",
623 QEMU_ARCH_ALL)
624 SRST
625 ``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]``
626 Specify boot order drives as a string of drive letters. Valid drive
627 letters depend on the target architecture. The x86 PC uses: a, b
628 (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p
629 (Etherboot from network adapter 1-4), hard disk boot is the default.
630 To apply a particular boot order only on the first startup, specify
631 it via ``once``. Note that the ``order`` or ``once`` parameter
632 should not be used together with the ``bootindex`` property of
633 devices, since the firmware implementations normally do not support
634 both at the same time.
636 Interactive boot menus/prompts can be enabled via ``menu=on`` as far
637 as firmware/BIOS supports them. The default is non-interactive boot.
639 A splash picture could be passed to bios, enabling user to show it
640 as logo, when option splash=sp\_name is given and menu=on, If
641 firmware/BIOS supports them. Currently Seabios for X86 system
642 support it. limitation: The splash file could be a jpeg file or a
643 BMP file in 24 BPP format(true color). The resolution should be
644 supported by the SVGA mode, so the recommended is 320x240, 640x480,
645 800x640.
647 A timeout could be passed to bios, guest will pause for rb\_timeout
648 ms when boot failed, then reboot. If rb\_timeout is '-1', guest will
649 not reboot, qemu passes '-1' to bios by default. Currently Seabios
650 for X86 system support it.
652 Do strict boot via ``strict=on`` as far as firmware/BIOS supports
653 it. This only effects when boot priority is changed by bootindex
654 options. The default is non-strict boot.
656 .. parsed-literal::
658 # try to boot from network first, then from hard disk
659 |qemu_system_x86| -boot order=nc
660 # boot from CD-ROM first, switch back to default order after reboot
661 |qemu_system_x86| -boot once=d
662 # boot with a splash picture for 5 seconds.
663 |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000
665 Note: The legacy format '-boot drives' is still supported but its
666 use is discouraged as it may be removed from future versions.
667 ERST
669 DEF("m", HAS_ARG, QEMU_OPTION_m,
670 "-m [size=]megs[,slots=n,maxmem=size]\n"
671 " configure guest RAM\n"
672 " size: initial amount of guest memory\n"
673 " slots: number of hotplug slots (default: none)\n"
674 " maxmem: maximum amount of guest memory (default: none)\n"
675 " Note: Some architectures might enforce a specific granularity\n",
676 QEMU_ARCH_ALL)
677 SRST
678 ``-m [size=]megs[,slots=n,maxmem=size]``
679 Sets guest startup RAM size to megs megabytes. Default is 128 MiB.
680 Optionally, a suffix of "M" or "G" can be used to signify a value in
681 megabytes or gigabytes respectively. Optional pair slots, maxmem
682 could be used to set amount of hotpluggable memory slots and maximum
683 amount of memory. Note that maxmem must be aligned to the page size.
685 For example, the following command-line sets the guest startup RAM
686 size to 1GB, creates 3 slots to hotplug additional memory and sets
687 the maximum memory the guest can reach to 4GB:
689 .. parsed-literal::
691 |qemu_system| -m 1G,slots=3,maxmem=4G
693 If slots and maxmem are not specified, memory hotplug won't be
694 enabled and the guest startup RAM will never increase.
695 ERST
697 DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath,
698 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL)
699 SRST
700 ``-mem-path path``
701 Allocate guest RAM from a temporarily created file in path.
702 ERST
704 DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc,
705 "-mem-prealloc preallocate guest memory (use with -mem-path)\n",
706 QEMU_ARCH_ALL)
707 SRST
708 ``-mem-prealloc``
709 Preallocate memory when using -mem-path.
710 ERST
712 DEF("k", HAS_ARG, QEMU_OPTION_k,
713 "-k language use keyboard layout (for example 'fr' for French)\n",
714 QEMU_ARCH_ALL)
715 SRST
716 ``-k language``
717 Use keyboard layout language (for example ``fr`` for French). This
718 option is only needed where it is not easy to get raw PC keycodes
719 (e.g. on Macs, with some X11 servers or with a VNC or curses
720 display). You don't normally need to use it on PC/Linux or
721 PC/Windows hosts.
723 The available layouts are:
727 ar de-ch es fo fr-ca hu ja mk no pt-br sv
728 da en-gb et fr fr-ch is lt nl pl ru th
729 de en-us fi fr-be hr it lv nl-be pt sl tr
731 The default is ``en-us``.
732 ERST
735 DEF("audio", HAS_ARG, QEMU_OPTION_audio,
736 "-audio [driver=]driver[,prop[=value][,...]]\n"
737 " specifies default audio backend when `audiodev` is not\n"
738 " used to create a machine or sound device;"
739 " options are the same as for -audiodev\n"
740 "-audio [driver=]driver,model=value[,prop[=value][,...]]\n"
741 " specifies the audio backend and device to use;\n"
742 " apart from 'model', options are the same as for -audiodev.\n"
743 " use '-audio model=help' to show possible devices.\n",
744 QEMU_ARCH_ALL)
745 SRST
746 ``-audio [driver=]driver[,model=value][,prop[=value][,...]]``
747 If the ``model`` option is specified, ``-audio`` is a shortcut
748 for configuring both the guest audio hardware and the host audio
749 backend in one go. The guest hardware model can be set with
750 ``model=modelname``. Use ``model=help`` to list the available
751 device types.
753 The following two example do exactly the same, to show how ``-audio``
754 can be used to shorten the command line length:
756 .. parsed-literal::
758 |qemu_system| -audiodev pa,id=pa -device sb16,audiodev=pa
759 |qemu_system| -audio pa,model=sb16
761 If the ``model`` option is not specified, ``-audio`` is used to
762 configure a default audio backend that will be used whenever the
763 ``audiodev`` property is not set on a device or machine. In
764 particular, ``-audio none`` ensures that no audio is produced even
765 for machines that have embedded sound hardware.
767 In both cases, the driver option is the same as with the corresponding
768 ``-audiodev`` option below. Use ``driver=help`` to list the available
769 drivers.
771 ERST
773 DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev,
774 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n"
775 " specifies the audio backend to use\n"
776 " Use ``-audiodev help`` to list the available drivers\n"
777 " id= identifier of the backend\n"
778 " timer-period= timer period in microseconds\n"
779 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n"
780 " in|out.fixed-settings= use fixed settings for host audio\n"
781 " in|out.frequency= frequency to use with fixed settings\n"
782 " in|out.channels= number of channels to use with fixed settings\n"
783 " in|out.format= sample format to use with fixed settings\n"
784 " valid values: s8, s16, s32, u8, u16, u32, f32\n"
785 " in|out.voices= number of voices to use\n"
786 " in|out.buffer-length= length of buffer in microseconds\n"
787 "-audiodev none,id=id,[,prop[=value][,...]]\n"
788 " dummy driver that discards all output\n"
789 #ifdef CONFIG_AUDIO_ALSA
790 "-audiodev alsa,id=id[,prop[=value][,...]]\n"
791 " in|out.dev= name of the audio device to use\n"
792 " in|out.period-length= length of period in microseconds\n"
793 " in|out.try-poll= attempt to use poll mode\n"
794 " threshold= threshold (in microseconds) when playback starts\n"
795 #endif
796 #ifdef CONFIG_AUDIO_COREAUDIO
797 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n"
798 " in|out.buffer-count= number of buffers\n"
799 #endif
800 #ifdef CONFIG_AUDIO_DSOUND
801 "-audiodev dsound,id=id[,prop[=value][,...]]\n"
802 " latency= add extra latency to playback in microseconds\n"
803 #endif
804 #ifdef CONFIG_AUDIO_OSS
805 "-audiodev oss,id=id[,prop[=value][,...]]\n"
806 " in|out.dev= path of the audio device to use\n"
807 " in|out.buffer-count= number of buffers\n"
808 " in|out.try-poll= attempt to use poll mode\n"
809 " try-mmap= try using memory mapped access\n"
810 " exclusive= open device in exclusive mode\n"
811 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n"
812 #endif
813 #ifdef CONFIG_AUDIO_PA
814 "-audiodev pa,id=id[,prop[=value][,...]]\n"
815 " server= PulseAudio server address\n"
816 " in|out.name= source/sink device name\n"
817 " in|out.latency= desired latency in microseconds\n"
818 #endif
819 #ifdef CONFIG_AUDIO_PIPEWIRE
820 "-audiodev pipewire,id=id[,prop[=value][,...]]\n"
821 " in|out.name= source/sink device name\n"
822 " in|out.stream-name= name of pipewire stream\n"
823 " in|out.latency= desired latency in microseconds\n"
824 #endif
825 #ifdef CONFIG_AUDIO_SDL
826 "-audiodev sdl,id=id[,prop[=value][,...]]\n"
827 " in|out.buffer-count= number of buffers\n"
828 #endif
829 #ifdef CONFIG_AUDIO_SNDIO
830 "-audiodev sndio,id=id[,prop[=value][,...]]\n"
831 #endif
832 #ifdef CONFIG_SPICE
833 "-audiodev spice,id=id[,prop[=value][,...]]\n"
834 #endif
835 #ifdef CONFIG_DBUS_DISPLAY
836 "-audiodev dbus,id=id[,prop[=value][,...]]\n"
837 #endif
838 "-audiodev wav,id=id[,prop[=value][,...]]\n"
839 " path= path of wav file to record\n",
840 QEMU_ARCH_ALL)
841 SRST
842 ``-audiodev [driver=]driver,id=id[,prop[=value][,...]]``
843 Adds a new audio backend driver identified by id. There are global
844 and driver specific properties. Some values can be set differently
845 for input and output, they're marked with ``in|out.``. You can set
846 the input's property with ``in.prop`` and the output's property with
847 ``out.prop``. For example:
851 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000
852 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified
854 NOTE: parameter validation is known to be incomplete, in many cases
855 specifying an invalid option causes QEMU to print an error message
856 and continue emulation without sound.
858 Valid global options are:
860 ``id=identifier``
861 Identifies the audio backend.
863 ``timer-period=period``
864 Sets the timer period used by the audio subsystem in
865 microseconds. Default is 10000 (10 ms).
867 ``in|out.mixing-engine=on|off``
868 Use QEMU's mixing engine to mix all streams inside QEMU and
869 convert audio formats when not supported by the backend. When
870 off, fixed-settings must be off too. Note that disabling this
871 option means that the selected backend must support multiple
872 streams and the audio formats used by the virtual cards,
873 otherwise you'll get no sound. It's not recommended to disable
874 this option unless you want to use 5.1 or 7.1 audio, as mixing
875 engine only supports mono and stereo audio. Default is on.
877 ``in|out.fixed-settings=on|off``
878 Use fixed settings for host audio. When off, it will change
879 based on how the guest opens the sound card. In this case you
880 must not specify frequency, channels or format. Default is on.
882 ``in|out.frequency=frequency``
883 Specify the frequency to use when using fixed-settings. Default
884 is 44100Hz.
886 ``in|out.channels=channels``
887 Specify the number of channels to use when using fixed-settings.
888 Default is 2 (stereo).
890 ``in|out.format=format``
891 Specify the sample format to use when using fixed-settings.
892 Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``,
893 ``u32``, ``f32``. Default is ``s16``.
895 ``in|out.voices=voices``
896 Specify the number of voices to use. Default is 1.
898 ``in|out.buffer-length=usecs``
899 Sets the size of the buffer in microseconds.
901 ``-audiodev none,id=id[,prop[=value][,...]]``
902 Creates a dummy backend that discards all outputs. This backend has
903 no backend specific properties.
905 ``-audiodev alsa,id=id[,prop[=value][,...]]``
906 Creates backend using the ALSA. This backend is only available on
907 Linux.
909 ALSA specific options are:
911 ``in|out.dev=device``
912 Specify the ALSA device to use for input and/or output. Default
913 is ``default``.
915 ``in|out.period-length=usecs``
916 Sets the period length in microseconds.
918 ``in|out.try-poll=on|off``
919 Attempt to use poll mode with the device. Default is on.
921 ``threshold=threshold``
922 Threshold (in microseconds) when playback starts. Default is 0.
924 ``-audiodev coreaudio,id=id[,prop[=value][,...]]``
925 Creates a backend using Apple's Core Audio. This backend is only
926 available on Mac OS and only supports playback.
928 Core Audio specific options are:
930 ``in|out.buffer-count=count``
931 Sets the count of the buffers.
933 ``-audiodev dsound,id=id[,prop[=value][,...]]``
934 Creates a backend using Microsoft's DirectSound. This backend is
935 only available on Windows and only supports playback.
937 DirectSound specific options are:
939 ``latency=usecs``
940 Add extra usecs microseconds latency to playback. Default is
941 10000 (10 ms).
943 ``-audiodev oss,id=id[,prop[=value][,...]]``
944 Creates a backend using OSS. This backend is available on most
945 Unix-like systems.
947 OSS specific options are:
949 ``in|out.dev=device``
950 Specify the file name of the OSS device to use. Default is
951 ``/dev/dsp``.
953 ``in|out.buffer-count=count``
954 Sets the count of the buffers.
956 ``in|out.try-poll=on|of``
957 Attempt to use poll mode with the device. Default is on.
959 ``try-mmap=on|off``
960 Try using memory mapped device access. Default is off.
962 ``exclusive=on|off``
963 Open the device in exclusive mode (vmix won't work in this
964 case). Default is off.
966 ``dsp-policy=policy``
967 Sets the timing policy (between 0 and 10, where smaller number
968 means smaller latency but higher CPU usage). Use -1 to use
969 buffer sizes specified by ``buffer`` and ``buffer-count``. This
970 option is ignored if you do not have OSS 4. Default is 5.
972 ``-audiodev pa,id=id[,prop[=value][,...]]``
973 Creates a backend using PulseAudio. This backend is available on
974 most systems.
976 PulseAudio specific options are:
978 ``server=server``
979 Sets the PulseAudio server to connect to.
981 ``in|out.name=sink``
982 Use the specified source/sink for recording/playback.
984 ``in|out.latency=usecs``
985 Desired latency in microseconds. The PulseAudio server will try
986 to honor this value but actual latencies may be lower or higher.
988 ``-audiodev pipewire,id=id[,prop[=value][,...]]``
989 Creates a backend using PipeWire. This backend is available on
990 most systems.
992 PipeWire specific options are:
994 ``in|out.latency=usecs``
995 Desired latency in microseconds.
997 ``in|out.name=sink``
998 Use the specified source/sink for recording/playback.
1000 ``in|out.stream-name``
1001 Specify the name of pipewire stream.
1003 ``-audiodev sdl,id=id[,prop[=value][,...]]``
1004 Creates a backend using SDL. This backend is available on most
1005 systems, but you should use your platform's native backend if
1006 possible.
1008 SDL specific options are:
1010 ``in|out.buffer-count=count``
1011 Sets the count of the buffers.
1013 ``-audiodev sndio,id=id[,prop[=value][,...]]``
1014 Creates a backend using SNDIO. This backend is available on
1015 OpenBSD and most other Unix-like systems.
1017 Sndio specific options are:
1019 ``in|out.dev=device``
1020 Specify the sndio device to use for input and/or output. Default
1021 is ``default``.
1023 ``in|out.latency=usecs``
1024 Sets the desired period length in microseconds.
1026 ``-audiodev spice,id=id[,prop[=value][,...]]``
1027 Creates a backend that sends audio through SPICE. This backend
1028 requires ``-spice`` and automatically selected in that case, so
1029 usually you can ignore this option. This backend has no backend
1030 specific properties.
1032 ``-audiodev wav,id=id[,prop[=value][,...]]``
1033 Creates a backend that writes audio to a WAV file.
1035 Backend specific options are:
1037 ``path=path``
1038 Write recorded audio into the specified file. Default is
1039 ``qemu.wav``.
1040 ERST
1042 DEF("device", HAS_ARG, QEMU_OPTION_device,
1043 "-device driver[,prop[=value][,...]]\n"
1044 " add device (based on driver)\n"
1045 " prop=value,... sets driver properties\n"
1046 " use '-device help' to print all possible drivers\n"
1047 " use '-device driver,help' to print all possible properties\n",
1048 QEMU_ARCH_ALL)
1049 SRST
1050 ``-device driver[,prop[=value][,...]]``
1051 Add device driver. prop=value sets driver properties. Valid
1052 properties depend on the driver. To get help on possible drivers and
1053 properties, use ``-device help`` and ``-device driver,help``.
1055 Some drivers are:
1057 ``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]``
1058 Add an IPMI BMC. This is a simulation of a hardware management
1059 interface processor that normally sits on a system. It provides a
1060 watchdog and the ability to reset and power control the system. You
1061 need to connect this to an IPMI interface to make it useful
1063 The IPMI slave address to use for the BMC. The default is 0x20. This
1064 address is the BMC's address on the I2C network of management
1065 controllers. If you don't know what this means, it is safe to ignore
1068 ``id=id``
1069 The BMC id for interfaces to use this device.
1071 ``slave_addr=val``
1072 Define slave address to use for the BMC. The default is 0x20.
1074 ``sdrfile=file``
1075 file containing raw Sensor Data Records (SDR) data. The default
1076 is none.
1078 ``fruareasize=val``
1079 size of a Field Replaceable Unit (FRU) area. The default is
1080 1024.
1082 ``frudatafile=file``
1083 file containing raw Field Replaceable Unit (FRU) inventory data.
1084 The default is none.
1086 ``guid=uuid``
1087 value for the GUID for the BMC, in standard UUID format. If this
1088 is set, get "Get GUID" command to the BMC will return it.
1089 Otherwise "Get GUID" will return an error.
1091 ``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]``
1092 Add a connection to an external IPMI BMC simulator. Instead of
1093 locally emulating the BMC like the above item, instead connect to an
1094 external entity that provides the IPMI services.
1096 A connection is made to an external BMC simulator. If you do this,
1097 it is strongly recommended that you use the "reconnect=" chardev
1098 option to reconnect to the simulator if the connection is lost. Note
1099 that if this is not used carefully, it can be a security issue, as
1100 the interface has the ability to send resets, NMIs, and power off
1101 the VM. It's best if QEMU makes a connection to an external
1102 simulator running on a secure port on localhost, so neither the
1103 simulator nor QEMU is exposed to any outside network.
1105 See the "lanserv/README.vm" file in the OpenIPMI library for more
1106 details on the external interface.
1108 ``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]``
1109 Add a KCS IPMI interface on the ISA bus. This also adds a
1110 corresponding ACPI and SMBIOS entries, if appropriate.
1112 ``bmc=id``
1113 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern
1114 above.
1116 ``ioport=val``
1117 Define the I/O address of the interface. The default is 0xca0
1118 for KCS.
1120 ``irq=val``
1121 Define the interrupt to use. The default is 5. To disable
1122 interrupts, set this to 0.
1124 ``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]``
1125 Like the KCS interface, but defines a BT interface. The default port
1126 is 0xe4 and the default interrupt is 5.
1128 ``-device pci-ipmi-kcs,bmc=id``
1129 Add a KCS IPMI interface on the PCI bus.
1131 ``bmc=id``
1132 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above.
1134 ``-device pci-ipmi-bt,bmc=id``
1135 Like the KCS interface, but defines a BT interface on the PCI bus.
1137 ``-device intel-iommu[,option=...]``
1138 This is only supported by ``-machine q35``, which will enable Intel VT-d
1139 emulation within the guest. It supports below options:
1141 ``intremap=on|off`` (default: auto)
1142 This enables interrupt remapping feature. It's required to enable
1143 complete x2apic. Currently it only supports kvm kernel-irqchip modes
1144 ``off`` or ``split``, while full kernel-irqchip is not yet supported.
1145 The default value is "auto", which will be decided by the mode of
1146 kernel-irqchip.
1148 ``caching-mode=on|off`` (default: off)
1149 This enables caching mode for the VT-d emulated device. When
1150 caching-mode is enabled, each guest DMA buffer mapping will generate an
1151 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in
1152 a synchronous way. It is required for ``-device vfio-pci`` to work
1153 with the VT-d device, because host assigned devices requires to setup
1154 the DMA mapping on the host before guest DMA starts.
1156 ``device-iotlb=on|off`` (default: off)
1157 This enables device-iotlb capability for the emulated VT-d device. So
1158 far virtio/vhost should be the only real user for this parameter,
1159 paired with ats=on configured for the device.
1161 ``aw-bits=39|48`` (default: 39)
1162 This decides the address width of IOVA address space. The address
1163 space has 39 bits width for 3-level IOMMU page tables, and 48 bits for
1164 4-level IOMMU page tables.
1166 Please also refer to the wiki page for general scenarios of VT-d
1167 emulation in QEMU: https://wiki.qemu.org/Features/VT-d.
1169 ERST
1171 DEF("name", HAS_ARG, QEMU_OPTION_name,
1172 "-name string1[,process=string2][,debug-threads=on|off]\n"
1173 " set the name of the guest\n"
1174 " string1 sets the window title and string2 the process name\n"
1175 " When debug-threads is enabled, individual threads are given a separate name\n"
1176 " NOTE: The thread names are for debugging and not a stable API.\n",
1177 QEMU_ARCH_ALL)
1178 SRST
1179 ``-name name``
1180 Sets the name of the guest. This name will be displayed in the SDL
1181 window caption. The name will also be used for the VNC server. Also
1182 optionally set the top visible process name in Linux. Naming of
1183 individual threads can also be enabled on Linux to aid debugging.
1184 ERST
1186 DEF("uuid", HAS_ARG, QEMU_OPTION_uuid,
1187 "-uuid %08x-%04x-%04x-%04x-%012x\n"
1188 " specify machine UUID\n", QEMU_ARCH_ALL)
1189 SRST
1190 ``-uuid uuid``
1191 Set system UUID.
1192 ERST
1194 DEFHEADING()
1196 DEFHEADING(Block device options:)
1198 SRST
1199 The QEMU block device handling options have a long history and
1200 have gone through several iterations as the feature set and complexity
1201 of the block layer have grown. Many online guides to QEMU often
1202 reference older and deprecated options, which can lead to confusion.
1204 The most explicit way to describe disks is to use a combination of
1205 ``-device`` to specify the hardware device and ``-blockdev`` to
1206 describe the backend. The device defines what the guest sees and the
1207 backend describes how QEMU handles the data. It is the only guaranteed
1208 stable interface for describing block devices and as such is
1209 recommended for management tools and scripting.
1211 The ``-drive`` option combines the device and backend into a single
1212 command line option which is a more human friendly. There is however no
1213 interface stability guarantee although some older board models still
1214 need updating to work with the modern blockdev forms.
1216 Older options like ``-hda`` are essentially macros which expand into
1217 ``-drive`` options for various drive interfaces. The original forms
1218 bake in a lot of assumptions from the days when QEMU was emulating a
1219 legacy PC, they are not recommended for modern configurations.
1221 ERST
1223 DEF("fda", HAS_ARG, QEMU_OPTION_fda,
1224 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
1225 DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
1226 SRST
1227 ``-fda file``
1229 ``-fdb file``
1230 Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in
1231 the System Emulation Users Guide).
1232 ERST
1234 DEF("hda", HAS_ARG, QEMU_OPTION_hda,
1235 "-hda/-hdb file use 'file' as hard disk 0/1 image\n", QEMU_ARCH_ALL)
1236 DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL)
1237 DEF("hdc", HAS_ARG, QEMU_OPTION_hdc,
1238 "-hdc/-hdd file use 'file' as hard disk 2/3 image\n", QEMU_ARCH_ALL)
1239 DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL)
1240 SRST
1241 ``-hda file``
1243 ``-hdb file``
1245 ``-hdc file``
1247 ``-hdd file``
1248 Use file as hard disk 0, 1, 2 or 3 image on the default bus of the
1249 emulated machine (this is for example the IDE bus on most x86 machines,
1250 but it can also be SCSI, virtio or something else on other target
1251 architectures). See also the :ref:`disk images` chapter in the System
1252 Emulation Users Guide.
1253 ERST
1255 DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom,
1256 "-cdrom file use 'file' as CD-ROM image\n",
1257 QEMU_ARCH_ALL)
1258 SRST
1259 ``-cdrom file``
1260 Use file as CD-ROM image on the default bus of the emulated machine
1261 (which is IDE1 master on x86, so you cannot use ``-hdc`` and ``-cdrom``
1262 at the same time there). On systems that support it, you can use the
1263 host CD-ROM by using ``/dev/cdrom`` as filename.
1264 ERST
1266 DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev,
1267 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n"
1268 " [,cache.direct=on|off][,cache.no-flush=on|off]\n"
1269 " [,read-only=on|off][,auto-read-only=on|off]\n"
1270 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n"
1271 " [,driver specific parameters...]\n"
1272 " configure a block backend\n", QEMU_ARCH_ALL)
1273 SRST
1274 ``-blockdev option[,option[,option[,...]]]``
1275 Define a new block driver node. Some of the options apply to all
1276 block drivers, other options are only accepted for a specific block
1277 driver. See below for a list of generic options and options for the
1278 most common block drivers.
1280 Options that expect a reference to another node (e.g. ``file``) can
1281 be given in two ways. Either you specify the node name of an already
1282 existing node (file=node-name), or you define a new node inline,
1283 adding options for the referenced node after a dot
1284 (file.filename=path,file.aio=native).
1286 A block driver node created with ``-blockdev`` can be used for a
1287 guest device by specifying its node name for the ``drive`` property
1288 in a ``-device`` argument that defines a block device.
1290 ``Valid options for any block driver node:``
1291 ``driver``
1292 Specifies the block driver to use for the given node.
1294 ``node-name``
1295 This defines the name of the block driver node by which it
1296 will be referenced later. The name must be unique, i.e. it
1297 must not match the name of a different block driver node, or
1298 (if you use ``-drive`` as well) the ID of a drive.
1300 If no node name is specified, it is automatically generated.
1301 The generated node name is not intended to be predictable
1302 and changes between QEMU invocations. For the top level, an
1303 explicit node name must be specified.
1305 ``read-only``
1306 Open the node read-only. Guest write attempts will fail.
1308 Note that some block drivers support only read-only access,
1309 either generally or in certain configurations. In this case,
1310 the default value ``read-only=off`` does not work and the
1311 option must be specified explicitly.
1313 ``auto-read-only``
1314 If ``auto-read-only=on`` is set, QEMU may fall back to
1315 read-only usage even when ``read-only=off`` is requested, or
1316 even switch between modes as needed, e.g. depending on
1317 whether the image file is writable or whether a writing user
1318 is attached to the node.
1320 ``force-share``
1321 Override the image locking system of QEMU by forcing the
1322 node to utilize weaker shared access for permissions where
1323 it would normally request exclusive access. When there is
1324 the potential for multiple instances to have the same file
1325 open (whether this invocation of QEMU is the first or the
1326 second instance), both instances must permit shared access
1327 for the second instance to succeed at opening the file.
1329 Enabling ``force-share=on`` requires ``read-only=on``.
1331 ``cache.direct``
1332 The host page cache can be avoided with ``cache.direct=on``.
1333 This will attempt to do disk IO directly to the guest's
1334 memory. QEMU may still perform an internal copy of the data.
1336 ``cache.no-flush``
1337 In case you don't care about data integrity over host
1338 failures, you can use ``cache.no-flush=on``. This option
1339 tells QEMU that it never needs to write any data to the disk
1340 but can instead keep things in cache. If anything goes
1341 wrong, like your host losing power, the disk storage getting
1342 disconnected accidentally, etc. your image will most
1343 probably be rendered unusable.
1345 ``discard=discard``
1346 discard is one of "ignore" (or "off") or "unmap" (or "on")
1347 and controls whether ``discard`` (also known as ``trim`` or
1348 ``unmap``) requests are ignored or passed to the filesystem.
1349 Some machine types may not support discard requests.
1351 ``detect-zeroes=detect-zeroes``
1352 detect-zeroes is "off", "on" or "unmap" and enables the
1353 automatic conversion of plain zero writes by the OS to
1354 driver specific optimized zero write commands. You may even
1355 choose "unmap" if discard is set to "unmap" to allow a zero
1356 write to be converted to an ``unmap`` operation.
1358 ``Driver-specific options for file``
1359 This is the protocol-level block driver for accessing regular
1360 files.
1362 ``filename``
1363 The path to the image file in the local filesystem
1365 ``aio``
1366 Specifies the AIO backend (threads/native/io_uring,
1367 default: threads)
1369 ``locking``
1370 Specifies whether the image file is protected with Linux OFD
1371 / POSIX locks. The default is to use the Linux Open File
1372 Descriptor API if available, otherwise no lock is applied.
1373 (auto/on/off, default: auto)
1375 Example:
1379 -blockdev driver=file,node-name=disk,filename=disk.img
1381 ``Driver-specific options for raw``
1382 This is the image format block driver for raw images. It is
1383 usually stacked on top of a protocol level block driver such as
1384 ``file``.
1386 ``file``
1387 Reference to or definition of the data source block driver
1388 node (e.g. a ``file`` driver node)
1390 Example 1:
1394 -blockdev driver=file,node-name=disk_file,filename=disk.img
1395 -blockdev driver=raw,node-name=disk,file=disk_file
1397 Example 2:
1401 -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img
1403 ``Driver-specific options for qcow2``
1404 This is the image format block driver for qcow2 images. It is
1405 usually stacked on top of a protocol level block driver such as
1406 ``file``.
1408 ``file``
1409 Reference to or definition of the data source block driver
1410 node (e.g. a ``file`` driver node)
1412 ``backing``
1413 Reference to or definition of the backing file block device
1414 (default is taken from the image file). It is allowed to
1415 pass ``null`` here in order to disable the default backing
1416 file.
1418 ``lazy-refcounts``
1419 Whether to enable the lazy refcounts feature (on/off;
1420 default is taken from the image file)
1422 ``cache-size``
1423 The maximum total size of the L2 table and refcount block
1424 caches in bytes (default: the sum of l2-cache-size and
1425 refcount-cache-size)
1427 ``l2-cache-size``
1428 The maximum size of the L2 table cache in bytes (default: if
1429 cache-size is not specified - 32M on Linux platforms, and 8M
1430 on non-Linux platforms; otherwise, as large as possible
1431 within the cache-size, while permitting the requested or the
1432 minimal refcount cache size)
1434 ``refcount-cache-size``
1435 The maximum size of the refcount block cache in bytes
1436 (default: 4 times the cluster size; or if cache-size is
1437 specified, the part of it which is not used for the L2
1438 cache)
1440 ``cache-clean-interval``
1441 Clean unused entries in the L2 and refcount caches. The
1442 interval is in seconds. The default value is 600 on
1443 supporting platforms, and 0 on other platforms. Setting it
1444 to 0 disables this feature.
1446 ``pass-discard-request``
1447 Whether discard requests to the qcow2 device should be
1448 forwarded to the data source (on/off; default: on if
1449 discard=unmap is specified, off otherwise)
1451 ``pass-discard-snapshot``
1452 Whether discard requests for the data source should be
1453 issued when a snapshot operation (e.g. deleting a snapshot)
1454 frees clusters in the qcow2 file (on/off; default: on)
1456 ``pass-discard-other``
1457 Whether discard requests for the data source should be
1458 issued on other occasions where a cluster gets freed
1459 (on/off; default: off)
1461 ``discard-no-unref``
1462 When enabled, data clusters will remain preallocated when they are
1463 no longer used, e.g. because they are discarded or converted to
1464 zero clusters. As usual, whether the old data is discarded or kept
1465 on the protocol level (i.e. in the image file) depends on the
1466 setting of the pass-discard-request option. Keeping the clusters
1467 preallocated prevents qcow2 fragmentation that would otherwise be
1468 caused by freeing and re-allocating them later. Besides potential
1469 performance degradation, such fragmentation can lead to increased
1470 allocation of clusters past the end of the image file,
1471 resulting in image files whose file length can grow much larger
1472 than their guest disk size would suggest.
1473 If image file length is of concern (e.g. when storing qcow2
1474 images directly on block devices), you should consider enabling
1475 this option.
1477 ``overlap-check``
1478 Which overlap checks to perform for writes to the image
1479 (none/constant/cached/all; default: cached). For details or
1480 finer granularity control refer to the QAPI documentation of
1481 ``blockdev-add``.
1483 Example 1:
1487 -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2
1488 -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216
1490 Example 2:
1494 -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2
1496 ``Driver-specific options for other drivers``
1497 Please refer to the QAPI documentation of the ``blockdev-add``
1498 QMP command.
1499 ERST
1501 DEF("drive", HAS_ARG, QEMU_OPTION_drive,
1502 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n"
1503 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n"
1504 " [,snapshot=on|off][,rerror=ignore|stop|report]\n"
1505 " [,werror=ignore|stop|report|enospc][,id=name]\n"
1506 " [,aio=threads|native|io_uring]\n"
1507 " [,readonly=on|off][,copy-on-read=on|off]\n"
1508 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n"
1509 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n"
1510 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n"
1511 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n"
1512 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n"
1513 " [[,iops_size=is]]\n"
1514 " [[,group=g]]\n"
1515 " use 'file' as a drive image\n", QEMU_ARCH_ALL)
1516 SRST
1517 ``-drive option[,option[,option[,...]]]``
1518 Define a new drive. This includes creating a block driver node (the
1519 backend) as well as a guest device, and is mostly a shortcut for
1520 defining the corresponding ``-blockdev`` and ``-device`` options.
1522 ``-drive`` accepts all options that are accepted by ``-blockdev``.
1523 In addition, it knows the following options:
1525 ``file=file``
1526 This option defines which disk image (see the :ref:`disk images`
1527 chapter in the System Emulation Users Guide) to use with this drive.
1528 If the filename contains comma, you must double it (for instance,
1529 "file=my,,file" to use file "my,file").
1531 Special files such as iSCSI devices can be specified using
1532 protocol specific URLs. See the section for "Device URL Syntax"
1533 for more information.
1535 ``if=interface``
1536 This option defines on which type on interface the drive is
1537 connected. Available types are: ide, scsi, sd, mtd, floppy,
1538 pflash, virtio, none.
1540 ``bus=bus,unit=unit``
1541 These options define where is connected the drive by defining
1542 the bus number and the unit id.
1544 ``index=index``
1545 This option defines where the drive is connected by using an
1546 index in the list of available connectors of a given interface
1547 type.
1549 ``media=media``
1550 This option defines the type of the media: disk or cdrom.
1552 ``snapshot=snapshot``
1553 snapshot is "on" or "off" and controls snapshot mode for the
1554 given drive (see ``-snapshot``).
1556 ``cache=cache``
1557 cache is "none", "writeback", "unsafe", "directsync" or
1558 "writethrough" and controls how the host cache is used to access
1559 block data. This is a shortcut that sets the ``cache.direct``
1560 and ``cache.no-flush`` options (as in ``-blockdev``), and
1561 additionally ``cache.writeback``, which provides a default for
1562 the ``write-cache`` option of block guest devices (as in
1563 ``-device``). The modes correspond to the following settings:
1565 ============= =============== ============ ==============
1566 \ cache.writeback cache.direct cache.no-flush
1567 ============= =============== ============ ==============
1568 writeback on off off
1569 none on on off
1570 writethrough off off off
1571 directsync off on off
1572 unsafe on off on
1573 ============= =============== ============ ==============
1575 The default mode is ``cache=writeback``.
1577 ``aio=aio``
1578 aio is "threads", "native", or "io_uring" and selects between pthread
1579 based disk I/O, native Linux AIO, or Linux io_uring API.
1581 ``format=format``
1582 Specify which disk format will be used rather than detecting the
1583 format. Can be used to specify format=raw to avoid interpreting
1584 an untrusted format header.
1586 ``werror=action,rerror=action``
1587 Specify which action to take on write and read errors. Valid
1588 actions are: "ignore" (ignore the error and try to continue),
1589 "stop" (pause QEMU), "report" (report the error to the guest),
1590 "enospc" (pause QEMU only if the host disk is full; report the
1591 error to the guest otherwise). The default setting is
1592 ``werror=enospc`` and ``rerror=report``.
1594 ``copy-on-read=copy-on-read``
1595 copy-on-read is "on" or "off" and enables whether to copy read
1596 backing file sectors into the image file.
1598 ``bps=b,bps_rd=r,bps_wr=w``
1599 Specify bandwidth throttling limits in bytes per second, either
1600 for all request types or for reads or writes only. Small values
1601 can lead to timeouts or hangs inside the guest. A safe minimum
1602 for disks is 2 MB/s.
1604 ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm``
1605 Specify bursts in bytes per second, either for all request types
1606 or for reads or writes only. Bursts allow the guest I/O to spike
1607 above the limit temporarily.
1609 ``iops=i,iops_rd=r,iops_wr=w``
1610 Specify request rate limits in requests per second, either for
1611 all request types or for reads or writes only.
1613 ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm``
1614 Specify bursts in requests per second, either for all request
1615 types or for reads or writes only. Bursts allow the guest I/O to
1616 spike above the limit temporarily.
1618 ``iops_size=is``
1619 Let every is bytes of a request count as a new request for iops
1620 throttling purposes. Use this option to prevent guests from
1621 circumventing iops limits by sending fewer but larger requests.
1623 ``group=g``
1624 Join a throttling quota group with given name g. All drives that
1625 are members of the same group are accounted for together. Use
1626 this option to prevent guests from circumventing throttling
1627 limits by using many small disks instead of a single larger
1628 disk.
1630 By default, the ``cache.writeback=on`` mode is used. It will report
1631 data writes as completed as soon as the data is present in the host
1632 page cache. This is safe as long as your guest OS makes sure to
1633 correctly flush disk caches where needed. If your guest OS does not
1634 handle volatile disk write caches correctly and your host crashes or
1635 loses power, then the guest may experience data corruption.
1637 For such guests, you should consider using ``cache.writeback=off``.
1638 This means that the host page cache will be used to read and write
1639 data, but write notification will be sent to the guest only after
1640 QEMU has made sure to flush each write to the disk. Be aware that
1641 this has a major impact on performance.
1643 When using the ``-snapshot`` option, unsafe caching is always used.
1645 Copy-on-read avoids accessing the same backing file sectors
1646 repeatedly and is useful when the backing file is over a slow
1647 network. By default copy-on-read is off.
1649 Instead of ``-cdrom`` you can use:
1651 .. parsed-literal::
1653 |qemu_system| -drive file=file,index=2,media=cdrom
1655 Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use:
1657 .. parsed-literal::
1659 |qemu_system| -drive file=file,index=0,media=disk
1660 |qemu_system| -drive file=file,index=1,media=disk
1661 |qemu_system| -drive file=file,index=2,media=disk
1662 |qemu_system| -drive file=file,index=3,media=disk
1664 You can open an image using pre-opened file descriptors from an fd
1665 set:
1667 .. parsed-literal::
1669 |qemu_system| \\
1670 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
1671 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
1672 -drive file=/dev/fdset/2,index=0,media=disk
1674 You can connect a CDROM to the slave of ide0:
1676 .. parsed-literal::
1678 |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom
1680 If you don't specify the "file=" argument, you define an empty
1681 drive:
1683 .. parsed-literal::
1685 |qemu_system_x86| -drive if=ide,index=1,media=cdrom
1687 Instead of ``-fda``, ``-fdb``, you can use:
1689 .. parsed-literal::
1691 |qemu_system_x86| -drive file=file,index=0,if=floppy
1692 |qemu_system_x86| -drive file=file,index=1,if=floppy
1694 By default, interface is "ide" and index is automatically
1695 incremented:
1697 .. parsed-literal::
1699 |qemu_system_x86| -drive file=a -drive file=b
1701 is interpreted like:
1703 .. parsed-literal::
1705 |qemu_system_x86| -hda a -hdb b
1706 ERST
1708 DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock,
1709 "-mtdblock file use 'file' as on-board Flash memory image\n",
1710 QEMU_ARCH_ALL)
1711 SRST
1712 ``-mtdblock file``
1713 Use file as on-board Flash memory image.
1714 ERST
1716 DEF("sd", HAS_ARG, QEMU_OPTION_sd,
1717 "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL)
1718 SRST
1719 ``-sd file``
1720 Use file as SecureDigital card image.
1721 ERST
1723 DEF("snapshot", 0, QEMU_OPTION_snapshot,
1724 "-snapshot write to temporary files instead of disk image files\n",
1725 QEMU_ARCH_ALL)
1726 SRST
1727 ``-snapshot``
1728 Write to temporary files instead of disk image files. In this case,
1729 the raw disk image you use is not written back. You can however
1730 force the write back by pressing C-a s (see the :ref:`disk images`
1731 chapter in the System Emulation Users Guide).
1733 .. warning::
1734 snapshot is incompatible with ``-blockdev`` (instead use qemu-img
1735 to manually create snapshot images to attach to your blockdev).
1736 If you have mixed ``-blockdev`` and ``-drive`` declarations you
1737 can use the 'snapshot' property on your drive declarations
1738 instead of this global option.
1740 ERST
1742 DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev,
1743 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1744 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n"
1745 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n"
1746 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n"
1747 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n"
1748 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n"
1749 " [[,throttling.iops-size=is]]\n"
1750 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n"
1751 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n"
1752 "-fsdev synth,id=id\n",
1753 QEMU_ARCH_ALL)
1755 SRST
1756 ``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]``
1758 ``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]``
1760 ``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]``
1762 ``-fsdev synth,id=id[,readonly=on]``
1763 Define a new file system device. Valid options are:
1765 ``local``
1766 Accesses to the filesystem are done by QEMU.
1768 ``proxy``
1769 Accesses to the filesystem are done by virtfs-proxy-helper(1). This
1770 option is deprecated (since QEMU 8.1) and will be removed in a future
1771 version of QEMU. Use ``local`` instead.
1773 ``synth``
1774 Synthetic filesystem, only used by QTests.
1776 ``id=id``
1777 Specifies identifier for this device.
1779 ``path=path``
1780 Specifies the export path for the file system device. Files
1781 under this path will be available to the 9p client on the guest.
1783 ``security_model=security_model``
1784 Specifies the security model to be used for this export path.
1785 Supported security models are "passthrough", "mapped-xattr",
1786 "mapped-file" and "none". In "passthrough" security model, files
1787 are stored using the same credentials as they are created on the
1788 guest. This requires QEMU to run as root. In "mapped-xattr"
1789 security model, some of the file attributes like uid, gid, mode
1790 bits and link target are stored as file attributes. For
1791 "mapped-file" these attributes are stored in the hidden
1792 .virtfs\_metadata directory. Directories exported by this
1793 security model cannot interact with other unix tools. "none"
1794 security model is same as passthrough except the sever won't
1795 report failures if it fails to set file attributes like
1796 ownership. Security model is mandatory only for local fsdriver.
1797 Other fsdrivers (like proxy) don't take security model as a
1798 parameter.
1800 ``writeout=writeout``
1801 This is an optional argument. The only supported value is
1802 "immediate". This means that host page cache will be used to
1803 read and write data but write notification will be sent to the
1804 guest only when the data has been reported as written by the
1805 storage subsystem.
1807 ``readonly=on``
1808 Enables exporting 9p share as a readonly mount for guests. By
1809 default read-write access is given.
1811 ``socket=socket``
1812 Enables proxy filesystem driver to use passed socket file for
1813 communicating with virtfs-proxy-helper(1).
1815 ``sock_fd=sock_fd``
1816 Enables proxy filesystem driver to use passed socket descriptor
1817 for communicating with virtfs-proxy-helper(1). Usually a helper
1818 like libvirt will create socketpair and pass one of the fds as
1819 sock\_fd.
1821 ``fmode=fmode``
1822 Specifies the default mode for newly created files on the host.
1823 Works only with security models "mapped-xattr" and
1824 "mapped-file".
1826 ``dmode=dmode``
1827 Specifies the default mode for newly created directories on the
1828 host. Works only with security models "mapped-xattr" and
1829 "mapped-file".
1831 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w``
1832 Specify bandwidth throttling limits in bytes per second, either
1833 for all request types or for reads or writes only.
1835 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm``
1836 Specify bursts in bytes per second, either for all request types
1837 or for reads or writes only. Bursts allow the guest I/O to spike
1838 above the limit temporarily.
1840 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w``
1841 Specify request rate limits in requests per second, either for
1842 all request types or for reads or writes only.
1844 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm``
1845 Specify bursts in requests per second, either for all request
1846 types or for reads or writes only. Bursts allow the guest I/O to
1847 spike above the limit temporarily.
1849 ``throttling.iops-size=is``
1850 Let every is bytes of a request count as a new request for iops
1851 throttling purposes.
1853 -fsdev option is used along with -device driver "virtio-9p-...".
1855 ``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag``
1856 Options for virtio-9p-... driver are:
1858 ``type``
1859 Specifies the variant to be used. Supported values are "pci",
1860 "ccw" or "device", depending on the machine type.
1862 ``fsdev=id``
1863 Specifies the id value specified along with -fsdev option.
1865 ``mount_tag=mount_tag``
1866 Specifies the tag name to be used by the guest to mount this
1867 export point.
1868 ERST
1870 DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs,
1871 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1872 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n"
1873 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n"
1874 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n"
1875 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n",
1876 QEMU_ARCH_ALL)
1878 SRST
1879 ``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]``
1881 ``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1883 ``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1885 ``-virtfs synth,mount_tag=mount_tag``
1886 Define a new virtual filesystem device and expose it to the guest using
1887 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain
1888 directory on host is made directly accessible by guest as a pass-through
1889 file system by using the 9P network protocol for communication between
1890 host and guests, if desired even accessible, shared by several guests
1891 simultaneously.
1893 Note that ``-virtfs`` is actually just a convenience shortcut for its
1894 generalized form ``-fsdev -device virtio-9p-pci``.
1896 The general form of pass-through file system options are:
1898 ``local``
1899 Accesses to the filesystem are done by QEMU.
1901 ``proxy``
1902 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1903 This option is deprecated (since QEMU 8.1) and will be removed in a
1904 future version of QEMU. Use ``local`` instead.
1906 ``synth``
1907 Synthetic filesystem, only used by QTests.
1909 ``id=id``
1910 Specifies identifier for the filesystem device
1912 ``path=path``
1913 Specifies the export path for the file system device. Files
1914 under this path will be available to the 9p client on the guest.
1916 ``security_model=security_model``
1917 Specifies the security model to be used for this export path.
1918 Supported security models are "passthrough", "mapped-xattr",
1919 "mapped-file" and "none". In "passthrough" security model, files
1920 are stored using the same credentials as they are created on the
1921 guest. This requires QEMU to run as root. In "mapped-xattr"
1922 security model, some of the file attributes like uid, gid, mode
1923 bits and link target are stored as file attributes. For
1924 "mapped-file" these attributes are stored in the hidden
1925 .virtfs\_metadata directory. Directories exported by this
1926 security model cannot interact with other unix tools. "none"
1927 security model is same as passthrough except the sever won't
1928 report failures if it fails to set file attributes like
1929 ownership. Security model is mandatory only for local fsdriver.
1930 Other fsdrivers (like proxy) don't take security model as a
1931 parameter.
1933 ``writeout=writeout``
1934 This is an optional argument. The only supported value is
1935 "immediate". This means that host page cache will be used to
1936 read and write data but write notification will be sent to the
1937 guest only when the data has been reported as written by the
1938 storage subsystem.
1940 ``readonly=on``
1941 Enables exporting 9p share as a readonly mount for guests. By
1942 default read-write access is given.
1944 ``socket=socket``
1945 Enables proxy filesystem driver to use passed socket file for
1946 communicating with virtfs-proxy-helper(1). Usually a helper like
1947 libvirt will create socketpair and pass one of the fds as
1948 sock\_fd.
1950 ``sock_fd``
1951 Enables proxy filesystem driver to use passed 'sock\_fd' as the
1952 socket descriptor for interfacing with virtfs-proxy-helper(1).
1954 ``fmode=fmode``
1955 Specifies the default mode for newly created files on the host.
1956 Works only with security models "mapped-xattr" and
1957 "mapped-file".
1959 ``dmode=dmode``
1960 Specifies the default mode for newly created directories on the
1961 host. Works only with security models "mapped-xattr" and
1962 "mapped-file".
1964 ``mount_tag=mount_tag``
1965 Specifies the tag name to be used by the guest to mount this
1966 export point.
1968 ``multidevs=multidevs``
1969 Specifies how to deal with multiple devices being shared with a
1970 9p export. Supported behaviours are either "remap", "forbid" or
1971 "warn". The latter is the default behaviour on which virtfs 9p
1972 expects only one device to be shared with the same export, and
1973 if more than one device is shared and accessed via the same 9p
1974 export then only a warning message is logged (once) by qemu on
1975 host side. In order to avoid file ID collisions on guest you
1976 should either create a separate virtfs export for each device to
1977 be shared with guests (recommended way) or you might use "remap"
1978 instead which allows you to share multiple devices with only one
1979 export instead, which is achieved by remapping the original
1980 inode numbers from host to guest in a way that would prevent
1981 such collisions. Remapping inodes in such use cases is required
1982 because the original device IDs from host are never passed and
1983 exposed on guest. Instead all files of an export shared with
1984 virtfs always share the same device id on guest. So two files
1985 with identical inode numbers but from actually different devices
1986 on host would otherwise cause a file ID collision and hence
1987 potential misbehaviours on guest. "forbid" on the other hand
1988 assumes like "warn" that only one device is shared by the same
1989 export, however it will not only log a warning message but also
1990 deny access to additional devices on guest. Note though that
1991 "forbid" does currently not block all possible file access
1992 operations (e.g. readdir() would still return entries from other
1993 devices).
1994 ERST
1996 DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi,
1997 "-iscsi [user=user][,password=password][,password-secret=secret-id]\n"
1998 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE]\n"
1999 " [,initiator-name=initiator-iqn][,id=target-iqn]\n"
2000 " [,timeout=timeout]\n"
2001 " iSCSI session parameters\n", QEMU_ARCH_ALL)
2003 SRST
2004 ``-iscsi``
2005 Configure iSCSI session parameters.
2006 ERST
2008 DEFHEADING()
2010 DEFHEADING(USB convenience options:)
2012 DEF("usb", 0, QEMU_OPTION_usb,
2013 "-usb enable on-board USB host controller (if not enabled by default)\n",
2014 QEMU_ARCH_ALL)
2015 SRST
2016 ``-usb``
2017 Enable USB emulation on machine types with an on-board USB host
2018 controller (if not enabled by default). Note that on-board USB host
2019 controllers may not support USB 3.0. In this case
2020 ``-device qemu-xhci`` can be used instead on machines with PCI.
2021 ERST
2023 DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice,
2024 "-usbdevice name add the host or guest USB device 'name'\n",
2025 QEMU_ARCH_ALL)
2026 SRST
2027 ``-usbdevice devname``
2028 Add the USB device devname, and enable an on-board USB controller
2029 if possible and necessary (just like it can be done via
2030 ``-machine usb=on``). Note that this option is mainly intended for
2031 the user's convenience only. More fine-grained control can be
2032 achieved by selecting a USB host controller (if necessary) and the
2033 desired USB device via the ``-device`` option instead. For example,
2034 instead of using ``-usbdevice mouse`` it is possible to use
2035 ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse
2036 to a USB 3.0 controller instead (at least on machines that support
2037 PCI and do not have an USB controller enabled by default yet).
2038 For more details, see the chapter about
2039 :ref:`Connecting USB devices` in the System Emulation Users Guide.
2040 Possible devices for devname are:
2042 ``braille``
2043 Braille device. This will use BrlAPI to display the braille
2044 output on a real or fake device (i.e. it also creates a
2045 corresponding ``braille`` chardev automatically beside the
2046 ``usb-braille`` USB device).
2048 ``keyboard``
2049 Standard USB keyboard. Will override the PS/2 keyboard (if present).
2051 ``mouse``
2052 Virtual Mouse. This will override the PS/2 mouse emulation when
2053 activated.
2055 ``tablet``
2056 Pointer device that uses absolute coordinates (like a
2057 touchscreen). This means QEMU is able to report the mouse
2058 position without having to grab the mouse. Also overrides the
2059 PS/2 mouse emulation when activated.
2061 ``wacom-tablet``
2062 Wacom PenPartner USB tablet.
2065 ERST
2067 DEFHEADING()
2069 DEFHEADING(Display options:)
2071 DEF("display", HAS_ARG, QEMU_OPTION_display,
2072 #if defined(CONFIG_SPICE)
2073 "-display spice-app[,gl=on|off]\n"
2074 #endif
2075 #if defined(CONFIG_SDL)
2076 "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n"
2077 " [,window-close=on|off]\n"
2078 #endif
2079 #if defined(CONFIG_GTK)
2080 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n"
2081 " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n"
2082 " [,show-menubar=on|off]\n"
2083 #endif
2084 #if defined(CONFIG_VNC)
2085 "-display vnc=<display>[,<optargs>]\n"
2086 #endif
2087 #if defined(CONFIG_CURSES)
2088 "-display curses[,charset=<encoding>]\n"
2089 #endif
2090 #if defined(CONFIG_COCOA)
2091 "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n"
2092 " [,show-cursor=on|off][,left-command-key=on|off]\n"
2093 " [,full-screen=on|off][,zoom-to-fit=on|off]\n"
2094 #endif
2095 #if defined(CONFIG_OPENGL)
2096 "-display egl-headless[,rendernode=<file>]\n"
2097 #endif
2098 #if defined(CONFIG_DBUS_DISPLAY)
2099 "-display dbus[,addr=<dbusaddr>]\n"
2100 " [,gl=on|core|es|off][,rendernode=<file>]\n"
2101 #endif
2102 "-display none\n"
2103 " select display backend type\n"
2104 " The default display is equivalent to\n "
2105 #if defined(CONFIG_GTK)
2106 "\"-display gtk\"\n"
2107 #elif defined(CONFIG_SDL)
2108 "\"-display sdl\"\n"
2109 #elif defined(CONFIG_COCOA)
2110 "\"-display cocoa\"\n"
2111 #elif defined(CONFIG_VNC)
2112 "\"-vnc localhost:0,to=99,id=default\"\n"
2113 #else
2114 "\"-display none\"\n"
2115 #endif
2116 , QEMU_ARCH_ALL)
2117 SRST
2118 ``-display type``
2119 Select type of display to use. Use ``-display help`` to list the available
2120 display types. Valid values for type are
2122 ``spice-app[,gl=on|off]``
2123 Start QEMU as a Spice server and launch the default Spice client
2124 application. The Spice server will redirect the serial consoles
2125 and QEMU monitors. (Since 4.0)
2127 ``dbus``
2128 Export the display over D-Bus interfaces. (Since 7.0)
2130 The connection is registered with the "org.qemu" name (and queued when
2131 already owned).
2133 ``addr=<dbusaddr>`` : D-Bus bus address to connect to.
2135 ``p2p=yes|no`` : Use peer-to-peer connection, accepted via QMP ``add_client``.
2137 ``gl=on|off|core|es`` : Use OpenGL for rendering (the D-Bus interface
2138 will share framebuffers with DMABUF file descriptors).
2140 ``sdl``
2141 Display video output via SDL (usually in a separate graphics
2142 window; see the SDL documentation for other possibilities).
2143 Valid parameters are:
2145 ``grab-mod=<mods>`` : Used to select the modifier keys for toggling
2146 the mouse grabbing in conjunction with the "g" key. ``<mods>`` can be
2147 either ``lshift-lctrl-lalt`` or ``rctrl``.
2149 ``gl=on|off|core|es`` : Use OpenGL for displaying
2151 ``show-cursor=on|off`` : Force showing the mouse cursor
2153 ``window-close=on|off`` : Allow to quit qemu with window close button
2155 ``gtk``
2156 Display video output in a GTK window. This interface provides
2157 drop-down menus and other UI elements to configure and control
2158 the VM during runtime. Valid parameters are:
2160 ``full-screen=on|off`` : Start in fullscreen mode
2162 ``gl=on|off`` : Use OpenGL for displaying
2164 ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover
2166 ``show-tabs=on|off`` : Display the tab bar for switching between the
2167 various graphical interfaces (e.g. VGA and
2168 virtual console character devices) by default.
2170 ``show-cursor=on|off`` : Force showing the mouse cursor
2172 ``window-close=on|off`` : Allow to quit qemu with window close button
2174 ``show-menubar=on|off`` : Display the main window menubar, defaults to "on"
2176 ``zoom-to-fit=on|off`` : Expand video output to the window size,
2177 defaults to "off"
2179 ``curses[,charset=<encoding>]``
2180 Display video output via curses. For graphics device models
2181 which support a text mode, QEMU can display this output using a
2182 curses/ncurses interface. Nothing is displayed when the graphics
2183 device is in graphical mode or if the graphics device does not
2184 support a text mode. Generally only the VGA device models
2185 support text mode. The font charset used by the guest can be
2186 specified with the ``charset`` option, for example
2187 ``charset=CP850`` for IBM CP850 encoding. The default is
2188 ``CP437``.
2190 ``cocoa``
2191 Display video output in a Cocoa window. Mac only. This interface
2192 provides drop-down menus and other UI elements to configure and
2193 control the VM during runtime. Valid parameters are:
2195 ``full-grab=on|off`` : Capture all key presses, including system combos.
2196 This requires accessibility permissions, since it
2197 performs a global grab on key events.
2198 (default: off) See
2199 https://support.apple.com/en-in/guide/mac-help/mh32356/mac
2201 ``swap-opt-cmd=on|off`` : Swap the Option and Command keys so that their
2202 key codes match their position on non-Mac
2203 keyboards and you can use Meta/Super and Alt
2204 where you expect them. (default: off)
2206 ``show-cursor=on|off`` : Force showing the mouse cursor
2208 ``left-command-key=on|off`` : Disable forwarding left command key to host
2210 ``full-screen=on|off`` : Start in fullscreen mode
2212 ``zoom-to-fit=on|off`` : Expand video output to the window size,
2213 defaults to "off"
2215 ``egl-headless[,rendernode=<file>]``
2216 Offload all OpenGL operations to a local DRI device. For any
2217 graphical display, this display needs to be paired with either
2218 VNC or SPICE displays.
2220 ``vnc=<display>``
2221 Start a VNC server on display <display>
2223 ``none``
2224 Do not display video output. The guest will still see an
2225 emulated graphics card, but its output will not be displayed to
2226 the QEMU user. This option differs from the -nographic option in
2227 that it only affects what is done with video output; -nographic
2228 also changes the destination of the serial and parallel port
2229 data.
2230 ERST
2232 DEF("nographic", 0, QEMU_OPTION_nographic,
2233 "-nographic disable graphical output and redirect serial I/Os to console\n",
2234 QEMU_ARCH_ALL)
2235 SRST
2236 ``-nographic``
2237 Normally, if QEMU is compiled with graphical window support, it
2238 displays output such as guest graphics, guest console, and the QEMU
2239 monitor in a window. With this option, you can totally disable
2240 graphical output so that QEMU is a simple command line application.
2241 The emulated serial port is redirected on the console and muxed with
2242 the monitor (unless redirected elsewhere explicitly). Therefore, you
2243 can still use QEMU to debug a Linux kernel with a serial console.
2244 Use C-a h for help on switching between the console and monitor.
2245 ERST
2247 #ifdef CONFIG_SPICE
2248 DEF("spice", HAS_ARG, QEMU_OPTION_spice,
2249 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n"
2250 " [,x509-key-file=<file>][,x509-key-password=<file>]\n"
2251 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n"
2252 " [,x509-dh-key-file=<file>][,addr=addr]\n"
2253 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n"
2254 " [,tls-ciphers=<list>]\n"
2255 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n"
2256 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n"
2257 " [,sasl=on|off][,disable-ticketing=on|off]\n"
2258 " [,password-secret=<secret-id>]\n"
2259 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n"
2260 " [,jpeg-wan-compression=[auto|never|always]]\n"
2261 " [,zlib-glz-wan-compression=[auto|never|always]]\n"
2262 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n"
2263 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n"
2264 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n"
2265 " [,gl=[on|off]][,rendernode=<file>]\n"
2266 " enable spice\n"
2267 " at least one of {port, tls-port} is mandatory\n",
2268 QEMU_ARCH_ALL)
2269 #endif
2270 SRST
2271 ``-spice option[,option[,...]]``
2272 Enable the spice remote desktop protocol. Valid options are
2274 ``port=<nr>``
2275 Set the TCP port spice is listening on for plaintext channels.
2277 ``addr=<addr>``
2278 Set the IP address spice is listening on. Default is any
2279 address.
2281 ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off``
2282 Force using the specified IP version.
2284 ``password-secret=<secret-id>``
2285 Set the ID of the ``secret`` object containing the password
2286 you need to authenticate.
2288 ``sasl=on|off``
2289 Require that the client use SASL to authenticate with the spice.
2290 The exact choice of authentication method used is controlled
2291 from the system / user's SASL configuration file for the 'qemu'
2292 service. This is typically found in /etc/sasl2/qemu.conf. If
2293 running QEMU as an unprivileged user, an environment variable
2294 SASL\_CONF\_PATH can be used to make it search alternate
2295 locations for the service config. While some SASL auth methods
2296 can also provide data encryption (eg GSSAPI), it is recommended
2297 that SASL always be combined with the 'tls' and 'x509' settings
2298 to enable use of SSL and server certificates. This ensures a
2299 data encryption preventing compromise of authentication
2300 credentials.
2302 ``disable-ticketing=on|off``
2303 Allow client connects without authentication.
2305 ``disable-copy-paste=on|off``
2306 Disable copy paste between the client and the guest.
2308 ``disable-agent-file-xfer=on|off``
2309 Disable spice-vdagent based file-xfer between the client and the
2310 guest.
2312 ``tls-port=<nr>``
2313 Set the TCP port spice is listening on for encrypted channels.
2315 ``x509-dir=<dir>``
2316 Set the x509 file directory. Expects same filenames as -vnc
2317 $display,x509=$dir
2319 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>``
2320 The x509 file names can also be configured individually.
2322 ``tls-ciphers=<list>``
2323 Specify which ciphers to use.
2325 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]``
2326 Force specific channel to be used with or without TLS
2327 encryption. The options can be specified multiple times to
2328 configure multiple channels. The special name "default" can be
2329 used to set the default mode. For channels which are not
2330 explicitly forced into one mode the spice client is allowed to
2331 pick tls/plaintext as he pleases.
2333 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]``
2334 Configure image compression (lossless). Default is auto\_glz.
2336 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]``
2337 Configure wan image compression (lossy for slow links). Default
2338 is auto.
2340 ``streaming-video=[off|all|filter]``
2341 Configure video stream detection. Default is off.
2343 ``agent-mouse=[on|off]``
2344 Enable/disable passing mouse events via vdagent. Default is on.
2346 ``playback-compression=[on|off]``
2347 Enable/disable audio stream compression (using celt 0.5.1).
2348 Default is on.
2350 ``seamless-migration=[on|off]``
2351 Enable/disable spice seamless migration. Default is off.
2353 ``gl=[on|off]``
2354 Enable/disable OpenGL context. Default is off.
2356 ``rendernode=<file>``
2357 DRM render node for OpenGL rendering. If not specified, it will
2358 pick the first available. (Since 2.9)
2359 ERST
2361 DEF("portrait", 0, QEMU_OPTION_portrait,
2362 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n",
2363 QEMU_ARCH_ALL)
2364 SRST
2365 ``-portrait``
2366 Rotate graphical output 90 deg left (only PXA LCD).
2367 ERST
2369 DEF("rotate", HAS_ARG, QEMU_OPTION_rotate,
2370 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n",
2371 QEMU_ARCH_ALL)
2372 SRST
2373 ``-rotate deg``
2374 Rotate graphical output some deg left (only PXA LCD).
2375 ERST
2377 DEF("vga", HAS_ARG, QEMU_OPTION_vga,
2378 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n"
2379 " select video card type\n", QEMU_ARCH_ALL)
2380 SRST
2381 ``-vga type``
2382 Select type of VGA card to emulate. Valid values for type are
2384 ``cirrus``
2385 Cirrus Logic GD5446 Video card. All Windows versions starting
2386 from Windows 95 should recognize and use this graphic card. For
2387 optimal performances, use 16 bit color depth in the guest and
2388 the host OS. (This card was the default before QEMU 2.2)
2390 ``std``
2391 Standard VGA card with Bochs VBE extensions. If your guest OS
2392 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if
2393 you want to use high resolution modes (>= 1280x1024x16) then you
2394 should use this option. (This card is the default since QEMU
2395 2.2)
2397 ``vmware``
2398 VMWare SVGA-II compatible adapter. Use it if you have
2399 sufficiently recent XFree86/XOrg server or Windows guest with a
2400 driver for this card.
2402 ``qxl``
2403 QXL paravirtual graphic card. It is VGA compatible (including
2404 VESA 2.0 VBE support). Works best with qxl guest drivers
2405 installed though. Recommended choice when using the spice
2406 protocol.
2408 ``tcx``
2409 (sun4m only) Sun TCX framebuffer. This is the default
2410 framebuffer for sun4m machines and offers both 8-bit and 24-bit
2411 colour depths at a fixed resolution of 1024x768.
2413 ``cg3``
2414 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit
2415 framebuffer for sun4m machines available in both 1024x768
2416 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people
2417 wishing to run older Solaris versions.
2419 ``virtio``
2420 Virtio VGA card.
2422 ``none``
2423 Disable VGA card.
2424 ERST
2426 DEF("full-screen", 0, QEMU_OPTION_full_screen,
2427 "-full-screen start in full screen\n", QEMU_ARCH_ALL)
2428 SRST
2429 ``-full-screen``
2430 Start in full screen.
2431 ERST
2433 DEF("g", HAS_ARG, QEMU_OPTION_g ,
2434 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n",
2435 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K)
2436 SRST
2437 ``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]``
2438 Set the initial graphical resolution and depth (PPC, SPARC only).
2440 For PPC the default is 800x600x32.
2442 For SPARC with the TCX graphics device, the default is 1024x768x8
2443 with the option of 1024x768x24. For cgthree, the default is
2444 1024x768x8 with the option of 1152x900x8 for people who wish to use
2445 OBP.
2446 ERST
2448 #ifdef CONFIG_VNC
2449 DEF("vnc", HAS_ARG, QEMU_OPTION_vnc ,
2450 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL)
2451 #endif
2452 SRST
2453 ``-vnc display[,option[,option[,...]]]``
2454 Normally, if QEMU is compiled with graphical window support, it
2455 displays output such as guest graphics, guest console, and the QEMU
2456 monitor in a window. With this option, you can have QEMU listen on
2457 VNC display display and redirect the VGA display over the VNC
2458 session. It is very useful to enable the usb tablet device when
2459 using this option (option ``-device usb-tablet``). When using the
2460 VNC display, you must use the ``-k`` parameter to set the keyboard
2461 layout if you are not using en-us. Valid syntax for the display is
2463 ``to=L``
2464 With this option, QEMU will try next available VNC displays,
2465 until the number L, if the origianlly defined "-vnc display" is
2466 not available, e.g. port 5900+display is already used by another
2467 application. By default, to=0.
2469 ``host:d``
2470 TCP connections will only be allowed from host on display d. By
2471 convention the TCP port is 5900+d. Optionally, host can be
2472 omitted in which case the server will accept connections from
2473 any host.
2475 ``unix:path``
2476 Connections will be allowed over UNIX domain sockets where path
2477 is the location of a unix socket to listen for connections on.
2479 ``none``
2480 VNC is initialized but not started. The monitor ``change``
2481 command can be used to later start the VNC server.
2483 Following the display value there may be one or more option flags
2484 separated by commas. Valid options are
2486 ``reverse=on|off``
2487 Connect to a listening VNC client via a "reverse" connection.
2488 The client is specified by the display. For reverse network
2489 connections (host:d,``reverse``), the d argument is a TCP port
2490 number, not a display number.
2492 ``websocket=on|off``
2493 Opens an additional TCP listening port dedicated to VNC
2494 Websocket connections. If a bare websocket option is given, the
2495 Websocket port is 5700+display. An alternative port can be
2496 specified with the syntax ``websocket``\ =port.
2498 If host is specified connections will only be allowed from this
2499 host. It is possible to control the websocket listen address
2500 independently, using the syntax ``websocket``\ =host:port.
2502 If no TLS credentials are provided, the websocket connection
2503 runs in unencrypted mode. If TLS credentials are provided, the
2504 websocket connection requires encrypted client connections.
2506 ``password=on|off``
2507 Require that password based authentication is used for client
2508 connections.
2510 The password must be set separately using the ``set_password``
2511 command in the :ref:`QEMU monitor`. The
2512 syntax to change your password is:
2513 ``set_password <protocol> <password>`` where <protocol> could be
2514 either "vnc" or "spice".
2516 If you would like to change <protocol> password expiration, you
2517 should use ``expire_password <protocol> <expiration-time>``
2518 where expiration time could be one of the following options:
2519 now, never, +seconds or UNIX time of expiration, e.g. +60 to
2520 make password expire in 60 seconds, or 1335196800 to make
2521 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for
2522 this date and time).
2524 You can also use keywords "now" or "never" for the expiration
2525 time to allow <protocol> password to expire immediately or never
2526 expire.
2528 ``password-secret=<secret-id>``
2529 Require that password based authentication is used for client
2530 connections, using the password provided by the ``secret``
2531 object identified by ``secret-id``.
2533 ``tls-creds=ID``
2534 Provides the ID of a set of TLS credentials to use to secure the
2535 VNC server. They will apply to both the normal VNC server socket
2536 and the websocket socket (if enabled). Setting TLS credentials
2537 will cause the VNC server socket to enable the VeNCrypt auth
2538 mechanism. The credentials should have been previously created
2539 using the ``-object tls-creds`` argument.
2541 ``tls-authz=ID``
2542 Provides the ID of the QAuthZ authorization object against which
2543 the client's x509 distinguished name will validated. This object
2544 is only resolved at time of use, so can be deleted and recreated
2545 on the fly while the VNC server is active. If missing, it will
2546 default to denying access.
2548 ``sasl=on|off``
2549 Require that the client use SASL to authenticate with the VNC
2550 server. The exact choice of authentication method used is
2551 controlled from the system / user's SASL configuration file for
2552 the 'qemu' service. This is typically found in
2553 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user,
2554 an environment variable SASL\_CONF\_PATH can be used to make it
2555 search alternate locations for the service config. While some
2556 SASL auth methods can also provide data encryption (eg GSSAPI),
2557 it is recommended that SASL always be combined with the 'tls'
2558 and 'x509' settings to enable use of SSL and server
2559 certificates. This ensures a data encryption preventing
2560 compromise of authentication credentials. See the
2561 :ref:`VNC security` section in the System Emulation Users Guide
2562 for details on using SASL authentication.
2564 ``sasl-authz=ID``
2565 Provides the ID of the QAuthZ authorization object against which
2566 the client's SASL username will validated. This object is only
2567 resolved at time of use, so can be deleted and recreated on the
2568 fly while the VNC server is active. If missing, it will default
2569 to denying access.
2571 ``acl=on|off``
2572 Legacy method for enabling authorization of clients against the
2573 x509 distinguished name and SASL username. It results in the
2574 creation of two ``authz-list`` objects with IDs of
2575 ``vnc.username`` and ``vnc.x509dname``. The rules for these
2576 objects must be configured with the HMP ACL commands.
2578 This option is deprecated and should no longer be used. The new
2579 ``sasl-authz`` and ``tls-authz`` options are a replacement.
2581 ``lossy=on|off``
2582 Enable lossy compression methods (gradient, JPEG, ...). If this
2583 option is set, VNC client may receive lossy framebuffer updates
2584 depending on its encoding settings. Enabling this option can
2585 save a lot of bandwidth at the expense of quality.
2587 ``non-adaptive=on|off``
2588 Disable adaptive encodings. Adaptive encodings are enabled by
2589 default. An adaptive encoding will try to detect frequently
2590 updated screen regions, and send updates in these regions using
2591 a lossy encoding (like JPEG). This can be really helpful to save
2592 bandwidth when playing videos. Disabling adaptive encodings
2593 restores the original static behavior of encodings like Tight.
2595 ``share=[allow-exclusive|force-shared|ignore]``
2596 Set display sharing policy. 'allow-exclusive' allows clients to
2597 ask for exclusive access. As suggested by the rfb spec this is
2598 implemented by dropping other connections. Connecting multiple
2599 clients in parallel requires all clients asking for a shared
2600 session (vncviewer: -shared switch). This is the default.
2601 'force-shared' disables exclusive client access. Useful for
2602 shared desktop sessions, where you don't want someone forgetting
2603 specify -shared disconnect everybody else. 'ignore' completely
2604 ignores the shared flag and allows everybody connect
2605 unconditionally. Doesn't conform to the rfb spec but is
2606 traditional QEMU behavior.
2608 ``key-delay-ms``
2609 Set keyboard delay, for key down and key up events, in
2610 milliseconds. Default is 10. Keyboards are low-bandwidth
2611 devices, so this slowdown can help the device and guest to keep
2612 up and not lose events in case events are arriving in bulk.
2613 Possible causes for the latter are flaky network connections, or
2614 scripts for automated testing.
2616 ``audiodev=audiodev``
2617 Use the specified audiodev when the VNC client requests audio
2618 transmission. When not using an -audiodev argument, this option
2619 must be omitted, otherwise is must be present and specify a
2620 valid audiodev.
2622 ``power-control=on|off``
2623 Permit the remote client to issue shutdown, reboot or reset power
2624 control requests.
2625 ERST
2627 ARCHHEADING(, QEMU_ARCH_I386)
2629 ARCHHEADING(i386 target only:, QEMU_ARCH_I386)
2631 DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack,
2632 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n",
2633 QEMU_ARCH_I386)
2634 SRST
2635 ``-win2k-hack``
2636 Use it when installing Windows 2000 to avoid a disk full bug. After
2637 Windows 2000 is installed, you no longer need this option (this
2638 option slows down the IDE transfers).
2639 ERST
2641 DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk,
2642 "-no-fd-bootchk disable boot signature checking for floppy disks\n",
2643 QEMU_ARCH_I386)
2644 SRST
2645 ``-no-fd-bootchk``
2646 Disable boot signature checking for floppy disks in BIOS. May be
2647 needed to boot from old floppy disks.
2648 ERST
2650 DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable,
2651 "-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"
2652 " ACPI table description\n", QEMU_ARCH_I386)
2653 SRST
2654 ``-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]...]``
2655 Add ACPI table with specified header fields and context from
2656 specified files. For file=, take whole ACPI table from the specified
2657 files, including all ACPI headers (possible overridden by other
2658 options). For data=, only data portion of the table is used, all
2659 header information is specified in the command line. If a SLIC table
2660 is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id
2661 fields will override the same in the RSDT and the FADT (a.k.a.
2662 FACP), in order to ensure the field matches required by the
2663 Microsoft SLIC spec and the ACPI spec.
2664 ERST
2666 DEF("smbios", HAS_ARG, QEMU_OPTION_smbios,
2667 "-smbios file=binary\n"
2668 " load SMBIOS entry from binary file\n"
2669 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n"
2670 " [,uefi=on|off]\n"
2671 " specify SMBIOS type 0 fields\n"
2672 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2673 " [,uuid=uuid][,sku=str][,family=str]\n"
2674 " specify SMBIOS type 1 fields\n"
2675 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2676 " [,asset=str][,location=str]\n"
2677 " specify SMBIOS type 2 fields\n"
2678 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n"
2679 " [,sku=str]\n"
2680 " specify SMBIOS type 3 fields\n"
2681 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n"
2682 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n"
2683 " [,processor-id=%d]\n"
2684 " specify SMBIOS type 4 fields\n"
2685 "-smbios type=8[,external_reference=str][,internal_reference=str][,connector_type=%d][,port_type=%d]\n"
2686 " specify SMBIOS type 8 fields\n"
2687 "-smbios type=11[,value=str][,path=filename]\n"
2688 " specify SMBIOS type 11 fields\n"
2689 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n"
2690 " [,asset=str][,part=str][,speed=%d]\n"
2691 " specify SMBIOS type 17 fields\n"
2692 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n"
2693 " specify SMBIOS type 41 fields\n",
2694 QEMU_ARCH_I386 | QEMU_ARCH_ARM | QEMU_ARCH_LOONGARCH)
2695 SRST
2696 ``-smbios file=binary``
2697 Load SMBIOS entry from binary file.
2699 ``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]``
2700 Specify SMBIOS type 0 fields
2702 ``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]``
2703 Specify SMBIOS type 1 fields
2705 ``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]``
2706 Specify SMBIOS type 2 fields
2708 ``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]``
2709 Specify SMBIOS type 3 fields
2711 ``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-id=%d]``
2712 Specify SMBIOS type 4 fields
2714 ``-smbios type=11[,value=str][,path=filename]``
2715 Specify SMBIOS type 11 fields
2717 This argument can be repeated multiple times, and values are added in the order they are parsed.
2718 Applications intending to use OEM strings data are encouraged to use their application name as
2719 a prefix for the value string. This facilitates passing information for multiple applications
2720 concurrently.
2722 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax
2723 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes.
2725 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to
2726 the SMBIOS table in the order in which they appear.
2728 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535
2729 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the
2730 guest. Instead it should be used as a indicator to inform the guest where to locate the real
2731 data set, for example, by specifying the serial ID of a block device.
2733 An example passing three strings is
2735 .. parsed-literal::
2737 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\
2738 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\
2739 path=/some/file/with/oemstringsdata.txt
2741 In the guest OS this is visible with the ``dmidecode`` command
2743 .. parsed-literal::
2745 $ dmidecode -t 11
2746 Handle 0x0E00, DMI type 11, 5 bytes
2747 OEM Strings
2748 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/
2749 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os
2750 String 3: myapp:some extra data
2753 ``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]``
2754 Specify SMBIOS type 17 fields
2756 ``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]``
2757 Specify SMBIOS type 41 fields
2759 This argument can be repeated multiple times. Its main use is to allow network interfaces be created
2760 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface
2761 position on the PCI bus.
2763 Here is an example of use:
2765 .. parsed-literal::
2767 -netdev user,id=internet \\
2768 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\
2769 -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev
2771 In the guest OS, the device should then appear as ``eno1``:
2773 ..parsed-literal::
2775 $ ip -brief l
2776 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP>
2777 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP>
2779 Currently, the PCI device has to be attached to the root bus.
2781 ERST
2783 DEFHEADING()
2785 DEFHEADING(Network options:)
2787 DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
2788 #ifdef CONFIG_SLIRP
2789 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n"
2790 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n"
2791 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n"
2792 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n"
2793 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]"
2794 #ifndef _WIN32
2795 "[,smb=dir[,smbserver=addr]]\n"
2796 #endif
2797 " configure a user mode network backend with ID 'str',\n"
2798 " its DHCP server and optional services\n"
2799 #endif
2800 #ifdef _WIN32
2801 "-netdev tap,id=str,ifname=name\n"
2802 " configure a host TAP network backend with ID 'str'\n"
2803 #else
2804 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n"
2805 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n"
2806 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n"
2807 " [,poll-us=n]\n"
2808 " configure a host TAP network backend with ID 'str'\n"
2809 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2810 " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n"
2811 " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n"
2812 " to deconfigure it\n"
2813 " use '[down]script=no' to disable script execution\n"
2814 " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n"
2815 " configure it\n"
2816 " use 'fd=h' to connect to an already opened TAP interface\n"
2817 " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n"
2818 " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n"
2819 " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n"
2820 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n"
2821 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n"
2822 " use vhost=on to enable experimental in kernel accelerator\n"
2823 " (only has effect for virtio guests which use MSIX)\n"
2824 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n"
2825 " use 'vhostfd=h' to connect to an already opened vhost net device\n"
2826 " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n"
2827 " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n"
2828 " use 'poll-us=n' to specify the maximum number of microseconds that could be\n"
2829 " spent on busy polling for vhost net\n"
2830 "-netdev bridge,id=str[,br=bridge][,helper=helper]\n"
2831 " configure a host TAP network backend with ID 'str' that is\n"
2832 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2833 " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n"
2834 #endif
2835 #ifdef __linux__
2836 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n"
2837 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n"
2838 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n"
2839 " [,rxcookie=rxcookie][,offset=offset]\n"
2840 " configure a network backend with ID 'str' connected to\n"
2841 " an Ethernet over L2TPv3 pseudowire.\n"
2842 " Linux kernel 3.3+ as well as most routers can talk\n"
2843 " L2TPv3. This transport allows connecting a VM to a VM,\n"
2844 " VM to a router and even VM to Host. It is a nearly-universal\n"
2845 " standard (RFC3931). Note - this implementation uses static\n"
2846 " pre-configured tunnels (same as the Linux kernel).\n"
2847 " use 'src=' to specify source address\n"
2848 " use 'dst=' to specify destination address\n"
2849 " use 'udp=on' to specify udp encapsulation\n"
2850 " use 'srcport=' to specify source udp port\n"
2851 " use 'dstport=' to specify destination udp port\n"
2852 " use 'ipv6=on' to force v6\n"
2853 " L2TPv3 uses cookies to prevent misconfiguration as\n"
2854 " well as a weak security measure\n"
2855 " use 'rxcookie=0x012345678' to specify a rxcookie\n"
2856 " use 'txcookie=0x012345678' to specify a txcookie\n"
2857 " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n"
2858 " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n"
2859 " use 'pincounter=on' to work around broken counter handling in peer\n"
2860 " use 'offset=X' to add an extra offset between header and data\n"
2861 #endif
2862 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n"
2863 " configure a network backend to connect to another network\n"
2864 " using a socket connection\n"
2865 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n"
2866 " configure a network backend to connect to a multicast maddr and port\n"
2867 " use 'localaddr=addr' to specify the host address to send packets from\n"
2868 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n"
2869 " configure a network backend to connect to another network\n"
2870 " using an UDP tunnel\n"
2871 "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port[,to=maxport][,numeric=on|off][,keep-alive=on|off][,mptcp=on|off][,addr.ipv4=on|off][,addr.ipv6=on|off][,reconnect=seconds]\n"
2872 "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off][,reconnect=seconds]\n"
2873 "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor[,reconnect=seconds]\n"
2874 " configure a network backend to connect to another network\n"
2875 " using a socket connection in stream mode.\n"
2876 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n"
2877 "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]\n"
2878 " configure a network backend to connect to a multicast maddr and port\n"
2879 " use ``local.host=addr`` to specify the host address to send packets from\n"
2880 "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n"
2881 "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n"
2882 "-netdev dgram,id=str,local.type=fd,local.str=file-descriptor\n"
2883 " configure a network backend to connect to another network\n"
2884 " using an UDP tunnel\n"
2885 #ifdef CONFIG_VDE
2886 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n"
2887 " configure a network backend to connect to port 'n' of a vde switch\n"
2888 " running on host and listening for incoming connections on 'socketpath'.\n"
2889 " Use group 'groupname' and mode 'octalmode' to change default\n"
2890 " ownership and permissions for communication port.\n"
2891 #endif
2892 #ifdef CONFIG_NETMAP
2893 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n"
2894 " attach to the existing netmap-enabled network interface 'name', or to a\n"
2895 " VALE port (created on the fly) called 'name' ('nmname' is name of the \n"
2896 " netmap device, defaults to '/dev/netmap')\n"
2897 #endif
2898 #ifdef CONFIG_AF_XDP
2899 "-netdev af-xdp,id=str,ifname=name[,mode=native|skb][,force-copy=on|off]\n"
2900 " [,queues=n][,start-queue=m][,inhibit=on|off][,sock-fds=x:y:...:z]\n"
2901 " attach to the existing network interface 'name' with AF_XDP socket\n"
2902 " use 'mode=MODE' to specify an XDP program attach mode\n"
2903 " use 'force-copy=on|off' to force XDP copy mode even if device supports zero-copy (default: off)\n"
2904 " use 'inhibit=on|off' to inhibit loading of a default XDP program (default: off)\n"
2905 " with inhibit=on,\n"
2906 " use 'sock-fds' to provide file descriptors for already open AF_XDP sockets\n"
2907 " added to a socket map in XDP program. One socket per queue.\n"
2908 " use 'queues=n' to specify how many queues of a multiqueue interface should be used\n"
2909 " use 'start-queue=m' to specify the first queue that should be used\n"
2910 #endif
2911 #ifdef CONFIG_POSIX
2912 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n"
2913 " configure a vhost-user network, backed by a chardev 'dev'\n"
2914 #endif
2915 #ifdef __linux__
2916 "-netdev vhost-vdpa,id=str[,vhostdev=/path/to/dev][,vhostfd=h]\n"
2917 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n"
2918 " use 'vhostdev=/path/to/dev' to open a vhost vdpa device\n"
2919 " use 'vhostfd=h' to connect to an already opened vhost vdpa device\n"
2920 #endif
2921 #ifdef CONFIG_VMNET
2922 "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n"
2923 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n"
2924 " configure a vmnet network backend in host mode with ID 'str',\n"
2925 " isolate this interface from others with 'isolated',\n"
2926 " configure the address range and choose a subnet mask,\n"
2927 " specify network UUID 'uuid' to disable DHCP and interact with\n"
2928 " vmnet-host interfaces within this isolated network\n"
2929 "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n"
2930 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n"
2931 " configure a vmnet network backend in shared mode with ID 'str',\n"
2932 " configure the address range and choose a subnet mask,\n"
2933 " set IPv6 ULA prefix (of length 64) to use for internal network,\n"
2934 " isolate this interface from others with 'isolated'\n"
2935 "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n"
2936 " configure a vmnet network backend in bridged mode with ID 'str',\n"
2937 " use 'ifname=name' to select a physical network interface to be bridged,\n"
2938 " isolate this interface from others with 'isolated'\n"
2939 #endif
2940 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n"
2941 " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL)
2942 DEF("nic", HAS_ARG, QEMU_OPTION_nic,
2943 "-nic [tap|bridge|"
2944 #ifdef CONFIG_SLIRP
2945 "user|"
2946 #endif
2947 #ifdef __linux__
2948 "l2tpv3|"
2949 #endif
2950 #ifdef CONFIG_VDE
2951 "vde|"
2952 #endif
2953 #ifdef CONFIG_NETMAP
2954 "netmap|"
2955 #endif
2956 #ifdef CONFIG_AF_XDP
2957 "af-xdp|"
2958 #endif
2959 #ifdef CONFIG_POSIX
2960 "vhost-user|"
2961 #endif
2962 #ifdef CONFIG_VMNET
2963 "vmnet-host|vmnet-shared|vmnet-bridged|"
2964 #endif
2965 "socket][,option][,...][mac=macaddr]\n"
2966 " initialize an on-board / default host NIC (using MAC address\n"
2967 " macaddr) and connect it to the given host network backend\n"
2968 "-nic none use it alone to have zero network devices (the default is to\n"
2969 " provided a 'user' network connection)\n",
2970 QEMU_ARCH_ALL)
2971 DEF("net", HAS_ARG, QEMU_OPTION_net,
2972 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n"
2973 " configure or create an on-board (or machine default) NIC and\n"
2974 " connect it to hub 0 (please use -nic unless you need a hub)\n"
2975 "-net ["
2976 #ifdef CONFIG_SLIRP
2977 "user|"
2978 #endif
2979 "tap|"
2980 "bridge|"
2981 #ifdef CONFIG_VDE
2982 "vde|"
2983 #endif
2984 #ifdef CONFIG_NETMAP
2985 "netmap|"
2986 #endif
2987 #ifdef CONFIG_AF_XDP
2988 "af-xdp|"
2989 #endif
2990 #ifdef CONFIG_VMNET
2991 "vmnet-host|vmnet-shared|vmnet-bridged|"
2992 #endif
2993 "socket][,option][,option][,...]\n"
2994 " old way to initialize a host network interface\n"
2995 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL)
2996 SRST
2997 ``-nic [tap|bridge|user|l2tpv3|vde|netmap|af-xdp|vhost-user|socket][,...][,mac=macaddr][,model=mn]``
2998 This option is a shortcut for configuring both the on-board
2999 (default) guest NIC hardware and the host network backend in one go.
3000 The host backend options are the same as with the corresponding
3001 ``-netdev`` options below. The guest NIC model can be set with
3002 ``model=modelname``. Use ``model=help`` to list the available device
3003 types. The hardware MAC address can be set with ``mac=macaddr``.
3005 The following two example do exactly the same, to show how ``-nic``
3006 can be used to shorten the command line length:
3008 .. parsed-literal::
3010 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32
3011 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32
3013 ``-nic none``
3014 Indicate that no network devices should be configured. It is used to
3015 override the default configuration (default NIC with "user" host
3016 network backend) which is activated if no other networking options
3017 are provided.
3019 ``-netdev user,id=id[,option][,option][,...]``
3020 Configure user mode host network backend which requires no
3021 administrator privilege to run. Valid options are:
3023 ``id=id``
3024 Assign symbolic name for use in monitor commands.
3026 ``ipv4=on|off and ipv6=on|off``
3027 Specify that either IPv4 or IPv6 must be enabled. If neither is
3028 specified both protocols are enabled.
3030 ``net=addr[/mask]``
3031 Set IP network address the guest will see. Optionally specify
3032 the netmask, either in the form a.b.c.d or as number of valid
3033 top-most bits. Default is 10.0.2.0/24.
3035 ``host=addr``
3036 Specify the guest-visible address of the host. Default is the
3037 2nd IP in the guest network, i.e. x.x.x.2.
3039 ``ipv6-net=addr[/int]``
3040 Set IPv6 network address the guest will see (default is
3041 fec0::/64). The network prefix is given in the usual hexadecimal
3042 IPv6 address notation. The prefix size is optional, and is given
3043 as the number of valid top-most bits (default is 64).
3045 ``ipv6-host=addr``
3046 Specify the guest-visible IPv6 address of the host. Default is
3047 the 2nd IPv6 in the guest network, i.e. xxxx::2.
3049 ``restrict=on|off``
3050 If this option is enabled, the guest will be isolated, i.e. it
3051 will not be able to contact the host and no guest IP packets
3052 will be routed over the host to the outside. This option does
3053 not affect any explicitly set forwarding rules.
3055 ``hostname=name``
3056 Specifies the client hostname reported by the built-in DHCP
3057 server.
3059 ``dhcpstart=addr``
3060 Specify the first of the 16 IPs the built-in DHCP server can
3061 assign. Default is the 15th to 31st IP in the guest network,
3062 i.e. x.x.x.15 to x.x.x.31.
3064 ``dns=addr``
3065 Specify the guest-visible address of the virtual nameserver. The
3066 address must be different from the host address. Default is the
3067 3rd IP in the guest network, i.e. x.x.x.3.
3069 ``ipv6-dns=addr``
3070 Specify the guest-visible address of the IPv6 virtual
3071 nameserver. The address must be different from the host address.
3072 Default is the 3rd IP in the guest network, i.e. xxxx::3.
3074 ``dnssearch=domain``
3075 Provides an entry for the domain-search list sent by the
3076 built-in DHCP server. More than one domain suffix can be
3077 transmitted by specifying this option multiple times. If
3078 supported, this will cause the guest to automatically try to
3079 append the given domain suffix(es) in case a domain name can not
3080 be resolved.
3082 Example:
3084 .. parsed-literal::
3086 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org
3088 ``domainname=domain``
3089 Specifies the client domain name reported by the built-in DHCP
3090 server.
3092 ``tftp=dir``
3093 When using the user mode network stack, activate a built-in TFTP
3094 server. The files in dir will be exposed as the root of a TFTP
3095 server. The TFTP client on the guest must be configured in
3096 binary mode (use the command ``bin`` of the Unix TFTP client).
3098 ``tftp-server-name=name``
3099 In BOOTP reply, broadcast name as the "TFTP server name"
3100 (RFC2132 option 66). This can be used to advise the guest to
3101 load boot files or configurations from a different server than
3102 the host address.
3104 ``bootfile=file``
3105 When using the user mode network stack, broadcast file as the
3106 BOOTP filename. In conjunction with ``tftp``, this can be used
3107 to network boot a guest from a local directory.
3109 Example (using pxelinux):
3111 .. parsed-literal::
3113 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\
3114 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0
3116 ``smb=dir[,smbserver=addr]``
3117 When using the user mode network stack, activate a built-in SMB
3118 server so that Windows OSes can access to the host files in
3119 ``dir`` transparently. The IP address of the SMB server can be
3120 set to addr. By default the 4th IP in the guest network is used,
3121 i.e. x.x.x.4.
3123 In the guest Windows OS, the line:
3127 10.0.2.4 smbserver
3129 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows
3130 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows
3131 NT/2000).
3133 Then ``dir`` can be accessed in ``\\smbserver\qemu``.
3135 Note that a SAMBA server must be installed on the host OS.
3137 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport``
3138 Redirect incoming TCP or UDP connections to the host port
3139 hostport to the guest IP address guestaddr on guest port
3140 guestport. If guestaddr is not specified, its value is x.x.x.15
3141 (default first address given by the built-in DHCP server). By
3142 specifying hostaddr, the rule can be bound to a specific host
3143 interface. If no connection type is set, TCP is used. This
3144 option can be given multiple times.
3146 For example, to redirect host X11 connection from screen 1 to
3147 guest screen 0, use the following:
3149 .. parsed-literal::
3151 # on the host
3152 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000
3153 # this host xterm should open in the guest X11 server
3154 xterm -display :1
3156 To redirect telnet connections from host port 5555 to telnet
3157 port on the guest, use the following:
3159 .. parsed-literal::
3161 # on the host
3162 |qemu_system| -nic user,hostfwd=tcp::5555-:23
3163 telnet localhost 5555
3165 Then when you use on the host ``telnet localhost 5555``, you
3166 connect to the guest telnet server.
3168 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command``
3169 Forward guest TCP connections to the IP address server on port
3170 port to the character device dev or to a program executed by
3171 cmd:command which gets spawned for each connection. This option
3172 can be given multiple times.
3174 You can either use a chardev directly and have that one used
3175 throughout QEMU's lifetime, like in the following example:
3177 .. parsed-literal::
3179 # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever
3180 # the guest accesses it
3181 |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321
3183 Or you can execute a command on every TCP connection established
3184 by the guest, so that QEMU behaves similar to an inetd process
3185 for that virtual server:
3187 .. parsed-literal::
3189 # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234
3190 # and connect the TCP stream to its stdin/stdout
3191 |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321'
3193 ``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]``
3194 Configure a host TAP network backend with ID id.
3196 Use the network script file to configure it and the network script
3197 dfile to deconfigure it. If name is not provided, the OS
3198 automatically provides one. The default network configure script is
3199 ``/etc/qemu-ifup`` and the default network deconfigure script is
3200 ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to
3201 disable script execution.
3203 If running QEMU as an unprivileged user, use the network helper
3204 to configure the TAP interface and attach it to the bridge.
3205 The default network helper executable is
3206 ``/path/to/qemu-bridge-helper`` and the default bridge device is
3207 ``br0``.
3209 ``fd``\ =h can be used to specify the handle of an already opened
3210 host TAP interface.
3212 Examples:
3214 .. parsed-literal::
3216 #launch a QEMU instance with the default network script
3217 |qemu_system| linux.img -nic tap
3219 .. parsed-literal::
3221 #launch a QEMU instance with two NICs, each one connected
3222 #to a TAP device
3223 |qemu_system| linux.img \\
3224 -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\
3225 -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1
3227 .. parsed-literal::
3229 #launch a QEMU instance with the default network helper to
3230 #connect a TAP device to bridge br0
3231 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\
3232 -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper"
3234 ``-netdev bridge,id=id[,br=bridge][,helper=helper]``
3235 Connect a host TAP network interface to a host bridge device.
3237 Use the network helper helper to configure the TAP interface and
3238 attach it to the bridge. The default network helper executable is
3239 ``/path/to/qemu-bridge-helper`` and the default bridge device is
3240 ``br0``.
3242 Examples:
3244 .. parsed-literal::
3246 #launch a QEMU instance with the default network helper to
3247 #connect a TAP device to bridge br0
3248 |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1
3250 .. parsed-literal::
3252 #launch a QEMU instance with the default network helper to
3253 #connect a TAP device to bridge qemubr0
3254 |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1
3256 ``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]``
3257 This host network backend can be used to connect the guest's network
3258 to another QEMU virtual machine using a TCP socket connection. If
3259 ``listen`` is specified, QEMU waits for incoming connections on port
3260 (host is optional). ``connect`` is used to connect to another QEMU
3261 instance using the ``listen`` option. ``fd``\ =h specifies an
3262 already opened TCP socket.
3264 Example:
3266 .. parsed-literal::
3268 # launch a first QEMU instance
3269 |qemu_system| linux.img \\
3270 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3271 -netdev socket,id=n1,listen=:1234
3272 # connect the network of this instance to the network of the first instance
3273 |qemu_system| linux.img \\
3274 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
3275 -netdev socket,id=n2,connect=127.0.0.1:1234
3277 ``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]``
3278 Configure a socket host network backend to share the guest's network
3279 traffic with another QEMU virtual machines using a UDP multicast
3280 socket, effectively making a bus for every QEMU with same multicast
3281 address maddr and port. NOTES:
3283 1. Several QEMU can be running on different hosts and share same bus
3284 (assuming correct multicast setup for these hosts).
3286 2. mcast support is compatible with User Mode Linux (argument
3287 ``ethN=mcast``), see http://user-mode-linux.sf.net.
3289 3. Use ``fd=h`` to specify an already opened UDP multicast socket.
3291 Example:
3293 .. parsed-literal::
3295 # launch one QEMU instance
3296 |qemu_system| linux.img \\
3297 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3298 -netdev socket,id=n1,mcast=230.0.0.1:1234
3299 # launch another QEMU instance on same "bus"
3300 |qemu_system| linux.img \\
3301 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
3302 -netdev socket,id=n2,mcast=230.0.0.1:1234
3303 # launch yet another QEMU instance on same "bus"
3304 |qemu_system| linux.img \\
3305 -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\
3306 -netdev socket,id=n3,mcast=230.0.0.1:1234
3308 Example (User Mode Linux compat.):
3310 .. parsed-literal::
3312 # launch QEMU instance (note mcast address selected is UML's default)
3313 |qemu_system| linux.img \\
3314 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3315 -netdev socket,id=n1,mcast=239.192.168.1:1102
3316 # launch UML
3317 /path/to/linux ubd0=/path/to/root_fs eth0=mcast
3319 Example (send packets from host's 1.2.3.4):
3321 .. parsed-literal::
3323 |qemu_system| linux.img \\
3324 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3325 -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4
3327 ``-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]``
3328 Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931)
3329 is a popular protocol to transport Ethernet (and other Layer 2) data
3330 frames between two systems. It is present in routers, firewalls and
3331 the Linux kernel (from version 3.3 onwards).
3333 This transport allows a VM to communicate to another VM, router or
3334 firewall directly.
3336 ``src=srcaddr``
3337 source address (mandatory)
3339 ``dst=dstaddr``
3340 destination address (mandatory)
3342 ``udp``
3343 select udp encapsulation (default is ip).
3345 ``srcport=srcport``
3346 source udp port.
3348 ``dstport=dstport``
3349 destination udp port.
3351 ``ipv6``
3352 force v6, otherwise defaults to v4.
3354 ``rxcookie=rxcookie``; \ ``txcookie=txcookie``
3355 Cookies are a weak form of security in the l2tpv3 specification.
3356 Their function is mostly to prevent misconfiguration. By default
3357 they are 32 bit.
3359 ``cookie64``
3360 Set cookie size to 64 bit instead of the default 32
3362 ``counter=off``
3363 Force a 'cut-down' L2TPv3 with no counter as in
3364 draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00
3366 ``pincounter=on``
3367 Work around broken counter handling in peer. This may also help
3368 on networks which have packet reorder.
3370 ``offset=offset``
3371 Add an extra offset between header and data
3373 For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to
3374 the bridge br-lan on the remote Linux host 1.2.3.4:
3376 .. parsed-literal::
3378 # Setup tunnel on linux host using raw ip as encapsulation
3379 # on 1.2.3.4
3380 ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\
3381 encap udp udp_sport 16384 udp_dport 16384
3382 ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\
3383 0xFFFFFFFF peer_session_id 0xFFFFFFFF
3384 ifconfig vmtunnel0 mtu 1500
3385 ifconfig vmtunnel0 up
3386 brctl addif br-lan vmtunnel0
3389 # on 4.3.2.1
3390 # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter
3392 |qemu_system| linux.img -device e1000,netdev=n1 \\
3393 -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter
3395 ``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]``
3396 Configure VDE backend to connect to PORT n of a vde switch running
3397 on host and listening for incoming connections on socketpath. Use
3398 GROUP groupname and MODE octalmode to change default ownership and
3399 permissions for communication port. This option is only available if
3400 QEMU has been compiled with vde support enabled.
3402 Example:
3404 .. parsed-literal::
3406 # launch vde switch
3407 vde_switch -F -sock /tmp/myswitch
3408 # launch QEMU instance
3409 |qemu_system| linux.img -nic vde,sock=/tmp/myswitch
3411 ``-netdev af-xdp,id=str,ifname=name[,mode=native|skb][,force-copy=on|off][,queues=n][,start-queue=m][,inhibit=on|off][,sock-fds=x:y:...:z]``
3412 Configure AF_XDP backend to connect to a network interface 'name'
3413 using AF_XDP socket. A specific program attach mode for a default
3414 XDP program can be forced with 'mode', defaults to best-effort,
3415 where the likely most performant mode will be in use. Number of queues
3416 'n' should generally match the number or queues in the interface,
3417 defaults to 1. Traffic arriving on non-configured device queues will
3418 not be delivered to the network backend.
3420 .. parsed-literal::
3422 # set number of queues to 4
3423 ethtool -L eth0 combined 4
3424 # launch QEMU instance
3425 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\
3426 -netdev af-xdp,id=n1,ifname=eth0,queues=4
3428 'start-queue' option can be specified if a particular range of queues
3429 [m, m + n] should be in use. For example, this is may be necessary in
3430 order to use certain NICs in native mode. Kernel allows the driver to
3431 create a separate set of XDP queues on top of regular ones, and only
3432 these queues can be used for AF_XDP sockets. NICs that work this way
3433 may also require an additional traffic redirection with ethtool to these
3434 special queues.
3436 .. parsed-literal::
3438 # set number of queues to 1
3439 ethtool -L eth0 combined 1
3440 # redirect all the traffic to the second queue (id: 1)
3441 # note: drivers may require non-empty key/mask pair.
3442 ethtool -N eth0 flow-type ether \\
3443 dst 00:00:00:00:00:00 m FF:FF:FF:FF:FF:FE action 1
3444 ethtool -N eth0 flow-type ether \\
3445 dst 00:00:00:00:00:01 m FF:FF:FF:FF:FF:FE action 1
3446 # launch QEMU instance
3447 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\
3448 -netdev af-xdp,id=n1,ifname=eth0,queues=1,start-queue=1
3450 XDP program can also be loaded externally. In this case 'inhibit' option
3451 should be set to 'on' and 'sock-fds' provided with file descriptors for
3452 already open but not bound XDP sockets already added to a socket map for
3453 corresponding queues. One socket per queue.
3455 .. parsed-literal::
3457 |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\
3458 -netdev af-xdp,id=n1,ifname=eth0,queues=3,inhibit=on,sock-fds=15:16:17
3460 ``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]``
3461 Establish a vhost-user netdev, backed by a chardev id. The chardev
3462 should be a unix domain socket backed one. The vhost-user uses a
3463 specifically defined protocol to pass vhost ioctl replacement
3464 messages to an application on the other end of the socket. On
3465 non-MSIX guests, the feature can be forced with vhostforce. Use
3466 'queues=n' to specify the number of queues to be created for
3467 multiqueue vhost-user.
3469 Example:
3473 qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \
3474 -numa node,memdev=mem \
3475 -chardev socket,id=chr0,path=/path/to/socket \
3476 -netdev type=vhost-user,id=net0,chardev=chr0 \
3477 -device virtio-net-pci,netdev=net0
3479 ``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]``
3480 Establish a vhost-vdpa netdev.
3482 vDPA device is a device that uses a datapath which complies with
3483 the virtio specifications with a vendor specific control path.
3484 vDPA devices can be both physically located on the hardware or
3485 emulated by software.
3487 ``-netdev hubport,id=id,hubid=hubid[,netdev=nd]``
3488 Create a hub port on the emulated hub with ID hubid.
3490 The hubport netdev lets you connect a NIC to a QEMU emulated hub
3491 instead of a single netdev. Alternatively, you can also connect the
3492 hubport to another netdev with ID nd by using the ``netdev=nd``
3493 option.
3495 ``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]``
3496 Legacy option to configure or create an on-board (or machine
3497 default) Network Interface Card(NIC) and connect it either to the
3498 emulated hub with ID 0 (i.e. the default hub), or to the netdev nd.
3499 If model is omitted, then the default NIC model associated with the
3500 machine type is used. Note that the default NIC model may change in
3501 future QEMU releases, so it is highly recommended to always specify
3502 a model. Optionally, the MAC address can be changed to mac, the
3503 device address set to addr (PCI cards only), and a name can be
3504 assigned for use in monitor commands. Optionally, for PCI cards, you
3505 can specify the number v of MSI-X vectors that the card should have;
3506 this option currently only affects virtio cards; set v = 0 to
3507 disable MSI-X. If no ``-net`` option is specified, a single NIC is
3508 created. QEMU can emulate several different models of network card.
3509 Use ``-net nic,model=help`` for a list of available devices for your
3510 target.
3512 ``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]``
3513 Configure a host network backend (with the options corresponding to
3514 the same ``-netdev`` option) and connect it to the emulated hub 0
3515 (the default hub). Use name to specify the name of the hub port.
3516 ERST
3518 DEFHEADING()
3520 DEFHEADING(Character device options:)
3522 DEF("chardev", HAS_ARG, QEMU_OPTION_chardev,
3523 "-chardev help\n"
3524 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3525 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n"
3526 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n"
3527 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n"
3528 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n"
3529 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n"
3530 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n"
3531 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n"
3532 " [,logfile=PATH][,logappend=on|off]\n"
3533 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3534 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n"
3535 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3536 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n"
3537 "-chardev file,id=id,path=path[,input-path=input-file][,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3538 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3539 #ifdef _WIN32
3540 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3541 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3542 #else
3543 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3544 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n"
3545 #endif
3546 #ifdef CONFIG_BRLAPI
3547 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3548 #endif
3549 #if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \
3550 || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__)
3551 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3552 #endif
3553 #if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__)
3554 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3555 #endif
3556 #if defined(CONFIG_SPICE)
3557 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3558 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3559 #endif
3560 , QEMU_ARCH_ALL
3563 SRST
3564 The general form of a character device option is:
3566 ``-chardev backend,id=id[,mux=on|off][,options]``
3567 Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``,
3568 ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``,
3569 ``pty``, ``stdio``, ``braille``, ``parallel``,
3570 ``spicevmc``, ``spiceport``. The specific backend will determine the
3571 applicable options.
3573 Use ``-chardev help`` to print all available chardev backend types.
3575 All devices must have an id, which can be any string up to 127
3576 characters long. It is used to uniquely identify this device in
3577 other command line directives.
3579 A character device may be used in multiplexing mode by multiple
3580 front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is
3581 a "1:N" device, and here the "1" end is your specified chardev
3582 backend, and the "N" end is the various parts of QEMU that can talk
3583 to a chardev. If you create a chardev with ``id=myid`` and
3584 ``mux=on``, QEMU will create a multiplexer with your specified ID,
3585 and you can then configure multiple front ends to use that chardev
3586 ID for their input/output. Up to four different front ends can be
3587 connected to a single multiplexed chardev. (Without multiplexing
3588 enabled, a chardev can only be used by a single front end.) For
3589 instance you could use this to allow a single stdio chardev to be
3590 used by two serial ports and the QEMU monitor:
3594 -chardev stdio,mux=on,id=char0 \
3595 -mon chardev=char0,mode=readline \
3596 -serial chardev:char0 \
3597 -serial chardev:char0
3599 You can have more than one multiplexer in a system configuration;
3600 for instance you could have a TCP port multiplexed between UART 0
3601 and UART 1, and stdio multiplexed between the QEMU monitor and a
3602 parallel port:
3606 -chardev stdio,mux=on,id=char0 \
3607 -mon chardev=char0,mode=readline \
3608 -parallel chardev:char0 \
3609 -chardev tcp,...,mux=on,id=char1 \
3610 -serial chardev:char1 \
3611 -serial chardev:char1
3613 When you're using a multiplexed character device, some escape
3614 sequences are interpreted in the input. See the chapter about
3615 :ref:`keys in the character backend multiplexer` in the
3616 System Emulation Users Guide for more details.
3618 Note that some other command line options may implicitly create
3619 multiplexed character backends; for instance ``-serial mon:stdio``
3620 creates a multiplexed stdio backend connected to the serial port and
3621 the QEMU monitor, and ``-nographic`` also multiplexes the console
3622 and the monitor to stdio.
3624 There is currently no support for multiplexing in the other
3625 direction (where a single QEMU front end takes input and output from
3626 multiple chardevs).
3628 Every backend supports the ``logfile`` option, which supplies the
3629 path to a file to record all data transmitted via the backend. The
3630 ``logappend`` option controls whether the log file will be truncated
3631 or appended to when opened.
3633 The available backends are:
3635 ``-chardev null,id=id``
3636 A void device. This device will not emit any data, and will drop any
3637 data it receives. The null backend does not take any options.
3639 ``-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]``
3640 Create a two-way stream socket, which can be either a TCP or a unix
3641 socket. A unix socket will be created if ``path`` is specified.
3642 Behaviour is undefined if TCP options are specified for a unix
3643 socket.
3645 ``server=on|off`` specifies that the socket shall be a listening socket.
3647 ``wait=on|off`` specifies that QEMU should not block waiting for a client
3648 to connect to a listening socket.
3650 ``telnet=on|off`` specifies that traffic on the socket should interpret
3651 telnet escape sequences.
3653 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for
3654 communication.
3656 ``reconnect`` sets the timeout for reconnecting on non-server
3657 sockets when the remote end goes away. qemu will delay this many
3658 seconds and then attempt to reconnect. Zero disables reconnecting,
3659 and is the default.
3661 ``tls-creds`` requests enablement of the TLS protocol for
3662 encryption, and specifies the id of the TLS credentials to use for
3663 the handshake. The credentials must be previously created with the
3664 ``-object tls-creds`` argument.
3666 ``tls-auth`` provides the ID of the QAuthZ authorization object
3667 against which the client's x509 distinguished name will be
3668 validated. This object is only resolved at time of use, so can be
3669 deleted and recreated on the fly while the chardev server is active.
3670 If missing, it will default to denying access.
3672 TCP and unix socket options are given below:
3674 ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]``
3675 ``host`` for a listening socket specifies the local address to
3676 be bound. For a connecting socket species the remote host to
3677 connect to. ``host`` is optional for listening sockets. If not
3678 specified it defaults to ``0.0.0.0``.
3680 ``port`` for a listening socket specifies the local port to be
3681 bound. For a connecting socket specifies the port on the remote
3682 host to connect to. ``port`` can be given as either a port
3683 number or a service name. ``port`` is required.
3685 ``to`` is only relevant to listening sockets. If it is
3686 specified, and ``port`` cannot be bound, QEMU will attempt to
3687 bind to subsequent ports up to and including ``to`` until it
3688 succeeds. ``to`` must be specified as a port number.
3690 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4
3691 or IPv6 must be used. If neither is specified the socket may
3692 use either protocol.
3694 ``nodelay=on|off`` disables the Nagle algorithm.
3696 ``unix options: path=path[,abstract=on|off][,tight=on|off]``
3697 ``path`` specifies the local path of the unix socket. ``path``
3698 is required.
3699 ``abstract=on|off`` specifies the use of the abstract socket namespace,
3700 rather than the filesystem. Optional, defaults to false.
3701 ``tight=on|off`` sets the socket length of abstract sockets to their minimum,
3702 rather than the full sun_path length. Optional, defaults to true.
3704 ``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]``
3705 Sends all traffic from the guest to a remote host over UDP.
3707 ``host`` specifies the remote host to connect to. If not specified
3708 it defaults to ``localhost``.
3710 ``port`` specifies the port on the remote host to connect to.
3711 ``port`` is required.
3713 ``localaddr`` specifies the local address to bind to. If not
3714 specified it defaults to ``0.0.0.0``.
3716 ``localport`` specifies the local port to bind to. If not specified
3717 any available local port will be used.
3719 ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used.
3720 If neither is specified the device may use either protocol.
3722 ``-chardev msmouse,id=id``
3723 Forward QEMU's emulated msmouse events to the guest. ``msmouse``
3724 does not take any options.
3726 ``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]``
3727 Connect to a QEMU text console. ``vc`` may optionally be given a
3728 specific size.
3730 ``width`` and ``height`` specify the width and height respectively
3731 of the console, in pixels.
3733 ``cols`` and ``rows`` specify that the console be sized to fit a
3734 text console with the given dimensions.
3736 ``-chardev ringbuf,id=id[,size=size]``
3737 Create a ring buffer with fixed size ``size``. size must be a power
3738 of two and defaults to ``64K``.
3740 ``-chardev file,id=id,path=path[,input-path=input-path]``
3741 Log all traffic received from the guest to a file.
3743 ``path`` specifies the path of the file to be opened. This file will
3744 be created if it does not already exist, and overwritten if it does.
3745 ``path`` is required.
3747 If ``input-path`` is specified, this is the path of a second file
3748 which will be used for input. If ``input-path`` is not specified,
3749 no input will be available from the chardev.
3751 Note that ``input-path`` is not supported on Windows hosts.
3753 ``-chardev pipe,id=id,path=path``
3754 Create a two-way connection to the guest. The behaviour differs
3755 slightly between Windows hosts and other hosts:
3757 On Windows, a single duplex pipe will be created at
3758 ``\\.pipe\path``.
3760 On other hosts, 2 pipes will be created called ``path.in`` and
3761 ``path.out``. Data written to ``path.in`` will be received by the
3762 guest. Data written by the guest can be read from ``path.out``. QEMU
3763 will not create these fifos, and requires them to be present.
3765 ``path`` forms part of the pipe path as described above. ``path`` is
3766 required.
3768 ``-chardev console,id=id``
3769 Send traffic from the guest to QEMU's standard output. ``console``
3770 does not take any options.
3772 ``console`` is only available on Windows hosts.
3774 ``-chardev serial,id=id,path=path``
3775 Send traffic from the guest to a serial device on the host.
3777 On Unix hosts serial will actually accept any tty device, not only
3778 serial lines.
3780 ``path`` specifies the name of the serial device to open.
3782 ``-chardev pty,id=id``
3783 Create a new pseudo-terminal on the host and connect to it. ``pty``
3784 does not take any options.
3786 ``pty`` is not available on Windows hosts.
3788 ``-chardev stdio,id=id[,signal=on|off]``
3789 Connect to standard input and standard output of the QEMU process.
3791 ``signal`` controls if signals are enabled on the terminal, that
3792 includes exiting QEMU with the key sequence Control-c. This option
3793 is enabled by default, use ``signal=off`` to disable it.
3795 ``-chardev braille,id=id``
3796 Connect to a local BrlAPI server. ``braille`` does not take any
3797 options.
3799 ``-chardev parallel,id=id,path=path``
3801 ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD
3802 hosts.
3804 Connect to a local parallel port.
3806 ``path`` specifies the path to the parallel port device. ``path`` is
3807 required.
3809 ``-chardev spicevmc,id=id,debug=debug,name=name``
3810 ``spicevmc`` is only available when spice support is built in.
3812 ``debug`` debug level for spicevmc
3814 ``name`` name of spice channel to connect to
3816 Connect to a spice virtual machine channel, such as vdiport.
3818 ``-chardev spiceport,id=id,debug=debug,name=name``
3819 ``spiceport`` is only available when spice support is built in.
3821 ``debug`` debug level for spicevmc
3823 ``name`` name of spice port to connect to
3825 Connect to a spice port, allowing a Spice client to handle the
3826 traffic identified by a name (preferably a fqdn).
3827 ERST
3829 DEFHEADING()
3831 #ifdef CONFIG_TPM
3832 DEFHEADING(TPM device options:)
3834 DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \
3835 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n"
3836 " use path to provide path to a character device; default is /dev/tpm0\n"
3837 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n"
3838 " not provided it will be searched for in /sys/class/misc/tpm?/device\n"
3839 "-tpmdev emulator,id=id,chardev=dev\n"
3840 " configure the TPM device using chardev backend\n",
3841 QEMU_ARCH_ALL)
3842 SRST
3843 The general form of a TPM device option is:
3845 ``-tpmdev backend,id=id[,options]``
3846 The specific backend type will determine the applicable options. The
3847 ``-tpmdev`` option creates the TPM backend and requires a
3848 ``-device`` option that specifies the TPM frontend interface model.
3850 Use ``-tpmdev help`` to print all available TPM backend types.
3852 The available backends are:
3854 ``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path``
3855 (Linux-host only) Enable access to the host's TPM using the
3856 passthrough driver.
3858 ``path`` specifies the path to the host's TPM device, i.e., on a
3859 Linux host this would be ``/dev/tpm0``. ``path`` is optional and by
3860 default ``/dev/tpm0`` is used.
3862 ``cancel-path`` specifies the path to the host TPM device's sysfs
3863 entry allowing for cancellation of an ongoing TPM command.
3864 ``cancel-path`` is optional and by default QEMU will search for the
3865 sysfs entry to use.
3867 Some notes about using the host's TPM with the passthrough driver:
3869 The TPM device accessed by the passthrough driver must not be used
3870 by any other application on the host.
3872 Since the host's firmware (BIOS/UEFI) has already initialized the
3873 TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize
3874 the TPM again and may therefore not show a TPM-specific menu that
3875 would otherwise allow the user to configure the TPM, e.g., allow the
3876 user to enable/disable or activate/deactivate the TPM. Further, if
3877 TPM ownership is released from within a VM then the host's TPM will
3878 get disabled and deactivated. To enable and activate the TPM again
3879 afterwards, the host has to be rebooted and the user is required to
3880 enter the firmware's menu to enable and activate the TPM. If the TPM
3881 is left disabled and/or deactivated most TPM commands will fail.
3883 To create a passthrough TPM use the following two options:
3887 -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0
3889 Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by
3890 ``tpmdev=tpm0`` in the device option.
3892 ``-tpmdev emulator,id=id,chardev=dev``
3893 (Linux-host only) Enable access to a TPM emulator using Unix domain
3894 socket based chardev backend.
3896 ``chardev`` specifies the unique ID of a character device backend
3897 that provides connection to the software TPM server.
3899 To create a TPM emulator backend device with chardev socket backend:
3903 -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0
3904 ERST
3906 DEFHEADING()
3908 #endif
3910 DEFHEADING(Boot Image or Kernel specific:)
3911 SRST
3912 There are broadly 4 ways you can boot a system with QEMU.
3914 - specify a firmware and let it control finding a kernel
3915 - specify a firmware and pass a hint to the kernel to boot
3916 - direct kernel image boot
3917 - manually load files into the guest's address space
3919 The third method is useful for quickly testing kernels but as there is
3920 no firmware to pass configuration information to the kernel the
3921 hardware must either be probeable, the kernel built for the exact
3922 configuration or passed some configuration data (e.g. a DTB blob)
3923 which tells the kernel what drivers it needs. This exact details are
3924 often hardware specific.
3926 The final method is the most generic way of loading images into the
3927 guest address space and used mostly for ``bare metal`` type
3928 development where the reset vectors of the processor are taken into
3929 account.
3931 ERST
3933 SRST
3935 For x86 machines and some other architectures ``-bios`` will generally
3936 do the right thing with whatever it is given. For other machines the
3937 more strict ``-pflash`` option needs an image that is sized for the
3938 flash device for the given machine type.
3940 Please see the :ref:`system-targets-ref` section of the manual for
3941 more detailed documentation.
3943 ERST
3945 DEF("bios", HAS_ARG, QEMU_OPTION_bios, \
3946 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL)
3947 SRST
3948 ``-bios file``
3949 Set the filename for the BIOS.
3950 ERST
3952 DEF("pflash", HAS_ARG, QEMU_OPTION_pflash,
3953 "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL)
3954 SRST
3955 ``-pflash file``
3956 Use file as a parallel flash image.
3957 ERST
3959 SRST
3961 The kernel options were designed to work with Linux kernels although
3962 other things (like hypervisors) can be packaged up as a kernel
3963 executable image. The exact format of a executable image is usually
3964 architecture specific.
3966 The way in which the kernel is started (what address it is loaded at,
3967 what if any information is passed to it via CPU registers, the state
3968 of the hardware when it is started, and so on) is also architecture
3969 specific. Typically it follows the specification laid down by the
3970 Linux kernel for how kernels for that architecture must be started.
3972 ERST
3974 DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \
3975 "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL)
3976 SRST
3977 ``-kernel bzImage``
3978 Use bzImage as kernel image. The kernel can be either a Linux kernel
3979 or in multiboot format.
3980 ERST
3982 DEF("append", HAS_ARG, QEMU_OPTION_append, \
3983 "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL)
3984 SRST
3985 ``-append cmdline``
3986 Use cmdline as kernel command line
3987 ERST
3989 DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \
3990 "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL)
3991 SRST
3993 ``-initrd file``
3994 Use file as initial ram disk.
3996 ``-initrd "file1 arg=foo,file2"``
3997 This syntax is only available with multiboot.
3999 Use file1 and file2 as modules and pass ``arg=foo`` as parameter to the
4000 first module. Commas can be provided in module parameters by doubling
4001 them on the command line to escape them:
4003 ``-initrd "bzImage earlyprintk=xen,,keep root=/dev/xvda1,initrd.img"``
4004 Multiboot only. Use bzImage as the first module with
4005 "``earlyprintk=xen,keep root=/dev/xvda1``" as its command line,
4006 and initrd.img as the second module.
4008 ERST
4010 DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \
4011 "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL)
4012 SRST
4013 ``-dtb file``
4014 Use file as a device tree binary (dtb) image and pass it to the
4015 kernel on boot.
4016 ERST
4018 SRST
4020 Finally you can also manually load images directly into the address
4021 space of the guest. This is most useful for developers who already
4022 know the layout of their guest and take care to ensure something sane
4023 will happen when the reset vector executes.
4025 The generic loader can be invoked by using the loader device:
4027 ``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]``
4029 there is also the guest loader which operates in a similar way but
4030 tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where
4031 the guest image is:
4033 ``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]``
4035 ERST
4037 DEFHEADING()
4039 DEFHEADING(Debug/Expert options:)
4041 DEF("compat", HAS_ARG, QEMU_OPTION_compat,
4042 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n"
4043 " Policy for handling deprecated management interfaces\n"
4044 "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n"
4045 " Policy for handling unstable management interfaces\n",
4046 QEMU_ARCH_ALL)
4047 SRST
4048 ``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]``
4049 Set policy for handling deprecated management interfaces (experimental):
4051 ``deprecated-input=accept`` (default)
4052 Accept deprecated commands and arguments
4053 ``deprecated-input=reject``
4054 Reject deprecated commands and arguments
4055 ``deprecated-input=crash``
4056 Crash on deprecated commands and arguments
4057 ``deprecated-output=accept`` (default)
4058 Emit deprecated command results and events
4059 ``deprecated-output=hide``
4060 Suppress deprecated command results and events
4062 Limitation: covers only syntactic aspects of QMP.
4064 ``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]``
4065 Set policy for handling unstable management interfaces (experimental):
4067 ``unstable-input=accept`` (default)
4068 Accept unstable commands and arguments
4069 ``unstable-input=reject``
4070 Reject unstable commands and arguments
4071 ``unstable-input=crash``
4072 Crash on unstable commands and arguments
4073 ``unstable-output=accept`` (default)
4074 Emit unstable command results and events
4075 ``unstable-output=hide``
4076 Suppress unstable command results and events
4078 Limitation: covers only syntactic aspects of QMP.
4079 ERST
4081 DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg,
4082 "-fw_cfg [name=]<name>,file=<file>\n"
4083 " add named fw_cfg entry with contents from file\n"
4084 "-fw_cfg [name=]<name>,string=<str>\n"
4085 " add named fw_cfg entry with contents from string\n",
4086 QEMU_ARCH_ALL)
4087 SRST
4088 ``-fw_cfg [name=]name,file=file``
4089 Add named fw\_cfg entry with contents from file file.
4090 If the filename contains comma, you must double it (for instance,
4091 "file=my,,file" to use file "my,file").
4093 ``-fw_cfg [name=]name,string=str``
4094 Add named fw\_cfg entry with contents from string str.
4095 If the string contains comma, you must double it (for instance,
4096 "string=my,,string" to use file "my,string").
4098 The terminating NUL character of the contents of str will not be
4099 included as part of the fw\_cfg item data. To insert contents with
4100 embedded NUL characters, you have to use the file parameter.
4102 The fw\_cfg entries are passed by QEMU through to the guest.
4104 Example:
4108 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin
4110 creates an fw\_cfg entry named opt/com.mycompany/blob with contents
4111 from ./my\_blob.bin.
4112 ERST
4114 DEF("serial", HAS_ARG, QEMU_OPTION_serial, \
4115 "-serial dev redirect the serial port to char device 'dev'\n",
4116 QEMU_ARCH_ALL)
4117 SRST
4118 ``-serial dev``
4119 Redirect the virtual serial port to host character device dev. The
4120 default device is ``vc`` in graphical mode and ``stdio`` in non
4121 graphical mode.
4123 This option can be used several times to simulate up to 4 serial
4124 ports.
4126 Use ``-serial none`` to disable all serial ports.
4128 Available character devices are:
4130 ``vc[:WxH]``
4131 Virtual console. Optionally, a width and height can be given in
4132 pixel with
4136 vc:800x600
4138 It is also possible to specify width or height in characters:
4142 vc:80Cx24C
4144 ``pty``
4145 [Linux only] Pseudo TTY (a new PTY is automatically allocated)
4147 ``none``
4148 No device is allocated.
4150 ``null``
4151 void device
4153 ``chardev:id``
4154 Use a named character device defined with the ``-chardev``
4155 option.
4157 ``/dev/XXX``
4158 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial
4159 port parameters are set according to the emulated ones.
4161 ``/dev/parportN``
4162 [Linux only, parallel port only] Use host parallel port N.
4163 Currently SPP and EPP parallel port features can be used.
4165 ``file:filename``
4166 Write output to filename. No character can be read.
4168 ``stdio``
4169 [Unix only] standard input/output
4171 ``pipe:filename``
4172 name pipe filename
4174 ``COMn``
4175 [Windows only] Use host serial port n
4177 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]``
4178 This implements UDP Net Console. When remote\_host or src\_ip
4179 are not specified they default to ``0.0.0.0``. When not using a
4180 specified src\_port a random port is automatically chosen.
4182 If you just want a simple readonly console you can use
4183 ``netcat`` or ``nc``, by starting QEMU with:
4184 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time
4185 QEMU writes something to that port it will appear in the
4186 netconsole session.
4188 If you plan to send characters back via netconsole or you want
4189 to stop and start QEMU a lot of times, you should have QEMU use
4190 the same source port each time by using something like ``-serial
4191 udp::4555@:4556`` to QEMU. Another approach is to use a patched
4192 version of netcat which can listen to a TCP port and send and
4193 receive characters via udp. If you have a patched version of
4194 netcat which activates telnet remote echo and single char
4195 transfer, then you can use the following options to set up a
4196 netcat redirector to allow telnet on port 5555 to access the
4197 QEMU port.
4199 ``QEMU Options:``
4200 -serial udp::4555@:4556
4202 ``netcat options:``
4203 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T
4205 ``telnet options:``
4206 localhost 5555
4208 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]``
4209 The TCP Net Console has two modes of operation. It can send the
4210 serial I/O to a location or wait for a connection from a
4211 location. By default the TCP Net Console is sent to host at the
4212 port. If you use the ``server=on`` option QEMU will wait for a client
4213 socket application to connect to the port before continuing,
4214 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off``
4215 option disables the Nagle buffering algorithm. The ``reconnect=on``
4216 option only applies if ``server=no`` is set, if the connection goes
4217 down it will attempt to reconnect at the given interval. If host
4218 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a
4219 time is accepted. You can use ``telnet=on`` to connect to the
4220 corresponding character device.
4222 ``Example to send tcp console to 192.168.0.2 port 4444``
4223 -serial tcp:192.168.0.2:4444
4225 ``Example to listen and wait on port 4444 for connection``
4226 -serial tcp::4444,server=on
4228 ``Example to not wait and listen on ip 192.168.0.100 port 4444``
4229 -serial tcp:192.168.0.100:4444,server=on,wait=off
4231 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]``
4232 The telnet protocol is used instead of raw tcp sockets. The
4233 options work the same as if you had specified ``-serial tcp``.
4234 The difference is that the port acts like a telnet server or
4235 client using telnet option negotiation. This will also allow you
4236 to send the MAGIC\_SYSRQ sequence if you use a telnet that
4237 supports sending the break sequence. Typically in unix telnet
4238 you do it with Control-] and then type "send break" followed by
4239 pressing the enter key.
4241 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]``
4242 The WebSocket protocol is used instead of raw tcp socket. The
4243 port acts as a WebSocket server. Client mode is not supported.
4245 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]``
4246 A unix domain socket is used instead of a tcp socket. The option
4247 works the same as if you had specified ``-serial tcp`` except
4248 the unix domain socket path is used for connections.
4250 ``mon:dev_string``
4251 This is a special option to allow the monitor to be multiplexed
4252 onto another serial port. The monitor is accessed with key
4253 sequence of Control-a and then pressing c. dev\_string should be
4254 any one of the serial devices specified above. An example to
4255 multiplex the monitor onto a telnet server listening on port
4256 4444 would be:
4258 ``-serial mon:telnet::4444,server=on,wait=off``
4260 When the monitor is multiplexed to stdio in this way, Ctrl+C
4261 will not terminate QEMU any more but will be passed to the guest
4262 instead.
4264 ``braille``
4265 Braille device. This will use BrlAPI to display the braille
4266 output on a real or fake device.
4268 ``msmouse``
4269 Three button serial mouse. Configure the guest to use Microsoft
4270 protocol.
4271 ERST
4273 DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \
4274 "-parallel dev redirect the parallel port to char device 'dev'\n",
4275 QEMU_ARCH_ALL)
4276 SRST
4277 ``-parallel dev``
4278 Redirect the virtual parallel port to host device dev (same devices
4279 as the serial port). On Linux hosts, ``/dev/parportN`` can be used
4280 to use hardware devices connected on the corresponding host parallel
4281 port.
4283 This option can be used several times to simulate up to 3 parallel
4284 ports.
4286 Use ``-parallel none`` to disable all parallel ports.
4287 ERST
4289 DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \
4290 "-monitor dev redirect the monitor to char device 'dev'\n",
4291 QEMU_ARCH_ALL)
4292 SRST
4293 ``-monitor dev``
4294 Redirect the monitor to host device dev (same devices as the serial
4295 port). The default device is ``vc`` in graphical mode and ``stdio``
4296 in non graphical mode. Use ``-monitor none`` to disable the default
4297 monitor.
4298 ERST
4299 DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \
4300 "-qmp dev like -monitor but opens in 'control' mode\n",
4301 QEMU_ARCH_ALL)
4302 SRST
4303 ``-qmp dev``
4304 Like ``-monitor`` but opens in 'control' mode. For example, to make
4305 QMP available on localhost port 4444::
4307 -qmp tcp:localhost:4444,server=on,wait=off
4309 Not all options are configurable via this syntax; for maximum
4310 flexibility use the ``-mon`` option and an accompanying ``-chardev``.
4312 ERST
4313 DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \
4314 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n",
4315 QEMU_ARCH_ALL)
4316 SRST
4317 ``-qmp-pretty dev``
4318 Like ``-qmp`` but uses pretty JSON formatting.
4319 ERST
4321 DEF("mon", HAS_ARG, QEMU_OPTION_mon, \
4322 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL)
4323 SRST
4324 ``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]``
4325 Set up a monitor connected to the chardev ``name``.
4326 QEMU supports two monitors: the Human Monitor Protocol
4327 (HMP; for human interaction), and the QEMU Monitor Protocol
4328 (QMP; a JSON RPC-style protocol).
4329 The default is HMP; ``mode=control`` selects QMP instead.
4330 ``pretty`` is only valid when ``mode=control``,
4331 turning on JSON pretty printing to ease
4332 human reading and debugging.
4334 For example::
4336 -chardev socket,id=mon1,host=localhost,port=4444,server=on,wait=off \
4337 -mon chardev=mon1,mode=control,pretty=on
4339 enables the QMP monitor on localhost port 4444 with pretty-printing.
4340 ERST
4342 DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \
4343 "-debugcon dev redirect the debug console to char device 'dev'\n",
4344 QEMU_ARCH_ALL)
4345 SRST
4346 ``-debugcon dev``
4347 Redirect the debug console to host device dev (same devices as the
4348 serial port). The debug console is an I/O port which is typically
4349 port 0xe9; writing to that I/O port sends output to this device. The
4350 default device is ``vc`` in graphical mode and ``stdio`` in non
4351 graphical mode.
4352 ERST
4354 DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \
4355 "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL)
4356 SRST
4357 ``-pidfile file``
4358 Store the QEMU process PID in file. It is useful if you launch QEMU
4359 from a script.
4360 ERST
4362 DEF("singlestep", 0, QEMU_OPTION_singlestep, \
4363 "-singlestep deprecated synonym for -accel tcg,one-insn-per-tb=on\n", QEMU_ARCH_ALL)
4364 SRST
4365 ``-singlestep``
4366 This is a deprecated synonym for the TCG accelerator property
4367 ``one-insn-per-tb``.
4368 ERST
4370 DEF("preconfig", 0, QEMU_OPTION_preconfig, \
4371 "--preconfig pause QEMU before machine is initialized (experimental)\n",
4372 QEMU_ARCH_ALL)
4373 SRST
4374 ``--preconfig``
4375 Pause QEMU for interactive configuration before the machine is
4376 created, which allows querying and configuring properties that will
4377 affect machine initialization. Use QMP command 'x-exit-preconfig' to
4378 exit the preconfig state and move to the next state (i.e. run guest
4379 if -S isn't used or pause the second time if -S is used). This
4380 option is experimental.
4381 ERST
4383 DEF("S", 0, QEMU_OPTION_S, \
4384 "-S freeze CPU at startup (use 'c' to start execution)\n",
4385 QEMU_ARCH_ALL)
4386 SRST
4387 ``-S``
4388 Do not start CPU at startup (you must type 'c' in the monitor).
4389 ERST
4391 DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit,
4392 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n"
4393 " run qemu with overcommit hints\n"
4394 " mem-lock=on|off controls memory lock support (default: off)\n"
4395 " cpu-pm=on|off controls cpu power management (default: off)\n",
4396 QEMU_ARCH_ALL)
4397 SRST
4398 ``-overcommit mem-lock=on|off``
4400 ``-overcommit cpu-pm=on|off``
4401 Run qemu with hints about host resource overcommit. The default is
4402 to assume that host overcommits all resources.
4404 Locking qemu and guest memory can be enabled via ``mem-lock=on``
4405 (disabled by default). This works when host memory is not
4406 overcommitted and reduces the worst-case latency for guest.
4408 Guest ability to manage power state of host cpus (increasing latency
4409 for other processes on the same host cpu, but decreasing latency for
4410 guest) can be enabled via ``cpu-pm=on`` (disabled by default). This
4411 works best when host CPU is not overcommitted. When used, host
4412 estimates of CPU cycle and power utilization will be incorrect, not
4413 taking into account guest idle time.
4414 ERST
4416 DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \
4417 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n"
4418 " the guest without waiting for gdb to connect; use -S too\n"
4419 " if you want it to not start execution.)\n",
4420 QEMU_ARCH_ALL)
4421 SRST
4422 ``-gdb dev``
4423 Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter
4424 in the System Emulation Users Guide). Note that this option does not pause QEMU
4425 execution -- if you want QEMU to not start the guest until you
4426 connect with gdb and issue a ``continue`` command, you will need to
4427 also pass the ``-S`` option to QEMU.
4429 The most usual configuration is to listen on a local TCP socket::
4431 -gdb tcp::3117
4433 but you can specify other backends; UDP, pseudo TTY, or even stdio
4434 are all reasonable use cases. For example, a stdio connection
4435 allows you to start QEMU from within gdb and establish the
4436 connection via a pipe:
4438 .. parsed-literal::
4440 (gdb) target remote | exec |qemu_system| -gdb stdio ...
4441 ERST
4443 DEF("s", 0, QEMU_OPTION_s, \
4444 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n",
4445 QEMU_ARCH_ALL)
4446 SRST
4447 ``-s``
4448 Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234
4449 (see the :ref:`GDB usage` chapter in the System Emulation Users Guide).
4450 ERST
4452 DEF("d", HAS_ARG, QEMU_OPTION_d, \
4453 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n",
4454 QEMU_ARCH_ALL)
4455 SRST
4456 ``-d item1[,...]``
4457 Enable logging of specified items. Use '-d help' for a list of log
4458 items.
4459 ERST
4461 DEF("D", HAS_ARG, QEMU_OPTION_D, \
4462 "-D logfile output log to logfile (default stderr)\n",
4463 QEMU_ARCH_ALL)
4464 SRST
4465 ``-D logfile``
4466 Output log in logfile instead of to stderr
4467 ERST
4469 DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \
4470 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n",
4471 QEMU_ARCH_ALL)
4472 SRST
4473 ``-dfilter range1[,...]``
4474 Filter debug output to that relevant to a range of target addresses.
4475 The filter spec can be either start+size, start-size or start..end
4476 where start end and size are the addresses and sizes required. For
4477 example:
4481 -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000
4483 Will dump output for any code in the 0x1000 sized block starting at
4484 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and
4485 another 0x1000 sized block starting at 0xffffffc00005f000.
4486 ERST
4488 DEF("seed", HAS_ARG, QEMU_OPTION_seed, \
4489 "-seed number seed the pseudo-random number generator\n",
4490 QEMU_ARCH_ALL)
4491 SRST
4492 ``-seed number``
4493 Force the guest to use a deterministic pseudo-random number
4494 generator, seeded with number. This does not affect crypto routines
4495 within the host.
4496 ERST
4498 DEF("L", HAS_ARG, QEMU_OPTION_L, \
4499 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n",
4500 QEMU_ARCH_ALL)
4501 SRST
4502 ``-L path``
4503 Set the directory for the BIOS, VGA BIOS and keymaps.
4505 To list all the data directories, use ``-L help``.
4506 ERST
4508 DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \
4509 "-enable-kvm enable KVM full virtualization support\n",
4510 QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC |
4511 QEMU_ARCH_RISCV | QEMU_ARCH_S390X)
4512 SRST
4513 ``-enable-kvm``
4514 Enable KVM full virtualization support. This option is only
4515 available if KVM support is enabled when compiling.
4516 ERST
4518 DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid,
4519 "-xen-domid id specify xen guest domain id\n",
4520 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4521 DEF("xen-attach", 0, QEMU_OPTION_xen_attach,
4522 "-xen-attach attach to existing xen domain\n"
4523 " libxl will use this when starting QEMU\n",
4524 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4525 DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict,
4526 "-xen-domid-restrict restrict set of available xen operations\n"
4527 " to specified domain id. (Does not affect\n"
4528 " xenpv machine type).\n",
4529 QEMU_ARCH_ARM | QEMU_ARCH_I386)
4530 SRST
4531 ``-xen-domid id``
4532 Specify xen guest domain id (XEN only).
4534 ``-xen-attach``
4535 Attach to existing xen domain. libxl will use this when starting
4536 QEMU (XEN only). Restrict set of available xen operations to
4537 specified domain id (XEN only).
4538 ERST
4540 DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \
4541 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL)
4542 SRST
4543 ``-no-reboot``
4544 Exit instead of rebooting.
4545 ERST
4547 DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \
4548 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL)
4549 SRST
4550 ``-no-shutdown``
4551 Don't exit QEMU on guest shutdown, but instead only stop the
4552 emulation. This allows for instance switching to monitor to commit
4553 changes to the disk image.
4554 ERST
4556 DEF("action", HAS_ARG, QEMU_OPTION_action,
4557 "-action reboot=reset|shutdown\n"
4558 " action when guest reboots [default=reset]\n"
4559 "-action shutdown=poweroff|pause\n"
4560 " action when guest shuts down [default=poweroff]\n"
4561 "-action panic=pause|shutdown|exit-failure|none\n"
4562 " action when guest panics [default=shutdown]\n"
4563 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n"
4564 " action when watchdog fires [default=reset]\n",
4565 QEMU_ARCH_ALL)
4566 SRST
4567 ``-action event=action``
4568 The action parameter serves to modify QEMU's default behavior when
4569 certain guest events occur. It provides a generic method for specifying the
4570 same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown``
4571 parameters.
4573 Examples:
4575 ``-action panic=none``
4576 ``-action reboot=shutdown,shutdown=pause``
4577 ``-device i6300esb -action watchdog=pause``
4579 ERST
4581 DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \
4582 "-loadvm [tag|id]\n" \
4583 " start right away with a saved state (loadvm in monitor)\n",
4584 QEMU_ARCH_ALL)
4585 SRST
4586 ``-loadvm file``
4587 Start right away with a saved state (``loadvm`` in monitor)
4588 ERST
4590 #ifndef _WIN32
4591 DEF("daemonize", 0, QEMU_OPTION_daemonize, \
4592 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL)
4593 #endif
4594 SRST
4595 ``-daemonize``
4596 Daemonize the QEMU process after initialization. QEMU will not
4597 detach from standard IO until it is ready to receive connections on
4598 any of its devices. This option is a useful way for external
4599 programs to launch QEMU without having to cope with initialization
4600 race conditions.
4601 ERST
4603 DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \
4604 "-option-rom rom load a file, rom, into the option ROM space\n",
4605 QEMU_ARCH_ALL)
4606 SRST
4607 ``-option-rom file``
4608 Load the contents of file as an option ROM. This option is useful to
4609 load things like EtherBoot.
4610 ERST
4612 DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \
4613 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \
4614 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n",
4615 QEMU_ARCH_ALL)
4617 SRST
4618 ``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]``
4619 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at
4620 the current UTC or local time, respectively. ``localtime`` is
4621 required for correct date in MS-DOS or Windows. To start at a
4622 specific point in time, provide datetime in the format
4623 ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC.
4625 By default the RTC is driven by the host system time. This allows
4626 using of the RTC as accurate reference clock inside the guest,
4627 specifically if the host time is smoothly following an accurate
4628 external reference clock, e.g. via NTP. If you want to isolate the
4629 guest time from the host, you can set ``clock`` to ``rt`` instead,
4630 which provides a host monotonic clock if host support it. To even
4631 prevent the RTC from progressing during suspension, you can set
4632 ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is
4633 recommended especially in icount mode in order to preserve
4634 determinism; however, note that in icount mode the speed of the
4635 virtual clock is variable and can in general differ from the host
4636 clock.
4638 Enable ``driftfix`` (i386 targets only) if you experience time drift
4639 problems, specifically with Windows' ACPI HAL. This option will try
4640 to figure out how many timer interrupts were not processed by the
4641 Windows guest and will re-inject them.
4642 ERST
4644 DEF("icount", HAS_ARG, QEMU_OPTION_icount, \
4645 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \
4646 " enable virtual instruction counter with 2^N clock ticks per\n" \
4647 " instruction, enable aligning the host and virtual clocks\n" \
4648 " or disable real time cpu sleeping, and optionally enable\n" \
4649 " record-and-replay mode\n", QEMU_ARCH_ALL)
4650 SRST
4651 ``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]``
4652 Enable virtual instruction counter. The virtual cpu will execute one
4653 instruction every 2^N ns of virtual time. If ``auto`` is specified
4654 then the virtual cpu speed will be automatically adjusted to keep
4655 virtual time within a few seconds of real time.
4657 Note that while this option can give deterministic behavior, it does
4658 not provide cycle accurate emulation. Modern CPUs contain
4659 superscalar out of order cores with complex cache hierarchies. The
4660 number of instructions executed often has little or no correlation
4661 with actual performance.
4663 When the virtual cpu is sleeping, the virtual time will advance at
4664 default speed unless ``sleep=on`` is specified. With
4665 ``sleep=on``, the virtual time will jump to the next timer
4666 deadline instantly whenever the virtual cpu goes to sleep mode and
4667 will not advance if no timer is enabled. This behavior gives
4668 deterministic execution times from the guest point of view.
4669 The default if icount is enabled is ``sleep=off``.
4670 ``sleep=on`` cannot be used together with either ``shift=auto``
4671 or ``align=on``.
4673 ``align=on`` will activate the delay algorithm which will try to
4674 synchronise the host clock and the virtual clock. The goal is to
4675 have a guest running at the real frequency imposed by the shift
4676 option. Whenever the guest clock is behind the host clock and if
4677 ``align=on`` is specified then we print a message to the user to
4678 inform about the delay. Currently this option does not work when
4679 ``shift`` is ``auto``. Note: The sync algorithm will work for those
4680 shift values for which the guest clock runs ahead of the host clock.
4681 Typically this happens when the shift value is high (how high
4682 depends on the host machine). The default if icount is enabled
4683 is ``align=off``.
4685 When the ``rr`` option is specified deterministic record/replay is
4686 enabled. The ``rrfile=`` option must also be provided to
4687 specify the path to the replay log. In record mode data is written
4688 to this file, and in replay mode it is read back.
4689 If the ``rrsnapshot`` option is given then it specifies a VM snapshot
4690 name. In record mode, a new VM snapshot with the given name is created
4691 at the start of execution recording. In replay mode this option
4692 specifies the snapshot name used to load the initial VM state.
4693 ERST
4695 DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \
4696 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \
4697 " action when watchdog fires [default=reset]\n",
4698 QEMU_ARCH_ALL)
4699 SRST
4700 ``-watchdog-action action``
4701 The action controls what QEMU will do when the watchdog timer
4702 expires. The default is ``reset`` (forcefully reset the guest).
4703 Other possible actions are: ``shutdown`` (attempt to gracefully
4704 shutdown the guest), ``poweroff`` (forcefully poweroff the guest),
4705 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the
4706 guest), ``debug`` (print a debug message and continue), or ``none``
4707 (do nothing).
4709 Note that the ``shutdown`` action requires that the guest responds
4710 to ACPI signals, which it may not be able to do in the sort of
4711 situations where the watchdog would have expired, and thus
4712 ``-watchdog-action shutdown`` is not recommended for production use.
4714 Examples:
4716 ``-device i6300esb -watchdog-action pause``
4718 ERST
4720 DEF("echr", HAS_ARG, QEMU_OPTION_echr, \
4721 "-echr chr set terminal escape character instead of ctrl-a\n",
4722 QEMU_ARCH_ALL)
4723 SRST
4724 ``-echr numeric_ascii_value``
4725 Change the escape character used for switching to the monitor when
4726 using monitor and serial sharing. The default is ``0x01`` when using
4727 the ``-nographic`` option. ``0x01`` is equal to pressing
4728 ``Control-a``. You can select a different character from the ascii
4729 control keys where 1 through 26 map to Control-a through Control-z.
4730 For instance you could use the either of the following to change the
4731 escape character to Control-t.
4733 ``-echr 0x14``; \ ``-echr 20``
4735 ERST
4737 DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \
4738 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \
4739 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \
4740 "-incoming unix:socketpath\n" \
4741 " prepare for incoming migration, listen on\n" \
4742 " specified protocol and socket address\n" \
4743 "-incoming fd:fd\n" \
4744 "-incoming file:filename[,offset=offset]\n" \
4745 "-incoming exec:cmdline\n" \
4746 " accept incoming migration on given file descriptor\n" \
4747 " or from given external command\n" \
4748 "-incoming defer\n" \
4749 " wait for the URI to be specified via migrate_incoming\n",
4750 QEMU_ARCH_ALL)
4751 SRST
4752 ``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]``
4754 ``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]``
4755 Prepare for incoming migration, listen on a given tcp port.
4757 ``-incoming unix:socketpath``
4758 Prepare for incoming migration, listen on a given unix socket.
4760 ``-incoming fd:fd``
4761 Accept incoming migration from a given file descriptor.
4763 ``-incoming file:filename[,offset=offset]``
4764 Accept incoming migration from a given file starting at offset.
4765 offset allows the common size suffixes, or a 0x prefix, but not both.
4767 ``-incoming exec:cmdline``
4768 Accept incoming migration as an output from specified external
4769 command.
4771 ``-incoming defer``
4772 Wait for the URI to be specified via migrate\_incoming. The monitor
4773 can be used to change settings (such as migration parameters) prior
4774 to issuing the migrate\_incoming to allow the migration to begin.
4775 ERST
4777 DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \
4778 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL)
4779 SRST
4780 ``-only-migratable``
4781 Only allow migratable devices. Devices will not be allowed to enter
4782 an unmigratable state.
4783 ERST
4785 DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \
4786 "-nodefaults don't create default devices\n", QEMU_ARCH_ALL)
4787 SRST
4788 ``-nodefaults``
4789 Don't create default devices. Normally, QEMU sets the default
4790 devices like serial port, parallel port, virtual console, monitor
4791 device, VGA adapter, floppy and CD-ROM drive and others. The
4792 ``-nodefaults`` option will disable all those default devices.
4793 ERST
4795 #ifndef _WIN32
4796 DEF("chroot", HAS_ARG, QEMU_OPTION_chroot, \
4797 "-chroot dir chroot to dir just before starting the VM (deprecated)\n",
4798 QEMU_ARCH_ALL)
4799 #endif
4800 SRST
4801 ``-chroot dir``
4802 Deprecated, use '-run-with chroot=...' instead.
4803 Immediately before starting guest execution, chroot to the specified
4804 directory. Especially useful in combination with -runas.
4805 ERST
4807 #ifndef _WIN32
4808 DEF("runas", HAS_ARG, QEMU_OPTION_runas, \
4809 "-runas user change to user id user just before starting the VM\n" \
4810 " user can be numeric uid:gid instead\n",
4811 QEMU_ARCH_ALL)
4812 #endif
4813 SRST
4814 ``-runas user``
4815 Immediately before starting guest execution, drop root privileges,
4816 switching to the specified user.
4817 ERST
4819 DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env,
4820 "-prom-env variable=value\n"
4821 " set OpenBIOS nvram variables\n",
4822 QEMU_ARCH_PPC | QEMU_ARCH_SPARC)
4823 SRST
4824 ``-prom-env variable=value``
4825 Set OpenBIOS nvram variable to given value (PPC, SPARC only).
4829 qemu-system-sparc -prom-env 'auto-boot?=false' \
4830 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single'
4834 qemu-system-ppc -prom-env 'auto-boot?=false' \
4835 -prom-env 'boot-device=hd:2,\yaboot' \
4836 -prom-env 'boot-args=conf=hd:2,\yaboot.conf'
4837 ERST
4838 DEF("semihosting", 0, QEMU_OPTION_semihosting,
4839 "-semihosting semihosting mode\n",
4840 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA |
4841 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4842 SRST
4843 ``-semihosting``
4844 Enable :ref:`Semihosting` mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only).
4846 .. warning::
4847 Note that this allows guest direct access to the host filesystem, so
4848 should only be used with a trusted guest OS.
4850 See the -semihosting-config option documentation for further
4851 information about the facilities this enables.
4852 ERST
4853 DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config,
4854 "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]\n" \
4855 " semihosting configuration\n",
4856 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA |
4857 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4858 SRST
4859 ``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]``
4860 Enable and configure :ref:`Semihosting` (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V
4861 only).
4863 .. warning::
4864 Note that this allows guest direct access to the host filesystem, so
4865 should only be used with a trusted guest OS.
4867 ``target=native|gdb|auto``
4868 Defines where the semihosting calls will be addressed, to QEMU
4869 (``native``) or to GDB (``gdb``). The default is ``auto``, which
4870 means ``gdb`` during debug sessions and ``native`` otherwise.
4872 ``chardev=str1``
4873 Send the output to a chardev backend output for native or auto
4874 output when not in gdb
4876 ``userspace=on|off``
4877 Allows code running in guest userspace to access the semihosting
4878 interface. The default is that only privileged guest code can
4879 make semihosting calls. Note that setting ``userspace=on`` should
4880 only be used if all guest code is trusted (for example, in
4881 bare-metal test case code).
4883 ``arg=str1,arg=str2,...``
4884 Allows the user to pass input arguments, and can be used
4885 multiple times to build up a list. The old-style
4886 ``-kernel``/``-append`` method of passing a command line is
4887 still supported for backward compatibility. If both the
4888 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are
4889 specified, the former is passed to semihosting as it always
4890 takes precedence.
4891 ERST
4892 DEF("old-param", 0, QEMU_OPTION_old_param,
4893 "-old-param old param mode\n", QEMU_ARCH_ARM)
4894 SRST
4895 ``-old-param``
4896 Old param mode (ARM only).
4897 ERST
4899 DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \
4900 "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \
4901 " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \
4902 " Enable seccomp mode 2 system call filter (default 'off').\n" \
4903 " use 'obsolete' to allow obsolete system calls that are provided\n" \
4904 " by the kernel, but typically no longer used by modern\n" \
4905 " C library implementations.\n" \
4906 " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \
4907 " to elevate privileges using set*uid|gid system calls.\n" \
4908 " The value 'children' will deny set*uid|gid system calls for\n" \
4909 " main QEMU process but will allow forks and execves to run unprivileged\n" \
4910 " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \
4911 " blocking *fork and execve\n" \
4912 " use 'resourcecontrol' to disable process affinity and schedular priority\n",
4913 QEMU_ARCH_ALL)
4914 SRST
4915 ``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]``
4916 Enable Seccomp mode 2 system call filter. 'on' will enable syscall
4917 filtering and 'off' will disable it. The default is 'off'.
4919 ``obsolete=string``
4920 Enable Obsolete system calls
4922 ``elevateprivileges=string``
4923 Disable set\*uid\|gid system calls
4925 ``spawn=string``
4926 Disable \*fork and execve
4928 ``resourcecontrol=string``
4929 Disable process affinity and schedular priority
4930 ERST
4932 DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig,
4933 "-readconfig <file>\n"
4934 " read config file\n", QEMU_ARCH_ALL)
4935 SRST
4936 ``-readconfig file``
4937 Read device configuration from file. This approach is useful when
4938 you want to spawn QEMU process with many command line options but
4939 you don't want to exceed the command line character limit.
4940 ERST
4942 DEF("no-user-config", 0, QEMU_OPTION_nouserconfig,
4943 "-no-user-config\n"
4944 " do not load default user-provided config files at startup\n",
4945 QEMU_ARCH_ALL)
4946 SRST
4947 ``-no-user-config``
4948 The ``-no-user-config`` option makes QEMU not load any of the
4949 user-provided config files on sysconfdir.
4950 ERST
4952 DEF("trace", HAS_ARG, QEMU_OPTION_trace,
4953 "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n"
4954 " specify tracing options\n",
4955 QEMU_ARCH_ALL)
4956 SRST
4957 ``-trace [[enable=]pattern][,events=file][,file=file]``
4958 .. include:: ../qemu-option-trace.rst.inc
4960 ERST
4961 DEF("plugin", HAS_ARG, QEMU_OPTION_plugin,
4962 "-plugin [file=]<file>[,<argname>=<argvalue>]\n"
4963 " load a plugin\n",
4964 QEMU_ARCH_ALL)
4965 SRST
4966 ``-plugin file=file[,argname=argvalue]``
4967 Load a plugin.
4969 ``file=file``
4970 Load the given plugin from a shared library file.
4972 ``argname=argvalue``
4973 Argument passed to the plugin. (Can be given multiple times.)
4974 ERST
4976 HXCOMM Internal use
4977 DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL)
4978 DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL)
4980 #ifdef __linux__
4981 DEF("async-teardown", 0, QEMU_OPTION_asyncteardown,
4982 "-async-teardown enable asynchronous teardown\n",
4983 QEMU_ARCH_ALL)
4984 SRST
4985 ``-async-teardown``
4986 This option is deprecated and should no longer be used. The new option
4987 ``-run-with async-teardown=on`` is a replacement.
4988 ERST
4989 #endif
4990 #ifdef CONFIG_POSIX
4991 DEF("run-with", HAS_ARG, QEMU_OPTION_run_with,
4992 "-run-with [async-teardown=on|off][,chroot=dir]\n"
4993 " Set miscellaneous QEMU process lifecycle options:\n"
4994 " async-teardown=on enables asynchronous teardown (Linux only)\n"
4995 " chroot=dir chroot to dir just before starting the VM\n",
4996 QEMU_ARCH_ALL)
4997 SRST
4998 ``-run-with [async-teardown=on|off][,chroot=dir]``
4999 Set QEMU process lifecycle options.
5001 ``async-teardown=on`` enables asynchronous teardown. A new process called
5002 "cleanup/<QEMU_PID>" will be created at startup sharing the address
5003 space with the main QEMU process, using clone. It will wait for the
5004 main QEMU process to terminate completely, and then exit. This allows
5005 QEMU to terminate very quickly even if the guest was huge, leaving the
5006 teardown of the address space to the cleanup process. Since the cleanup
5007 process shares the same cgroups as the main QEMU process, accounting is
5008 performed correctly. This only works if the cleanup process is not
5009 forcefully killed with SIGKILL before the main QEMU process has
5010 terminated completely.
5012 ``chroot=dir`` can be used for doing a chroot to the specified directory
5013 immediately before starting the guest execution. This is especially useful
5014 in combination with -runas.
5015 ERST
5016 #endif
5018 DEF("msg", HAS_ARG, QEMU_OPTION_msg,
5019 "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n"
5020 " control error message format\n"
5021 " timestamp=on enables timestamps (default: off)\n"
5022 " guest-name=on enables guest name prefix but only if\n"
5023 " -name guest option is set (default: off)\n",
5024 QEMU_ARCH_ALL)
5025 SRST
5026 ``-msg [timestamp[=on|off]][,guest-name[=on|off]]``
5027 Control error message format.
5029 ``timestamp=on|off``
5030 Prefix messages with a timestamp. Default is off.
5032 ``guest-name=on|off``
5033 Prefix messages with guest name but only if -name guest option is set
5034 otherwise the option is ignored. Default is off.
5035 ERST
5037 DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate,
5038 "-dump-vmstate <file>\n"
5039 " Output vmstate information in JSON format to file.\n"
5040 " Use the scripts/vmstate-static-checker.py file to\n"
5041 " check for possible regressions in migration code\n"
5042 " by comparing two such vmstate dumps.\n",
5043 QEMU_ARCH_ALL)
5044 SRST
5045 ``-dump-vmstate file``
5046 Dump json-encoded vmstate information for current machine type to
5047 file in file
5048 ERST
5050 DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile,
5051 "-enable-sync-profile\n"
5052 " enable synchronization profiling\n",
5053 QEMU_ARCH_ALL)
5054 SRST
5055 ``-enable-sync-profile``
5056 Enable synchronization profiling.
5057 ERST
5059 #if defined(CONFIG_TCG) && defined(CONFIG_LINUX)
5060 DEF("perfmap", 0, QEMU_OPTION_perfmap,
5061 "-perfmap generate a /tmp/perf-${pid}.map file for perf\n",
5062 QEMU_ARCH_ALL)
5063 SRST
5064 ``-perfmap``
5065 Generate a map file for Linux perf tools that will allow basic profiling
5066 information to be broken down into basic blocks.
5067 ERST
5069 DEF("jitdump", 0, QEMU_OPTION_jitdump,
5070 "-jitdump generate a jit-${pid}.dump file for perf\n",
5071 QEMU_ARCH_ALL)
5072 SRST
5073 ``-jitdump``
5074 Generate a dump file for Linux perf tools that maps basic blocks to symbol
5075 names, line numbers and JITted code.
5076 ERST
5077 #endif
5079 DEFHEADING()
5081 DEFHEADING(Generic object creation:)
5083 DEF("object", HAS_ARG, QEMU_OPTION_object,
5084 "-object TYPENAME[,PROP1=VALUE1,...]\n"
5085 " create a new object of type TYPENAME setting properties\n"
5086 " in the order they are specified. Note that the 'id'\n"
5087 " property must be set. These objects are placed in the\n"
5088 " '/objects' path.\n",
5089 QEMU_ARCH_ALL)
5090 SRST
5091 ``-object typename[,prop1=value1,...]``
5092 Create a new object of type typename setting properties in the order
5093 they are specified. Note that the 'id' property must be set. These
5094 objects are placed in the '/objects' path.
5096 ``-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,offset=offset,readonly=on|off,rom=on|off|auto``
5097 Creates a memory file backend object, which can be used to back
5098 the guest RAM with huge pages.
5100 The ``id`` parameter is a unique ID that will be used to
5101 reference this memory region in other parameters, e.g. ``-numa``,
5102 ``-device nvdimm``, etc.
5104 The ``size`` option provides the size of the memory region, and
5105 accepts common suffixes, e.g. ``500M``.
5107 The ``mem-path`` provides the path to either a shared memory or
5108 huge page filesystem mount.
5110 The ``share`` boolean option determines whether the memory
5111 region is marked as private to QEMU, or shared. The latter
5112 allows a co-operating external process to access the QEMU memory
5113 region.
5115 The ``share`` is also required for pvrdma devices due to
5116 limitations in the RDMA API provided by Linux.
5118 Setting share=on might affect the ability to configure NUMA
5119 bindings for the memory backend under some circumstances, see
5120 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel
5121 source tree for additional details.
5123 Setting the ``discard-data`` boolean option to on indicates that
5124 file contents can be destroyed when QEMU exits, to avoid
5125 unnecessarily flushing data to the backing file. Note that
5126 ``discard-data`` is only an optimization, and QEMU might not
5127 discard file contents if it aborts unexpectedly or is terminated
5128 using SIGKILL.
5130 The ``merge`` boolean option enables memory merge, also known as
5131 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider
5132 the pages for memory deduplication.
5134 Setting the ``dump`` boolean option to off excludes the memory
5135 from core dumps. This feature is also known as MADV\_DONTDUMP.
5137 The ``prealloc`` boolean option enables memory preallocation.
5139 The ``host-nodes`` option binds the memory range to a list of
5140 NUMA host nodes.
5142 The ``policy`` option sets the NUMA policy to one of the
5143 following values:
5145 ``default``
5146 default host policy
5148 ``preferred``
5149 prefer the given host node list for allocation
5151 ``bind``
5152 restrict memory allocation to the given host node list
5154 ``interleave``
5155 interleave memory allocations across the given host node
5156 list
5158 The ``align`` option specifies the base address alignment when
5159 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg
5160 ``2M``. Some backend store specified by ``mem-path`` requires an
5161 alignment different than the default one used by QEMU, eg the
5162 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
5163 such cases, users can specify the required alignment via this
5164 option.
5166 The ``offset`` option specifies the offset into the target file
5167 that the region starts at. You can use this parameter to back
5168 multiple regions with a single file.
5170 The ``pmem`` option specifies whether the backing file specified
5171 by ``mem-path`` is in host persistent memory that can be
5172 accessed using the SNIA NVM programming model (e.g. Intel
5173 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary
5174 operations to guarantee the persistence of its own writes to
5175 ``mem-path`` (e.g. in vNVDIMM label emulation and live
5176 migration). Also, we will map the backend-file with MAP\_SYNC
5177 flag, which ensures the file metadata is in sync for
5178 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC
5179 requires support from both the host kernel (since Linux kernel
5180 4.15) and the filesystem of ``mem-path`` mounted with DAX
5181 option.
5183 The ``readonly`` option specifies whether the backing file is opened
5184 read-only or read-write (default).
5186 The ``rom`` option specifies whether to create Read Only Memory
5187 (ROM) that cannot be modified by the VM. Any write attempts to such
5188 ROM will be denied. Most use cases want proper RAM instead of ROM.
5189 However, selected use cases, like R/O NVDIMMs, can benefit from
5190 ROM. If set to ``on``, create ROM; if set to ``off``, create
5191 writable RAM; if set to ``auto`` (default), the value of the
5192 ``readonly`` option is used. This option is primarily helpful when
5193 we want to have writable RAM in configurations that would
5194 traditionally create ROM before the ``rom`` option was introduced:
5195 VM templating, where we want to open a file readonly
5196 (``readonly=on``) and mark the memory to be private for QEMU
5197 (``share=off``). For this use case, we need writable RAM instead
5198 of ROM, and want to also set ``rom=off``.
5200 ``-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``
5201 Creates a memory backend object, which can be used to back the
5202 guest RAM. Memory backend objects offer more control than the
5203 ``-m`` option that is traditionally used to define guest RAM.
5204 Please refer to ``memory-backend-file`` for a description of the
5205 options.
5207 ``-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``
5208 Creates an anonymous memory file backend object, which allows
5209 QEMU to share the memory with an external process (e.g. when
5210 using vhost-user). The memory is allocated with memfd and
5211 optional sealing. (Linux only)
5213 The ``seal`` option creates a sealed-file, that will block
5214 further resizing the memory ('on' by default).
5216 The ``hugetlb`` option specify the file to be created resides in
5217 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction
5218 with the ``hugetlb`` option, the ``hugetlbsize`` option specify
5219 the hugetlb page size on systems that support multiple hugetlb
5220 page sizes (it must be a power of 2 value supported by the
5221 system).
5223 In some versions of Linux, the ``hugetlb`` option is
5224 incompatible with the ``seal`` option (requires at least Linux
5225 4.16).
5227 Please refer to ``memory-backend-file`` for a description of the
5228 other options.
5230 The ``share`` boolean option is on by default with memfd.
5232 ``-object iommufd,id=id[,fd=fd]``
5233 Creates an iommufd backend which allows control of DMA mapping
5234 through the ``/dev/iommu`` device.
5236 The ``id`` parameter is a unique ID which frontends (such as
5237 vfio-pci of vdpa) will use to connect with the iommufd backend.
5239 The ``fd`` parameter is an optional pre-opened file descriptor
5240 resulting from ``/dev/iommu`` opening. Usually the iommufd is shared
5241 across all subsystems, bringing the benefit of centralized
5242 reference counting.
5244 ``-object rng-builtin,id=id``
5245 Creates a random number generator backend which obtains entropy
5246 from QEMU builtin functions. The ``id`` parameter is a unique ID
5247 that will be used to reference this entropy backend from the
5248 ``virtio-rng`` device. By default, the ``virtio-rng`` device
5249 uses this RNG backend.
5251 ``-object rng-random,id=id,filename=/dev/random``
5252 Creates a random number generator backend which obtains entropy
5253 from a device on the host. The ``id`` parameter is a unique ID
5254 that will be used to reference this entropy backend from the
5255 ``virtio-rng`` device. The ``filename`` parameter specifies
5256 which file to obtain entropy from and if omitted defaults to
5257 ``/dev/urandom``.
5259 ``-object rng-egd,id=id,chardev=chardevid``
5260 Creates a random number generator backend which obtains entropy
5261 from an external daemon running on the host. The ``id``
5262 parameter is a unique ID that will be used to reference this
5263 entropy backend from the ``virtio-rng`` device. The ``chardev``
5264 parameter is the unique ID of a character device backend that
5265 provides the connection to the RNG daemon.
5267 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off``
5268 Creates a TLS anonymous credentials object, which can be used to
5269 provide TLS support on network backends. The ``id`` parameter is
5270 a unique ID which network backends will use to access the
5271 credentials. The ``endpoint`` is either ``server`` or ``client``
5272 depending on whether the QEMU network backend that uses the
5273 credentials will be acting as a client or as a server. If
5274 ``verify-peer`` is enabled (the default) then once the handshake
5275 is completed, the peer credentials will be verified, though this
5276 is a no-op for anonymous credentials.
5278 The dir parameter tells QEMU where to find the credential files.
5279 For server endpoints, this directory may contain a file
5280 dh-params.pem providing diffie-hellman parameters to use for the
5281 TLS server. If the file is missing, QEMU will generate a set of
5282 DH parameters at startup. This is a computationally expensive
5283 operation that consumes random pool entropy, so it is
5284 recommended that a persistent set of parameters be generated
5285 upfront and saved.
5287 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]``
5288 Creates a TLS Pre-Shared Keys (PSK) credentials object, which
5289 can be used to provide TLS support on network backends. The
5290 ``id`` parameter is a unique ID which network backends will use
5291 to access the credentials. The ``endpoint`` is either ``server``
5292 or ``client`` depending on whether the QEMU network backend that
5293 uses the credentials will be acting as a client or as a server.
5294 For clients only, ``username`` is the username which will be
5295 sent to the server. If omitted it defaults to "qemu".
5297 The dir parameter tells QEMU where to find the keys file. It is
5298 called "dir/keys.psk" and contains "username:key" pairs. This
5299 file can most easily be created using the GnuTLS ``psktool``
5300 program.
5302 For server endpoints, dir may also contain a file dh-params.pem
5303 providing diffie-hellman parameters to use for the TLS server.
5304 If the file is missing, QEMU will generate a set of DH
5305 parameters at startup. This is a computationally expensive
5306 operation that consumes random pool entropy, so it is
5307 recommended that a persistent set of parameters be generated up
5308 front and saved.
5310 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id``
5311 Creates a TLS anonymous credentials object, which can be used to
5312 provide TLS support on network backends. The ``id`` parameter is
5313 a unique ID which network backends will use to access the
5314 credentials. The ``endpoint`` is either ``server`` or ``client``
5315 depending on whether the QEMU network backend that uses the
5316 credentials will be acting as a client or as a server. If
5317 ``verify-peer`` is enabled (the default) then once the handshake
5318 is completed, the peer credentials will be verified. With x509
5319 certificates, this implies that the clients must be provided
5320 with valid client certificates too.
5322 The dir parameter tells QEMU where to find the credential files.
5323 For server endpoints, this directory may contain a file
5324 dh-params.pem providing diffie-hellman parameters to use for the
5325 TLS server. If the file is missing, QEMU will generate a set of
5326 DH parameters at startup. This is a computationally expensive
5327 operation that consumes random pool entropy, so it is
5328 recommended that a persistent set of parameters be generated
5329 upfront and saved.
5331 For x509 certificate credentials the directory will contain
5332 further files providing the x509 certificates. The certificates
5333 must be stored in PEM format, in filenames ca-cert.pem,
5334 ca-crl.pem (optional), server-cert.pem (only servers),
5335 server-key.pem (only servers), client-cert.pem (only clients),
5336 and client-key.pem (only clients).
5338 For the server-key.pem and client-key.pem files which contain
5339 sensitive private keys, it is possible to use an encrypted
5340 version by providing the passwordid parameter. This provides the
5341 ID of a previously created ``secret`` object containing the
5342 password for decryption.
5344 The priority parameter allows to override the global default
5345 priority used by gnutls. This can be useful if the system
5346 administrator needs to use a weaker set of crypto priorities for
5347 QEMU without potentially forcing the weakness onto all
5348 applications. Or conversely if one wants wants a stronger
5349 default for QEMU than for all other applications, they can do
5350 this through this parameter. Its format is a gnutls priority
5351 string as described at
5352 https://gnutls.org/manual/html_node/Priority-Strings.html.
5354 ``-object tls-cipher-suites,id=id,priority=priority``
5355 Creates a TLS cipher suites object, which can be used to control
5356 the TLS cipher/protocol algorithms that applications are permitted
5357 to use.
5359 The ``id`` parameter is a unique ID which frontends will use to
5360 access the ordered list of permitted TLS cipher suites from the
5361 host.
5363 The ``priority`` parameter allows to override the global default
5364 priority used by gnutls. This can be useful if the system
5365 administrator needs to use a weaker set of crypto priorities for
5366 QEMU without potentially forcing the weakness onto all
5367 applications. Or conversely if one wants wants a stronger
5368 default for QEMU than for all other applications, they can do
5369 this through this parameter. Its format is a gnutls priority
5370 string as described at
5371 https://gnutls.org/manual/html_node/Priority-Strings.html.
5373 An example of use of this object is to control UEFI HTTPS Boot.
5374 The tls-cipher-suites object exposes the ordered list of permitted
5375 TLS cipher suites from the host side to the guest firmware, via
5376 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER
5377 objects. The firmware uses the IANA_TLS_CIPHER array for configuring
5378 guest-side TLS.
5380 In the following example, the priority at which the host-side policy
5381 is retrieved is given by the ``priority`` property.
5382 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to
5383 refer to /etc/crypto-policies/back-ends/gnutls.config.
5385 .. parsed-literal::
5387 # |qemu_system| \\
5388 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\
5389 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0
5391 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]``
5392 Interval t can't be 0, this filter batches the packet delivery:
5393 all packets arriving in a given interval on netdev netdevid are
5394 delayed until the end of the interval. Interval is in
5395 microseconds. ``status`` is optional that indicate whether the
5396 netfilter is on (enabled) or off (disabled), the default status
5397 for netfilter will be 'on'.
5399 queue all\|rx\|tx is an option that can be applied to any
5400 netfilter.
5402 ``all``: the filter is attached both to the receive and the
5403 transmit queue of the netdev (default).
5405 ``rx``: the filter is attached to the receive queue of the
5406 netdev, where it will receive packets sent to the netdev.
5408 ``tx``: the filter is attached to the transmit queue of the
5409 netdev, where it will receive packets sent by the netdev.
5411 position head\|tail\|id=<id> is an option to specify where the
5412 filter should be inserted in the filter list. It can be applied
5413 to any netfilter.
5415 ``head``: the filter is inserted at the head of the filter list,
5416 before any existing filters.
5418 ``tail``: the filter is inserted at the tail of the filter list,
5419 behind any existing filters (default).
5421 ``id=<id>``: the filter is inserted before or behind the filter
5422 specified by <id>, see the insert option below.
5424 insert behind\|before is an option to specify where to insert
5425 the new filter relative to the one specified with
5426 position=id=<id>. It can be applied to any netfilter.
5428 ``before``: insert before the specified filter.
5430 ``behind``: insert behind the specified filter (default).
5432 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
5433 filter-mirror on netdev netdevid,mirror net packet to
5434 chardevchardevid, if it has the vnet\_hdr\_support flag,
5435 filter-mirror will mirror packet with vnet\_hdr\_len.
5437 ``-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]``
5438 filter-redirector on netdev netdevid,redirect filter's net
5439 packet to chardev chardevid,and redirect indev's packet to
5440 filter.if it has the vnet\_hdr\_support flag, filter-redirector
5441 will redirect packet with vnet\_hdr\_len. Create a
5442 filter-redirector we need to differ outdev id from indev id, id
5443 can not be the same. we can just use indev or outdev, but at
5444 least one of indev or outdev need to be specified.
5446 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
5447 Filter-rewriter is a part of COLO project.It will rewrite tcp
5448 packet to secondary from primary to keep secondary tcp
5449 connection,and rewrite tcp packet to primary from secondary make
5450 tcp packet can be handled by client.if it has the
5451 vnet\_hdr\_support flag, we can parse packet with vnet header.
5453 usage: colo secondary: -object
5454 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object
5455 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object
5456 filter-rewriter,id=rew0,netdev=hn0,queue=all
5458 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]``
5459 Dump the network traffic on netdev dev to the file specified by
5460 filename. At most len bytes (64k by default) per packet are
5461 stored. The file format is libpcap, so it can be analyzed with
5462 tools such as tcpdump or Wireshark.
5464 ``-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}]``
5465 Colo-compare gets packet from primary\_in chardevid and
5466 secondary\_in, then compare whether the payload of primary packet
5467 and secondary packet are the same. If same, it will output
5468 primary packet to out\_dev, else it will notify COLO-framework to do
5469 checkpoint and send primary packet to out\_dev. In order to
5470 improve efficiency, we need to put the task of comparison in
5471 another iothread. If it has the vnet\_hdr\_support flag,
5472 colo compare will send/recv packet with vnet\_hdr\_len.
5473 The compare\_timeout=@var{ms} determines the maximum time of the
5474 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms}
5475 is to set the period of scanning expired primary node network packets.
5476 The max\_queue\_size=@var{size} is to set the max compare queue
5477 size depend on user environment.
5478 If user want to use Xen COLO, need to add the notify\_dev to
5479 notify Xen colo-frame to do checkpoint.
5481 COLO-compare must be used with the help of filter-mirror,
5482 filter-redirector and filter-rewriter.
5486 KVM COLO
5488 primary:
5489 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
5490 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
5491 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
5492 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
5493 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
5494 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
5495 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
5496 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
5497 -object iothread,id=iothread1
5498 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
5499 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
5500 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
5501 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1
5503 secondary:
5504 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
5505 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
5506 -chardev socket,id=red0,host=3.3.3.3,port=9003
5507 -chardev socket,id=red1,host=3.3.3.3,port=9004
5508 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
5509 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
5512 Xen COLO
5514 primary:
5515 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
5516 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
5517 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
5518 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
5519 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
5520 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
5521 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
5522 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
5523 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off
5524 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
5525 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
5526 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
5527 -object iothread,id=iothread1
5528 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1
5530 secondary:
5531 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
5532 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
5533 -chardev socket,id=red0,host=3.3.3.3,port=9003
5534 -chardev socket,id=red1,host=3.3.3.3,port=9004
5535 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
5536 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
5538 If you want to know the detail of above command line, you can
5539 read the colo-compare git log.
5541 ``-object cryptodev-backend-builtin,id=id[,queues=queues]``
5542 Creates a cryptodev backend which executes crypto operations from
5543 the QEMU cipher APIs. The id parameter is a unique ID that will
5544 be used to reference this cryptodev backend from the
5545 ``virtio-crypto`` device. The queues parameter is optional,
5546 which specify the queue number of cryptodev backend, the default
5547 of queues is 1.
5549 .. parsed-literal::
5551 # |qemu_system| \\
5552 [...] \\
5553 -object cryptodev-backend-builtin,id=cryptodev0 \\
5554 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
5555 [...]
5557 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]``
5558 Creates a vhost-user cryptodev backend, backed by a chardev
5559 chardevid. The id parameter is a unique ID that will be used to
5560 reference this cryptodev backend from the ``virtio-crypto``
5561 device. The chardev should be a unix domain socket backed one.
5562 The vhost-user uses a specifically defined protocol to pass
5563 vhost ioctl replacement messages to an application on the other
5564 end of the socket. The queues parameter is optional, which
5565 specify the queue number of cryptodev backend for multiqueue
5566 vhost-user, the default of queues is 1.
5568 .. parsed-literal::
5570 # |qemu_system| \\
5571 [...] \\
5572 -chardev socket,id=chardev0,path=/path/to/socket \\
5573 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\
5574 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
5575 [...]
5577 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]``
5579 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]``
5580 Defines a secret to store a password, encryption key, or some
5581 other sensitive data. The sensitive data can either be passed
5582 directly via the data parameter, or indirectly via the file
5583 parameter. Using the data parameter is insecure unless the
5584 sensitive data is encrypted.
5586 The sensitive data can be provided in raw format (the default),
5587 or base64. When encoded as JSON, the raw format only supports
5588 valid UTF-8 characters, so base64 is recommended for sending
5589 binary data. QEMU will convert from which ever format is
5590 provided to the format it needs internally. eg, an RBD password
5591 can be provided in raw format, even though it will be base64
5592 encoded when passed onto the RBD sever.
5594 For added protection, it is possible to encrypt the data
5595 associated with a secret using the AES-256-CBC cipher. Use of
5596 encryption is indicated by providing the keyid and iv
5597 parameters. The keyid parameter provides the ID of a previously
5598 defined secret that contains the AES-256 decryption key. This
5599 key should be 32-bytes long and be base64 encoded. The iv
5600 parameter provides the random initialization vector used for
5601 encryption of this particular secret and should be a base64
5602 encrypted string of the 16-byte IV.
5604 The simplest (insecure) usage is to provide the secret inline
5606 .. parsed-literal::
5608 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw
5610 The simplest secure usage is to provide the secret via a file
5612 # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object
5613 secret,id=sec0,file=mypasswd.txt,format=raw
5615 For greater security, AES-256-CBC should be used. To illustrate
5616 usage, consider the openssl command line tool which can encrypt
5617 the data. Note that when encrypting, the plaintext must be
5618 padded to the cipher block size (32 bytes) using the standard
5619 PKCS#5/6 compatible padding algorithm.
5621 First a master key needs to be created in base64 encoding:
5625 # openssl rand -base64 32 > key.b64
5626 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"')
5628 Each secret to be encrypted needs to have a random
5629 initialization vector generated. These do not need to be kept
5630 secret
5634 # openssl rand -base64 16 > iv.b64
5635 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"')
5637 The secret to be defined can now be encrypted, in this case
5638 we're telling openssl to base64 encode the result, but it could
5639 be left as raw bytes if desired.
5643 # SECRET=$(printf "letmein" |
5644 openssl enc -aes-256-cbc -a -K $KEY -iv $IV)
5646 When launching QEMU, create a master secret pointing to
5647 ``key.b64`` and specify that to be used to decrypt the user
5648 password. Pass the contents of ``iv.b64`` to the second secret
5650 .. parsed-literal::
5652 # |qemu_system| \\
5653 -object secret,id=secmaster0,format=base64,file=key.b64 \\
5654 -object secret,id=sec0,keyid=secmaster0,format=base64,\\
5655 data=$SECRET,iv=$(<iv.b64)
5657 ``-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]``
5658 Create a Secure Encrypted Virtualization (SEV) guest object,
5659 which can be used to provide the guest memory encryption support
5660 on AMD processors.
5662 When memory encryption is enabled, one of the physical address
5663 bit (aka the C-bit) is utilized to mark if a memory page is
5664 protected. The ``cbitpos`` is used to provide the C-bit
5665 position. The C-bit position is Host family dependent hence user
5666 must provide this value. On EPYC, the value should be 47.
5668 When memory encryption is enabled, we loose certain bits in
5669 physical address space. The ``reduced-phys-bits`` is used to
5670 provide the number of bits we loose in physical address space.
5671 Similar to C-bit, the value is Host family dependent. On EPYC,
5672 a guest will lose a maximum of 1 bit, so the value should be 1.
5674 The ``sev-device`` provides the device file to use for
5675 communicating with the SEV firmware running inside AMD Secure
5676 Processor. The default device is '/dev/sev'. If hardware
5677 supports memory encryption then /dev/sev devices are created by
5678 CCP driver.
5680 The ``policy`` provides the guest policy to be enforced by the
5681 SEV firmware and restrict what configuration and operational
5682 commands can be performed on this guest by the hypervisor. The
5683 policy should be provided by the guest owner and is bound to the
5684 guest and cannot be changed throughout the lifetime of the
5685 guest. The default is 0.
5687 If guest ``policy`` allows sharing the key with another SEV
5688 guest then ``handle`` can be use to provide handle of the guest
5689 from which to share the key.
5691 The ``dh-cert-file`` and ``session-file`` provides the guest
5692 owner's Public Diffie-Hillman key defined in SEV spec. The PDH
5693 and session parameters are used for establishing a cryptographic
5694 session with the guest owner to negotiate keys used for
5695 attestation. The file must be encoded in base64.
5697 The ``kernel-hashes`` adds the hashes of given kernel/initrd/
5698 cmdline to a designated guest firmware page for measured Linux
5699 boot with -kernel. The default is off. (Since 6.2)
5701 e.g to launch a SEV guest
5703 .. parsed-literal::
5705 # |qemu_system_x86| \\
5706 ...... \\
5707 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=1 \\
5708 -machine ...,memory-encryption=sev0 \\
5709 .....
5711 ``-object authz-simple,id=id,identity=string``
5712 Create an authorization object that will control access to
5713 network services.
5715 The ``identity`` parameter is identifies the user and its format
5716 depends on the network service that authorization object is
5717 associated with. For authorizing based on TLS x509 certificates,
5718 the identity must be the x509 distinguished name. Note that care
5719 must be taken to escape any commas in the distinguished name.
5721 An example authorization object to validate a x509 distinguished
5722 name would look like:
5724 .. parsed-literal::
5726 # |qemu_system| \\
5727 ... \\
5728 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\
5731 Note the use of quotes due to the x509 distinguished name
5732 containing whitespace, and escaping of ','.
5734 ``-object authz-listfile,id=id,filename=path,refresh=on|off``
5735 Create an authorization object that will control access to
5736 network services.
5738 The ``filename`` parameter is the fully qualified path to a file
5739 containing the access control list rules in JSON format.
5741 An example set of rules that match against SASL usernames might
5742 look like:
5747 "rules": [
5748 { "match": "fred", "policy": "allow", "format": "exact" },
5749 { "match": "bob", "policy": "allow", "format": "exact" },
5750 { "match": "danb", "policy": "deny", "format": "glob" },
5751 { "match": "dan*", "policy": "allow", "format": "exact" },
5753 "policy": "deny"
5756 When checking access the object will iterate over all the rules
5757 and the first rule to match will have its ``policy`` value
5758 returned as the result. If no rules match, then the default
5759 ``policy`` value is returned.
5761 The rules can either be an exact string match, or they can use
5762 the simple UNIX glob pattern matching to allow wildcards to be
5763 used.
5765 If ``refresh`` is set to true the file will be monitored and
5766 automatically reloaded whenever its content changes.
5768 As with the ``authz-simple`` object, the format of the identity
5769 strings being matched depends on the network service, but is
5770 usually a TLS x509 distinguished name, or a SASL username.
5772 An example authorization object to validate a SASL username
5773 would look like:
5775 .. parsed-literal::
5777 # |qemu_system| \\
5778 ... \\
5779 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\
5782 ``-object authz-pam,id=id,service=string``
5783 Create an authorization object that will control access to
5784 network services.
5786 The ``service`` parameter provides the name of a PAM service to
5787 use for authorization. It requires that a file
5788 ``/etc/pam.d/service`` exist to provide the configuration for
5789 the ``account`` subsystem.
5791 An example authorization object to validate a TLS x509
5792 distinguished name would look like:
5794 .. parsed-literal::
5796 # |qemu_system| \\
5797 ... \\
5798 -object authz-pam,id=auth0,service=qemu-vnc \\
5801 There would then be a corresponding config file for PAM at
5802 ``/etc/pam.d/qemu-vnc`` that contains:
5806 account requisite pam_listfile.so item=user sense=allow \
5807 file=/etc/qemu/vnc.allow
5809 Finally the ``/etc/qemu/vnc.allow`` file would contain the list
5810 of x509 distinguished names that are permitted access
5814 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB
5816 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch``
5817 Creates a dedicated event loop thread that devices can be
5818 assigned to. This is known as an IOThread. By default device
5819 emulation happens in vCPU threads or the main event loop thread.
5820 This can become a scalability bottleneck. IOThreads allow device
5821 emulation and I/O to run on other host CPUs.
5823 The ``id`` parameter is a unique ID that will be used to
5824 reference this IOThread from ``-device ...,iothread=id``.
5825 Multiple devices can be assigned to an IOThread. Note that not
5826 all devices support an ``iothread`` parameter.
5828 The ``query-iothreads`` QMP command lists IOThreads and reports
5829 their thread IDs so that the user can configure host CPU
5830 pinning/affinity.
5832 IOThreads use an adaptive polling algorithm to reduce event loop
5833 latency. Instead of entering a blocking system call to monitor
5834 file descriptors and then pay the cost of being woken up when an
5835 event occurs, the polling algorithm spins waiting for events for
5836 a short time. The algorithm's default parameters are suitable
5837 for many cases but can be adjusted based on knowledge of the
5838 workload and/or host device latency.
5840 The ``poll-max-ns`` parameter is the maximum number of
5841 nanoseconds to busy wait for events. Polling can be disabled by
5842 setting this value to 0.
5844 The ``poll-grow`` parameter is the multiplier used to increase
5845 the polling time when the algorithm detects it is missing events
5846 due to not polling long enough.
5848 The ``poll-shrink`` parameter is the divisor used to decrease
5849 the polling time when the algorithm detects it is spending too
5850 long polling without encountering events.
5852 The ``aio-max-batch`` parameter is the maximum number of requests
5853 in a batch for the AIO engine, 0 means that the engine will use
5854 its default.
5856 The IOThread parameters can be modified at run-time using the
5857 ``qom-set`` command (where ``iothread1`` is the IOThread's
5858 ``id``):
5862 (qemu) qom-set /objects/iothread1 poll-max-ns 100000
5863 ERST
5866 HXCOMM This is the last statement. Insert new options before this line!
5868 #undef DEF
5869 #undef DEFHEADING
5870 #undef ARCHHEADING