1 HXCOMM Use
DEFHEADING() to define headings
in both help text and rST
.
2 HXCOMM Text between SRST and ERST is copied to the rST version and
3 HXCOMM discarded from C version
.
4 HXCOMM
DEF(option
, HAS_ARG
/0, opt_enum
, opt_help
, arch_mask
) is used to
5 HXCOMM construct option structures
, enums and help message
for specified
7 HXCOMM HXCOMM can be used
for comments
, discarded from both rST and C
.
9 DEFHEADING(Standard options
:)
11 DEF("help", 0, QEMU_OPTION_h
,
12 "-h or -help display this help and exit\n", QEMU_ARCH_ALL
)
18 DEF("version", 0, QEMU_OPTION_version
,
19 "-version display version information and exit\n", QEMU_ARCH_ALL
)
22 Display version information and exit
25 DEF("machine", HAS_ARG
, QEMU_OPTION_machine
, \
26 "-machine [type=]name[,prop[=value][,...]]\n"
27 " selects emulated machine ('-machine help' for list)\n"
28 " property accel=accel1[:accel2[:...]] selects accelerator\n"
29 " supported accelerators are kvm, xen, hax, hvf, nvmm, whpx or tcg (default: tcg)\n"
30 " vmport=on|off|auto controls emulation of vmport (default: auto)\n"
31 " dump-guest-core=on|off include guest memory in a core dump (default=on)\n"
32 " mem-merge=on|off controls memory merge support (default: on)\n"
33 " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n"
34 " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n"
35 " suppress-vmdesc=on|off disables self-describing migration (default=off)\n"
36 " nvdimm=on|off controls NVDIMM support (default=off)\n"
37 " memory-encryption=@var{} memory encryption object to use (default=none)\n"
38 " hmat=on|off controls ACPI HMAT support (default=off)\n"
39 " memory-backend='backend-id' specifies explicitly provided backend for main RAM (default=none)\n"
40 " cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]\n",
43 ``
-machine
[type
=]name
[,prop
=value
[,...]]``
44 Select the emulated machine by name
. Use ``
-machine help`` to list
47 For architectures which aim to support live migration compatibility
48 across releases
, each release will introduce a
new versioned machine
49 type
. For example
, the
2.8.0 release introduced machine types
50 "pc-i440fx-2.8" and
"pc-q35-2.8" for the x86\_64
/i686 architectures
.
52 To allow live migration of guests from QEMU version
2.8.0, to QEMU
53 version
2.9.0, the
2.9.0 version must support the
"pc-i440fx-2.8"
54 and
"pc-q35-2.8" machines too
. To allow users live migrating VMs to
55 skip multiple intermediate releases when upgrading
, new releases of
56 QEMU will support machine types from many previous versions
.
58 Supported machine properties are
:
60 ``accel
=accels1
[:accels2
[:...]]``
61 This is used to enable an accelerator
. Depending on the target
62 architecture
, kvm
, xen
, hax
, hvf
, nvmm
, whpx or tcg can be available
.
63 By
default, tcg is used
. If there is more than one accelerator
64 specified
, the next one is used
if the previous one fails to
67 ``vmport
=on|off|auto``
68 Enables emulation of VMWare IO port
, for vmmouse etc
. auto says
69 to select the value based on accel
. For accel
=xen the
default is
70 off otherwise the
default is on
.
72 ``dump
-guest
-core
=on|off``
73 Include guest memory
in a core dump
. The
default is on
.
76 Enables or disables memory merge support
. This feature
, when
77 supported by the host
, de
-duplicates identical memory pages
78 among VMs
instances (enabled by
default).
80 ``aes
-key
-wrap
=on|off``
81 Enables or disables AES key wrapping support on s390
-ccw hosts
.
82 This feature controls whether AES wrapping keys will be created
83 to allow execution of AES cryptographic functions
. The
default
86 ``dea
-key
-wrap
=on|off``
87 Enables or disables DEA key wrapping support on s390
-ccw hosts
.
88 This feature controls whether DEA wrapping keys will be created
89 to allow execution of DEA cryptographic functions
. The
default
93 Enables or disables NVDIMM support
. The
default is off
.
95 ``memory
-encryption
=``
96 Memory encryption object to use
. The
default is none
.
99 Enables or disables ACPI Heterogeneous Memory Attribute Table
100 (HMAT
) support
. The
default is off
.
102 ``memory
-backend
='id'``
103 An alternative to legacy ``
-mem
-path`` and ``mem
-prealloc`` options
.
104 Allows to use a memory backend as main RAM
.
109 -object memory
-backend
-file
,id
=pc
.ram
,size
=512M
,mem
-path
=/hugetlbfs
,prealloc
=on
,share
=on
110 -machine memory
-backend
=pc
.ram
113 Migration compatibility note
:
115 * as backend id one shall use value of
'default-ram-id', advertised by
116 machine
type (available via ``query
-machines`` QMP command
), if migration
117 to
/from old
QEMU (<5.0) is expected
.
118 * for machine types
4.0 and older
, user shall
119 use ``x
-use
-canonical
-path
-for-ramblock
-id
=off`` backend option
120 if migration to
/from old
QEMU (<5.0) is expected
.
125 -object memory
-backend
-ram
,id
=pc
.ram
,size
=512M
,x
-use
-canonical
-path
-for-ramblock
-id
=off
126 -machine memory
-backend
=pc
.ram
129 ``cxl
-fmw
.0.targets
.0=firsttarget
,cxl
-fmw
.0.targets
.1=secondtarget
,cxl
-fmw
.0.size
=size
[,cxl
-fmw
.0.interleave
-granularity
=granularity
]``
130 Define a CXL Fixed Memory
Window (CFMW
).
132 Described
in the CXL
2.0 ECN
: CEDT CFMWS
& QTG _DSM
.
134 They are regions of Host Physical
Addresses (HPA
) on a system which
135 may be interleaved across one or more CXL host bridges
. The system
136 software will assign particular devices into these windows and
137 configure the downstream Host
-managed Device
Memory (HDM
) decoders
138 in root ports
, switch ports and devices appropriately to meet the
139 interleave requirements before enabling the memory devices
.
141 ``targets
.X
=target`` provides the mapping to CXL host bridges
142 which may be identified by the id provied
in the
-device entry
.
143 Multiple entries are needed to specify all the targets when
144 the fixed memory window represents interleaved memory
. X is the
147 ``size
=size`` sets the size of the CFMW
. This must be a multiple of
148 256MiB
. The region will be aligned to
256MiB but the location is
149 platform and configuration dependent
.
151 ``interleave
-granularity
=granularity`` sets the granularity of
152 interleave
. Default
256KiB
. Only
256KiB
, 512KiB
, 1024KiB
, 2048KiB
153 4096KiB
, 8192KiB and
16384KiB granularities supported
.
159 -machine cxl
-fmw
.0.targets
.0=cxl
.0,cxl
-fmw
.0.targets
.1=cxl
.1,cxl
-fmw
.0.size
=128G
,cxl
-fmw
.0.interleave
-granularity
=512k
162 DEF("M", HAS_ARG
, QEMU_OPTION_M
,
163 " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n",
167 ``sgx
-epc
.0.memdev
=@
var{memid
},sgx
-epc
.0.node
=@
var{numaid
}``
168 Define an SGX EPC section
.
171 DEF("cpu", HAS_ARG
, QEMU_OPTION_cpu
,
172 "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL
)
175 Select CPU
model (``
-cpu help``
for list and additional feature
179 DEF("accel", HAS_ARG
, QEMU_OPTION_accel
,
180 "-accel [accel=]accelerator[,prop[=value][,...]]\n"
181 " select accelerator (kvm, xen, hax, hvf, nvmm, whpx or tcg; use 'help' for a list)\n"
182 " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n"
183 " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n"
184 " kvm-shadow-mem=size of KVM shadow MMU in bytes\n"
185 " split-wx=on|off (enable TCG split w^x mapping)\n"
186 " tb-size=n (TCG translation block cache size)\n"
187 " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n"
188 " thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL
)
190 ``
-accel name
[,prop
=value
[,...]]``
191 This is used to enable an accelerator
. Depending on the target
192 architecture
, kvm
, xen
, hax
, hvf
, nvmm
, whpx or tcg can be available
. By
193 default, tcg is used
. If there is more than one accelerator
194 specified
, the next one is used
if the previous one fails to
197 ``igd
-passthru
=on|off``
198 When Xen is
in use
, this option controls whether Intel
199 integrated graphics devices can be passed through to the guest
202 ``kernel
-irqchip
=on|off|split``
203 Controls KVM
in-kernel irqchip support
. The
default is full
204 acceleration of the interrupt controllers
. On x86
, split irqchip
205 reduces the kernel attack surface
, at a performance cost
for
206 non
-MSI interrupts
. Disabling the
in-kernel irqchip completely
207 is not recommended except
for debugging purposes
.
209 ``kvm
-shadow
-mem
=size``
210 Defines the size of the KVM shadow MMU
.
213 Controls the use of split w^x mapping
for the TCG code generation
214 buffer
. Some operating systems require
this to be enabled
, and
in
215 such a
case this will
default on
. On other operating systems
, this
216 will
default off
, but one may enable
this for testing or debugging
.
219 Controls the
size (in MiB
) of the TCG translation block cache
.
221 ``thread
=single|multi``
222 Controls number of TCG threads
. When the TCG is multi
-threaded
223 there will be one thread per vCPU therefore taking advantage of
224 additional host cores
. The
default is to enable multi
-threading
225 where both the back
-end and front
-ends support it and no
226 incompatible TCG features have been
enabled (e
.g
.
229 ``dirty
-ring
-size
=n``
230 When the KVM accelerator is used
, it controls the size of the per
-vCPU
231 dirty page ring
buffer (number of entries
for each vCPU
). It should
232 be a value that is power of two
, and it should be
1024 or
bigger (but
233 still less than the maximum value that the kernel supports
). 4096
234 could be a good initial value
if you have no idea which is the best
.
235 Set
this value to
0 to disable the feature
. By
default, this feature
236 is
disabled (dirty
-ring
-size
=0). When enabled
, KVM will instead
237 record dirty pages
in a bitmap
.
241 DEF("smp", HAS_ARG
, QEMU_OPTION_smp
,
242 "-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]\n"
243 " set the number of initial CPUs to 'n' [default=1]\n"
244 " maxcpus= maximum number of total CPUs, including\n"
245 " offline CPUs for hotplug, etc\n"
246 " sockets= number of sockets on the machine board\n"
247 " dies= number of dies in one socket\n"
248 " clusters= number of clusters in one die\n"
249 " cores= number of cores in one cluster\n"
250 " threads= number of threads in one core\n"
251 "Note: Different machines may have different subsets of the CPU topology\n"
252 " parameters supported, so the actual meaning of the supported parameters\n"
253 " will vary accordingly. For example, for a machine type that supports a\n"
254 " three-level CPU hierarchy of sockets/cores/threads, the parameters will\n"
255 " sequentially mean as below:\n"
256 " sockets means the number of sockets on the machine board\n"
257 " cores means the number of cores in one socket\n"
258 " threads means the number of threads in one core\n"
259 " For a particular machine type board, an expected CPU topology hierarchy\n"
260 " can be defined through the supported sub-option. Unsupported parameters\n"
261 " can also be provided in addition to the sub-option, but their values\n"
262 " must be set as 1 in the purpose of correct parsing.\n",
265 ``
-smp
[[cpus
=]n
][,maxcpus
=maxcpus
][,sockets
=sockets
][,dies
=dies
][,clusters
=clusters
][,cores
=cores
][,threads
=threads
]``
266 Simulate a SMP system with
'\ ``n``\ ' CPUs initially present on
267 the machine type board
. On boards supporting CPU hotplug
, the optional
268 '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be
269 added at runtime
. When both parameters are omitted
, the maximum number
270 of CPUs will be calculated from the provided topology members and the
271 initial CPU count will match the maximum number
. When only one of them
272 is given then the omitted one will be set to its counterpart
's value.
273 Both parameters may be specified, but the maximum number of CPUs must
274 be equal to or greater than the initial CPU count. Product of the
275 CPU topology hierarchy must be equal to the maximum number of CPUs.
276 Both parameters are subject to an upper limit that is determined by
277 the specific machine type chosen.
279 To control reporting of CPU topology information, values of the topology
280 parameters can be specified. Machines may only support a subset of the
281 parameters and different machines may have different subsets supported
282 which vary depending on capacity of the corresponding CPU targets. So
283 for a particular machine type board, an expected topology hierarchy can
284 be defined through the supported sub-option. Unsupported parameters can
285 also be provided in addition to the sub-option, but their values must be
286 set as 1 in the purpose of correct parsing.
288 Either the initial CPU count, or at least one of the topology parameters
289 must be specified. The specified parameters must be greater than zero,
290 explicit configuration like "cpus=0" is not allowed. Values for any
291 omitted parameters will be computed from those which are given.
293 For example, the following sub-option defines a CPU topology hierarchy
294 (2 sockets totally on the machine, 2 cores per socket, 2 threads per
295 core) for a machine that only supports sockets/cores/threads.
296 Some members of the option can be omitted but their values will be
297 automatically computed:
301 -smp 8,sockets=2,cores=2,threads=2,maxcpus=8
303 The following sub-option defines a CPU topology hierarchy (2 sockets
304 totally on the machine, 2 dies per socket, 2 cores per die, 2 threads
305 per core) for PC machines which support sockets/dies/cores/threads.
306 Some members of the option can be omitted but their values will be
307 automatically computed:
311 -smp 16,sockets=2,dies=2,cores=2,threads=2,maxcpus=16
313 The following sub-option defines a CPU topology hierarchy (2 sockets
314 totally on the machine, 2 clusters per socket, 2 cores per cluster,
315 2 threads per core) for ARM virt machines which support sockets/clusters
316 /cores/threads. Some members of the option can be omitted but their values
317 will be automatically computed:
321 -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16
323 Historically preference was given to the coarsest topology parameters
324 when computing missing values (ie sockets preferred over cores, which
325 were preferred over threads), however, this behaviour is considered
326 liable to change. Prior to 6.2 the preference was sockets over cores
327 over threads. Since 6.2 the preference is cores over sockets over threads.
329 For example, the following option defines a machine board with 2 sockets
330 of 1 core before 6.2 and 1 socket of 2 cores after 6.2:
337 DEF("numa", HAS_ARG, QEMU_OPTION_numa,
338 "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
339 "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n"
340 "-numa dist,src=source,dst=destination,val=distance\n"
341 "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n"
342 "-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"
343 "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n",
346 ``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
348 ``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]``
350 ``-numa dist,src=source,dst=destination,val=distance``
352 ``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]``
354 ``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=tpye[,latency=lat][,bandwidth=bw]``
356 ``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]``
357 Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA
358 distance from a source node to a destination node. Set the ACPI
359 Heterogeneous Memory Attributes for the given nodes.
361 Legacy VCPU assignment uses '\ ``cpus``\
' option where firstcpu and
362 lastcpu are CPU indexes. Each '\ ``cpus``\
' option represent a
363 contiguous range of CPU indexes (or a single VCPU if lastcpu is
364 omitted). A non-contiguous set of VCPUs can be represented by
365 providing multiple '\ ``cpus``\
' options. If '\ ``cpus``\
' is
366 omitted on all nodes, VCPUs are automatically split between them.
368 For example, the following option assigns VCPUs 0, 1, 2 and 5 to a
373 -numa node,cpus=0-2,cpus=5
375 '\ ``cpu``\
' option is a new alternative to '\ ``cpus``\
' option
376 which uses '\ ``socket
-id|core
-id|thread
-id``\
' properties to
377 assign CPU objects to a node using topology layout properties of
378 CPU. The set of properties is machine specific, and depends on used
379 machine type/'\ ``smp``\
' options. It could be queried with
380 '\ ``hotpluggable
-cpus``\
' monitor command. '\ ``node
-id``\
'
381 property specifies node to which CPU object will be assigned, it's
382 required
for node to be declared with
'\ ``node``\ ' option before
383 it
's used with '\ ``cpu``\
' option.
390 -smp 1,sockets=2,maxcpus=2 \
391 -numa node,nodeid=0 -numa node,nodeid=1 \
392 -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1
394 Legacy '\ ``mem``\
' assigns a given RAM amount to a node (not supported
395 for 5.1 and newer machine types). '\ ``memdev``\
' assigns RAM from
396 a given memory backend device to a node. If '\ ``mem``\
' and
397 '\ ``memdev``\
' are omitted in all nodes, RAM is split equally between them.
400 '\ ``mem``\
' and '\ ``memdev``\
' are mutually exclusive.
401 Furthermore, if one node uses '\ ``memdev``\
', all of them have to
404 '\ ``initiator``\
' is an additional option that points to an
405 initiator NUMA node that has best performance (the lowest latency or
406 largest bandwidth) to this NUMA node. Note that this option can be
407 set only when the machine property 'hmat
' is set to 'on
'.
409 Following example creates a machine with 2 NUMA nodes, node 0 has
410 CPU. node 1 has only memory, and its initiator is node 0. Note that
411 because node 0 has CPU, by default the initiator of node 0 is itself
417 -m 2G,slots=2,maxmem=4G \
418 -object memory-backend-ram,size=1G,id=m0 \
419 -object memory-backend-ram,size=1G,id=m1 \
420 -numa node,nodeid=0,memdev=m0 \
421 -numa node,nodeid=1,memdev=m1,initiator=0 \
422 -smp 2,sockets=2,maxcpus=2 \
423 -numa cpu,node-id=0,socket-id=0 \
424 -numa cpu,node-id=0,socket-id=1
426 source and destination are NUMA node IDs. distance is the NUMA
427 distance from source to destination. The distance from a node to
428 itself is always 10. If any pair of nodes is given a distance, then
429 all pairs must be given distances. Although, when distances are only
430 given in one direction for each pair of nodes, then the distances in
431 the opposite directions are assumed to be the same. If, however, an
432 asymmetrical pair of distances is given for even one node pair, then
433 all node pairs must be provided distance values for both directions,
434 even when they are symmetrical. When a node is unreachable from
435 another node, set the pair's distance to
255.
437 Note that the
-``numa`` option doesn
't allocate any of the specified
438 resources, it just assigns existing resources to NUMA nodes. This
439 means that one still has to use the ``-m``, ``-smp`` options to
440 allocate RAM and VCPUs respectively.
442 Use '\ ``hmat
-lb``\
' to set System Locality Latency and Bandwidth
443 Information between initiator and target NUMA nodes in ACPI
444 Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can
445 create memory requests, usually it has one or more processors.
446 Target NUMA node contains addressable memory.
448 In '\ ``hmat
-lb``\
' option, node are NUMA node IDs. hierarchy is
449 the memory hierarchy of the target NUMA node: if hierarchy is
450 'memory
', the structure represents the memory performance; if
451 hierarchy is 'first
-level\|second
-level\|third
-level
', this
452 structure represents aggregated performance of memory side caches
453 for each domain. type of 'data
-type
' is type of data represented by
454 this structure instance: if 'hierarchy
' is 'memory
', 'data
-type
' is
455 'access\|read\|write
' latency or 'access\|read\|write
' bandwidth of
456 the target memory; if 'hierarchy
' is
457 'first
-level\|second
-level\|third
-level
', 'data
-type
' is
458 'access\|read\|write
' hit latency or 'access\|read\|write
' hit
459 bandwidth of the target memory side cache.
461 lat is latency value in nanoseconds. bw is bandwidth value, the
462 possible value and units are NUM[M\|G\|T], mean that the bandwidth
463 value are NUM byte per second (or MB/s, GB/s or TB/s depending on
464 used suffix). Note that if latency or bandwidth value is 0, means
465 the corresponding latency or bandwidth information is not provided.
467 In '\ ``hmat
-cache``\
' option, node-id is the NUMA-id of the memory
468 belongs. size is the size of memory side cache in bytes. level is
469 the cache level described in this structure, note that the cache
470 level 0 should not be used with '\ ``hmat
-cache``\
' option.
471 associativity is the cache associativity, the possible value is
472 'none
/direct(direct
-mapped
)/complex(complex cache indexing
)'. policy
473 is the write policy. line is the cache Line size in bytes.
475 For example, the following options describe 2 NUMA nodes. Node 0 has
476 2 cpus and a ram, node 1 has only a ram. The processors in node 0
477 access memory in node 0 with access-latency 5 nanoseconds,
478 access-bandwidth is 200 MB/s; The processors in NUMA node 0 access
479 memory in NUMA node 1 with access-latency 10 nanoseconds,
480 access-bandwidth is 100 MB/s. And for memory side cache information,
481 NUMA node 0 and 1 both have 1 level memory cache, size is 10KB,
482 policy is write-back, the cache Line size is 8 bytes:
488 -object memory-backend-ram,size=1G,id=m0 \
489 -object memory-backend-ram,size=1G,id=m1 \
490 -smp 2,sockets=2,maxcpus=2 \
491 -numa node,nodeid=0,memdev=m0 \
492 -numa node,nodeid=1,memdev=m1,initiator=0 \
493 -numa cpu,node-id=0,socket-id=0 \
494 -numa cpu,node-id=0,socket-id=1 \
495 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \
496 -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \
497 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \
498 -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \
499 -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \
500 -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8
503 DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd,
504 "-add-fd fd=fd,set=set[,opaque=opaque]\n"
505 " Add 'fd
' to fd 'set
'\n", QEMU_ARCH_ALL)
507 ``-add-fd fd=fd,set=set[,opaque=opaque]``
508 Add a file descriptor to an fd set. Valid options are:
511 This option defines the file descriptor of which a duplicate is
512 added to fd set. The file descriptor cannot be stdin, stdout, or
516 This option defines the ID of the fd set to add the file
520 This option defines a free-form string that can be used to
523 You can open an image using pre-opened file descriptors from an fd
529 -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\
530 -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\
531 -drive file=/dev/fdset/2,index=0,media=disk
534 DEF("set", HAS_ARG, QEMU_OPTION_set,
535 "-set group.id.arg=value\n"
536 " set <arg> parameter for item <id> of type <group>\n"
537 " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL)
539 ``-set group.id.arg=value``
540 Set parameter arg for item id of type group
543 DEF("global", HAS_ARG, QEMU_OPTION_global,
544 "-global driver.property=value\n"
545 "-global driver=driver,property=property,value=value\n"
546 " set a global default for a driver property\n",
549 ``-global driver.prop=value``
551 ``-global driver=driver,property=property,value=value``
552 Set default value of driver's property prop to value
, e
.g
.:
556 |qemu_system_x86|
-global ide
-hd
.physical_block_size
=4096 disk
-image
.img
558 In particular
, you can use
this to set driver properties
for devices
559 which are created automatically by the machine model
. To create a
560 device which is not created automatically and set properties on it
,
563 -global driver
.prop
=value is shorthand
for -global
564 driver
=driver
,property
=prop
,value
=value
. The longhand syntax works
565 even when driver contains a dot
.
568 DEF("boot", HAS_ARG
, QEMU_OPTION_boot
,
569 "-boot [order=drives][,once=drives][,menu=on|off]\n"
570 " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n"
571 " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n"
572 " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n"
573 " 'sp_time': the period that splash picture last if menu=on, unit is ms\n"
574 " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n",
577 ``
-boot
[order
=drives
][,once
=drives
][,menu
=on|off
][,splash
=sp_name
][,splash
-time
=sp_time
][,reboot
-timeout
=rb_timeout
][,strict
=on|off
]``
578 Specify boot order drives as a string of drive letters
. Valid drive
579 letters depend on the target architecture
. The x86 PC uses
: a
, b
580 (floppy
1 and
2), c (first hard disk
), d (first CD
-ROM
), n
-p
581 (Etherboot from network adapter
1-4), hard disk boot is the
default.
582 To apply a particular boot order only on the first startup
, specify
583 it via ``once``
. Note that the ``order`` or ``once`` parameter
584 should not be used together with the ``bootindex`` property of
585 devices
, since the firmware implementations normally
do not support
586 both at the same time
.
588 Interactive boot menus
/prompts can be enabled via ``menu
=on`` as far
589 as firmware
/BIOS supports them
. The
default is non
-interactive boot
.
591 A splash picture could be passed to bios
, enabling user to show it
592 as logo
, when option splash
=sp\_name is given and menu
=on
, If
593 firmware
/BIOS supports them
. Currently Seabios
for X86 system
594 support it
. limitation
: The splash file could be a jpeg file or a
595 BMP file
in 24 BPP
format(true color
). The resolution should be
596 supported by the SVGA mode
, so the recommended is
320x240
, 640x480
,
599 A timeout could be passed to bios
, guest will pause
for rb\_timeout
600 ms when boot failed
, then reboot
. If rb\_timeout is
'-1', guest will
601 not reboot
, qemu passes
'-1' to bios by
default. Currently Seabios
602 for X86 system support it
.
604 Do strict boot via ``strict
=on`` as far as firmware
/BIOS supports
605 it
. This only effects when boot priority is changed by bootindex
606 options
. The
default is non
-strict boot
.
610 #
try to boot from network first
, then from hard disk
611 |qemu_system_x86|
-boot order
=nc
612 # boot from CD
-ROM first
, switch back to
default order after reboot
613 |qemu_system_x86|
-boot once
=d
614 # boot with a splash picture
for 5 seconds
.
615 |qemu_system_x86|
-boot menu
=on
,splash
=/root
/boot
.bmp
,splash
-time
=5000
617 Note
: The legacy format
'-boot drives' is still supported but its
618 use is discouraged as it may be removed from future versions
.
621 DEF("m", HAS_ARG
, QEMU_OPTION_m
,
622 "-m [size=]megs[,slots=n,maxmem=size]\n"
623 " configure guest RAM\n"
624 " size: initial amount of guest memory\n"
625 " slots: number of hotplug slots (default: none)\n"
626 " maxmem: maximum amount of guest memory (default: none)\n"
627 "NOTE: Some architectures might enforce a specific granularity\n",
630 ``
-m
[size
=]megs
[,slots
=n
,maxmem
=size
]``
631 Sets guest startup RAM size to megs megabytes
. Default is
128 MiB
.
632 Optionally
, a suffix of
"M" or
"G" can be used to signify a value
in
633 megabytes or gigabytes respectively
. Optional pair slots
, maxmem
634 could be used to set amount of hotpluggable memory slots and maximum
635 amount of memory
. Note that maxmem must be aligned to the page size
.
637 For example
, the following command
-line sets the guest startup RAM
638 size to
1GB
, creates
3 slots to hotplug additional memory and sets
639 the maximum memory the guest can reach to
4GB
:
643 |qemu_system|
-m
1G
,slots
=3,maxmem
=4G
645 If slots and maxmem are not specified
, memory hotplug won
't be
646 enabled and the guest startup RAM will never increase.
649 DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath,
650 "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL)
653 Allocate guest RAM from a temporarily created file in path.
656 DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc,
657 "-mem-prealloc preallocate guest memory (use with -mem-path)\n",
661 Preallocate memory when using -mem-path.
664 DEF("k", HAS_ARG, QEMU_OPTION_k,
665 "-k language use keyboard layout (for example 'fr
' for French)\n",
669 Use keyboard layout language (for example ``fr`` for French). This
670 option is only needed where it is not easy to get raw PC keycodes
671 (e.g. on Macs, with some X11 servers or with a VNC or curses
672 display). You don't normally need to use it on PC
/Linux or
675 The available layouts are
:
679 ar de
-ch es fo fr
-ca hu ja mk no pt
-br sv
680 da en
-gb et fr fr
-ch is lt nl pl ru th
681 de en
-us fi fr
-be hr it lv nl
-be pt sl tr
683 The
default is ``en
-us``
.
687 HXCOMM Deprecated by
-audiodev
688 DEF("audio-help", 0, QEMU_OPTION_audio_help
,
689 "-audio-help show -audiodev equivalent of the currently specified audio settings\n",
693 Will show the
-audiodev equivalent of the currently specified
694 (deprecated
) environment variables
.
697 DEF("audio", HAS_ARG
, QEMU_OPTION_audio
,
698 "-audio [driver=]driver,model=value[,prop[=value][,...]]\n"
699 " specifies the audio backend and device to use;\n"
700 " apart from 'model', options are the same as for -audiodev.\n"
701 " use '-audio model=help' to show possible devices.\n",
704 ``
-audio
[driver
=]driver
,model
=value
[,prop
[=value
][,...]]``
705 This option is a shortcut
for configuring both the guest audio
706 hardware and the host audio backend
in one go
.
707 The host backend options are the same as with the corresponding
708 ``
-audiodev`` options below
. The guest hardware model can be set with
709 ``model
=modelname``
. Use ``model
=help`` to list the available device
712 The following two example
do exactly the same
, to show how ``
-audio``
713 can be used to shorten the command line length
:
717 |qemu_system|
-audiodev pa
,id
=pa
-device sb16
,audiodev
=pa
718 |qemu_system|
-audio pa
,model
=sb16
721 DEF("audiodev", HAS_ARG
, QEMU_OPTION_audiodev
,
722 "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n"
723 " specifies the audio backend to use\n"
724 " id= identifier of the backend\n"
725 " timer-period= timer period in microseconds\n"
726 " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n"
727 " in|out.fixed-settings= use fixed settings for host audio\n"
728 " in|out.frequency= frequency to use with fixed settings\n"
729 " in|out.channels= number of channels to use with fixed settings\n"
730 " in|out.format= sample format to use with fixed settings\n"
731 " valid values: s8, s16, s32, u8, u16, u32, f32\n"
732 " in|out.voices= number of voices to use\n"
733 " in|out.buffer-length= length of buffer in microseconds\n"
734 "-audiodev none,id=id,[,prop[=value][,...]]\n"
735 " dummy driver that discards all output\n"
736 #ifdef CONFIG_AUDIO_ALSA
737 "-audiodev alsa,id=id[,prop[=value][,...]]\n"
738 " in|out.dev= name of the audio device to use\n"
739 " in|out.period-length= length of period in microseconds\n"
740 " in|out.try-poll= attempt to use poll mode\n"
741 " threshold= threshold (in microseconds) when playback starts\n"
743 #ifdef CONFIG_AUDIO_COREAUDIO
744 "-audiodev coreaudio,id=id[,prop[=value][,...]]\n"
745 " in|out.buffer-count= number of buffers\n"
747 #ifdef CONFIG_AUDIO_DSOUND
748 "-audiodev dsound,id=id[,prop[=value][,...]]\n"
749 " latency= add extra latency to playback in microseconds\n"
751 #ifdef CONFIG_AUDIO_OSS
752 "-audiodev oss,id=id[,prop[=value][,...]]\n"
753 " in|out.dev= path of the audio device to use\n"
754 " in|out.buffer-count= number of buffers\n"
755 " in|out.try-poll= attempt to use poll mode\n"
756 " try-mmap= try using memory mapped access\n"
757 " exclusive= open device in exclusive mode\n"
758 " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n"
760 #ifdef CONFIG_AUDIO_PA
761 "-audiodev pa,id=id[,prop[=value][,...]]\n"
762 " server= PulseAudio server address\n"
763 " in|out.name= source/sink device name\n"
764 " in|out.latency= desired latency in microseconds\n"
766 #ifdef CONFIG_AUDIO_SDL
767 "-audiodev sdl,id=id[,prop[=value][,...]]\n"
768 " in|out.buffer-count= number of buffers\n"
771 "-audiodev spice,id=id[,prop[=value][,...]]\n"
773 #ifdef CONFIG_DBUS_DISPLAY
774 "-audiodev dbus,id=id[,prop[=value][,...]]\n"
776 "-audiodev wav,id=id[,prop[=value][,...]]\n"
777 " path= path of wav file to record\n",
780 ``
-audiodev
[driver
=]driver
,id
=id
[,prop
[=value
][,...]]``
781 Adds a
new audio backend driver identified by id
. There are global
782 and driver specific properties
. Some values can be set differently
783 for input and output
, they
're marked with ``in|out.``. You can set
784 the input's property with ``
in.prop`` and the output
's property with
785 ``out.prop``. For example:
789 -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000
790 -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified
792 NOTE: parameter validation is known to be incomplete, in many cases
793 specifying an invalid option causes QEMU to print an error message
794 and continue emulation without sound.
796 Valid global options are:
799 Identifies the audio backend.
801 ``timer-period=period``
802 Sets the timer period used by the audio subsystem in
803 microseconds. Default is 10000 (10 ms).
805 ``in|out.mixing-engine=on|off``
806 Use QEMU's mixing engine to mix all streams inside QEMU and
807 convert audio formats when not supported by the backend
. When
808 off
, fixed
-settings must be off too
. Note that disabling
this
809 option means that the selected backend must support multiple
810 streams and the audio formats used by the virtual cards
,
811 otherwise you
'll get no sound. It's not recommended to disable
812 this option unless you want to use
5.1 or
7.1 audio
, as mixing
813 engine only supports mono and stereo audio
. Default is on
.
815 ``
in|out
.fixed
-settings
=on|off``
816 Use fixed settings
for host audio
. When off
, it will change
817 based on how the guest opens the sound card
. In
this case you
818 must not specify frequency
, channels or format
. Default is on
.
820 ``
in|out
.frequency
=frequency``
821 Specify the frequency to use when
using fixed
-settings
. Default
824 ``
in|out
.channels
=channels``
825 Specify the number of channels to use when
using fixed
-settings
.
826 Default is
2 (stereo
).
828 ``
in|out
.format
=format``
829 Specify the sample format to use when
using fixed
-settings
.
830 Valid values are
: ``s8``
, ``s16``
, ``s32``
, ``u8``
, ``u16``
,
831 ``u32``
, ``f32``
. Default is ``s16``
.
833 ``
in|out
.voices
=voices``
834 Specify the number of voices to use
. Default is
1.
836 ``
in|out
.buffer
-length
=usecs``
837 Sets the size of the buffer
in microseconds
.
839 ``
-audiodev none
,id
=id
[,prop
[=value
][,...]]``
840 Creates a dummy backend that discards all outputs
. This backend has
841 no backend specific properties
.
843 ``
-audiodev alsa
,id
=id
[,prop
[=value
][,...]]``
844 Creates backend
using the ALSA
. This backend is only available on
847 ALSA specific options are
:
849 ``
in|out
.dev
=device``
850 Specify the ALSA device to use
for input and
/or output
. Default
853 ``
in|out
.period
-length
=usecs``
854 Sets the period length
in microseconds
.
856 ``
in|out
.try-poll
=on|off``
857 Attempt to use poll mode with the device
. Default is on
.
859 ``threshold
=threshold``
860 Threshold (in microseconds
) when playback starts
. Default is
0.
862 ``
-audiodev coreaudio
,id
=id
[,prop
[=value
][,...]]``
863 Creates a backend
using Apple
's Core Audio. This backend is only
864 available on Mac OS and only supports playback.
866 Core Audio specific options are:
868 ``in|out.buffer-count=count``
869 Sets the count of the buffers.
871 ``-audiodev dsound,id=id[,prop[=value][,...]]``
872 Creates a backend using Microsoft's DirectSound
. This backend is
873 only available on Windows and only supports playback
.
875 DirectSound specific options are
:
878 Add extra usecs microseconds latency to playback
. Default is
881 ``
-audiodev oss
,id
=id
[,prop
[=value
][,...]]``
882 Creates a backend
using OSS
. This backend is available on most
885 OSS specific options are
:
887 ``
in|out
.dev
=device``
888 Specify the file name of the OSS device to use
. Default is
891 ``
in|out
.buffer
-count
=count``
892 Sets the count of the buffers
.
894 ``
in|out
.try-poll
=on|of``
895 Attempt to use poll mode with the device
. Default is on
.
898 Try
using memory mapped device access
. Default is off
.
901 Open the device
in exclusive
mode (vmix won
't work in this
902 case). Default is off.
904 ``dsp-policy=policy``
905 Sets the timing policy (between 0 and 10, where smaller number
906 means smaller latency but higher CPU usage). Use -1 to use
907 buffer sizes specified by ``buffer`` and ``buffer-count``. This
908 option is ignored if you do not have OSS 4. Default is 5.
910 ``-audiodev pa,id=id[,prop[=value][,...]]``
911 Creates a backend using PulseAudio. This backend is available on
914 PulseAudio specific options are:
917 Sets the PulseAudio server to connect to.
920 Use the specified source/sink for recording/playback.
922 ``in|out.latency=usecs``
923 Desired latency in microseconds. The PulseAudio server will try
924 to honor this value but actual latencies may be lower or higher.
926 ``-audiodev sdl,id=id[,prop[=value][,...]]``
927 Creates a backend using SDL. This backend is available on most
928 systems, but you should use your platform's native backend
if
931 SDL specific options are
:
933 ``
in|out
.buffer
-count
=count``
934 Sets the count of the buffers
.
936 ``
-audiodev spice
,id
=id
[,prop
[=value
][,...]]``
937 Creates a backend that sends audio through SPICE
. This backend
938 requires ``
-spice`` and automatically selected
in that
case, so
939 usually you can ignore
this option
. This backend has no backend
942 ``
-audiodev wav
,id
=id
[,prop
[=value
][,...]]``
943 Creates a backend that writes audio to a WAV file
.
945 Backend specific options are
:
948 Write recorded audio into the specified file
. Default is
952 DEF("device", HAS_ARG
, QEMU_OPTION_device
,
953 "-device driver[,prop[=value][,...]]\n"
954 " add device (based on driver)\n"
955 " prop=value,... sets driver properties\n"
956 " use '-device help' to print all possible drivers\n"
957 " use '-device driver,help' to print all possible properties\n",
960 ``
-device driver
[,prop
[=value
][,...]]``
961 Add device driver
. prop
=value sets driver properties
. Valid
962 properties depend on the driver
. To get help on possible drivers and
963 properties
, use ``
-device help`` and ``
-device driver
,help``
.
967 ``
-device ipmi
-bmc
-sim
,id
=id
[,prop
[=value
][,...]]``
968 Add an IPMI BMC
. This is a simulation of a hardware management
969 interface processor that normally sits on a system
. It provides a
970 watchdog and the ability to reset and power control the system
. You
971 need to connect
this to an IPMI
interface to make it useful
973 The IPMI slave address to use
for the BMC
. The
default is
0x20. This
974 address is the BMC
's address on the I2C network of management
975 controllers. If you don't know what
this means
, it is safe to ignore
979 The BMC id
for interfaces to use
this device
.
982 Define slave address to use
for the BMC
. The
default is
0x20.
985 file containing raw Sensor Data
Records (SDR
) data
. The
default
989 size of a Field Replaceable
Unit (FRU
) area
. The
default is
993 file containing raw Field Replaceable
Unit (FRU
) inventory data
.
997 value
for the GUID
for the BMC
, in standard UUID format
. If
this
998 is set
, get
"Get GUID" command to the BMC will
return it
.
999 Otherwise
"Get GUID" will
return an error
.
1001 ``
-device ipmi
-bmc
-extern,id
=id
,chardev
=id
[,slave_addr
=val
]``
1002 Add a connection to an external IPMI BMC simulator
. Instead of
1003 locally emulating the BMC like the above item
, instead connect to an
1004 external entity that provides the IPMI services
.
1006 A connection is made to an external BMC simulator
. If you
do this,
1007 it is strongly recommended that you use the
"reconnect=" chardev
1008 option to reconnect to the simulator
if the connection is lost
. Note
1009 that
if this is not used carefully
, it can be a security issue
, as
1010 the
interface has the ability to send resets
, NMIs
, and power off
1011 the VM
. It
's best if QEMU makes a connection to an external
1012 simulator running on a secure port on localhost, so neither the
1013 simulator nor QEMU is exposed to any outside network.
1015 See the "lanserv/README.vm" file in the OpenIPMI library for more
1016 details on the external interface.
1018 ``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]``
1019 Add a KCS IPMI interafce on the ISA bus. This also adds a
1020 corresponding ACPI and SMBIOS entries, if appropriate.
1023 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern
1027 Define the I/O address of the interface. The default is 0xca0
1031 Define the interrupt to use. The default is 5. To disable
1032 interrupts, set this to 0.
1034 ``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]``
1035 Like the KCS interface, but defines a BT interface. The default port
1036 is 0xe4 and the default interrupt is 5.
1038 ``-device pci-ipmi-kcs,bmc=id``
1039 Add a KCS IPMI interafce on the PCI bus.
1042 The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above.
1044 ``-device pci-ipmi-bt,bmc=id``
1045 Like the KCS interface, but defines a BT interface on the PCI bus.
1047 ``-device intel-iommu[,option=...]``
1048 This is only supported by ``-machine q35``, which will enable Intel VT-d
1049 emulation within the guest. It supports below options:
1051 ``intremap=on|off`` (default: auto)
1052 This enables interrupt remapping feature. It's required to enable
1053 complete x2apic
. Currently it only supports kvm kernel
-irqchip modes
1054 ``off`` or ``split``
, while full kernel
-irqchip is not yet supported
.
1055 The
default value is
"auto", which will be decided by the mode of
1058 ``caching
-mode
=on|off``
(default: off
)
1059 This enables caching mode
for the VT
-d emulated device
. When
1060 caching
-mode is enabled
, each guest DMA buffer mapping will generate an
1061 IOTLB invalidation from the guest IOMMU driver to the vIOMMU device
in
1062 a synchronous way
. It is required
for ``
-device vfio
-pci`` to work
1063 with the VT
-d device
, because host assigned devices requires to setup
1064 the DMA mapping on the host before guest DMA starts
.
1066 ``device
-iotlb
=on|off``
(default: off
)
1067 This enables device
-iotlb capability
for the emulated VT
-d device
. So
1068 far virtio
/vhost should be the only real user
for this parameter
,
1069 paired with ats
=on configured
for the device
.
1071 ``aw
-bits
=39|
48``
(default: 39)
1072 This decides the address width of IOVA address space
. The address
1073 space has
39 bits width
for 3-level IOMMU page tables
, and
48 bits
for
1074 4-level IOMMU page tables
.
1076 Please also refer to the wiki page
for general scenarios of VT
-d
1077 emulation
in QEMU
: https
://wiki.qemu.org/Features/VT-d.
1081 DEF("name", HAS_ARG
, QEMU_OPTION_name
,
1082 "-name string1[,process=string2][,debug-threads=on|off]\n"
1083 " set the name of the guest\n"
1084 " string1 sets the window title and string2 the process name\n"
1085 " When debug-threads is enabled, individual threads are given a separate name\n"
1086 " NOTE: The thread names are for debugging and not a stable API.\n",
1090 Sets the name of the guest
. This name will be displayed
in the SDL
1091 window caption
. The name will also be used
for the VNC server
. Also
1092 optionally set the top visible process name
in Linux
. Naming of
1093 individual threads can also be enabled on Linux to aid debugging
.
1096 DEF("uuid", HAS_ARG
, QEMU_OPTION_uuid
,
1097 "-uuid %08x-%04x-%04x-%04x-%012x\n"
1098 " specify machine UUID\n", QEMU_ARCH_ALL
)
1106 DEFHEADING(Block device options
:)
1108 DEF("fda", HAS_ARG
, QEMU_OPTION_fda
,
1109 "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL
)
1110 DEF("fdb", HAS_ARG
, QEMU_OPTION_fdb
, "", QEMU_ARCH_ALL
)
1115 Use file as floppy disk
0/1 image (see the
:ref
:`disk images` chapter
in
1116 the System Emulation Users Guide
).
1119 DEF("hda", HAS_ARG
, QEMU_OPTION_hda
,
1120 "-hda/-hdb file use 'file' as IDE hard disk 0/1 image\n", QEMU_ARCH_ALL
)
1121 DEF("hdb", HAS_ARG
, QEMU_OPTION_hdb
, "", QEMU_ARCH_ALL
)
1122 DEF("hdc", HAS_ARG
, QEMU_OPTION_hdc
,
1123 "-hdc/-hdd file use 'file' as IDE hard disk 2/3 image\n", QEMU_ARCH_ALL
)
1124 DEF("hdd", HAS_ARG
, QEMU_OPTION_hdd
, "", QEMU_ARCH_ALL
)
1133 Use file as hard disk
0, 1, 2 or
3 image (see the
:ref
:`disk images`
1134 chapter
in the System Emulation Users Guide
).
1137 DEF("cdrom", HAS_ARG
, QEMU_OPTION_cdrom
,
1138 "-cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master)\n",
1142 Use file as CD
-ROM
image (you cannot use ``
-hdc`` and ``
-cdrom`` at
1143 the same time
). You can use the host CD
-ROM by
using ``
/dev
/cdrom``
1147 DEF("blockdev", HAS_ARG
, QEMU_OPTION_blockdev
,
1148 "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n"
1149 " [,cache.direct=on|off][,cache.no-flush=on|off]\n"
1150 " [,read-only=on|off][,auto-read-only=on|off]\n"
1151 " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n"
1152 " [,driver specific parameters...]\n"
1153 " configure a block backend\n", QEMU_ARCH_ALL
)
1155 ``
-blockdev option
[,option
[,option
[,...]]]``
1156 Define a
new block driver node
. Some of the options apply to all
1157 block drivers
, other options are only accepted
for a specific block
1158 driver
. See below
for a list of generic options and options
for the
1159 most common block drivers
.
1161 Options that expect a reference to another
node (e
.g
. ``file``
) can
1162 be given
in two ways
. Either you specify the node name of an already
1163 existing
node (file
=node
-name
), or you define a
new node
inline,
1164 adding options
for the referenced node after a dot
1165 (file
.filename
=path
,file
.aio
=native
).
1167 A block driver node created with ``
-blockdev`` can be used
for a
1168 guest device by specifying its node name
for the ``drive`` property
1169 in a ``
-device`` argument that defines a block device
.
1171 ``Valid options
for any block driver node
:``
1173 Specifies the block driver to use
for the given node
.
1176 This defines the name of the block driver node by which it
1177 will be referenced later
. The name must be unique
, i
.e
. it
1178 must not match the name of a different block driver node
, or
1179 (if you use ``
-drive`` as well
) the ID of a drive
.
1181 If no node name is specified
, it is automatically generated
.
1182 The generated node name is not intended to be predictable
1183 and changes between QEMU invocations
. For the top level
, an
1184 explicit node name must be specified
.
1187 Open the node read
-only
. Guest write attempts will fail
.
1189 Note that some block drivers support only read
-only access
,
1190 either generally or
in certain configurations
. In
this case,
1191 the
default value ``read
-only
=off`` does not work and the
1192 option must be specified explicitly
.
1195 If ``auto
-read
-only
=on`` is set
, QEMU may fall back to
1196 read
-only usage even when ``read
-only
=off`` is requested
, or
1197 even
switch between modes as needed
, e
.g
. depending on
1198 whether the image file is writable or whether a writing user
1199 is attached to the node
.
1202 Override the image locking system of QEMU by forcing the
1203 node to utilize weaker shared access
for permissions where
1204 it would normally request exclusive access
. When there is
1205 the potential
for multiple instances to have the same file
1206 open (whether
this invocation of QEMU is the first or the
1207 second instance
), both instances must permit shared access
1208 for the second instance to succeed at opening the file
.
1210 Enabling ``force
-share
=on`` requires ``read
-only
=on``
.
1213 The host page cache can be avoided with ``cache
.direct
=on``
.
1214 This will attempt to
do disk IO directly to the guest
's
1215 memory. QEMU may still perform an internal copy of the data.
1218 In case you don't care about data integrity over host
1219 failures
, you can use ``cache
.no
-flush
=on``
. This option
1220 tells QEMU that it
never needs to write any data to the disk
1221 but can instead keep things
in cache
. If anything goes
1222 wrong
, like your host losing power
, the disk storage getting
1223 disconnected accidentally
, etc
. your image will most
1224 probably be rendered unusable
.
1227 discard is one of
"ignore" (or
"off") or
"unmap" (or
"on")
1228 and controls whether ``discard``
(also known as ``trim`` or
1229 ``unmap``
) requests are ignored or passed to the filesystem
.
1230 Some machine types may not support discard requests
.
1232 ``detect
-zeroes
=detect
-zeroes``
1233 detect
-zeroes is
"off", "on" or
"unmap" and enables the
1234 automatic conversion of plain zero writes by the OS to
1235 driver specific optimized zero write commands
. You may even
1236 choose
"unmap" if discard is set to
"unmap" to allow a zero
1237 write to be converted to an ``unmap`` operation
.
1239 ``Driver
-specific options
for file``
1240 This is the protocol
-level block driver
for accessing regular
1244 The path to the image file
in the local filesystem
1247 Specifies the AIO
backend (threads
/native
/io_uring
,
1251 Specifies whether the image file is protected with Linux OFD
1252 / POSIX locks
. The
default is to use the Linux Open File
1253 Descriptor API
if available
, otherwise no lock is applied
.
1254 (auto
/on
/off
, default: auto
)
1260 -blockdev driver
=file
,node
-name
=disk
,filename
=disk
.img
1262 ``Driver
-specific options
for raw``
1263 This is the image format block driver
for raw images
. It is
1264 usually stacked on top of a protocol level block driver such as
1268 Reference to or definition of the data source block driver
1269 node (e
.g
. a ``file`` driver node
)
1275 -blockdev driver
=file
,node
-name
=disk_file
,filename
=disk
.img
1276 -blockdev driver
=raw
,node
-name
=disk
,file
=disk_file
1282 -blockdev driver
=raw
,node
-name
=disk
,file
.driver
=file
,file
.filename
=disk
.img
1284 ``Driver
-specific options
for qcow2``
1285 This is the image format block driver
for qcow2 images
. It is
1286 usually stacked on top of a protocol level block driver such as
1290 Reference to or definition of the data source block driver
1291 node (e
.g
. a ``file`` driver node
)
1294 Reference to or definition of the backing file block device
1295 (default is taken from the image file
). It is allowed to
1296 pass ``
null``
here in order to disable the
default backing
1300 Whether to enable the lazy refcounts
feature (on
/off
;
1301 default is taken from the image file
)
1304 The maximum total size of the L2 table and refcount block
1305 caches
in bytes (default: the sum of l2
-cache
-size and
1306 refcount
-cache
-size
)
1309 The maximum size of the L2 table cache
in bytes (default: if
1310 cache
-size is not specified
- 32M on Linux platforms
, and
8M
1311 on non
-Linux platforms
; otherwise
, as large as possible
1312 within the cache
-size
, while permitting the requested or the
1313 minimal refcount cache size
)
1315 ``refcount
-cache
-size``
1316 The maximum size of the refcount block cache
in bytes
1317 (default: 4 times the cluster size
; or
if cache
-size is
1318 specified
, the part of it which is not used
for the L2
1321 ``cache
-clean
-interval``
1322 Clean unused entries
in the L2 and refcount caches
. The
1323 interval is
in seconds
. The
default value is
600 on
1324 supporting platforms
, and
0 on other platforms
. Setting it
1325 to
0 disables
this feature
.
1327 ``pass
-discard
-request``
1328 Whether discard requests to the qcow2 device should be
1329 forwarded to the data
source (on
/off
; default: on
if
1330 discard
=unmap is specified
, off otherwise
)
1332 ``pass
-discard
-snapshot``
1333 Whether discard requests
for the data source should be
1334 issued when a snapshot
operation (e
.g
. deleting a snapshot
)
1335 frees clusters
in the qcow2
file (on
/off
; default: on
)
1337 ``pass
-discard
-other``
1338 Whether discard requests
for the data source should be
1339 issued on other occasions where a cluster gets freed
1340 (on
/off
; default: off
)
1343 Which overlap checks to perform
for writes to the image
1344 (none
/constant
/cached
/all
; default: cached
). For details or
1345 finer granularity control refer to the QAPI documentation of
1352 -blockdev driver
=file
,node
-name
=my_file
,filename
=/tmp
/disk
.qcow2
1353 -blockdev driver
=qcow2
,node
-name
=hda
,file
=my_file
,overlap
-check
=none
,cache
-size
=16777216
1359 -blockdev driver
=qcow2
,node
-name
=disk
,file
.driver
=http
,file
.filename
=http
://example.com/image.qcow2
1361 ``Driver
-specific options
for other drivers``
1362 Please refer to the QAPI documentation of the ``blockdev
-add``
1366 DEF("drive", HAS_ARG
, QEMU_OPTION_drive
,
1367 "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n"
1368 " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n"
1369 " [,snapshot=on|off][,rerror=ignore|stop|report]\n"
1370 " [,werror=ignore|stop|report|enospc][,id=name]\n"
1371 " [,aio=threads|native|io_uring]\n"
1372 " [,readonly=on|off][,copy-on-read=on|off]\n"
1373 " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n"
1374 " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n"
1375 " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n"
1376 " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n"
1377 " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n"
1378 " [[,iops_size=is]]\n"
1380 " use 'file' as a drive image\n", QEMU_ARCH_ALL
)
1382 ``
-drive option
[,option
[,option
[,...]]]``
1383 Define a
new drive
. This includes creating a block driver
node (the
1384 backend
) as well as a guest device
, and is mostly a shortcut
for
1385 defining the corresponding ``
-blockdev`` and ``
-device`` options
.
1387 ``
-drive`` accepts all options that are accepted by ``
-blockdev``
.
1388 In addition
, it knows the following options
:
1391 This option defines which disk
image (see the
:ref
:`disk images`
1392 chapter
in the System Emulation Users Guide
) to use with
this drive
.
1393 If the filename contains comma
, you must double
it (for instance
,
1394 "file=my,,file" to use file
"my,file").
1396 Special files such as iSCSI devices can be specified
using
1397 protocol specific URLs
. See the section
for "Device URL Syntax"
1398 for more information
.
1401 This option defines on which type on
interface the drive is
1402 connected
. Available types are
: ide
, scsi
, sd
, mtd
, floppy
,
1403 pflash
, virtio
, none
.
1405 ``bus
=bus
,unit
=unit``
1406 These options define where is connected the drive by defining
1407 the bus number and the unit id
.
1410 This option defines where the drive is connected by
using an
1411 index
in the list of available connectors of a given
interface
1415 This option defines the type of the media
: disk or cdrom
.
1417 ``snapshot
=snapshot``
1418 snapshot is
"on" or
"off" and controls snapshot mode
for the
1419 given
drive (see ``
-snapshot``
).
1422 cache is
"none", "writeback", "unsafe", "directsync" or
1423 "writethrough" and controls how the host cache is used to access
1424 block data
. This is a shortcut that sets the ``cache
.direct``
1425 and ``cache
.no
-flush``
options (as
in ``
-blockdev``
), and
1426 additionally ``cache
.writeback``
, which provides a
default for
1427 the ``write
-cache`` option of block guest
devices (as
in
1428 ``
-device``
). The modes correspond to the following settings
:
1430 ============= =============== ============ ==============
1431 \ cache
.writeback cache
.direct cache
.no
-flush
1432 ============= =============== ============ ==============
1433 writeback on off off
1435 writethrough off off off
1436 directsync off on off
1438 ============= =============== ============ ==============
1440 The
default mode is ``cache
=writeback``
.
1443 aio is
"threads", "native", or
"io_uring" and selects between pthread
1444 based disk I
/O
, native Linux AIO
, or Linux io_uring API
.
1447 Specify which disk format will be used rather than detecting the
1448 format
. Can be used to specify format
=raw to avoid interpreting
1449 an untrusted format header
.
1451 ``werror
=action
,rerror
=action``
1452 Specify which action to take on write and read errors
. Valid
1453 actions are
: "ignore" (ignore the error and
try to
continue),
1454 "stop" (pause QEMU
), "report" (report the error to the guest
),
1455 "enospc" (pause QEMU only
if the host disk is full
; report the
1456 error to the guest otherwise
). The
default setting is
1457 ``werror
=enospc`` and ``rerror
=report``
.
1459 ``copy
-on
-read
=copy
-on
-read``
1460 copy
-on
-read is
"on" or
"off" and enables whether to copy read
1461 backing file sectors into the image file
.
1463 ``bps
=b
,bps_rd
=r
,bps_wr
=w``
1464 Specify bandwidth throttling limits
in bytes per second
, either
1465 for all request types or
for reads or writes only
. Small values
1466 can lead to timeouts or hangs inside the guest
. A safe minimum
1467 for disks is
2 MB
/s
.
1469 ``bps_max
=bm
,bps_rd_max
=rm
,bps_wr_max
=wm``
1470 Specify bursts
in bytes per second
, either
for all request types
1471 or
for reads or writes only
. Bursts allow the guest I
/O to spike
1472 above the limit temporarily
.
1474 ``iops
=i
,iops_rd
=r
,iops_wr
=w``
1475 Specify request rate limits
in requests per second
, either
for
1476 all request types or
for reads or writes only
.
1478 ``iops_max
=bm
,iops_rd_max
=rm
,iops_wr_max
=wm``
1479 Specify bursts
in requests per second
, either
for all request
1480 types or
for reads or writes only
. Bursts allow the guest I
/O to
1481 spike above the limit temporarily
.
1484 Let every is bytes of a request count as a
new request
for iops
1485 throttling purposes
. Use
this option to prevent guests from
1486 circumventing iops limits by sending fewer but larger requests
.
1489 Join a throttling quota group with given name g
. All drives that
1490 are members of the same group are accounted
for together
. Use
1491 this option to prevent guests from circumventing throttling
1492 limits by
using many small disks instead of a single larger
1495 By
default, the ``cache
.writeback
=on`` mode is used
. It will report
1496 data writes as completed as soon as the data is present
in the host
1497 page cache
. This is safe as long as your guest OS makes sure to
1498 correctly flush disk caches where needed
. If your guest OS does not
1499 handle volatile disk write caches correctly and your host crashes or
1500 loses power
, then the guest may experience data corruption
.
1502 For such guests
, you should consider
using ``cache
.writeback
=off``
.
1503 This means that the host page cache will be used to read and write
1504 data
, but write notification will be sent to the guest only after
1505 QEMU has made sure to flush each write to the disk
. Be aware that
1506 this has a major impact on performance
.
1508 When
using the ``
-snapshot`` option
, unsafe caching is always used
.
1510 Copy
-on
-read avoids accessing the same backing file sectors
1511 repeatedly and is useful when the backing file is over a slow
1512 network
. By
default copy
-on
-read is off
.
1514 Instead of ``
-cdrom`` you can use
:
1518 |qemu_system|
-drive file
=file
,index
=2,media
=cdrom
1520 Instead of ``
-hda``
, ``
-hdb``
, ``
-hdc``
, ``
-hdd``
, you can use
:
1524 |qemu_system|
-drive file
=file
,index
=0,media
=disk
1525 |qemu_system|
-drive file
=file
,index
=1,media
=disk
1526 |qemu_system|
-drive file
=file
,index
=2,media
=disk
1527 |qemu_system|
-drive file
=file
,index
=3,media
=disk
1529 You can open an image
using pre
-opened file descriptors from an fd
1535 -add
-fd fd
=3,set
=2,opaque
="rdwr:/path/to/file" \\
1536 -add
-fd fd
=4,set
=2,opaque
="rdonly:/path/to/file" \\
1537 -drive file
=/dev
/fdset
/2,index
=0,media
=disk
1539 You can connect a CDROM to the slave of ide0
:
1543 |qemu_system_x86|
-drive file
=file
,if=ide
,index
=1,media
=cdrom
1545 If you don
't specify the "file=" argument, you define an empty
1550 |qemu_system_x86| -drive if=ide,index=1,media=cdrom
1552 Instead of ``-fda``, ``-fdb``, you can use:
1556 |qemu_system_x86| -drive file=file,index=0,if=floppy
1557 |qemu_system_x86| -drive file=file,index=1,if=floppy
1559 By default, interface is "ide" and index is automatically
1564 |qemu_system_x86| -drive file=a -drive file=b"
1566 is interpreted like:
1570 |qemu_system_x86| -hda a -hdb b
1573 DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock,
1574 "-mtdblock file use 'file
' as on-board Flash memory image\n",
1578 Use file as on-board Flash memory image.
1581 DEF("sd", HAS_ARG, QEMU_OPTION_sd,
1582 "-sd file use 'file
' as SecureDigital card image\n", QEMU_ARCH_ALL)
1585 Use file as SecureDigital card image.
1588 DEF("snapshot", 0, QEMU_OPTION_snapshot,
1589 "-snapshot write to temporary files instead of disk image files\n",
1593 Write to temporary files instead of disk image files. In this case,
1594 the raw disk image you use is not written back. You can however
1595 force the write back by pressing C-a s (see the :ref:`disk images`
1596 chapter in the System Emulation Users Guide).
1599 DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev,
1600 "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1601 " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n"
1602 " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n"
1603 " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n"
1604 " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n"
1605 " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n"
1606 " [[,throttling.iops-size=is]]\n"
1607 "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n"
1608 "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n"
1609 "-fsdev synth,id=id\n",
1613 ``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]``
1615 ``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]``
1617 ``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]``
1619 ``-fsdev synth,id=id[,readonly=on]``
1620 Define a new file system device. Valid options are:
1623 Accesses to the filesystem are done by QEMU.
1626 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1629 Synthetic filesystem, only used by QTests.
1632 Specifies identifier for this device.
1635 Specifies the export path for the file system device. Files
1636 under this path will be available to the 9p client on the guest.
1638 ``security_model=security_model``
1639 Specifies the security model to be used for this export path.
1640 Supported security models are "passthrough", "mapped-xattr",
1641 "mapped-file" and "none". In "passthrough" security model, files
1642 are stored using the same credentials as they are created on the
1643 guest. This requires QEMU to run as root. In "mapped-xattr"
1644 security model, some of the file attributes like uid, gid, mode
1645 bits and link target are stored as file attributes. For
1646 "mapped-file" these attributes are stored in the hidden
1647 .virtfs\_metadata directory. Directories exported by this
1648 security model cannot interact with other unix tools. "none"
1649 security model is same as passthrough except the sever won't
1650 report failures
if it fails to set file attributes like
1651 ownership
. Security model is mandatory only
for local fsdriver
.
1652 Other
fsdrivers (like proxy
) don
't take security model as a
1655 ``writeout=writeout``
1656 This is an optional argument. The only supported value is
1657 "immediate". This means that host page cache will be used to
1658 read and write data but write notification will be sent to the
1659 guest only when the data has been reported as written by the
1663 Enables exporting 9p share as a readonly mount for guests. By
1664 default read-write access is given.
1667 Enables proxy filesystem driver to use passed socket file for
1668 communicating with virtfs-proxy-helper(1).
1671 Enables proxy filesystem driver to use passed socket descriptor
1672 for communicating with virtfs-proxy-helper(1). Usually a helper
1673 like libvirt will create socketpair and pass one of the fds as
1677 Specifies the default mode for newly created files on the host.
1678 Works only with security models "mapped-xattr" and
1682 Specifies the default mode for newly created directories on the
1683 host. Works only with security models "mapped-xattr" and
1686 ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w``
1687 Specify bandwidth throttling limits in bytes per second, either
1688 for all request types or for reads or writes only.
1690 ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm``
1691 Specify bursts in bytes per second, either for all request types
1692 or for reads or writes only. Bursts allow the guest I/O to spike
1693 above the limit temporarily.
1695 ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w``
1696 Specify request rate limits in requests per second, either for
1697 all request types or for reads or writes only.
1699 ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm``
1700 Specify bursts in requests per second, either for all request
1701 types or for reads or writes only. Bursts allow the guest I/O to
1702 spike above the limit temporarily.
1704 ``throttling.iops-size=is``
1705 Let every is bytes of a request count as a new request for iops
1706 throttling purposes.
1708 -fsdev option is used along with -device driver "virtio-9p-...".
1710 ``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag``
1711 Options for virtio-9p-... driver are:
1714 Specifies the variant to be used. Supported values are "pci",
1715 "ccw" or "device", depending on the machine type.
1718 Specifies the id value specified along with -fsdev option.
1720 ``mount_tag=mount_tag``
1721 Specifies the tag name to be used by the guest to mount this
1725 DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs,
1726 "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n"
1727 " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n"
1728 "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n"
1729 "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n"
1730 "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n",
1734 ``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]``
1736 ``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1738 ``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]``
1740 ``-virtfs synth,mount_tag=mount_tag``
1741 Define a new virtual filesystem device and expose it to the guest using
1742 a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain
1743 directory on host is made directly accessible by guest as a pass-through
1744 file system by using the 9P network protocol for communication between
1745 host and guests, if desired even accessible, shared by several guests
1748 Note that ``-virtfs`` is actually just a convenience shortcut for its
1749 generalized form ``-fsdev -device virtio-9p-pci``.
1751 The general form of pass-through file system options are:
1754 Accesses to the filesystem are done by QEMU.
1757 Accesses to the filesystem are done by virtfs-proxy-helper(1).
1760 Synthetic filesystem, only used by QTests.
1763 Specifies identifier for the filesystem device
1766 Specifies the export path for the file system device. Files
1767 under this path will be available to the 9p client on the guest.
1769 ``security_model=security_model``
1770 Specifies the security model to be used for this export path.
1771 Supported security models are "passthrough", "mapped-xattr",
1772 "mapped-file" and "none". In "passthrough" security model, files
1773 are stored using the same credentials as they are created on the
1774 guest. This requires QEMU to run as root. In "mapped-xattr"
1775 security model, some of the file attributes like uid, gid, mode
1776 bits and link target are stored as file attributes. For
1777 "mapped-file" these attributes are stored in the hidden
1778 .virtfs\_metadata directory. Directories exported by this
1779 security model cannot interact with other unix tools. "none"
1780 security model is same as passthrough except the sever won't
1781 report failures
if it fails to set file attributes like
1782 ownership
. Security model is mandatory only
for local fsdriver
.
1783 Other
fsdrivers (like proxy
) don
't take security model as a
1786 ``writeout=writeout``
1787 This is an optional argument. The only supported value is
1788 "immediate". This means that host page cache will be used to
1789 read and write data but write notification will be sent to the
1790 guest only when the data has been reported as written by the
1794 Enables exporting 9p share as a readonly mount for guests. By
1795 default read-write access is given.
1798 Enables proxy filesystem driver to use passed socket file for
1799 communicating with virtfs-proxy-helper(1). Usually a helper like
1800 libvirt will create socketpair and pass one of the fds as
1804 Enables proxy filesystem driver to use passed 'sock\_fd
' as the
1805 socket descriptor for interfacing with virtfs-proxy-helper(1).
1808 Specifies the default mode for newly created files on the host.
1809 Works only with security models "mapped-xattr" and
1813 Specifies the default mode for newly created directories on the
1814 host. Works only with security models "mapped-xattr" and
1817 ``mount_tag=mount_tag``
1818 Specifies the tag name to be used by the guest to mount this
1821 ``multidevs=multidevs``
1822 Specifies how to deal with multiple devices being shared with a
1823 9p export. Supported behaviours are either "remap", "forbid" or
1824 "warn". The latter is the default behaviour on which virtfs 9p
1825 expects only one device to be shared with the same export, and
1826 if more than one device is shared and accessed via the same 9p
1827 export then only a warning message is logged (once) by qemu on
1828 host side. In order to avoid file ID collisions on guest you
1829 should either create a separate virtfs export for each device to
1830 be shared with guests (recommended way) or you might use "remap"
1831 instead which allows you to share multiple devices with only one
1832 export instead, which is achieved by remapping the original
1833 inode numbers from host to guest in a way that would prevent
1834 such collisions. Remapping inodes in such use cases is required
1835 because the original device IDs from host are never passed and
1836 exposed on guest. Instead all files of an export shared with
1837 virtfs always share the same device id on guest. So two files
1838 with identical inode numbers but from actually different devices
1839 on host would otherwise cause a file ID collision and hence
1840 potential misbehaviours on guest. "forbid" on the other hand
1841 assumes like "warn" that only one device is shared by the same
1842 export, however it will not only log a warning message but also
1843 deny access to additional devices on guest. Note though that
1844 "forbid" does currently not block all possible file access
1845 operations (e.g. readdir() would still return entries from other
1849 DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi,
1850 "-iscsi [user=user][,password=password]\n"
1851 " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE\n"
1852 " [,initiator-name=initiator-iqn][,id=target-iqn]\n"
1853 " [,timeout=timeout]\n"
1854 " iSCSI session parameters\n", QEMU_ARCH_ALL)
1858 Configure iSCSI session parameters.
1863 DEFHEADING(USB convenience options:)
1865 DEF("usb", 0, QEMU_OPTION_usb,
1866 "-usb enable on-board USB host controller (if not enabled by default)\n",
1870 Enable USB emulation on machine types with an on-board USB host
1871 controller (if not enabled by default). Note that on-board USB host
1872 controllers may not support USB 3.0. In this case
1873 ``-device qemu-xhci`` can be used instead on machines with PCI.
1876 DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice,
1877 "-usbdevice name add the host or guest USB device 'name
'\n",
1880 ``-usbdevice devname``
1881 Add the USB device devname, and enable an on-board USB controller
1882 if possible and necessary (just like it can be done via
1883 ``-machine usb=on``). Note that this option is mainly intended for
1884 the user's convenience only
. More fine
-grained control can be
1885 achieved by selecting a USB host
controller (if necessary
) and the
1886 desired USB device via the ``
-device`` option instead
. For example
,
1887 instead of
using ``
-usbdevice mouse`` it is possible to use
1888 ``
-device qemu
-xhci
-device usb
-mouse`` to connect the USB mouse
1889 to a USB
3.0 controller
instead (at least on machines that support
1890 PCI and
do not have an USB controller enabled by
default yet
).
1891 For more details
, see the chapter about
1892 :ref
:`Connecting USB devices`
in the System Emulation Users Guide
.
1893 Possible devices
for devname are
:
1896 Braille device
. This will use BrlAPI to display the braille
1897 output on a real or fake
device (i
.e
. it also creates a
1898 corresponding ``braille`` chardev automatically beside the
1899 ``usb
-braille`` USB device
).
1902 Standard USB keyboard
. Will
override the PS
/2 keyboard (if present
).
1905 Virtual Mouse
. This will
override the PS
/2 mouse emulation when
1909 Pointer device that uses absolute
coordinates (like a
1910 touchscreen
). This means QEMU is able to report the mouse
1911 position without having to grab the mouse
. Also overrides the
1912 PS
/2 mouse emulation when activated
.
1915 Wacom PenPartner USB tablet
.
1922 DEFHEADING(Display options
:)
1924 DEF("display", HAS_ARG
, QEMU_OPTION_display
,
1925 #
if defined(CONFIG_SPICE
)
1926 "-display spice-app[,gl=on|off]\n"
1928 #
if defined(CONFIG_SDL
)
1929 "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n"
1930 " [,window-close=on|off]\n"
1932 #
if defined(CONFIG_GTK
)
1933 "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n"
1934 " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n"
1936 #
if defined(CONFIG_VNC
)
1937 "-display vnc=<display>[,<optargs>]\n"
1939 #
if defined(CONFIG_CURSES
)
1940 "-display curses[,charset=<encoding>]\n"
1942 #
if defined(CONFIG_COCOA
)
1943 "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n"
1945 #
if defined(CONFIG_OPENGL
)
1946 "-display egl-headless[,rendernode=<file>]\n"
1948 #
if defined(CONFIG_DBUS_DISPLAY
)
1949 "-display dbus[,addr=<dbusaddr>]\n"
1950 " [,gl=on|core|es|off][,rendernode=<file>]\n"
1952 #
if defined(CONFIG_COCOA
)
1953 "-display cocoa[,show-cursor=on|off][,left-command-key=on|off]\n"
1956 " select display backend type\n"
1957 " The default display is equivalent to\n "
1958 #
if defined(CONFIG_GTK
)
1959 "\"-display gtk\"\n"
1960 #elif
defined(CONFIG_SDL
)
1961 "\"-display sdl\"\n"
1962 #elif
defined(CONFIG_COCOA
)
1963 "\"-display cocoa\"\n"
1964 #elif
defined(CONFIG_VNC
)
1965 "\"-vnc localhost:0,to=99,id=default\"\n"
1967 "\"-display none\"\n"
1972 Select type of display to use
. Use ``
-display help`` to list the available
1973 display types
. Valid values
for type are
1975 ``spice
-app
[,gl
=on|off
]``
1976 Start QEMU as a Spice server and launch the
default Spice client
1977 application
. The Spice server will redirect the serial consoles
1978 and QEMU monitors
. (Since
4.0)
1981 Export the display over D
-Bus interfaces
. (Since
7.0)
1983 The connection is registered with the
"org.qemu" name (and queued when
1986 ``addr
=<dbusaddr
>``
: D
-Bus bus address to connect to
.
1988 ``p2p
=yes|no``
: Use peer
-to
-peer connection
, accepted via QMP ``add_client``
.
1990 ``gl
=on|off|core|es``
: Use OpenGL
for rendering (the D
-Bus
interface
1991 will share framebuffers with DMABUF file descriptors
).
1994 Display video output via
SDL (usually
in a separate graphics
1995 window
; see the SDL documentation
for other possibilities
).
1996 Valid parameters are
:
1998 ``grab
-mod
=<mods
>``
: Used to select the modifier keys
for toggling
1999 the mouse grabbing
in conjunction with the
"g" key
. ``
<mods
>`` can be
2000 either ``lshift
-lctrl
-lalt`` or ``rctrl``
.
2002 ``gl
=on|off|core|es``
: Use OpenGL
for displaying
2004 ``show
-cursor
=on|off``
: Force showing the mouse cursor
2006 ``window
-close
=on|off``
: Allow to quit qemu with window close button
2009 Display video output
in a GTK window
. This
interface provides
2010 drop
-down menus and other UI elements to configure and control
2011 the VM during runtime
. Valid parameters are
:
2013 ``full
-screen
=on|off``
: Start
in fullscreen mode
2015 ``gl
=on|off``
: Use OpenGL
for displaying
2017 ``grab
-on
-hover
=on|off``
: Grab keyboard input on mouse hover
2019 ``show
-tabs
=on|off``
: Display the tab bar
for switching between the
2020 various graphical
interfaces (e
.g
. VGA and
2021 virtual console character devices
) by
default.
2023 ``show
-cursor
=on|off``
: Force showing the mouse cursor
2025 ``window
-close
=on|off``
: Allow to quit qemu with window close button
2027 ``curses
[,charset
=<encoding
>]``
2028 Display video output via curses
. For graphics device models
2029 which support a text mode
, QEMU can display
this output
using a
2030 curses
/ncurses
interface. Nothing is displayed when the graphics
2031 device is
in graphical mode or
if the graphics device does not
2032 support a text mode
. Generally only the VGA device models
2033 support text mode
. The font charset used by the guest can be
2034 specified with the ``charset`` option
, for example
2035 ``charset
=CP850``
for IBM CP850 encoding
. The
default is
2039 Display video output
in a Cocoa window
. Mac only
. This
interface
2040 provides drop
-down menus and other UI elements to configure and
2041 control the VM during runtime
. Valid parameters are
:
2043 ``show
-cursor
=on|off``
: Force showing the mouse cursor
2045 ``left
-command
-key
=on|off``
: Disable forwarding left command key to host
2047 ``egl
-headless
[,rendernode
=<file
>]``
2048 Offload all OpenGL operations to a local DRI device
. For any
2049 graphical display
, this display needs to be paired with either
2050 VNC or SPICE displays
.
2053 Start a VNC server on display
<display
>
2056 Do not display video output
. The guest will still see an
2057 emulated graphics card
, but its output will not be displayed to
2058 the QEMU user
. This option differs from the
-nographic option
in
2059 that it only affects what is done with video output
; -nographic
2060 also changes the destination of the serial and parallel port
2064 DEF("nographic", 0, QEMU_OPTION_nographic
,
2065 "-nographic disable graphical output and redirect serial I/Os to console\n",
2069 Normally
, if QEMU is compiled with graphical window support
, it
2070 displays output such as guest graphics
, guest console
, and the QEMU
2071 monitor
in a window
. With
this option
, you can totally disable
2072 graphical output so that QEMU is a simple command line application
.
2073 The emulated serial port is redirected on the console and muxed with
2074 the
monitor (unless redirected elsewhere explicitly
). Therefore
, you
2075 can still use QEMU to debug a Linux kernel with a serial console
.
2076 Use C
-a h
for help on switching between the console and monitor
.
2080 DEF("spice", HAS_ARG
, QEMU_OPTION_spice
,
2081 "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n"
2082 " [,x509-key-file=<file>][,x509-key-password=<file>]\n"
2083 " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n"
2084 " [,x509-dh-key-file=<file>][,addr=addr]\n"
2085 " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n"
2086 " [,tls-ciphers=<list>]\n"
2087 " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n"
2088 " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n"
2089 " [,sasl=on|off][,disable-ticketing=on|off]\n"
2090 " [,password=<string>][,password-secret=<secret-id>]\n"
2091 " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n"
2092 " [,jpeg-wan-compression=[auto|never|always]]\n"
2093 " [,zlib-glz-wan-compression=[auto|never|always]]\n"
2094 " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n"
2095 " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n"
2096 " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n"
2097 " [,gl=[on|off]][,rendernode=<file>]\n"
2099 " at least one of {port, tls-port} is mandatory\n",
2103 ``
-spice option
[,option
[,...]]``
2104 Enable the spice remote desktop protocol
. Valid options are
2107 Set the TCP port spice is listening on
for plaintext channels
.
2110 Set the IP address spice is listening on
. Default is any
2113 ``ipv4
=on|off``
; \ ``ipv6
=on|off``
; \ ``unix
=on|off``
2114 Force
using the specified IP version
.
2116 ``password
=<string
>``
2117 Set the password you need to authenticate
.
2119 This option is deprecated and insecure because it leaves the
2120 password visible
in the process listing
. Use ``password
-secret``
2123 ``password
-secret
=<secret
-id
>``
2124 Set the ID of the ``secret`` object containing the password
2125 you need to authenticate
.
2128 Require that the client use SASL to authenticate with the spice
.
2129 The exact choice of authentication method used is controlled
2130 from the system
/ user
's SASL configuration file for the 'qemu
'
2131 service. This is typically found in /etc/sasl2/qemu.conf. If
2132 running QEMU as an unprivileged user, an environment variable
2133 SASL\_CONF\_PATH can be used to make it search alternate
2134 locations for the service config. While some SASL auth methods
2135 can also provide data encryption (eg GSSAPI), it is recommended
2136 that SASL always be combined with the 'tls
' and 'x509
' settings
2137 to enable use of SSL and server certificates. This ensures a
2138 data encryption preventing compromise of authentication
2141 ``disable-ticketing=on|off``
2142 Allow client connects without authentication.
2144 ``disable-copy-paste=on|off``
2145 Disable copy paste between the client and the guest.
2147 ``disable-agent-file-xfer=on|off``
2148 Disable spice-vdagent based file-xfer between the client and the
2152 Set the TCP port spice is listening on for encrypted channels.
2155 Set the x509 file directory. Expects same filenames as -vnc
2158 ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>``
2159 The x509 file names can also be configured individually.
2161 ``tls-ciphers=<list>``
2162 Specify which ciphers to use.
2164 ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]``
2165 Force specific channel to be used with or without TLS
2166 encryption. The options can be specified multiple times to
2167 configure multiple channels. The special name "default" can be
2168 used to set the default mode. For channels which are not
2169 explicitly forced into one mode the spice client is allowed to
2170 pick tls/plaintext as he pleases.
2172 ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]``
2173 Configure image compression (lossless). Default is auto\_glz.
2175 ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]``
2176 Configure wan image compression (lossy for slow links). Default
2179 ``streaming-video=[off|all|filter]``
2180 Configure video stream detection. Default is off.
2182 ``agent-mouse=[on|off]``
2183 Enable/disable passing mouse events via vdagent. Default is on.
2185 ``playback-compression=[on|off]``
2186 Enable/disable audio stream compression (using celt 0.5.1).
2189 ``seamless-migration=[on|off]``
2190 Enable/disable spice seamless migration. Default is off.
2193 Enable/disable OpenGL context. Default is off.
2195 ``rendernode=<file>``
2196 DRM render node for OpenGL rendering. If not specified, it will
2197 pick the first available. (Since 2.9)
2200 DEF("portrait", 0, QEMU_OPTION_portrait,
2201 "-portrait rotate graphical output 90 deg left (only PXA LCD)\n",
2205 Rotate graphical output 90 deg left (only PXA LCD).
2208 DEF("rotate", HAS_ARG, QEMU_OPTION_rotate,
2209 "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n",
2213 Rotate graphical output some deg left (only PXA LCD).
2216 DEF("vga", HAS_ARG, QEMU_OPTION_vga,
2217 "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n"
2218 " select video card type\n", QEMU_ARCH_ALL)
2221 Select type of VGA card to emulate. Valid values for type are
2224 Cirrus Logic GD5446 Video card. All Windows versions starting
2225 from Windows 95 should recognize and use this graphic card. For
2226 optimal performances, use 16 bit color depth in the guest and
2227 the host OS. (This card was the default before QEMU 2.2)
2230 Standard VGA card with Bochs VBE extensions. If your guest OS
2231 supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if
2232 you want to use high resolution modes (>= 1280x1024x16) then you
2233 should use this option. (This card is the default since QEMU
2237 VMWare SVGA-II compatible adapter. Use it if you have
2238 sufficiently recent XFree86/XOrg server or Windows guest with a
2239 driver for this card.
2242 QXL paravirtual graphic card. It is VGA compatible (including
2243 VESA 2.0 VBE support). Works best with qxl guest drivers
2244 installed though. Recommended choice when using the spice
2248 (sun4m only) Sun TCX framebuffer. This is the default
2249 framebuffer for sun4m machines and offers both 8-bit and 24-bit
2250 colour depths at a fixed resolution of 1024x768.
2253 (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit
2254 framebuffer for sun4m machines available in both 1024x768
2255 (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people
2256 wishing to run older Solaris versions.
2265 DEF("full-screen", 0, QEMU_OPTION_full_screen,
2266 "-full-screen start in full screen\n", QEMU_ARCH_ALL)
2269 Start in full screen.
2272 DEF("g", HAS_ARG, QEMU_OPTION_g ,
2273 "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n",
2274 QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K)
2276 ``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]``
2277 Set the initial graphical resolution and depth (PPC, SPARC only).
2279 For PPC the default is 800x600x32.
2281 For SPARC with the TCX graphics device, the default is 1024x768x8
2282 with the option of 1024x768x24. For cgthree, the default is
2283 1024x768x8 with the option of 1152x900x8 for people who wish to use
2287 DEF("vnc", HAS_ARG, QEMU_OPTION_vnc ,
2288 "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL)
2290 ``-vnc display[,option[,option[,...]]]``
2291 Normally, if QEMU is compiled with graphical window support, it
2292 displays output such as guest graphics, guest console, and the QEMU
2293 monitor in a window. With this option, you can have QEMU listen on
2294 VNC display display and redirect the VGA display over the VNC
2295 session. It is very useful to enable the usb tablet device when
2296 using this option (option ``-device usb-tablet``). When using the
2297 VNC display, you must use the ``-k`` parameter to set the keyboard
2298 layout if you are not using en-us. Valid syntax for the display is
2301 With this option, QEMU will try next available VNC displays,
2302 until the number L, if the origianlly defined "-vnc display" is
2303 not available, e.g. port 5900+display is already used by another
2304 application. By default, to=0.
2307 TCP connections will only be allowed from host on display d. By
2308 convention the TCP port is 5900+d. Optionally, host can be
2309 omitted in which case the server will accept connections from
2313 Connections will be allowed over UNIX domain sockets where path
2314 is the location of a unix socket to listen for connections on.
2317 VNC is initialized but not started. The monitor ``change``
2318 command can be used to later start the VNC server.
2320 Following the display value there may be one or more option flags
2321 separated by commas. Valid options are
2324 Connect to a listening VNC client via a "reverse" connection.
2325 The client is specified by the display. For reverse network
2326 connections (host:d,``reverse``), the d argument is a TCP port
2327 number, not a display number.
2329 ``websocket=on|off``
2330 Opens an additional TCP listening port dedicated to VNC
2331 Websocket connections. If a bare websocket option is given, the
2332 Websocket port is 5700+display. An alternative port can be
2333 specified with the syntax ``websocket``\ =port.
2335 If host is specified connections will only be allowed from this
2336 host. It is possible to control the websocket listen address
2337 independently, using the syntax ``websocket``\ =host:port.
2339 If no TLS credentials are provided, the websocket connection
2340 runs in unencrypted mode. If TLS credentials are provided, the
2341 websocket connection requires encrypted client connections.
2344 Require that password based authentication is used for client
2347 The password must be set separately using the ``set_password``
2348 command in the :ref:`QEMU monitor`. The
2349 syntax to change your password is:
2350 ``set_password <protocol> <password>`` where <protocol> could be
2351 either "vnc" or "spice".
2353 If you would like to change <protocol> password expiration, you
2354 should use ``expire_password <protocol> <expiration-time>``
2355 where expiration time could be one of the following options:
2356 now, never, +seconds or UNIX time of expiration, e.g. +60 to
2357 make password expire in 60 seconds, or 1335196800 to make
2358 password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for
2359 this date and time).
2361 You can also use keywords "now" or "never" for the expiration
2362 time to allow <protocol> password to expire immediately or never
2365 ``password-secret=<secret-id>``
2366 Require that password based authentication is used for client
2367 connections, using the password provided by the ``secret``
2368 object identified by ``secret-id``.
2371 Provides the ID of a set of TLS credentials to use to secure the
2372 VNC server. They will apply to both the normal VNC server socket
2373 and the websocket socket (if enabled). Setting TLS credentials
2374 will cause the VNC server socket to enable the VeNCrypt auth
2375 mechanism. The credentials should have been previously created
2376 using the ``-object tls-creds`` argument.
2379 Provides the ID of the QAuthZ authorization object against which
2380 the client's x509 distinguished name will validated
. This object
2381 is only resolved at time of use
, so can be deleted and recreated
2382 on the fly
while the VNC server is active
. If missing
, it will
2383 default to denying access
.
2386 Require that the client use SASL to authenticate with the VNC
2387 server
. The exact choice of authentication method used is
2388 controlled from the system
/ user
's SASL configuration file for
2389 the 'qemu
' service. This is typically found in
2390 /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user,
2391 an environment variable SASL\_CONF\_PATH can be used to make it
2392 search alternate locations for the service config. While some
2393 SASL auth methods can also provide data encryption (eg GSSAPI),
2394 it is recommended that SASL always be combined with the 'tls
'
2395 and 'x509
' settings to enable use of SSL and server
2396 certificates. This ensures a data encryption preventing
2397 compromise of authentication credentials. See the
2398 :ref:`VNC security` section in the System Emulation Users Guide
2399 for details on using SASL authentication.
2402 Provides the ID of the QAuthZ authorization object against which
2403 the client's SASL username will validated
. This object is only
2404 resolved at time of use
, so can be deleted and recreated on the
2405 fly
while the VNC server is active
. If missing
, it will
default
2409 Legacy method
for enabling authorization of clients against the
2410 x509 distinguished name and SASL username
. It results
in the
2411 creation of two ``authz
-list`` objects with IDs of
2412 ``vnc
.username`` and ``vnc
.x509dname``
. The rules
for these
2413 objects must be configured with the HMP ACL commands
.
2415 This option is deprecated and should no longer be used
. The
new
2416 ``sasl
-authz`` and ``tls
-authz`` options are a replacement
.
2419 Enable lossy compression
methods (gradient
, JPEG
, ...). If
this
2420 option is set
, VNC client may receive lossy framebuffer updates
2421 depending on its encoding settings
. Enabling
this option can
2422 save a lot of bandwidth at the expense of quality
.
2424 ``non
-adaptive
=on|off``
2425 Disable adaptive encodings
. Adaptive encodings are enabled by
2426 default. An adaptive encoding will
try to detect frequently
2427 updated screen regions
, and send updates
in these regions
using
2428 a lossy
encoding (like JPEG
). This can be really helpful to save
2429 bandwidth when playing videos
. Disabling adaptive encodings
2430 restores the original
static behavior of encodings like Tight
.
2432 ``share
=[allow
-exclusive|force
-shared|ignore
]``
2433 Set display sharing policy
. 'allow-exclusive' allows clients to
2434 ask
for exclusive access
. As suggested by the rfb spec
this is
2435 implemented by dropping other connections
. Connecting multiple
2436 clients
in parallel requires all clients asking
for a shared
2437 session (vncviewer
: -shared
switch). This is the
default.
2438 'force-shared' disables exclusive client access
. Useful
for
2439 shared desktop sessions
, where you don
't want someone forgetting
2440 specify -shared disconnect everybody else. 'ignore
' completely
2441 ignores the shared flag and allows everybody connect
2442 unconditionally. Doesn't conform to the rfb spec but is
2443 traditional QEMU behavior
.
2446 Set keyboard delay
, for key down and key up events
, in
2447 milliseconds
. Default is
10. Keyboards are low
-bandwidth
2448 devices
, so
this slowdown can help the device and guest to keep
2449 up and not lose events
in case events are arriving
in bulk
.
2450 Possible causes
for the latter are flaky network connections
, or
2451 scripts
for automated testing
.
2453 ``audiodev
=audiodev``
2454 Use the specified audiodev when the VNC client requests audio
2455 transmission
. When not
using an
-audiodev argument
, this option
2456 must be omitted
, otherwise is must be present and specify a
2459 ``power
-control
=on|off``
2460 Permit the remote client to issue shutdown
, reboot or reset power
2464 ARCHHEADING(, QEMU_ARCH_I386
)
2466 ARCHHEADING(i386 target only
:, QEMU_ARCH_I386
)
2468 DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack
,
2469 "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n",
2473 Use it when installing Windows
2000 to avoid a disk full bug
. After
2474 Windows
2000 is installed
, you no longer need
this option (this
2475 option slows down the IDE transfers
).
2478 DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk
,
2479 "-no-fd-bootchk disable boot signature checking for floppy disks\n",
2483 Disable boot signature checking
for floppy disks
in BIOS
. May be
2484 needed to boot from old floppy disks
.
2487 DEF("no-acpi", 0, QEMU_OPTION_no_acpi
,
2488 "-no-acpi disable ACPI\n", QEMU_ARCH_I386 | QEMU_ARCH_ARM
)
2491 Disable
ACPI (Advanced Configuration and Power Interface
) support
.
2492 Use it
if your guest OS complains about ACPI
problems (PC target
2496 DEF("no-hpet", 0, QEMU_OPTION_no_hpet
,
2497 "-no-hpet disable HPET\n", QEMU_ARCH_I386
)
2500 Disable HPET support
.
2503 DEF("acpitable", HAS_ARG
, QEMU_OPTION_acpitable
,
2504 "-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"
2505 " ACPI table description\n", QEMU_ARCH_I386
)
2507 ``
-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
]...]``
2508 Add ACPI table with specified header fields and context from
2509 specified files
. For file
=, take whole ACPI table from the specified
2510 files
, including all ACPI
headers (possible overridden by other
2511 options
). For data
=, only data portion of the table is used
, all
2512 header information is specified
in the command line
. If a SLIC table
2513 is supplied to QEMU
, then the SLIC
's oem\_id and oem\_table\_id
2514 fields will override the same in the RSDT and the FADT (a.k.a.
2515 FACP), in order to ensure the field matches required by the
2516 Microsoft SLIC spec and the ACPI spec.
2519 DEF("smbios", HAS_ARG, QEMU_OPTION_smbios,
2520 "-smbios file=binary\n"
2521 " load SMBIOS entry from binary file\n"
2522 "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n"
2524 " specify SMBIOS type 0 fields\n"
2525 "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2526 " [,uuid=uuid][,sku=str][,family=str]\n"
2527 " specify SMBIOS type 1 fields\n"
2528 "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n"
2529 " [,asset=str][,location=str]\n"
2530 " specify SMBIOS type 2 fields\n"
2531 "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n"
2533 " specify SMBIOS type 3 fields\n"
2534 "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n"
2535 " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n"
2536 " [,processor-id=%d]\n"
2537 " specify SMBIOS type 4 fields\n"
2538 "-smbios type=11[,value=str][,path=filename]\n"
2539 " specify SMBIOS type 11 fields\n"
2540 "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n"
2541 " [,asset=str][,part=str][,speed=%d]\n"
2542 " specify SMBIOS type 17 fields\n"
2543 "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n"
2544 " specify SMBIOS type 41 fields\n",
2545 QEMU_ARCH_I386 | QEMU_ARCH_ARM)
2547 ``-smbios file=binary``
2548 Load SMBIOS entry from binary file.
2550 ``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]``
2551 Specify SMBIOS type 0 fields
2553 ``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]``
2554 Specify SMBIOS type 1 fields
2556 ``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]``
2557 Specify SMBIOS type 2 fields
2559 ``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]``
2560 Specify SMBIOS type 3 fields
2562 ``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-id=%d]``
2563 Specify SMBIOS type 4 fields
2565 ``-smbios type=11[,value=str][,path=filename]``
2566 Specify SMBIOS type 11 fields
2568 This argument can be repeated multiple times, and values are added in the order they are parsed.
2569 Applications intending to use OEM strings data are encouraged to use their application name as
2570 a prefix for the value string. This facilitates passing information for multiple applications
2573 The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax
2574 loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes.
2576 Both the ``value`` and ``path`` options can be repeated multiple times and will be added to
2577 the SMBIOS table in the order in which they appear.
2579 Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535
2580 bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the
2581 guest. Instead it should be used as a indicator to inform the guest where to locate the real
2582 data set, for example, by specifying the serial ID of a block device.
2584 An example passing three strings is
2588 -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\
2589 value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\
2590 path=/some/file/with/oemstringsdata.txt
2592 In the guest OS this is visible with the ``dmidecode`` command
2597 Handle 0x0E00, DMI type 11, 5 bytes
2599 String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/
2600 String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os
2601 String 3: myapp:some extra data
2604 ``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]``
2605 Specify SMBIOS type 17 fields
2607 ``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]``
2608 Specify SMBIOS type 41 fields
2610 This argument can be repeated multiple times. Its main use is to allow network interfaces be created
2611 as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface
2612 position on the PCI bus.
2614 Here is an example of use:
2618 -netdev user,id=internet \\
2619 -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\
2620 -smbios type=41,designation='Onboard LAN
',instance=1,kind=ethernet,pcidev=internet-dev
2622 In the guest OS, the device should then appear as ``eno1``:
2627 lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP>
2628 eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP>
2630 Currently, the PCI device has to be attached to the root bus.
2636 DEFHEADING(Network options:)
2638 DEF("netdev", HAS_ARG, QEMU_OPTION_netdev,
2640 "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n"
2641 " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n"
2642 " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n"
2643 " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n"
2644 " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]"
2646 "[,smb=dir[,smbserver=addr]]\n"
2648 " configure a user mode network backend with ID 'str
',\n"
2649 " its DHCP server and optional services\n"
2652 "-netdev tap,id=str,ifname=name\n"
2653 " configure a host TAP network backend with ID 'str
'\n"
2655 "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n"
2656 " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n"
2657 " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n"
2659 " configure a host TAP network backend with ID 'str
'\n"
2660 " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2661 " use network scripts 'file
' (default=" DEFAULT_NETWORK_SCRIPT ")\n"
2662 " to configure it and 'dfile
' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n"
2663 " to deconfigure it\n"
2664 " use '[down
]script
=no
' to disable script execution\n"
2665 " use network helper 'helper
' (default=" DEFAULT_BRIDGE_HELPER ") to\n"
2667 " use 'fd
=h
' to connect to an already opened TAP interface\n"
2668 " use 'fds
=x
:y
:...:z
' to connect to already opened multiqueue capable TAP interfaces\n"
2669 " use 'sndbuf
=nbytes
' to limit the size of the send buffer (the\n"
2670 " default is disabled 'sndbuf
=0' to enable flow control set 'sndbuf
=1048576')\n"
2671 " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n"
2672 " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n"
2673 " use vhost=on to enable experimental in kernel accelerator\n"
2674 " (only has effect for virtio guests which use MSIX)\n"
2675 " use vhostforce=on to force vhost on for non-MSIX virtio guests\n"
2676 " use 'vhostfd
=h
' to connect to an already opened vhost net device\n"
2677 " use 'vhostfds
=x
:y
:...:z to connect to multiple already opened vhost net devices
\n"
2678 " use
'queues=n' to specify the number of queues to be created
for multiqueue TAP
\n"
2679 " use
'poll-us=n' to specify the maximum number of microseconds that could be
\n"
2680 " spent on busy polling
for vhost net
\n"
2681 "-netdev bridge
,id
=str
[,br
=bridge
][,helper
=helper
]\n"
2682 " configure a host TAP network backend with ID
'str' that is
\n"
2683 " connected to a
bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n"
2684 " using the program
'helper (default=" DEFAULT_BRIDGE_HELPER ")\n"
2687 "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n"
2688 " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n"
2689 " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n"
2690 " [,rxcookie=rxcookie][,offset=offset]\n"
2691 " configure a network backend with ID 'str
' connected to\n"
2692 " an Ethernet over L2TPv3 pseudowire.\n"
2693 " Linux kernel 3.3+ as well as most routers can talk\n"
2694 " L2TPv3. This transport allows connecting a VM to a VM,\n"
2695 " VM to a router and even VM to Host. It is a nearly-universal\n"
2696 " standard (RFC3931). Note - this implementation uses static\n"
2697 " pre-configured tunnels (same as the Linux kernel).\n"
2698 " use 'src
=' to specify source address\n"
2699 " use 'dst
=' to specify destination address\n"
2700 " use 'udp
=on
' to specify udp encapsulation\n"
2701 " use 'srcport
=' to specify source udp port\n"
2702 " use 'dstport
=' to specify destination udp port\n"
2703 " use 'ipv6
=on
' to force v6\n"
2704 " L2TPv3 uses cookies to prevent misconfiguration as\n"
2705 " well as a weak security measure\n"
2706 " use 'rxcookie
=0x012345678' to specify a rxcookie\n"
2707 " use 'txcookie
=0x012345678' to specify a txcookie\n"
2708 " use 'cookie64
=on
' to set cookie size to 64 bit, otherwise 32\n"
2709 " use 'counter
=off
' to force a 'cut
-down
' L2TPv3 with no counter\n"
2710 " use 'pincounter
=on
' to work around broken counter handling in peer\n"
2711 " use 'offset
=X
' to add an extra offset between header and data\n"
2713 "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n"
2714 " configure a network backend to connect to another network\n"
2715 " using a socket connection\n"
2716 "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n"
2717 " configure a network backend to connect to a multicast maddr and port\n"
2718 " use 'localaddr
=addr
' to specify the host address to send packets from\n"
2719 "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n"
2720 " configure a network backend to connect to another network\n"
2721 " using an UDP tunnel\n"
2723 "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n"
2724 " configure a network backend to connect to port 'n
' of a vde switch\n"
2725 " running on host and listening for incoming connections on 'socketpath
'.\n"
2726 " Use group 'groupname
' and mode 'octalmode
' to change default\n"
2727 " ownership and permissions for communication port.\n"
2729 #ifdef CONFIG_NETMAP
2730 "-netdev netmap,id=str,ifname=name[,devname=nmname]\n"
2731 " attach to the existing netmap-enabled network interface 'name
', or to a\n"
2732 " VALE port (created on the fly) called 'name
' ('nmname
' is name of the \n"
2733 " netmap device, defaults to '/dev
/netmap
')\n"
2736 "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n"
2737 " configure a vhost-user network, backed by a chardev 'dev
'\n"
2740 "-netdev vhost-vdpa,id=str,vhostdev=/path/to/dev\n"
2741 " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n"
2744 "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n"
2745 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n"
2746 " configure a vmnet network backend in host mode with ID 'str
',\n"
2747 " isolate this interface from others with 'isolated
',\n"
2748 " configure the address range and choose a subnet mask,\n"
2749 " specify network UUID 'uuid
' to disable DHCP and interact with\n"
2750 " vmnet-host interfaces within this isolated network\n"
2751 "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n"
2752 " [,start-address=addr,end-address=addr,subnet-mask=mask]\n"
2753 " configure a vmnet network backend in shared mode with ID 'str
',\n"
2754 " configure the address range and choose a subnet mask,\n"
2755 " set IPv6 ULA prefix (of length 64) to use for internal network,\n"
2756 " isolate this interface from others with 'isolated
'\n"
2757 "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n"
2758 " configure a vmnet network backend in bridged mode with ID 'str
',\n"
2759 " use 'ifname
=name
' to select a physical network interface to be bridged,\n"
2760 " isolate this interface from others with 'isolated
'\n"
2762 "-netdev hubport,id=str,hubid=n[,netdev=nd]\n"
2763 " configure a hub port on the hub with ID 'n
'\n", QEMU_ARCH_ALL)
2764 DEF("nic", HAS_ARG, QEMU_OPTION_nic,
2775 #ifdef CONFIG_NETMAP
2782 "vmnet-host|vmnet-shared|vmnet-bridged|"
2784 "socket][,option][,...][mac=macaddr]\n"
2785 " initialize an on-board / default host NIC (using MAC address\n"
2786 " macaddr) and connect it to the given host network backend\n"
2787 "-nic none use it alone to have zero network devices (the default is to\n"
2788 " provided a 'user
' network connection)\n",
2790 DEF("net", HAS_ARG, QEMU_OPTION_net,
2791 "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n"
2792 " configure or create an on-board (or machine default) NIC and\n"
2793 " connect it to hub 0 (please use -nic unless you need a hub)\n"
2803 #ifdef CONFIG_NETMAP
2807 "vmnet-host|vmnet-shared|vmnet-bridged|"
2809 "socket][,option][,option][,...]\n"
2810 " old way to initialize a host network interface\n"
2811 " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL)
2813 ``-nic [tap|bridge|user|l2tpv3|vde|netmap|vhost-user|socket][,...][,mac=macaddr][,model=mn]``
2814 This option is a shortcut for configuring both the on-board
2815 (default) guest NIC hardware and the host network backend in one go.
2816 The host backend options are the same as with the corresponding
2817 ``-netdev`` options below. The guest NIC model can be set with
2818 ``model=modelname``. Use ``model=help`` to list the available device
2819 types. The hardware MAC address can be set with ``mac=macaddr``.
2821 The following two example do exactly the same, to show how ``-nic``
2822 can be used to shorten the command line length:
2826 |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32
2827 |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32
2830 Indicate that no network devices should be configured. It is used to
2831 override the default configuration (default NIC with "user" host
2832 network backend) which is activated if no other networking options
2835 ``-netdev user,id=id[,option][,option][,...]``
2836 Configure user mode host network backend which requires no
2837 administrator privilege to run. Valid options are:
2840 Assign symbolic name for use in monitor commands.
2842 ``ipv4=on|off and ipv6=on|off``
2843 Specify that either IPv4 or IPv6 must be enabled. If neither is
2844 specified both protocols are enabled.
2847 Set IP network address the guest will see. Optionally specify
2848 the netmask, either in the form a.b.c.d or as number of valid
2849 top-most bits. Default is 10.0.2.0/24.
2852 Specify the guest-visible address of the host. Default is the
2853 2nd IP in the guest network, i.e. x.x.x.2.
2855 ``ipv6-net=addr[/int]``
2856 Set IPv6 network address the guest will see (default is
2857 fec0::/64). The network prefix is given in the usual hexadecimal
2858 IPv6 address notation. The prefix size is optional, and is given
2859 as the number of valid top-most bits (default is 64).
2862 Specify the guest-visible IPv6 address of the host. Default is
2863 the 2nd IPv6 in the guest network, i.e. xxxx::2.
2866 If this option is enabled, the guest will be isolated, i.e. it
2867 will not be able to contact the host and no guest IP packets
2868 will be routed over the host to the outside. This option does
2869 not affect any explicitly set forwarding rules.
2872 Specifies the client hostname reported by the built-in DHCP
2876 Specify the first of the 16 IPs the built-in DHCP server can
2877 assign. Default is the 15th to 31st IP in the guest network,
2878 i.e. x.x.x.15 to x.x.x.31.
2881 Specify the guest-visible address of the virtual nameserver. The
2882 address must be different from the host address. Default is the
2883 3rd IP in the guest network, i.e. x.x.x.3.
2886 Specify the guest-visible address of the IPv6 virtual
2887 nameserver. The address must be different from the host address.
2888 Default is the 3rd IP in the guest network, i.e. xxxx::3.
2890 ``dnssearch=domain``
2891 Provides an entry for the domain-search list sent by the
2892 built-in DHCP server. More than one domain suffix can be
2893 transmitted by specifying this option multiple times. If
2894 supported, this will cause the guest to automatically try to
2895 append the given domain suffix(es) in case a domain name can not
2902 |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org
2904 ``domainname=domain``
2905 Specifies the client domain name reported by the built-in DHCP
2909 When using the user mode network stack, activate a built-in TFTP
2910 server. The files in dir will be exposed as the root of a TFTP
2911 server. The TFTP client on the guest must be configured in
2912 binary mode (use the command ``bin`` of the Unix TFTP client).
2914 ``tftp-server-name=name``
2915 In BOOTP reply, broadcast name as the "TFTP server name"
2916 (RFC2132 option 66). This can be used to advise the guest to
2917 load boot files or configurations from a different server than
2921 When using the user mode network stack, broadcast file as the
2922 BOOTP filename. In conjunction with ``tftp``, this can be used
2923 to network boot a guest from a local directory.
2925 Example (using pxelinux):
2929 |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\
2930 -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0
2932 ``smb=dir[,smbserver=addr]``
2933 When using the user mode network stack, activate a built-in SMB
2934 server so that Windows OSes can access to the host files in
2935 ``dir`` transparently. The IP address of the SMB server can be
2936 set to addr. By default the 4th IP in the guest network is used,
2939 In the guest Windows OS, the line:
2945 must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows
2946 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows
2949 Then ``dir`` can be accessed in ``\\smbserver\qemu``.
2951 Note that a SAMBA server must be installed on the host OS.
2953 ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport``
2954 Redirect incoming TCP or UDP connections to the host port
2955 hostport to the guest IP address guestaddr on guest port
2956 guestport. If guestaddr is not specified, its value is x.x.x.15
2957 (default first address given by the built-in DHCP server). By
2958 specifying hostaddr, the rule can be bound to a specific host
2959 interface. If no connection type is set, TCP is used. This
2960 option can be given multiple times.
2962 For example, to redirect host X11 connection from screen 1 to
2963 guest screen 0, use the following:
2968 |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000
2969 # this host xterm should open in the guest X11 server
2972 To redirect telnet connections from host port 5555 to telnet
2973 port on the guest, use the following:
2978 |qemu_system| -nic user,hostfwd=tcp::5555-:23
2979 telnet localhost 5555
2981 Then when you use on the host ``telnet localhost 5555``, you
2982 connect to the guest telnet server.
2984 ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command``
2985 Forward guest TCP connections to the IP address server on port
2986 port to the character device dev or to a program executed by
2987 cmd:command which gets spawned for each connection. This option
2988 can be given multiple times.
2990 You can either use a chardev directly and have that one used
2991 throughout QEMU's lifetime
, like
in the following example
:
2995 # open
10.10.1.1:4321 on bootup
, connect
10.0.2.100:1234 to it whenever
2996 # the guest accesses it
2997 |qemu_system|
-nic user
,guestfwd
=tcp
:10.0.2.100:1234-tcp
:10.10.1.1:4321
2999 Or you can execute a command on every TCP connection established
3000 by the guest
, so that QEMU behaves similar to an inetd process
3001 for that virtual server
:
3005 # call
"netcat 10.10.1.1 4321" on every TCP connection to
10.0.2.100:1234
3006 # and connect the TCP stream to its stdin
/stdout
3007 |qemu_system|
-nic
'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321'
3009 ``
-netdev tap
,id
=id
[,fd
=h
][,ifname
=name
][,script
=file
][,downscript
=dfile
][,br
=bridge
][,helper
=helper
]``
3010 Configure a host TAP network backend with ID id
.
3012 Use the network script file to configure it and the network script
3013 dfile to deconfigure it
. If name is not provided
, the OS
3014 automatically provides one
. The
default network configure script is
3015 ``
/etc
/qemu
-ifup`` and the
default network deconfigure script is
3016 ``
/etc
/qemu
-ifdown``
. Use ``script
=no`` or ``downscript
=no`` to
3017 disable script execution
.
3019 If running QEMU as an unprivileged user
, use the network helper
3020 to configure the TAP
interface and attach it to the bridge
.
3021 The
default network helper executable is
3022 ``
/path
/to
/qemu
-bridge
-helper`` and the
default bridge device is
3025 ``fd``\
=h can be used to specify the handle of an already opened
3032 #launch a QEMU instance with the
default network script
3033 |qemu_system| linux
.img
-nic tap
3037 #launch a QEMU instance with two NICs
, each one connected
3039 |qemu_system| linux
.img
\\
3040 -netdev tap
,id
=nd0
,ifname
=tap0
-device e1000
,netdev
=nd0
\\
3041 -netdev tap
,id
=nd1
,ifname
=tap1
-device rtl8139
,netdev
=nd1
3045 #launch a QEMU instance with the
default network helper to
3046 #connect a TAP device to bridge br0
3047 |qemu_system| linux
.img
-device virtio
-net
-pci
,netdev
=n1
\\
3048 -netdev tap
,id
=n1
,"helper=/path/to/qemu-bridge-helper"
3050 ``
-netdev bridge
,id
=id
[,br
=bridge
][,helper
=helper
]``
3051 Connect a host TAP network
interface to a host bridge device
.
3053 Use the network helper helper to configure the TAP
interface and
3054 attach it to the bridge
. The
default network helper executable is
3055 ``
/path
/to
/qemu
-bridge
-helper`` and the
default bridge device is
3062 #launch a QEMU instance with the
default network helper to
3063 #connect a TAP device to bridge br0
3064 |qemu_system| linux
.img
-netdev bridge
,id
=n1
-device virtio
-net
,netdev
=n1
3068 #launch a QEMU instance with the
default network helper to
3069 #connect a TAP device to bridge qemubr0
3070 |qemu_system| linux
.img
-netdev bridge
,br
=qemubr0
,id
=n1
-device virtio
-net
,netdev
=n1
3072 ``
-netdev socket
,id
=id
[,fd
=h
][,listen
=[host
]:port
][,connect
=host
:port
]``
3073 This host network backend can be used to connect the guest
's network
3074 to another QEMU virtual machine using a TCP socket connection. If
3075 ``listen`` is specified, QEMU waits for incoming connections on port
3076 (host is optional). ``connect`` is used to connect to another QEMU
3077 instance using the ``listen`` option. ``fd``\ =h specifies an
3078 already opened TCP socket.
3084 # launch a first QEMU instance
3085 |qemu_system| linux.img \\
3086 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3087 -netdev socket,id=n1,listen=:1234
3088 # connect the network of this instance to the network of the first instance
3089 |qemu_system| linux.img \\
3090 -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\
3091 -netdev socket,id=n2,connect=127.0.0.1:1234
3093 ``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]``
3094 Configure a socket host network backend to share the guest's network
3095 traffic with another QEMU virtual machines
using a UDP multicast
3096 socket
, effectively making a bus
for every QEMU with same multicast
3097 address maddr and port
. NOTES
:
3099 1. Several QEMU can be running on different hosts and share same bus
3100 (assuming correct multicast setup
for these hosts
).
3102 2. mcast support is compatible with User Mode
Linux (argument
3103 ``ethN
=mcast``
), see http
://user-mode-linux.sf.net.
3105 3. Use ``fd
=h`` to specify an already opened UDP multicast socket
.
3111 # launch one QEMU instance
3112 |qemu_system| linux
.img
\\
3113 -device e1000
,netdev
=n1
,mac
=52:54:00:12:34:56 \\
3114 -netdev socket
,id
=n1
,mcast
=230.0.0.1:1234
3115 # launch another QEMU instance on same
"bus"
3116 |qemu_system| linux
.img
\\
3117 -device e1000
,netdev
=n2
,mac
=52:54:00:12:34:57 \\
3118 -netdev socket
,id
=n2
,mcast
=230.0.0.1:1234
3119 # launch yet another QEMU instance on same
"bus"
3120 |qemu_system| linux
.img
\\
3121 -device e1000
,netdev
=n3
,mac
=52:54:00:12:34:58 \\
3122 -netdev socket
,id
=n3
,mcast
=230.0.0.1:1234
3124 Example (User Mode Linux compat
.):
3128 # launch QEMU
instance (note mcast address selected is UML
's default)
3129 |qemu_system| linux.img \\
3130 -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\
3131 -netdev socket,id=n1,mcast=239.192.168.1:1102
3133 /path/to/linux ubd0=/path/to/root_fs eth0=mcast
3135 Example (send packets from host's
1.2.3.4):
3139 |qemu_system| linux
.img
\\
3140 -device e1000
,netdev
=n1
,mac
=52:54:00:12:34:56 \\
3141 -netdev socket
,id
=n1
,mcast
=239.192.168.1:1102,localaddr
=1.2.3.4
3143 ``
-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
]``
3144 Configure a L2TPv3 pseudowire host network backend
. L2TPv3 (RFC3931
)
3145 is a popular protocol to transport
Ethernet (and other Layer
2) data
3146 frames between two systems
. It is present
in routers
, firewalls and
3147 the Linux
kernel (from version
3.3 onwards
).
3149 This transport allows a VM to communicate to another VM
, router or
3153 source
address (mandatory
)
3156 destination
address (mandatory
)
3159 select udp
encapsulation (default is ip
).
3165 destination udp port
.
3168 force v6
, otherwise defaults to v4
.
3170 ``rxcookie
=rxcookie``
; \ ``txcookie
=txcookie``
3171 Cookies are a weak form of security
in the l2tpv3 specification
.
3172 Their
function is mostly to prevent misconfiguration
. By
default
3176 Set cookie size to
64 bit instead of the
default 32
3179 Force a
'cut-down' L2TPv3 with no counter as
in
3180 draft
-mkonstan
-l2tpext
-keyed
-ipv6
-tunnel
-00
3183 Work around broken counter handling
in peer
. This may also help
3184 on networks which have packet reorder
.
3187 Add an extra offset between header and data
3189 For example
, to attach a VM running on host
4.3.2.1 via L2TPv3 to
3190 the bridge br
-lan on the remote Linux host
1.2.3.4:
3194 # Setup tunnel on linux host
using raw ip as encapsulation
3196 ip l2tp add tunnel remote
4.3.2.1 local
1.2.3.4 tunnel_id
1 peer_tunnel_id
1 \\
3197 encap udp udp_sport
16384 udp_dport
16384
3198 ip l2tp add session tunnel_id
1 name vmtunnel0 session_id
\\
3199 0xFFFFFFFF peer_session_id
0xFFFFFFFF
3200 ifconfig vmtunnel0 mtu
1500
3201 ifconfig vmtunnel0 up
3202 brctl addif br
-lan vmtunnel0
3206 # launch QEMU instance
- if your network has reorder or is very lossy add
,pincounter
3208 |qemu_system| linux
.img
-device e1000
,netdev
=n1
\\
3209 -netdev l2tpv3
,id
=n1
,src
=4.2.3.1,dst
=1.2.3.4,udp
,srcport
=16384,dstport
=16384,rxsession
=0xffffffff,txsession
=0xffffffff,counter
3211 ``
-netdev vde
,id
=id
[,sock
=socketpath
][,port
=n
][,group
=groupname
][,mode
=octalmode
]``
3212 Configure VDE backend to connect to PORT n of a vde
switch running
3213 on host and listening
for incoming connections on socketpath
. Use
3214 GROUP groupname and MODE octalmode to change
default ownership and
3215 permissions
for communication port
. This option is only available
if
3216 QEMU has been compiled with vde support enabled
.
3223 vde_switch
-F
-sock
/tmp
/myswitch
3224 # launch QEMU instance
3225 |qemu_system| linux
.img
-nic vde
,sock
=/tmp
/myswitch
3227 ``
-netdev vhost
-user
,chardev
=id
[,vhostforce
=on|off
][,queues
=n
]``
3228 Establish a vhost
-user netdev
, backed by a chardev id
. The chardev
3229 should be a unix domain socket backed one
. The vhost
-user uses a
3230 specifically defined protocol to pass vhost ioctl replacement
3231 messages to an application on the other end of the socket
. On
3232 non
-MSIX guests
, the feature can be forced with vhostforce
. Use
3233 'queues=n' to specify the number of queues to be created
for
3234 multiqueue vhost
-user
.
3240 qemu
-m
512 -object memory
-backend
-file
,id
=mem
,size
=512M
,mem
-path
=/hugetlbfs
,share
=on \
3241 -numa node
,memdev
=mem \
3242 -chardev socket
,id
=chr0
,path
=/path
/to
/socket \
3243 -netdev type
=vhost
-user
,id
=net0
,chardev
=chr0 \
3244 -device virtio
-net
-pci
,netdev
=net0
3246 ``
-netdev vhost
-vdpa
,vhostdev
=/path
/to
/dev``
3247 Establish a vhost
-vdpa netdev
.
3249 vDPA device is a device that uses a datapath which complies with
3250 the virtio specifications with a vendor specific control path
.
3251 vDPA devices can be both physically located on the hardware or
3252 emulated by software
.
3254 ``
-netdev hubport
,id
=id
,hubid
=hubid
[,netdev
=nd
]``
3255 Create a hub port on the emulated hub with ID hubid
.
3257 The hubport netdev lets you connect a NIC to a QEMU emulated hub
3258 instead of a single netdev
. Alternatively
, you can also connect the
3259 hubport to another netdev with ID nd by
using the ``netdev
=nd``
3262 ``
-net nic
[,netdev
=nd
][,macaddr
=mac
][,model
=type
] [,name
=name
][,addr
=addr
][,vectors
=v
]``
3263 Legacy option to configure or create an on
-board (or machine
3264 default) Network Interface
Card(NIC
) and connect it either to the
3265 emulated hub with ID
0 (i
.e
. the
default hub
), or to the netdev nd
.
3266 If model is omitted
, then the
default NIC model associated with the
3267 machine type is used
. Note that the
default NIC model may change
in
3268 future QEMU releases
, so it is highly recommended to always specify
3269 a model
. Optionally
, the MAC address can be changed to mac
, the
3270 device address set to
addr (PCI cards only
), and a name can be
3271 assigned
for use
in monitor commands
. Optionally
, for PCI cards
, you
3272 can specify the number v of MSI
-X vectors that the card should have
;
3273 this option currently only affects virtio cards
; set v
= 0 to
3274 disable MSI
-X
. If no ``
-net`` option is specified
, a single NIC is
3275 created
. QEMU can emulate several different models of network card
.
3276 Use ``
-net nic
,model
=help``
for a list of available devices
for your
3279 ``
-net user|tap|bridge|socket|l2tpv3|vde
[,...][,name
=name
]``
3280 Configure a host network
backend (with the options corresponding to
3281 the same ``
-netdev`` option
) and connect it to the emulated hub
0
3282 (the
default hub
). Use name to specify the name of the hub port
.
3287 DEFHEADING(Character device options
:)
3289 DEF("chardev", HAS_ARG
, QEMU_OPTION_chardev
,
3291 "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3292 "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n"
3293 " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n"
3294 " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n"
3295 "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n"
3296 " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n"
3297 "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n"
3298 " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n"
3299 " [,logfile=PATH][,logappend=on|off]\n"
3300 "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3301 "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n"
3302 " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3303 "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n"
3304 "-chardev file,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3305 "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3307 "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3308 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3310 "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3311 "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n"
3313 #ifdef CONFIG_BRLAPI
3314 "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3316 #
if defined(__linux__
) ||
defined(__sun__
) ||
defined(__FreeBSD__
) \
3317 ||
defined(__NetBSD__
) ||
defined(__OpenBSD__
) ||
defined(__DragonFly__
)
3318 "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3319 "-chardev tty,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3321 #
if defined(__linux__
) ||
defined(__FreeBSD__
) ||
defined(__DragonFly__
)
3322 "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3323 "-chardev parport,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n"
3325 #
if defined(CONFIG_SPICE
)
3326 "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3327 "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n"
3333 The general form of a character device option is
:
3335 ``
-chardev backend
,id
=id
[,mux
=on|off
][,options
]``
3336 Backend is one of
: ``
null``
, ``socket``
, ``udp``
, ``msmouse``
,
3337 ``vc``
, ``ringbuf``
, ``file``
, ``pipe``
, ``console``
, ``serial``
,
3338 ``pty``
, ``stdio``
, ``braille``
, ``tty``
, ``parallel``
, ``parport``
,
3339 ``spicevmc``
, ``spiceport``
. The specific backend will determine the
3342 Use ``
-chardev help`` to print all available chardev backend types
.
3344 All devices must have an id
, which can be any string up to
127
3345 characters long
. It is used to uniquely identify
this device
in
3346 other command line directives
.
3348 A character device may be used
in multiplexing mode by multiple
3349 front
-ends
. Specify ``mux
=on`` to enable
this mode
. A multiplexer is
3350 a
"1:N" device
, and
here the
"1" end is your specified chardev
3351 backend
, and the
"N" end is the various parts of QEMU that can talk
3352 to a chardev
. If you create a chardev with ``id
=myid`` and
3353 ``mux
=on``
, QEMU will create a multiplexer with your specified ID
,
3354 and you can then configure multiple front ends to use that chardev
3355 ID
for their input
/output
. Up to four different front ends can be
3356 connected to a single multiplexed chardev
. (Without multiplexing
3357 enabled
, a chardev can only be used by a single front end
.) For
3358 instance you could use
this to allow a single stdio chardev to be
3359 used by two serial ports and the QEMU monitor
:
3363 -chardev stdio
,mux
=on
,id
=char0 \
3364 -mon chardev
=char0
,mode
=readline \
3365 -serial chardev
:char0 \
3366 -serial chardev
:char0
3368 You can have more than one multiplexer
in a system configuration
;
3369 for instance you could have a TCP port multiplexed between UART
0
3370 and UART
1, and stdio multiplexed between the QEMU monitor and a
3375 -chardev stdio
,mux
=on
,id
=char0 \
3376 -mon chardev
=char0
,mode
=readline \
3377 -parallel chardev
:char0 \
3378 -chardev tcp
,...,mux
=on
,id
=char1 \
3379 -serial chardev
:char1 \
3380 -serial chardev
:char1
3382 When you
're using a multiplexed character device, some escape
3383 sequences are interpreted in the input. See the chapter about
3384 :ref:`keys in the character backend multiplexer` in the
3385 System Emulation Users Guide for more details.
3387 Note that some other command line options may implicitly create
3388 multiplexed character backends; for instance ``-serial mon:stdio``
3389 creates a multiplexed stdio backend connected to the serial port and
3390 the QEMU monitor, and ``-nographic`` also multiplexes the console
3391 and the monitor to stdio.
3393 There is currently no support for multiplexing in the other
3394 direction (where a single QEMU front end takes input and output from
3397 Every backend supports the ``logfile`` option, which supplies the
3398 path to a file to record all data transmitted via the backend. The
3399 ``logappend`` option controls whether the log file will be truncated
3400 or appended to when opened.
3402 The available backends are:
3404 ``-chardev null,id=id``
3405 A void device. This device will not emit any data, and will drop any
3406 data it receives. The null backend does not take any options.
3408 ``-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]``
3409 Create a two-way stream socket, which can be either a TCP or a unix
3410 socket. A unix socket will be created if ``path`` is specified.
3411 Behaviour is undefined if TCP options are specified for a unix
3414 ``server=on|off`` specifies that the socket shall be a listening socket.
3416 ``wait=on|off`` specifies that QEMU should not block waiting for a client
3417 to connect to a listening socket.
3419 ``telnet=on|off`` specifies that traffic on the socket should interpret
3420 telnet escape sequences.
3422 ``websocket=on|off`` specifies that the socket uses WebSocket protocol for
3425 ``reconnect`` sets the timeout for reconnecting on non-server
3426 sockets when the remote end goes away. qemu will delay this many
3427 seconds and then attempt to reconnect. Zero disables reconnecting,
3430 ``tls-creds`` requests enablement of the TLS protocol for
3431 encryption, and specifies the id of the TLS credentials to use for
3432 the handshake. The credentials must be previously created with the
3433 ``-object tls-creds`` argument.
3435 ``tls-auth`` provides the ID of the QAuthZ authorization object
3436 against which the client's x509 distinguished name will be
3437 validated
. This object is only resolved at time of use
, so can be
3438 deleted and recreated on the fly
while the chardev server is active
.
3439 If missing
, it will
default to denying access
.
3441 TCP and unix socket options are given below
:
3443 ``TCP options
: port
=port
[,host
=host
][,to
=to
][,ipv4
=on|off
][,ipv6
=on|off
][,nodelay
=on|off
]``
3444 ``host``
for a listening socket specifies the local address to
3445 be bound
. For a connecting socket species the remote host to
3446 connect to
. ``host`` is optional
for listening sockets
. If not
3447 specified it defaults to ``
0.0.0.0``
.
3449 ``port``
for a listening socket specifies the local port to be
3450 bound
. For a connecting socket specifies the port on the remote
3451 host to connect to
. ``port`` can be given as either a port
3452 number or a service name
. ``port`` is required
.
3454 ``to`` is only relevant to listening sockets
. If it is
3455 specified
, and ``port`` cannot be bound
, QEMU will attempt to
3456 bind to subsequent ports up to and including ``to`` until it
3457 succeeds
. ``to`` must be specified as a port number
.
3459 ``ipv4
=on|off`` and ``ipv6
=on|off`` specify that either IPv4
3460 or IPv6 must be used
. If neither is specified the socket may
3461 use either protocol
.
3463 ``nodelay
=on|off`` disables the Nagle algorithm
.
3465 ``unix options
: path
=path
[,abstract
=on|off
][,tight
=on|off
]``
3466 ``path`` specifies the local path of the unix socket
. ``path``
3468 ``abstract
=on|off`` specifies the use of the abstract socket namespace
,
3469 rather than the filesystem
. Optional
, defaults to
false.
3470 ``tight
=on|off`` sets the socket length of abstract sockets to their minimum
,
3471 rather than the full sun_path length
. Optional
, defaults to
true.
3473 ``
-chardev udp
,id
=id
[,host
=host
],port
=port
[,localaddr
=localaddr
][,localport
=localport
][,ipv4
=on|off
][,ipv6
=on|off
]``
3474 Sends all traffic from the guest to a remote host over UDP
.
3476 ``host`` specifies the remote host to connect to
. If not specified
3477 it defaults to ``localhost``
.
3479 ``port`` specifies the port on the remote host to connect to
.
3480 ``port`` is required
.
3482 ``localaddr`` specifies the local address to bind to
. If not
3483 specified it defaults to ``
0.0.0.0``
.
3485 ``localport`` specifies the local port to bind to
. If not specified
3486 any available local port will be used
.
3488 ``ipv4
=on|off`` and ``ipv6
=on|off`` specify that either IPv4 or IPv6 must be used
.
3489 If neither is specified the device may use either protocol
.
3491 ``
-chardev msmouse
,id
=id``
3492 Forward QEMU
's emulated msmouse events to the guest. ``msmouse``
3493 does not take any options.
3495 ``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]``
3496 Connect to a QEMU text console. ``vc`` may optionally be given a
3499 ``width`` and ``height`` specify the width and height respectively
3500 of the console, in pixels.
3502 ``cols`` and ``rows`` specify that the console be sized to fit a
3503 text console with the given dimensions.
3505 ``-chardev ringbuf,id=id[,size=size]``
3506 Create a ring buffer with fixed size ``size``. size must be a power
3507 of two and defaults to ``64K``.
3509 ``-chardev file,id=id,path=path``
3510 Log all traffic received from the guest to a file.
3512 ``path`` specifies the path of the file to be opened. This file will
3513 be created if it does not already exist, and overwritten if it does.
3514 ``path`` is required.
3516 ``-chardev pipe,id=id,path=path``
3517 Create a two-way connection to the guest. The behaviour differs
3518 slightly between Windows hosts and other hosts:
3520 On Windows, a single duplex pipe will be created at
3523 On other hosts, 2 pipes will be created called ``path.in`` and
3524 ``path.out``. Data written to ``path.in`` will be received by the
3525 guest. Data written by the guest can be read from ``path.out``. QEMU
3526 will not create these fifos, and requires them to be present.
3528 ``path`` forms part of the pipe path as described above. ``path`` is
3531 ``-chardev console,id=id``
3532 Send traffic from the guest to QEMU's standard output
. ``console``
3533 does not take any options
.
3535 ``console`` is only available on Windows hosts
.
3537 ``
-chardev serial
,id
=id
,path
=path``
3538 Send traffic from the guest to a serial device on the host
.
3540 On Unix hosts serial will actually accept any tty device
, not only
3543 ``path`` specifies the name of the serial device to open
.
3545 ``
-chardev pty
,id
=id``
3546 Create a
new pseudo
-terminal on the host and connect to it
. ``pty``
3547 does not take any options
.
3549 ``pty`` is not available on Windows hosts
.
3551 ``
-chardev stdio
,id
=id
[,signal
=on|off
]``
3552 Connect to standard input and standard output of the QEMU process
.
3554 ``signal`` controls
if signals are enabled on the terminal
, that
3555 includes exiting QEMU with the key sequence Control
-c
. This option
3556 is enabled by
default, use ``signal
=off`` to disable it
.
3558 ``
-chardev braille
,id
=id``
3559 Connect to a local BrlAPI server
. ``braille`` does not take any
3562 ``
-chardev tty
,id
=id
,path
=path``
3563 ``tty`` is only available on Linux
, Sun
, FreeBSD
, NetBSD
, OpenBSD
3564 and DragonFlyBSD hosts
. It is an alias
for ``serial``
.
3566 ``path`` specifies the path to the tty
. ``path`` is required
.
3568 ``
-chardev parallel
,id
=id
,path
=path``
3570 ``
-chardev parport
,id
=id
,path
=path``
3571 ``parallel`` is only available on Linux
, FreeBSD and DragonFlyBSD
3574 Connect to a local parallel port
.
3576 ``path`` specifies the path to the parallel port device
. ``path`` is
3579 ``
-chardev spicevmc
,id
=id
,debug
=debug
,name
=name``
3580 ``spicevmc`` is only available when spice support is built
in.
3582 ``debug`` debug level
for spicevmc
3584 ``name`` name of spice channel to connect to
3586 Connect to a spice virtual machine channel
, such as vdiport
.
3588 ``
-chardev spiceport
,id
=id
,debug
=debug
,name
=name``
3589 ``spiceport`` is only available when spice support is built
in.
3591 ``debug`` debug level
for spicevmc
3593 ``name`` name of spice port to connect to
3595 Connect to a spice port
, allowing a Spice client to handle the
3596 traffic identified by a
name (preferably a fqdn
).
3602 DEFHEADING(TPM device options
:)
3604 DEF("tpmdev", HAS_ARG
, QEMU_OPTION_tpmdev
, \
3605 "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n"
3606 " use path to provide path to a character device; default is /dev/tpm0\n"
3607 " use cancel-path to provide path to TPM's cancel sysfs entry; if\n"
3608 " not provided it will be searched for in /sys/class/misc/tpm?/device\n"
3609 "-tpmdev emulator,id=id,chardev=dev\n"
3610 " configure the TPM device using chardev backend\n",
3613 The general form of a TPM device option is
:
3615 ``
-tpmdev backend
,id
=id
[,options
]``
3616 The specific backend type will determine the applicable options
. The
3617 ``
-tpmdev`` option creates the TPM backend and requires a
3618 ``
-device`` option that specifies the TPM frontend
interface model
.
3620 Use ``
-tpmdev help`` to print all available TPM backend types
.
3622 The available backends are
:
3624 ``
-tpmdev passthrough
,id
=id
,path
=path
,cancel
-path
=cancel
-path``
3625 (Linux
-host only
) Enable access to the host
's TPM using the
3628 ``path`` specifies the path to the host's TPM device
, i
.e
., on a
3629 Linux host
this would be ``
/dev
/tpm0``
. ``path`` is optional and by
3630 default ``
/dev
/tpm0`` is used
.
3632 ``cancel
-path`` specifies the path to the host TPM device
's sysfs
3633 entry allowing for cancellation of an ongoing TPM command.
3634 ``cancel-path`` is optional and by default QEMU will search for the
3637 Some notes about using the host's TPM with the passthrough driver
:
3639 The TPM device accessed by the passthrough driver must not be used
3640 by any other application on the host
.
3642 Since the host
's firmware (BIOS/UEFI) has already initialized the
3643 TPM, the VM's
firmware (BIOS
/UEFI
) will not be able to initialize
3644 the TPM again and may therefore not show a TPM
-specific menu that
3645 would otherwise allow the user to configure the TPM
, e
.g
., allow the
3646 user to enable
/disable or activate
/deactivate the TPM
. Further
, if
3647 TPM ownership is released from within a VM then the host
's TPM will
3648 get disabled and deactivated. To enable and activate the TPM again
3649 afterwards, the host has to be rebooted and the user is required to
3650 enter the firmware's menu to enable and activate the TPM
. If the TPM
3651 is left disabled and
/or deactivated most TPM commands will fail
.
3653 To create a passthrough TPM use the following two options
:
3657 -tpmdev passthrough
,id
=tpm0
-device tpm
-tis
,tpmdev
=tpm0
3659 Note that the ``
-tpmdev`` id is ``tpm0`` and is referenced by
3660 ``tpmdev
=tpm0``
in the device option
.
3662 ``
-tpmdev emulator
,id
=id
,chardev
=dev``
3663 (Linux
-host only
) Enable access to a TPM emulator
using Unix domain
3664 socket based chardev backend
.
3666 ``chardev`` specifies the unique ID of a character device backend
3667 that provides connection to the software TPM server
.
3669 To create a TPM emulator backend device with chardev socket backend
:
3673 -chardev socket
,id
=chrtpm
,path
=/tmp
/swtpm
-sock
-tpmdev emulator
,id
=tpm0
,chardev
=chrtpm
-device tpm
-tis
,tpmdev
=tpm0
3680 DEFHEADING(Boot Image or Kernel specific
:)
3682 There are broadly
4 ways you can boot a system with QEMU
.
3684 - specify a firmware and let it control finding a kernel
3685 - specify a firmware and pass a hint to the kernel to boot
3686 - direct kernel image boot
3687 - manually load files into the guest
's address space
3689 The third method is useful for quickly testing kernels but as there is
3690 no firmware to pass configuration information to the kernel the
3691 hardware must either be probeable, the kernel built for the exact
3692 configuration or passed some configuration data (e.g. a DTB blob)
3693 which tells the kernel what drivers it needs. This exact details are
3694 often hardware specific.
3696 The final method is the most generic way of loading images into the
3697 guest address space and used mostly for ``bare metal`` type
3698 development where the reset vectors of the processor are taken into
3705 For x86 machines and some other architectures ``-bios`` will generally
3706 do the right thing with whatever it is given. For other machines the
3707 more strict ``-pflash`` option needs an image that is sized for the
3708 flash device for the given machine type.
3710 Please see the :ref:`system-targets-ref` section of the manual for
3711 more detailed documentation.
3715 DEF("bios", HAS_ARG, QEMU_OPTION_bios, \
3716 "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL)
3719 Set the filename for the BIOS.
3722 DEF("pflash", HAS_ARG, QEMU_OPTION_pflash,
3723 "-pflash file use 'file
' as a parallel flash image\n", QEMU_ARCH_ALL)
3726 Use file as a parallel flash image.
3731 The kernel options were designed to work with Linux kernels although
3732 other things (like hypervisors) can be packaged up as a kernel
3733 executable image. The exact format of a executable image is usually
3734 architecture specific.
3736 The way in which the kernel is started (what address it is loaded at,
3737 what if any information is passed to it via CPU registers, the state
3738 of the hardware when it is started, and so on) is also architecture
3739 specific. Typically it follows the specification laid down by the
3740 Linux kernel for how kernels for that architecture must be started.
3744 DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \
3745 "-kernel bzImage use 'bzImage
' as kernel image\n", QEMU_ARCH_ALL)
3748 Use bzImage as kernel image. The kernel can be either a Linux kernel
3749 or in multiboot format.
3752 DEF("append", HAS_ARG, QEMU_OPTION_append, \
3753 "-append cmdline use 'cmdline
' as kernel command line\n", QEMU_ARCH_ALL)
3756 Use cmdline as kernel command line
3759 DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \
3760 "-initrd file use 'file
' as initial ram disk\n", QEMU_ARCH_ALL)
3763 Use file as initial ram disk.
3765 ``-initrd "file1 arg=foo,file2"``
3766 This syntax is only available with multiboot.
3768 Use file1 and file2 as modules and pass arg=foo as parameter to the
3772 DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \
3773 "-dtb file use 'file
' as device tree image\n", QEMU_ARCH_ALL)
3776 Use file as a device tree binary (dtb) image and pass it to the
3782 Finally you can also manually load images directly into the address
3783 space of the guest. This is most useful for developers who already
3784 know the layout of their guest and take care to ensure something sane
3785 will happen when the reset vector executes.
3787 The generic loader can be invoked by using the loader device:
3789 ``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]``
3791 there is also the guest loader which operates in a similar way but
3792 tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where
3795 ``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]``
3801 DEFHEADING(Debug/Expert options:)
3803 DEF("compat", HAS_ARG, QEMU_OPTION_compat,
3804 "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n"
3805 " Policy for handling deprecated management interfaces\n"
3806 "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n"
3807 " Policy for handling unstable management interfaces\n",
3810 ``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]``
3811 Set policy for handling deprecated management interfaces (experimental):
3813 ``deprecated-input=accept`` (default)
3814 Accept deprecated commands and arguments
3815 ``deprecated-input=reject``
3816 Reject deprecated commands and arguments
3817 ``deprecated-input=crash``
3818 Crash on deprecated commands and arguments
3819 ``deprecated-output=accept`` (default)
3820 Emit deprecated command results and events
3821 ``deprecated-output=hide``
3822 Suppress deprecated command results and events
3824 Limitation: covers only syntactic aspects of QMP.
3826 ``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]``
3827 Set policy for handling unstable management interfaces (experimental):
3829 ``unstable-input=accept`` (default)
3830 Accept unstable commands and arguments
3831 ``unstable-input=reject``
3832 Reject unstable commands and arguments
3833 ``unstable-input=crash``
3834 Crash on unstable commands and arguments
3835 ``unstable-output=accept`` (default)
3836 Emit unstable command results and events
3837 ``unstable-output=hide``
3838 Suppress unstable command results and events
3840 Limitation: covers only syntactic aspects of QMP.
3843 DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg,
3844 "-fw_cfg [name=]<name>,file=<file>\n"
3845 " add named fw_cfg entry with contents from file\n"
3846 "-fw_cfg [name=]<name>,string=<str>\n"
3847 " add named fw_cfg entry with contents from string\n",
3850 ``-fw_cfg [name=]name,file=file``
3851 Add named fw\_cfg entry with contents from file file.
3853 ``-fw_cfg [name=]name,string=str``
3854 Add named fw\_cfg entry with contents from string str.
3856 The terminating NUL character of the contents of str will not be
3857 included as part of the fw\_cfg item data. To insert contents with
3858 embedded NUL characters, you have to use the file parameter.
3860 The fw\_cfg entries are passed by QEMU through to the guest.
3866 -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin
3868 creates an fw\_cfg entry named opt/com.mycompany/blob with contents
3869 from ./my\_blob.bin.
3872 DEF("serial", HAS_ARG, QEMU_OPTION_serial, \
3873 "-serial dev redirect the serial port to char device 'dev
'\n",
3877 Redirect the virtual serial port to host character device dev. The
3878 default device is ``vc`` in graphical mode and ``stdio`` in non
3881 This option can be used several times to simulate up to 4 serial
3884 Use ``-serial none`` to disable all serial ports.
3886 Available character devices are:
3889 Virtual console. Optionally, a width and height can be given in
3896 It is also possible to specify width or height in characters:
3903 [Linux only] Pseudo TTY (a new PTY is automatically allocated)
3906 No device is allocated.
3912 Use a named character device defined with the ``-chardev``
3916 [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial
3917 port parameters are set according to the emulated ones.
3920 [Linux only, parallel port only] Use host parallel port N.
3921 Currently SPP and EPP parallel port features can be used.
3924 Write output to filename. No character can be read.
3927 [Unix only] standard input/output
3933 [Windows only] Use host serial port n
3935 ``udp:[remote_host]:remote_port[@[src_ip]:src_port]``
3936 This implements UDP Net Console. When remote\_host or src\_ip
3937 are not specified they default to ``0.0.0.0``. When not using a
3938 specified src\_port a random port is automatically chosen.
3940 If you just want a simple readonly console you can use
3941 ``netcat`` or ``nc``, by starting QEMU with:
3942 ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time
3943 QEMU writes something to that port it will appear in the
3946 If you plan to send characters back via netconsole or you want
3947 to stop and start QEMU a lot of times, you should have QEMU use
3948 the same source port each time by using something like ``-serial
3949 udp::4555@:4556`` to QEMU. Another approach is to use a patched
3950 version of netcat which can listen to a TCP port and send and
3951 receive characters via udp. If you have a patched version of
3952 netcat which activates telnet remote echo and single char
3953 transfer, then you can use the following options to set up a
3954 netcat redirector to allow telnet on port 5555 to access the
3958 -serial udp::4555@:4556
3961 -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T
3966 ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]``
3967 The TCP Net Console has two modes of operation. It can send the
3968 serial I/O to a location or wait for a connection from a
3969 location. By default the TCP Net Console is sent to host at the
3970 port. If you use the ``server=on`` option QEMU will wait for a client
3971 socket application to connect to the port before continuing,
3972 unless the ``wait=on|off`` option was specified. The ``nodelay=on|off``
3973 option disables the Nagle buffering algorithm. The ``reconnect=on``
3974 option only applies if ``server=no`` is set, if the connection goes
3975 down it will attempt to reconnect at the given interval. If host
3976 is omitted, 0.0.0.0 is assumed. Only one TCP connection at a
3977 time is accepted. You can use ``telnet=on`` to connect to the
3978 corresponding character device.
3980 ``Example to send tcp console to 192.168.0.2 port 4444``
3981 -serial tcp:192.168.0.2:4444
3983 ``Example to listen and wait on port 4444 for connection``
3984 -serial tcp::4444,server=on
3986 ``Example to not wait and listen on ip 192.168.0.100 port 4444``
3987 -serial tcp:192.168.0.100:4444,server=on,wait=off
3989 ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]``
3990 The telnet protocol is used instead of raw tcp sockets. The
3991 options work the same as if you had specified ``-serial tcp``.
3992 The difference is that the port acts like a telnet server or
3993 client using telnet option negotiation. This will also allow you
3994 to send the MAGIC\_SYSRQ sequence if you use a telnet that
3995 supports sending the break sequence. Typically in unix telnet
3996 you do it with Control-] and then type "send break" followed by
3997 pressing the enter key.
3999 ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]``
4000 The WebSocket protocol is used instead of raw tcp socket. The
4001 port acts as a WebSocket server. Client mode is not supported.
4003 ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]``
4004 A unix domain socket is used instead of a tcp socket. The option
4005 works the same as if you had specified ``-serial tcp`` except
4006 the unix domain socket path is used for connections.
4009 This is a special option to allow the monitor to be multiplexed
4010 onto another serial port. The monitor is accessed with key
4011 sequence of Control-a and then pressing c. dev\_string should be
4012 any one of the serial devices specified above. An example to
4013 multiplex the monitor onto a telnet server listening on port
4016 ``-serial mon:telnet::4444,server=on,wait=off``
4018 When the monitor is multiplexed to stdio in this way, Ctrl+C
4019 will not terminate QEMU any more but will be passed to the guest
4023 Braille device. This will use BrlAPI to display the braille
4024 output on a real or fake device.
4027 Three button serial mouse. Configure the guest to use Microsoft
4031 DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \
4032 "-parallel dev redirect the parallel port to char device 'dev
'\n",
4036 Redirect the virtual parallel port to host device dev (same devices
4037 as the serial port). On Linux hosts, ``/dev/parportN`` can be used
4038 to use hardware devices connected on the corresponding host parallel
4041 This option can be used several times to simulate up to 3 parallel
4044 Use ``-parallel none`` to disable all parallel ports.
4047 DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \
4048 "-monitor dev redirect the monitor to char device 'dev
'\n",
4052 Redirect the monitor to host device dev (same devices as the serial
4053 port). The default device is ``vc`` in graphical mode and ``stdio``
4054 in non graphical mode. Use ``-monitor none`` to disable the default
4057 DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \
4058 "-qmp dev like -monitor but opens in 'control
' mode\n",
4062 Like -monitor but opens in 'control
' mode.
4064 DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \
4065 "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n",
4069 Like -qmp but uses pretty JSON formatting.
4072 DEF("mon", HAS_ARG, QEMU_OPTION_mon, \
4073 "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL)
4075 ``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]``
4076 Setup monitor on chardev name. ``mode=control`` configures
4077 a QMP monitor (a JSON RPC-style protocol) and it is not the
4078 same as HMP, the human monitor that has a "(qemu)" prompt.
4079 ``pretty`` is only valid when ``mode=control``,
4080 turning on JSON pretty printing to ease
4081 human reading and debugging.
4084 DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \
4085 "-debugcon dev redirect the debug console to char device 'dev
'\n",
4089 Redirect the debug console to host device dev (same devices as the
4090 serial port). The debug console is an I/O port which is typically
4091 port 0xe9; writing to that I/O port sends output to this device. The
4092 default device is ``vc`` in graphical mode and ``stdio`` in non
4096 DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \
4097 "-pidfile file write PID to 'file
'\n", QEMU_ARCH_ALL)
4100 Store the QEMU process PID in file. It is useful if you launch QEMU
4104 DEF("singlestep", 0, QEMU_OPTION_singlestep, \
4105 "-singlestep always run in singlestep mode\n", QEMU_ARCH_ALL)
4108 Run the emulation in single step mode.
4111 DEF("preconfig", 0, QEMU_OPTION_preconfig, \
4112 "--preconfig pause QEMU before machine is initialized (experimental)\n",
4116 Pause QEMU for interactive configuration before the machine is
4117 created, which allows querying and configuring properties that will
4118 affect machine initialization. Use QMP command 'x
-exit
-preconfig
' to
4119 exit the preconfig state and move to the next state (i.e. run guest
4120 if -S isn't used or pause the second time
if -S is used
). This
4121 option is experimental
.
4124 DEF("S", 0, QEMU_OPTION_S
, \
4125 "-S freeze CPU at startup (use 'c' to start execution)\n",
4129 Do not start CPU at
startup (you must type
'c' in the monitor
).
4132 DEF("overcommit", HAS_ARG
, QEMU_OPTION_overcommit
,
4133 "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n"
4134 " run qemu with overcommit hints\n"
4135 " mem-lock=on|off controls memory lock support (default: off)\n"
4136 " cpu-pm=on|off controls cpu power management (default: off)\n",
4139 ``
-overcommit mem
-lock
=on|off``
4141 ``
-overcommit cpu
-pm
=on|off``
4142 Run qemu with hints about host resource overcommit
. The
default is
4143 to assume that host overcommits all resources
.
4145 Locking qemu and guest memory can be enabled via ``mem
-lock
=on``
4146 (disabled by
default). This works when host memory is not
4147 overcommitted and reduces the worst
-case latency
for guest
.
4149 Guest ability to manage power state of host
cpus (increasing latency
4150 for other processes on the same host cpu
, but decreasing latency
for
4151 guest
) can be enabled via ``cpu
-pm
=on``
(disabled by
default). This
4152 works best when host CPU is not overcommitted
. When used
, host
4153 estimates of CPU cycle and power utilization will be incorrect
, not
4154 taking into account guest idle time
.
4157 DEF("gdb", HAS_ARG
, QEMU_OPTION_gdb
, \
4158 "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n"
4159 " the guest without waiting for gdb to connect; use -S too\n"
4160 " if you want it to not start execution.)\n",
4164 Accept a gdb connection on device
dev (see the
:ref
:`GDB usage` chapter
4165 in the System Emulation Users Guide
). Note that
this option does not pause QEMU
4166 execution
-- if you want QEMU to not start the guest until you
4167 connect with gdb and issue a ``
continue`` command
, you will need to
4168 also pass the ``
-S`` option to QEMU
.
4170 The most usual configuration is to listen on a local TCP socket
::
4174 but you can specify other backends
; UDP
, pseudo TTY
, or even stdio
4175 are all reasonable use cases
. For example
, a stdio connection
4176 allows you to start QEMU from within gdb and establish the
4177 connection via a pipe
:
4181 (gdb
) target remote | exec |qemu_system|
-gdb stdio
...
4184 DEF("s", 0, QEMU_OPTION_s
, \
4185 "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT
"\n",
4189 Shorthand
for -gdb tcp
::1234, i
.e
. open a gdbserver on TCP port
1234
4190 (see the
:ref
:`GDB usage` chapter
in the System Emulation Users Guide
).
4193 DEF("d", HAS_ARG
, QEMU_OPTION_d
, \
4194 "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n",
4198 Enable logging of specified items
. Use
'-d help' for a list of log
4202 DEF("D", HAS_ARG
, QEMU_OPTION_D
, \
4203 "-D logfile output log to logfile (default stderr)\n",
4207 Output log
in logfile instead of to stderr
4210 DEF("dfilter", HAS_ARG
, QEMU_OPTION_DFILTER
, \
4211 "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n",
4214 ``
-dfilter range1
[,...]``
4215 Filter debug output to that relevant to a range of target addresses
.
4216 The filter spec can be either start
+size
, start
-size or start
..end
4217 where start end and size are the addresses and sizes required
. For
4222 -dfilter
0x8000..0x8fff
,0xffffffc000080000+0x200,0xffffffc000060000-0x1000
4224 Will dump output
for any code
in the
0x1000 sized block starting at
4225 0x8000 and the
0x200 sized block starting at
0xffffffc000080000 and
4226 another
0x1000 sized block starting at
0xffffffc00005f000.
4229 DEF("seed", HAS_ARG
, QEMU_OPTION_seed
, \
4230 "-seed number seed the pseudo-random number generator\n",
4234 Force the guest to use a deterministic pseudo
-random number
4235 generator
, seeded with number
. This does not affect crypto routines
4239 DEF("L", HAS_ARG
, QEMU_OPTION_L
, \
4240 "-L path set the directory for the BIOS, VGA BIOS and keymaps\n",
4244 Set the directory
for the BIOS
, VGA BIOS and keymaps
.
4246 To list all the data directories
, use ``
-L help``
.
4249 DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm
, \
4250 "-enable-kvm enable KVM full virtualization support\n",
4251 QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC |
4252 QEMU_ARCH_RISCV | QEMU_ARCH_S390X
)
4255 Enable KVM full virtualization support
. This option is only
4256 available
if KVM support is enabled when compiling
.
4259 DEF("xen-domid", HAS_ARG
, QEMU_OPTION_xen_domid
,
4260 "-xen-domid id specify xen guest domain id\n",
4261 QEMU_ARCH_ARM | QEMU_ARCH_I386
)
4262 DEF("xen-attach", 0, QEMU_OPTION_xen_attach
,
4263 "-xen-attach attach to existing xen domain\n"
4264 " libxl will use this when starting QEMU\n",
4265 QEMU_ARCH_ARM | QEMU_ARCH_I386
)
4266 DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict
,
4267 "-xen-domid-restrict restrict set of available xen operations\n"
4268 " to specified domain id. (Does not affect\n"
4269 " xenpv machine type).\n",
4270 QEMU_ARCH_ARM | QEMU_ARCH_I386
)
4273 Specify xen guest domain
id (XEN only
).
4276 Attach to existing xen domain
. libxl will use
this when starting
4277 QEMU (XEN only
). Restrict set of available xen operations to
4278 specified domain
id (XEN only
).
4281 DEF("no-reboot", 0, QEMU_OPTION_no_reboot
, \
4282 "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL
)
4285 Exit instead of rebooting
.
4288 DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown
, \
4289 "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL
)
4292 Don
't exit QEMU on guest shutdown, but instead only stop the
4293 emulation. This allows for instance switching to monitor to commit
4294 changes to the disk image.
4297 DEF("action", HAS_ARG, QEMU_OPTION_action,
4298 "-action reboot=reset|shutdown\n"
4299 " action when guest reboots [default=reset]\n"
4300 "-action shutdown=poweroff|pause\n"
4301 " action when guest shuts down [default=poweroff]\n"
4302 "-action panic=pause|shutdown|exit-failure|none\n"
4303 " action when guest panics [default=shutdown]\n"
4304 "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n"
4305 " action when watchdog fires [default=reset]\n",
4308 ``-action event=action``
4309 The action parameter serves to modify QEMU's
default behavior when
4310 certain guest events occur
. It provides a generic method
for specifying the
4311 same behaviors that are modified by the ``
-no
-reboot`` and ``
-no
-shutdown``
4316 ``
-action panic
=none``
4317 ``
-action reboot
=shutdown
,shutdown
=pause``
4318 ``
-watchdog i6300esb
-action watchdog
=pause``
4322 DEF("loadvm", HAS_ARG
, QEMU_OPTION_loadvm
, \
4323 "-loadvm [tag|id]\n" \
4324 " start right away with a saved state (loadvm in monitor)\n",
4328 Start right away with a saved
state (``loadvm``
in monitor
)
4332 DEF("daemonize", 0, QEMU_OPTION_daemonize
, \
4333 "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL
)
4337 Daemonize the QEMU process after initialization
. QEMU will not
4338 detach from standard IO until it is ready to receive connections on
4339 any of its devices
. This option is a useful way
for external
4340 programs to launch QEMU without having to cope with initialization
4344 DEF("option-rom", HAS_ARG
, QEMU_OPTION_option_rom
, \
4345 "-option-rom rom load a file, rom, into the option ROM space\n",
4348 ``
-option
-rom file``
4349 Load the contents of file as an option ROM
. This option is useful to
4350 load things like EtherBoot
.
4353 DEF("rtc", HAS_ARG
, QEMU_OPTION_rtc
, \
4354 "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \
4355 " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n",
4359 ``
-rtc
[base
=utc|localtime|datetime
][,clock
=host|rt|vm
][,driftfix
=none|slew
]``
4360 Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at
4361 the current UTC or local time
, respectively
. ``localtime`` is
4362 required
for correct date
in MS
-DOS or Windows
. To start at a
4363 specific point
in time
, provide datetime
in the format
4364 ``
2006-06-17T16
:01:21`` or ``
2006-06-17``
. The
default base is UTC
.
4366 By
default the RTC is driven by the host system time
. This allows
4367 using of the RTC as accurate reference clock inside the guest
,
4368 specifically
if the host time is smoothly following an accurate
4369 external reference clock
, e
.g
. via NTP
. If you want to isolate the
4370 guest time from the host
, you can set ``clock`` to ``rt`` instead
,
4371 which provides a host monotonic clock
if host support it
. To even
4372 prevent the RTC from progressing during suspension
, you can set
4373 ``clock`` to ``vm``
(virtual clock
). '\ ``clock=vm``\ ' is
4374 recommended especially
in icount mode
in order to preserve
4375 determinism
; however
, note that
in icount mode the speed of the
4376 virtual clock is variable and can
in general differ from the host
4379 Enable ``driftfix``
(i386 targets only
) if you experience time drift
4380 problems
, specifically with Windows
' ACPI HAL. This option will try
4381 to figure out how many timer interrupts were not processed by the
4382 Windows guest and will re-inject them.
4385 DEF("icount", HAS_ARG, QEMU_OPTION_icount, \
4386 "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \
4387 " enable virtual instruction counter with 2^N clock ticks per\n" \
4388 " instruction, enable aligning the host and virtual clocks\n" \
4389 " or disable real time cpu sleeping, and optionally enable\n" \
4390 " record-and-replay mode\n", QEMU_ARCH_ALL)
4392 ``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]``
4393 Enable virtual instruction counter. The virtual cpu will execute one
4394 instruction every 2^N ns of virtual time. If ``auto`` is specified
4395 then the virtual cpu speed will be automatically adjusted to keep
4396 virtual time within a few seconds of real time.
4398 Note that while this option can give deterministic behavior, it does
4399 not provide cycle accurate emulation. Modern CPUs contain
4400 superscalar out of order cores with complex cache hierarchies. The
4401 number of instructions executed often has little or no correlation
4402 with actual performance.
4404 When the virtual cpu is sleeping, the virtual time will advance at
4405 default speed unless ``sleep=on`` is specified. With
4406 ``sleep=on``, the virtual time will jump to the next timer
4407 deadline instantly whenever the virtual cpu goes to sleep mode and
4408 will not advance if no timer is enabled. This behavior gives
4409 deterministic execution times from the guest point of view.
4410 The default if icount is enabled is ``sleep=off``.
4411 ``sleep=on`` cannot be used together with either ``shift=auto``
4414 ``align=on`` will activate the delay algorithm which will try to
4415 synchronise the host clock and the virtual clock. The goal is to
4416 have a guest running at the real frequency imposed by the shift
4417 option. Whenever the guest clock is behind the host clock and if
4418 ``align=on`` is specified then we print a message to the user to
4419 inform about the delay. Currently this option does not work when
4420 ``shift`` is ``auto``. Note: The sync algorithm will work for those
4421 shift values for which the guest clock runs ahead of the host clock.
4422 Typically this happens when the shift value is high (how high
4423 depends on the host machine). The default if icount is enabled
4426 When the ``rr`` option is specified deterministic record/replay is
4427 enabled. The ``rrfile=`` option must also be provided to
4428 specify the path to the replay log. In record mode data is written
4429 to this file, and in replay mode it is read back.
4430 If the ``rrsnapshot`` option is given then it specifies a VM snapshot
4431 name. In record mode, a new VM snapshot with the given name is created
4432 at the start of execution recording. In replay mode this option
4433 specifies the snapshot name used to load the initial VM state.
4436 DEF("watchdog", HAS_ARG, QEMU_OPTION_watchdog, \
4437 "-watchdog model\n" \
4438 " enable virtual hardware watchdog [default=none]\n",
4442 Create a virtual hardware watchdog device. Once enabled (by a guest
4443 action), the watchdog must be periodically polled by an agent inside
4444 the guest or else the guest will be restarted. Choose a model for
4445 which your guest has drivers.
4447 The model is the model of hardware watchdog to emulate. Use
4448 ``-watchdog help`` to list available hardware models. Only one
4449 watchdog can be enabled for a guest.
4451 The following models may be available:
4454 iBASE 700 is a very simple ISA watchdog with a single timer.
4457 Intel 6300ESB I/O controller hub is a much more featureful
4458 PCI-based dual-timer watchdog.
4461 A virtual watchdog for s390x backed by the diagnose 288
4462 hypercall (currently KVM only).
4465 DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \
4466 "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \
4467 " action when watchdog fires [default=reset]\n",
4470 ``-watchdog-action action``
4471 The action controls what QEMU will do when the watchdog timer
4472 expires. The default is ``reset`` (forcefully reset the guest).
4473 Other possible actions are: ``shutdown`` (attempt to gracefully
4474 shutdown the guest), ``poweroff`` (forcefully poweroff the guest),
4475 ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the
4476 guest), ``debug`` (print a debug message and continue), or ``none``
4479 Note that the ``shutdown`` action requires that the guest responds
4480 to ACPI signals, which it may not be able to do in the sort of
4481 situations where the watchdog would have expired, and thus
4482 ``-watchdog-action shutdown`` is not recommended for production use.
4486 ``-watchdog i6300esb -watchdog-action pause``; \ ``-watchdog ib700``
4490 DEF("echr", HAS_ARG, QEMU_OPTION_echr, \
4491 "-echr chr set terminal escape character instead of ctrl-a\n",
4494 ``-echr numeric_ascii_value``
4495 Change the escape character used for switching to the monitor when
4496 using monitor and serial sharing. The default is ``0x01`` when using
4497 the ``-nographic`` option. ``0x01`` is equal to pressing
4498 ``Control-a``. You can select a different character from the ascii
4499 control keys where 1 through 26 map to Control-a through Control-z.
4500 For instance you could use the either of the following to change the
4501 escape character to Control-t.
4503 ``-echr 0x14``; \ ``-echr 20``
4507 DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \
4508 "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \
4509 "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \
4510 "-incoming unix:socketpath\n" \
4511 " prepare for incoming migration, listen on\n" \
4512 " specified protocol and socket address\n" \
4513 "-incoming fd:fd\n" \
4514 "-incoming exec:cmdline\n" \
4515 " accept incoming migration on given file descriptor\n" \
4516 " or from given external command\n" \
4517 "-incoming defer\n" \
4518 " wait for the URI to be specified via migrate_incoming\n",
4521 ``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]``
4523 ``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]``
4524 Prepare for incoming migration, listen on a given tcp port.
4526 ``-incoming unix:socketpath``
4527 Prepare for incoming migration, listen on a given unix socket.
4530 Accept incoming migration from a given filedescriptor.
4532 ``-incoming exec:cmdline``
4533 Accept incoming migration as an output from specified external
4537 Wait for the URI to be specified via migrate\_incoming. The monitor
4538 can be used to change settings (such as migration parameters) prior
4539 to issuing the migrate\_incoming to allow the migration to begin.
4542 DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \
4543 "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL)
4545 ``-only-migratable``
4546 Only allow migratable devices. Devices will not be allowed to enter
4547 an unmigratable state.
4550 DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \
4551 "-nodefaults don't create
default devices
\n", QEMU_ARCH_ALL)
4554 Don't create default devices. Normally, QEMU sets the default
4555 devices like serial port, parallel port, virtual console, monitor
4556 device, VGA adapter, floppy and CD-ROM drive and others. The
4557 ``-nodefaults`` option will disable all those default devices.
4561 DEF("chroot
", HAS_ARG, QEMU_OPTION_chroot, \
4562 "-chroot dir chroot to dir just before starting the VM
\n",
4567 Immediately before starting guest execution, chroot to the specified
4568 directory. Especially useful in combination with -runas.
4569 This option is not supported for Windows hosts.
4573 DEF("runas
", HAS_ARG, QEMU_OPTION_runas, \
4574 "-runas user change to user id user just before starting the VM
\n" \
4575 " user can be numeric uid
:gid instead
\n",
4580 Immediately before starting guest execution, drop root privileges,
4581 switching to the specified user.
4584 DEF("prom
-env
", HAS_ARG, QEMU_OPTION_prom_env,
4585 "-prom
-env variable
=value
\n"
4586 " set OpenBIOS nvram variables
\n",
4587 QEMU_ARCH_PPC | QEMU_ARCH_SPARC)
4589 ``-prom-env variable=value``
4590 Set OpenBIOS nvram variable to given value (PPC, SPARC only).
4594 qemu-system-sparc -prom-env 'auto-boot?=false' \
4595 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single'
4599 qemu-system-ppc -prom-env 'auto-boot?=false' \
4600 -prom-env 'boot-device=hd:2,\yaboot' \
4601 -prom-env 'boot-args=conf=hd:2,\yaboot.conf'
4603 DEF("semihosting
", 0, QEMU_OPTION_semihosting,
4604 "-semihosting semihosting mode
\n",
4605 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA |
4606 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4609 Enable semihosting mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only).
4611 Note that this allows guest direct access to the host filesystem, so
4612 should only be used with a trusted guest OS.
4614 See the -semihosting-config option documentation for further
4615 information about the facilities this enables.
4617 DEF("semihosting
-config
", HAS_ARG, QEMU_OPTION_semihosting_config,
4618 "-semihosting
-config
[enable
=on|off
][,target
=native|gdb|auto
][,chardev
=id
][,arg
=str
[,...]]\n" \
4619 " semihosting configuration
\n",
4620 QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA |
4621 QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV)
4623 ``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,arg=str[,...]]``
4624 Enable and configure semihosting (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V
4627 Note that this allows guest direct access to the host filesystem, so
4628 should only be used with a trusted guest OS.
4630 On Arm this implements the standard semihosting API, version 2.0.
4632 On M68K this implements the "ColdFire GDB
" interface used by
4635 Xtensa semihosting provides basic file IO calls, such as
4636 open/read/write/seek/select. Tensilica baremetal libc for ISS and
4637 linux platform "sim
" use this interface.
4639 On RISC-V this implements the standard semihosting API, version 0.2.
4641 ``target=native|gdb|auto``
4642 Defines where the semihosting calls will be addressed, to QEMU
4643 (``native``) or to GDB (``gdb``). The default is ``auto``, which
4644 means ``gdb`` during debug sessions and ``native`` otherwise.
4647 Send the output to a chardev backend output for native or auto
4648 output when not in gdb
4650 ``arg=str1,arg=str2,...``
4651 Allows the user to pass input arguments, and can be used
4652 multiple times to build up a list. The old-style
4653 ``-kernel``/``-append`` method of passing a command line is
4654 still supported for backward compatibility. If both the
4655 ``--semihosting-config arg`` and the ``-kernel``/``-append`` are
4656 specified, the former is passed to semihosting as it always
4659 DEF("old
-param
", 0, QEMU_OPTION_old_param,
4660 "-old
-param old param mode
\n", QEMU_ARCH_ARM)
4663 Old param mode (ARM only).
4666 DEF("sandbox
", HAS_ARG, QEMU_OPTION_sandbox, \
4667 "-sandbox on
[,obsolete
=allow|deny
][,elevateprivileges
=allow|deny|children
]\n" \
4668 " [,spawn
=allow|deny
][,resourcecontrol
=allow|deny
]\n" \
4669 " Enable seccomp mode
2 system call
filter (default 'off').\n" \
4670 " use
'obsolete' to allow obsolete system calls that are provided
\n" \
4671 " by the kernel
, but typically no longer used by modern
\n" \
4672 " C library implementations
.\n" \
4673 " use
'elevateprivileges' to allow or deny the QEMU process ability
\n" \
4674 " to elevate privileges
using set
*uid|gid system calls
.\n" \
4675 " The value
'children' will deny set
*uid|gid system calls
for\n" \
4676 " main QEMU process but will allow forks and execves to run unprivileged
\n" \
4677 " use
'spawn' to avoid QEMU to spawn
new threads or processes by
\n" \
4678 " blocking
*fork and execve
\n" \
4679 " use
'resourcecontrol' to disable process affinity and schedular priority
\n",
4682 ``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]``
4683 Enable Seccomp mode 2 system call filter. 'on' will enable syscall
4684 filtering and 'off' will disable it. The default is 'off'.
4687 Enable Obsolete system calls
4689 ``elevateprivileges=string``
4690 Disable set\*uid\|gid system calls
4693 Disable \*fork and execve
4695 ``resourcecontrol=string``
4696 Disable process affinity and schedular priority
4699 DEF("readconfig
", HAS_ARG, QEMU_OPTION_readconfig,
4700 "-readconfig
<file
>\n"
4701 " read config file
\n", QEMU_ARCH_ALL)
4703 ``-readconfig file``
4704 Read device configuration from file. This approach is useful when
4705 you want to spawn QEMU process with many command line options but
4706 you don't want to exceed the command line character limit.
4709 DEF("no
-user
-config
", 0, QEMU_OPTION_nouserconfig,
4711 " do not load
default user
-provided config files at startup
\n",
4715 The ``-no-user-config`` option makes QEMU not load any of the
4716 user-provided config files on sysconfdir.
4719 DEF("trace-unassigned
", 0, QEMU_OPTION_trace_unassigned,
4720 "-trace-unassigned
\n"
4721 " Trace unassigned memory or i
/o accesses
\n",
4724 ``-trace-unassigned``
4725 Trace unassigned memory or i/o accesses to stderr.
4728 DEF("trace", HAS_ARG, QEMU_OPTION_trace,
4729 "-trace [[enable
=]<pattern
>][,events
=<file
>][,file
=<file
>]\n"
4730 " specify tracing options
\n",
4733 ``-trace [[enable=]pattern][,events=file][,file=file]``
4734 .. include:: ../qemu-option-trace.rst.inc
4737 DEF("plugin
", HAS_ARG, QEMU_OPTION_plugin,
4738 "-plugin
[file
=]<file
>[,<argname
>=<argvalue
>]\n"
4742 ``-plugin file=file[,argname=argvalue]``
4746 Load the given plugin from a shared library file.
4748 ``argname=argvalue``
4749 Argument passed to the plugin. (Can be given multiple times.)
4753 DEF("qtest
", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL)
4754 DEF("qtest
-log
", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL)
4756 DEF("msg
", HAS_ARG, QEMU_OPTION_msg,
4757 "-msg
[timestamp
[=on|off
]][,guest
-name
=[on|off
]]\n"
4758 " control error message format
\n"
4759 " timestamp
=on enables
timestamps (default: off
)\n"
4760 " guest
-name
=on enables guest name prefix but only
if\n"
4761 " -name guest option is
set (default: off
)\n",
4764 ``-msg [timestamp[=on|off]][,guest-name[=on|off]]``
4765 Control error message format.
4767 ``timestamp=on|off``
4768 Prefix messages with a timestamp. Default is off.
4770 ``guest-name=on|off``
4771 Prefix messages with guest name but only if -name guest option is set
4772 otherwise the option is ignored. Default is off.
4775 DEF("dump
-vmstate
", HAS_ARG, QEMU_OPTION_dump_vmstate,
4776 "-dump
-vmstate
<file
>\n"
4777 " Output vmstate information
in JSON format to file
.\n"
4778 " Use the scripts
/vmstate
-static-checker
.py file to
\n"
4779 " check
for possible regressions
in migration code
\n"
4780 " by comparing two such vmstate dumps
.\n",
4783 ``-dump-vmstate file``
4784 Dump json-encoded vmstate information for current machine type to
4788 DEF("enable
-sync
-profile
", 0, QEMU_OPTION_enable_sync_profile,
4789 "-enable
-sync
-profile
\n"
4790 " enable synchronization profiling
\n",
4793 ``-enable-sync-profile``
4794 Enable synchronization profiling.
4799 DEFHEADING(Generic object creation:)
4801 DEF("object
", HAS_ARG, QEMU_OPTION_object,
4802 "-object TYPENAME
[,PROP1
=VALUE1
,...]\n"
4803 " create a
new object of type TYPENAME setting properties
\n"
4804 " in the order they are specified
. Note that the
'id'\n"
4805 " property must be set
. These objects are placed
in the
\n"
4806 " '/objects' path
.\n",
4809 ``-object typename[,prop1=value1,...]``
4810 Create a new object of type typename setting properties in the order
4811 they are specified. Note that the 'id' property must be set. These
4812 objects are placed in the '/objects' path.
4814 ``-object memory-backend-file,id=id,size=size,mem-path=dir,share=on|off,discard-data=on|off,merge=on|off,dump=on|off,prealloc=on|off,host-nodes=host-nodes,policy=default|preferred|bind|interleave,align=align,readonly=on|off``
4815 Creates a memory file backend object, which can be used to back
4816 the guest RAM with huge pages.
4818 The ``id`` parameter is a unique ID that will be used to
4819 reference this memory region in other parameters, e.g. ``-numa``,
4820 ``-device nvdimm``, etc.
4822 The ``size`` option provides the size of the memory region, and
4823 accepts common suffixes, e.g. ``500M``.
4825 The ``mem-path`` provides the path to either a shared memory or
4826 huge page filesystem mount.
4828 The ``share`` boolean option determines whether the memory
4829 region is marked as private to QEMU, or shared. The latter
4830 allows a co-operating external process to access the QEMU memory
4833 The ``share`` is also required for pvrdma devices due to
4834 limitations in the RDMA API provided by Linux.
4836 Setting share=on might affect the ability to configure NUMA
4837 bindings for the memory backend under some circumstances, see
4838 Documentation/vm/numa\_memory\_policy.txt on the Linux kernel
4839 source tree for additional details.
4841 Setting the ``discard-data`` boolean option to on indicates that
4842 file contents can be destroyed when QEMU exits, to avoid
4843 unnecessarily flushing data to the backing file. Note that
4844 ``discard-data`` is only an optimization, and QEMU might not
4845 discard file contents if it aborts unexpectedly or is terminated
4848 The ``merge`` boolean option enables memory merge, also known as
4849 MADV\_MERGEABLE, so that Kernel Samepage Merging will consider
4850 the pages for memory deduplication.
4852 Setting the ``dump`` boolean option to off excludes the memory
4853 from core dumps. This feature is also known as MADV\_DONTDUMP.
4855 The ``prealloc`` boolean option enables memory preallocation.
4857 The ``host-nodes`` option binds the memory range to a list of
4860 The ``policy`` option sets the NUMA policy to one of the
4867 prefer the given host node list for allocation
4870 restrict memory allocation to the given host node list
4873 interleave memory allocations across the given host node
4876 The ``align`` option specifies the base address alignment when
4877 QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg
4878 ``2M``. Some backend store specified by ``mem-path`` requires an
4879 alignment different than the default one used by QEMU, eg the
4880 device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In
4881 such cases, users can specify the required alignment via this
4884 The ``pmem`` option specifies whether the backing file specified
4885 by ``mem-path`` is in host persistent memory that can be
4886 accessed using the SNIA NVM programming model (e.g. Intel
4887 NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary
4888 operations to guarantee the persistence of its own writes to
4889 ``mem-path`` (e.g. in vNVDIMM label emulation and live
4890 migration). Also, we will map the backend-file with MAP\_SYNC
4891 flag, which ensures the file metadata is in sync for
4892 ``mem-path`` in case of host crash or a power failure. MAP\_SYNC
4893 requires support from both the host kernel (since Linux kernel
4894 4.15) and the filesystem of ``mem-path`` mounted with DAX
4897 The ``readonly`` option specifies whether the backing file is opened
4898 read-only or read-write (default).
4900 ``-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``
4901 Creates a memory backend object, which can be used to back the
4902 guest RAM. Memory backend objects offer more control than the
4903 ``-m`` option that is traditionally used to define guest RAM.
4904 Please refer to ``memory-backend-file`` for a description of the
4907 ``-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``
4908 Creates an anonymous memory file backend object, which allows
4909 QEMU to share the memory with an external process (e.g. when
4910 using vhost-user). The memory is allocated with memfd and
4911 optional sealing. (Linux only)
4913 The ``seal`` option creates a sealed-file, that will block
4914 further resizing the memory ('on' by default).
4916 The ``hugetlb`` option specify the file to be created resides in
4917 the hugetlbfs filesystem (since Linux 4.14). Used in conjunction
4918 with the ``hugetlb`` option, the ``hugetlbsize`` option specify
4919 the hugetlb page size on systems that support multiple hugetlb
4920 page sizes (it must be a power of 2 value supported by the
4923 In some versions of Linux, the ``hugetlb`` option is
4924 incompatible with the ``seal`` option (requires at least Linux
4927 Please refer to ``memory-backend-file`` for a description of the
4930 The ``share`` boolean option is on by default with memfd.
4932 ``-object rng-builtin,id=id``
4933 Creates a random number generator backend which obtains entropy
4934 from QEMU builtin functions. The ``id`` parameter is a unique ID
4935 that will be used to reference this entropy backend from the
4936 ``virtio-rng`` device. By default, the ``virtio-rng`` device
4937 uses this RNG backend.
4939 ``-object rng-random,id=id,filename=/dev/random``
4940 Creates a random number generator backend which obtains entropy
4941 from a device on the host. The ``id`` parameter is a unique ID
4942 that will be used to reference this entropy backend from the
4943 ``virtio-rng`` device. The ``filename`` parameter specifies
4944 which file to obtain entropy from and if omitted defaults to
4947 ``-object rng-egd,id=id,chardev=chardevid``
4948 Creates a random number generator backend which obtains entropy
4949 from an external daemon running on the host. The ``id``
4950 parameter is a unique ID that will be used to reference this
4951 entropy backend from the ``virtio-rng`` device. The ``chardev``
4952 parameter is the unique ID of a character device backend that
4953 provides the connection to the RNG daemon.
4955 ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off``
4956 Creates a TLS anonymous credentials object, which can be used to
4957 provide TLS support on network backends. The ``id`` parameter is
4958 a unique ID which network backends will use to access the
4959 credentials. The ``endpoint`` is either ``server`` or ``client``
4960 depending on whether the QEMU network backend that uses the
4961 credentials will be acting as a client or as a server. If
4962 ``verify-peer`` is enabled (the default) then once the handshake
4963 is completed, the peer credentials will be verified, though this
4964 is a no-op for anonymous credentials.
4966 The dir parameter tells QEMU where to find the credential files.
4967 For server endpoints, this directory may contain a file
4968 dh-params.pem providing diffie-hellman parameters to use for the
4969 TLS server. If the file is missing, QEMU will generate a set of
4970 DH parameters at startup. This is a computationally expensive
4971 operation that consumes random pool entropy, so it is
4972 recommended that a persistent set of parameters be generated
4975 ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]``
4976 Creates a TLS Pre-Shared Keys (PSK) credentials object, which
4977 can be used to provide TLS support on network backends. The
4978 ``id`` parameter is a unique ID which network backends will use
4979 to access the credentials. The ``endpoint`` is either ``server``
4980 or ``client`` depending on whether the QEMU network backend that
4981 uses the credentials will be acting as a client or as a server.
4982 For clients only, ``username`` is the username which will be
4983 sent to the server. If omitted it defaults to "qemu
".
4985 The dir parameter tells QEMU where to find the keys file. It is
4986 called "dir
/keys
.psk
" and contains "username
:key
" pairs. This
4987 file can most easily be created using the GnuTLS ``psktool``
4990 For server endpoints, dir may also contain a file dh-params.pem
4991 providing diffie-hellman parameters to use for the TLS server.
4992 If the file is missing, QEMU will generate a set of DH
4993 parameters at startup. This is a computationally expensive
4994 operation that consumes random pool entropy, so it is
4995 recommended that a persistent set of parameters be generated up
4998 ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id``
4999 Creates a TLS anonymous credentials object, which can be used to
5000 provide TLS support on network backends. The ``id`` parameter is
5001 a unique ID which network backends will use to access the
5002 credentials. The ``endpoint`` is either ``server`` or ``client``
5003 depending on whether the QEMU network backend that uses the
5004 credentials will be acting as a client or as a server. If
5005 ``verify-peer`` is enabled (the default) then once the handshake
5006 is completed, the peer credentials will be verified. With x509
5007 certificates, this implies that the clients must be provided
5008 with valid client certificates too.
5010 The dir parameter tells QEMU where to find the credential files.
5011 For server endpoints, this directory may contain a file
5012 dh-params.pem providing diffie-hellman parameters to use for the
5013 TLS server. If the file is missing, QEMU will generate a set of
5014 DH parameters at startup. This is a computationally expensive
5015 operation that consumes random pool entropy, so it is
5016 recommended that a persistent set of parameters be generated
5019 For x509 certificate credentials the directory will contain
5020 further files providing the x509 certificates. The certificates
5021 must be stored in PEM format, in filenames ca-cert.pem,
5022 ca-crl.pem (optional), server-cert.pem (only servers),
5023 server-key.pem (only servers), client-cert.pem (only clients),
5024 and client-key.pem (only clients).
5026 For the server-key.pem and client-key.pem files which contain
5027 sensitive private keys, it is possible to use an encrypted
5028 version by providing the passwordid parameter. This provides the
5029 ID of a previously created ``secret`` object containing the
5030 password for decryption.
5032 The priority parameter allows to override the global default
5033 priority used by gnutls. This can be useful if the system
5034 administrator needs to use a weaker set of crypto priorities for
5035 QEMU without potentially forcing the weakness onto all
5036 applications. Or conversely if one wants wants a stronger
5037 default for QEMU than for all other applications, they can do
5038 this through this parameter. Its format is a gnutls priority
5039 string as described at
5040 https://gnutls.org/manual/html_node/Priority-Strings.html.
5042 ``-object tls-cipher-suites,id=id,priority=priority``
5043 Creates a TLS cipher suites object, which can be used to control
5044 the TLS cipher/protocol algorithms that applications are permitted
5047 The ``id`` parameter is a unique ID which frontends will use to
5048 access the ordered list of permitted TLS cipher suites from the
5051 The ``priority`` parameter allows to override the global default
5052 priority used by gnutls. This can be useful if the system
5053 administrator needs to use a weaker set of crypto priorities for
5054 QEMU without potentially forcing the weakness onto all
5055 applications. Or conversely if one wants wants a stronger
5056 default for QEMU than for all other applications, they can do
5057 this through this parameter. Its format is a gnutls priority
5058 string as described at
5059 https://gnutls.org/manual/html_node/Priority-Strings.html.
5061 An example of use of this object is to control UEFI HTTPS Boot.
5062 The tls-cipher-suites object exposes the ordered list of permitted
5063 TLS cipher suites from the host side to the guest firmware, via
5064 fw_cfg. The list is represented as an array of IANA_TLS_CIPHER
5065 objects. The firmware uses the IANA_TLS_CIPHER array for configuring
5068 In the following example, the priority at which the host-side policy
5069 is retrieved is given by the ``priority`` property.
5070 Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to
5071 refer to /etc/crypto-policies/back-ends/gnutls.config.
5076 -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\
5077 -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0
5079 ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]``
5080 Interval t can't be 0, this filter batches the packet delivery:
5081 all packets arriving in a given interval on netdev netdevid are
5082 delayed until the end of the interval. Interval is in
5083 microseconds. ``status`` is optional that indicate whether the
5084 netfilter is on (enabled) or off (disabled), the default status
5085 for netfilter will be 'on'.
5087 queue all\|rx\|tx is an option that can be applied to any
5090 ``all``: the filter is attached both to the receive and the
5091 transmit queue of the netdev (default).
5093 ``rx``: the filter is attached to the receive queue of the
5094 netdev, where it will receive packets sent to the netdev.
5096 ``tx``: the filter is attached to the transmit queue of the
5097 netdev, where it will receive packets sent by the netdev.
5099 position head\|tail\|id=<id> is an option to specify where the
5100 filter should be inserted in the filter list. It can be applied
5103 ``head``: the filter is inserted at the head of the filter list,
5104 before any existing filters.
5106 ``tail``: the filter is inserted at the tail of the filter list,
5107 behind any existing filters (default).
5109 ``id=<id>``: the filter is inserted before or behind the filter
5110 specified by <id>, see the insert option below.
5112 insert behind\|before is an option to specify where to insert
5113 the new filter relative to the one specified with
5114 position=id=<id>. It can be applied to any netfilter.
5116 ``before``: insert before the specified filter.
5118 ``behind``: insert behind the specified filter (default).
5120 ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
5121 filter-mirror on netdev netdevid,mirror net packet to
5122 chardevchardevid, if it has the vnet\_hdr\_support flag,
5123 filter-mirror will mirror packet with vnet\_hdr\_len.
5125 ``-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]``
5126 filter-redirector on netdev netdevid,redirect filter's net
5127 packet to chardev chardevid,and redirect indev's packet to
5128 filter.if it has the vnet\_hdr\_support flag, filter-redirector
5129 will redirect packet with vnet\_hdr\_len. Create a
5130 filter-redirector we need to differ outdev id from indev id, id
5131 can not be the same. we can just use indev or outdev, but at
5132 least one of indev or outdev need to be specified.
5134 ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]``
5135 Filter-rewriter is a part of COLO project.It will rewrite tcp
5136 packet to secondary from primary to keep secondary tcp
5137 connection,and rewrite tcp packet to primary from secondary make
5138 tcp packet can be handled by client.if it has the
5139 vnet\_hdr\_support flag, we can parse packet with vnet header.
5141 usage: colo secondary: -object
5142 filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object
5143 filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object
5144 filter-rewriter,id=rew0,netdev=hn0,queue=all
5146 ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]``
5147 Dump the network traffic on netdev dev to the file specified by
5148 filename. At most len bytes (64k by default) per packet are
5149 stored. The file format is libpcap, so it can be analyzed with
5150 tools such as tcpdump or Wireshark.
5152 ``-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}]``
5153 Colo-compare gets packet from primary\_in chardevid and
5154 secondary\_in, then compare whether the payload of primary packet
5155 and secondary packet are the same. If same, it will output
5156 primary packet to out\_dev, else it will notify COLO-framework to do
5157 checkpoint and send primary packet to out\_dev. In order to
5158 improve efficiency, we need to put the task of comparison in
5159 another iothread. If it has the vnet\_hdr\_support flag,
5160 colo compare will send/recv packet with vnet\_hdr\_len.
5161 The compare\_timeout=@var{ms} determines the maximum time of the
5162 colo-compare hold the packet. The expired\_scan\_cycle=@var{ms}
5163 is to set the period of scanning expired primary node network packets.
5164 The max\_queue\_size=@var{size} is to set the max compare queue
5165 size depend on user environment.
5166 If user want to use Xen COLO, need to add the notify\_dev to
5167 notify Xen colo-frame to do checkpoint.
5169 COLO-compare must be used with the help of filter-mirror,
5170 filter-redirector and filter-rewriter.
5177 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
5178 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
5179 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
5180 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
5181 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
5182 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
5183 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
5184 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
5185 -object iothread,id=iothread1
5186 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
5187 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
5188 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
5189 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1
5192 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
5193 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
5194 -chardev socket,id=red0,host=3.3.3.3,port=9003
5195 -chardev socket,id=red1,host=3.3.3.3,port=9004
5196 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
5197 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
5203 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,downscript=/etc/qemu-ifdown
5204 -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66
5205 -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off
5206 -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off
5207 -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off
5208 -chardev socket,id=compare0-0,host=3.3.3.3,port=9001
5209 -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off
5210 -chardev socket,id=compare_out0,host=3.3.3.3,port=9005
5211 -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off
5212 -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0
5213 -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out
5214 -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0
5215 -object iothread,id=iothread1
5216 -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=nofity_way,iothread=iothread1
5219 -netdev tap,id=hn0,vhost=off,script=/etc/qemu-ifup,down script=/etc/qemu-ifdown
5220 -device e1000,netdev=hn0,mac=52:a4:00:12:78:66
5221 -chardev socket,id=red0,host=3.3.3.3,port=9003
5222 -chardev socket,id=red1,host=3.3.3.3,port=9004
5223 -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0
5224 -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1
5226 If you want to know the detail of above command line, you can
5227 read the colo-compare git log.
5229 ``-object cryptodev-backend-builtin,id=id[,queues=queues]``
5230 Creates a cryptodev backend which executes crypto opreation from
5231 the QEMU cipher APIS. The id parameter is a unique ID that will
5232 be used to reference this cryptodev backend from the
5233 ``virtio-crypto`` device. The queues parameter is optional,
5234 which specify the queue number of cryptodev backend, the default
5241 -object cryptodev-backend-builtin,id=cryptodev0 \\
5242 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
5245 ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]``
5246 Creates a vhost-user cryptodev backend, backed by a chardev
5247 chardevid. The id parameter is a unique ID that will be used to
5248 reference this cryptodev backend from the ``virtio-crypto``
5249 device. The chardev should be a unix domain socket backed one.
5250 The vhost-user uses a specifically defined protocol to pass
5251 vhost ioctl replacement messages to an application on the other
5252 end of the socket. The queues parameter is optional, which
5253 specify the queue number of cryptodev backend for multiqueue
5254 vhost-user, the default of queues is 1.
5260 -chardev socket,id=chardev0,path=/path/to/socket \\
5261 -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\
5262 -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\
5265 ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]``
5267 ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]``
5268 Defines a secret to store a password, encryption key, or some
5269 other sensitive data. The sensitive data can either be passed
5270 directly via the data parameter, or indirectly via the file
5271 parameter. Using the data parameter is insecure unless the
5272 sensitive data is encrypted.
5274 The sensitive data can be provided in raw format (the default),
5275 or base64. When encoded as JSON, the raw format only supports
5276 valid UTF-8 characters, so base64 is recommended for sending
5277 binary data. QEMU will convert from which ever format is
5278 provided to the format it needs internally. eg, an RBD password
5279 can be provided in raw format, even though it will be base64
5280 encoded when passed onto the RBD sever.
5282 For added protection, it is possible to encrypt the data
5283 associated with a secret using the AES-256-CBC cipher. Use of
5284 encryption is indicated by providing the keyid and iv
5285 parameters. The keyid parameter provides the ID of a previously
5286 defined secret that contains the AES-256 decryption key. This
5287 key should be 32-bytes long and be base64 encoded. The iv
5288 parameter provides the random initialization vector used for
5289 encryption of this particular secret and should be a base64
5290 encrypted string of the 16-byte IV.
5292 The simplest (insecure) usage is to provide the secret inline
5296 # |qemu_system| -object secret,id=sec0,data=letmein,format=raw
5298 The simplest secure usage is to provide the secret via a file
5300 # printf "letmein
" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object
5301 secret,id=sec0,file=mypasswd.txt,format=raw
5303 For greater security, AES-256-CBC should be used. To illustrate
5304 usage, consider the openssl command line tool which can encrypt
5305 the data. Note that when encrypting, the plaintext must be
5306 padded to the cipher block size (32 bytes) using the standard
5307 PKCS#5/6 compatible padding algorithm.
5309 First a master key needs to be created in base64 encoding:
5313 # openssl rand -base64 32 > key.b64
5314 # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X
"')
5316 Each secret to be encrypted needs to have a random
5317 initialization vector generated. These do not need to be kept
5322 # openssl rand -base64 16 > iv.b64
5323 # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X
"')
5325 The secret to be defined can now be encrypted, in this case
5326 we're telling openssl to base64 encode the result, but it could
5327 be left as raw bytes if desired.
5331 # SECRET=$(printf "letmein
" |
5332 openssl enc -aes-256-cbc -a -K $KEY -iv $IV)
5334 When launching QEMU, create a master secret pointing to
5335 ``key.b64`` and specify that to be used to decrypt the user
5336 password. Pass the contents of ``iv.b64`` to the second secret
5341 -object secret,id=secmaster0,format=base64,file=key.b64 \\
5342 -object secret,id=sec0,keyid=secmaster0,format=base64,\\
5343 data=$SECRET,iv=$(<iv.b64)
5345 ``-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]``
5346 Create a Secure Encrypted Virtualization (SEV) guest object,
5347 which can be used to provide the guest memory encryption support
5350 When memory encryption is enabled, one of the physical address
5351 bit (aka the C-bit) is utilized to mark if a memory page is
5352 protected. The ``cbitpos`` is used to provide the C-bit
5353 position. The C-bit position is Host family dependent hence user
5354 must provide this value. On EPYC, the value should be 47.
5356 When memory encryption is enabled, we loose certain bits in
5357 physical address space. The ``reduced-phys-bits`` is used to
5358 provide the number of bits we loose in physical address space.
5359 Similar to C-bit, the value is Host family dependent. On EPYC,
5360 the value should be 5.
5362 The ``sev-device`` provides the device file to use for
5363 communicating with the SEV firmware running inside AMD Secure
5364 Processor. The default device is '/dev/sev'. If hardware
5365 supports memory encryption then /dev/sev devices are created by
5368 The ``policy`` provides the guest policy to be enforced by the
5369 SEV firmware and restrict what configuration and operational
5370 commands can be performed on this guest by the hypervisor. The
5371 policy should be provided by the guest owner and is bound to the
5372 guest and cannot be changed throughout the lifetime of the
5373 guest. The default is 0.
5375 If guest ``policy`` allows sharing the key with another SEV
5376 guest then ``handle`` can be use to provide handle of the guest
5377 from which to share the key.
5379 The ``dh-cert-file`` and ``session-file`` provides the guest
5380 owner's Public Diffie-Hillman key defined in SEV spec. The PDH
5381 and session parameters are used for establishing a cryptographic
5382 session with the guest owner to negotiate keys used for
5383 attestation. The file must be encoded in base64.
5385 The ``kernel-hashes`` adds the hashes of given kernel/initrd/
5386 cmdline to a designated guest firmware page for measured Linux
5387 boot with -kernel. The default is off. (Since 6.2)
5389 e.g to launch a SEV guest
5393 # |qemu_system_x86| \\
5395 -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=5 \\
5396 -machine ...,memory-encryption=sev0 \\
5399 ``-object authz-simple,id=id,identity=string``
5400 Create an authorization object that will control access to
5403 The ``identity`` parameter is identifies the user and its format
5404 depends on the network service that authorization object is
5405 associated with. For authorizing based on TLS x509 certificates,
5406 the identity must be the x509 distinguished name. Note that care
5407 must be taken to escape any commas in the distinguished name.
5409 An example authorization object to validate a x509 distinguished
5410 name would look like:
5416 -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\
5419 Note the use of quotes due to the x509 distinguished name
5420 containing whitespace, and escaping of ','.
5422 ``-object authz-listfile,id=id,filename=path,refresh=on|off``
5423 Create an authorization object that will control access to
5426 The ``filename`` parameter is the fully qualified path to a file
5427 containing the access control list rules in JSON format.
5429 An example set of rules that match against SASL usernames might
5436 { "match
": "fred
", "policy
": "allow
", "format
": "exact
" },
5437 { "match
": "bob
", "policy
": "allow
", "format
": "exact
" },
5438 { "match
": "danb
", "policy
": "deny
", "format
": "glob
" },
5439 { "match
": "dan
*", "policy
": "allow
", "format
": "exact
" },
5444 When checking access the object will iterate over all the rules
5445 and the first rule to match will have its ``policy`` value
5446 returned as the result. If no rules match, then the default
5447 ``policy`` value is returned.
5449 The rules can either be an exact string match, or they can use
5450 the simple UNIX glob pattern matching to allow wildcards to be
5453 If ``refresh`` is set to true the file will be monitored and
5454 automatically reloaded whenever its content changes.
5456 As with the ``authz-simple`` object, the format of the identity
5457 strings being matched depends on the network service, but is
5458 usually a TLS x509 distinguished name, or a SASL username.
5460 An example authorization object to validate a SASL username
5467 -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\
5470 ``-object authz-pam,id=id,service=string``
5471 Create an authorization object that will control access to
5474 The ``service`` parameter provides the name of a PAM service to
5475 use for authorization. It requires that a file
5476 ``/etc/pam.d/service`` exist to provide the configuration for
5477 the ``account`` subsystem.
5479 An example authorization object to validate a TLS x509
5480 distinguished name would look like:
5486 -object authz-pam,id=auth0,service=qemu-vnc \\
5489 There would then be a corresponding config file for PAM at
5490 ``/etc/pam.d/qemu-vnc`` that contains:
5494 account requisite pam_listfile.so item=user sense=allow \
5495 file=/etc/qemu/vnc.allow
5497 Finally the ``/etc/qemu/vnc.allow`` file would contain the list
5498 of x509 distingished names that are permitted access
5502 CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB
5504 ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch``
5505 Creates a dedicated event loop thread that devices can be
5506 assigned to. This is known as an IOThread. By default device
5507 emulation happens in vCPU threads or the main event loop thread.
5508 This can become a scalability bottleneck. IOThreads allow device
5509 emulation and I/O to run on other host CPUs.
5511 The ``id`` parameter is a unique ID that will be used to
5512 reference this IOThread from ``-device ...,iothread=id``.
5513 Multiple devices can be assigned to an IOThread. Note that not
5514 all devices support an ``iothread`` parameter.
5516 The ``query-iothreads`` QMP command lists IOThreads and reports
5517 their thread IDs so that the user can configure host CPU
5520 IOThreads use an adaptive polling algorithm to reduce event loop
5521 latency. Instead of entering a blocking system call to monitor
5522 file descriptors and then pay the cost of being woken up when an
5523 event occurs, the polling algorithm spins waiting for events for
5524 a short time. The algorithm's default parameters are suitable
5525 for many cases but can be adjusted based on knowledge of the
5526 workload and/or host device latency.
5528 The ``poll-max-ns`` parameter is the maximum number of
5529 nanoseconds to busy wait for events. Polling can be disabled by
5530 setting this value to 0.
5532 The ``poll-grow`` parameter is the multiplier used to increase
5533 the polling time when the algorithm detects it is missing events
5534 due to not polling long enough.
5536 The ``poll-shrink`` parameter is the divisor used to decrease
5537 the polling time when the algorithm detects it is spending too
5538 long polling without encountering events.
5540 The ``aio-max-batch`` parameter is the maximum number of requests
5541 in a batch for the AIO engine, 0 means that the engine will use
5544 The IOThread parameters can be modified at run-time using the
5545 ``qom-set`` command (where ``iothread1`` is the IOThread's
5550 (qemu) qom-set /objects/iothread1 poll-max-ns 100000
5554 HXCOMM This is the last statement. Insert new options before this line!