Peter Maydell | d95f260 | 2023-12-12 16:23:13 +0000 | [diff] [blame] | 1 | HXCOMM See docs/devel/docs.rst for the format of this file. |
| 2 | HXCOMM |
Peter Maydell | 3c95fde | 2020-03-06 17:17:44 +0000 | [diff] [blame] | 3 | HXCOMM Use DEFHEADING() to define headings in both help text and rST. |
| 4 | HXCOMM Text between SRST and ERST is copied to the rST version and |
| 5 | HXCOMM discarded from C version. |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 6 | HXCOMM DEF(option, HAS_ARG/0, opt_enum, opt_help, arch_mask) is used to |
| 7 | HXCOMM construct option structures, enums and help message for specified |
| 8 | HXCOMM architectures. |
Peter Maydell | 3c95fde | 2020-03-06 17:17:44 +0000 | [diff] [blame] | 9 | HXCOMM HXCOMM can be used for comments, discarded from both rST and C. |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 10 | |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 11 | DEFHEADING(Standard options:) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 12 | |
| 13 | DEF("help", 0, QEMU_OPTION_h, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 14 | "-h or -help display this help and exit\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 15 | SRST |
| 16 | ``-h`` |
| 17 | Display help and exit |
| 18 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 19 | |
pbrook | 9bd7e6d | 2009-04-07 22:58:45 +0000 | [diff] [blame] | 20 | DEF("version", 0, QEMU_OPTION_version, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 21 | "-version display version information and exit\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 22 | SRST |
| 23 | ``-version`` |
| 24 | Display version information and exit |
| 25 | ERST |
pbrook | 9bd7e6d | 2009-04-07 22:58:45 +0000 | [diff] [blame] | 26 | |
Jan Kiszka | 80f52a6 | 2011-07-23 12:39:46 +0200 | [diff] [blame] | 27 | DEF("machine", HAS_ARG, QEMU_OPTION_machine, \ |
| 28 | "-machine [type=]name[,prop[=value][,...]]\n" |
Peter Maydell | 585f603 | 2012-10-04 16:22:01 +0100 | [diff] [blame] | 29 | " selects emulated machine ('-machine help' for list)\n" |
Jan Kiszka | 80f52a6 | 2011-07-23 12:39:46 +0200 | [diff] [blame] | 30 | " property accel=accel1[:accel2[:...]] selects accelerator\n" |
Philippe Mathieu-Daudé | b91b0fc | 2023-06-24 00:52:29 +0200 | [diff] [blame] | 31 | " supported accelerators are kvm, xen, hvf, nvmm, whpx or tcg (default: tcg)\n" |
Don Slutz | d1048be | 2014-11-21 11:18:52 -0500 | [diff] [blame] | 32 | " vmport=on|off|auto controls emulation of vmport (default: auto)\n" |
Luiz Capitulino | 8490fc7 | 2012-09-05 16:50:16 -0300 | [diff] [blame] | 33 | " dump-guest-core=on|off include guest memory in a core dump (default=on)\n" |
Le Tan | a52a7fd | 2014-08-16 13:55:40 +0800 | [diff] [blame] | 34 | " mem-merge=on|off controls memory merge support (default: on)\n" |
Tony Krowiak | 2eb1cd0 | 2015-03-12 13:53:51 +0100 | [diff] [blame] | 35 | " aes-key-wrap=on|off controls support for AES key wrapping (default=on)\n" |
Alexander Graf | 9850c60 | 2015-02-23 13:56:42 +0100 | [diff] [blame] | 36 | " dea-key-wrap=on|off controls support for DEA key wrapping (default=on)\n" |
Xiao Guangrong | 87252e1 | 2015-12-02 15:20:58 +0800 | [diff] [blame] | 37 | " suppress-vmdesc=on|off disables self-describing migration (default=off)\n" |
Greg Kurz | 902c053 | 2016-02-18 12:32:25 +0100 | [diff] [blame] | 38 | " nvdimm=on|off controls NVDIMM support (default=off)\n" |
Tao Xu | 244b3f4 | 2019-12-13 09:19:22 +0800 | [diff] [blame] | 39 | " memory-encryption=@var{} memory encryption object to use (default=none)\n" |
Igor Mammedov | 8db0b20 | 2021-01-21 11:15:04 -0500 | [diff] [blame] | 40 | " hmat=on|off controls ACPI HMAT support (default=off)\n" |
Jonathan Cameron | 03b39fc | 2022-06-08 15:54:33 +0100 | [diff] [blame] | 41 | " memory-backend='backend-id' specifies explicitly provided backend for main RAM (default=none)\n" |
Cédric Le Goater | 5770289 | 2022-11-07 17:13:48 +0100 | [diff] [blame] | 42 | " cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]\n", |
Jan Kiszka | 80f52a6 | 2011-07-23 12:39:46 +0200 | [diff] [blame] | 43 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 44 | SRST |
| 45 | ``-machine [type=]name[,prop=value[,...]]`` |
| 46 | Select the emulated machine by name. Use ``-machine help`` to list |
| 47 | available machines. |
| 48 | |
| 49 | For architectures which aim to support live migration compatibility |
| 50 | across releases, each release will introduce a new versioned machine |
| 51 | type. For example, the 2.8.0 release introduced machine types |
| 52 | "pc-i440fx-2.8" and "pc-q35-2.8" for the x86\_64/i686 architectures. |
| 53 | |
| 54 | To allow live migration of guests from QEMU version 2.8.0, to QEMU |
| 55 | version 2.9.0, the 2.9.0 version must support the "pc-i440fx-2.8" |
| 56 | and "pc-q35-2.8" machines too. To allow users live migrating VMs to |
| 57 | skip multiple intermediate releases when upgrading, new releases of |
| 58 | QEMU will support machine types from many previous versions. |
| 59 | |
| 60 | Supported machine properties are: |
| 61 | |
| 62 | ``accel=accels1[:accels2[:...]]`` |
| 63 | This is used to enable an accelerator. Depending on the target |
Philippe Mathieu-Daudé | b91b0fc | 2023-06-24 00:52:29 +0200 | [diff] [blame] | 64 | architecture, kvm, xen, hvf, nvmm, whpx or tcg can be available. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 65 | By default, tcg is used. If there is more than one accelerator |
| 66 | specified, the next one is used if the previous one fails to |
| 67 | initialize. |
| 68 | |
| 69 | ``vmport=on|off|auto`` |
| 70 | Enables emulation of VMWare IO port, for vmmouse etc. auto says |
| 71 | to select the value based on accel. For accel=xen the default is |
| 72 | off otherwise the default is on. |
| 73 | |
| 74 | ``dump-guest-core=on|off`` |
| 75 | Include guest memory in a core dump. The default is on. |
| 76 | |
| 77 | ``mem-merge=on|off`` |
| 78 | Enables or disables memory merge support. This feature, when |
| 79 | supported by the host, de-duplicates identical memory pages |
| 80 | among VMs instances (enabled by default). |
| 81 | |
| 82 | ``aes-key-wrap=on|off`` |
| 83 | Enables or disables AES key wrapping support on s390-ccw hosts. |
| 84 | This feature controls whether AES wrapping keys will be created |
| 85 | to allow execution of AES cryptographic functions. The default |
| 86 | is on. |
| 87 | |
| 88 | ``dea-key-wrap=on|off`` |
| 89 | Enables or disables DEA key wrapping support on s390-ccw hosts. |
| 90 | This feature controls whether DEA wrapping keys will be created |
| 91 | to allow execution of DEA cryptographic functions. The default |
| 92 | is on. |
| 93 | |
| 94 | ``nvdimm=on|off`` |
| 95 | Enables or disables NVDIMM support. The default is off. |
| 96 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 97 | ``memory-encryption=`` |
| 98 | Memory encryption object to use. The default is none. |
| 99 | |
| 100 | ``hmat=on|off`` |
| 101 | Enables or disables ACPI Heterogeneous Memory Attribute Table |
| 102 | (HMAT) support. The default is off. |
Igor Mammedov | 8db0b20 | 2021-01-21 11:15:04 -0500 | [diff] [blame] | 103 | |
Peter Maydell | 9535582 | 2021-07-19 11:52:57 +0100 | [diff] [blame] | 104 | ``memory-backend='id'`` |
Igor Mammedov | 8db0b20 | 2021-01-21 11:15:04 -0500 | [diff] [blame] | 105 | An alternative to legacy ``-mem-path`` and ``mem-prealloc`` options. |
| 106 | Allows to use a memory backend as main RAM. |
| 107 | |
| 108 | For example: |
| 109 | :: |
Peter Maydell | 9535582 | 2021-07-19 11:52:57 +0100 | [diff] [blame] | 110 | |
| 111 | -object memory-backend-file,id=pc.ram,size=512M,mem-path=/hugetlbfs,prealloc=on,share=on |
| 112 | -machine memory-backend=pc.ram |
| 113 | -m 512M |
Igor Mammedov | 8db0b20 | 2021-01-21 11:15:04 -0500 | [diff] [blame] | 114 | |
| 115 | Migration compatibility note: |
Peter Maydell | 9535582 | 2021-07-19 11:52:57 +0100 | [diff] [blame] | 116 | |
| 117 | * as backend id one shall use value of 'default-ram-id', advertised by |
| 118 | machine type (available via ``query-machines`` QMP command), if migration |
| 119 | to/from old QEMU (<5.0) is expected. |
| 120 | * for machine types 4.0 and older, user shall |
| 121 | use ``x-use-canonical-path-for-ramblock-id=off`` backend option |
| 122 | if migration to/from old QEMU (<5.0) is expected. |
| 123 | |
Igor Mammedov | 8db0b20 | 2021-01-21 11:15:04 -0500 | [diff] [blame] | 124 | For example: |
| 125 | :: |
Peter Maydell | 9535582 | 2021-07-19 11:52:57 +0100 | [diff] [blame] | 126 | |
| 127 | -object memory-backend-ram,id=pc.ram,size=512M,x-use-canonical-path-for-ramblock-id=off |
| 128 | -machine memory-backend=pc.ram |
| 129 | -m 512M |
Jonathan Cameron | 03b39fc | 2022-06-08 15:54:33 +0100 | [diff] [blame] | 130 | |
| 131 | ``cxl-fmw.0.targets.0=firsttarget,cxl-fmw.0.targets.1=secondtarget,cxl-fmw.0.size=size[,cxl-fmw.0.interleave-granularity=granularity]`` |
| 132 | Define a CXL Fixed Memory Window (CFMW). |
| 133 | |
| 134 | Described in the CXL 2.0 ECN: CEDT CFMWS & QTG _DSM. |
| 135 | |
| 136 | They are regions of Host Physical Addresses (HPA) on a system which |
| 137 | may be interleaved across one or more CXL host bridges. The system |
| 138 | software will assign particular devices into these windows and |
| 139 | configure the downstream Host-managed Device Memory (HDM) decoders |
| 140 | in root ports, switch ports and devices appropriately to meet the |
| 141 | interleave requirements before enabling the memory devices. |
| 142 | |
| 143 | ``targets.X=target`` provides the mapping to CXL host bridges |
Stefan Weil | 2cb40d4 | 2022-11-10 20:08:25 +0100 | [diff] [blame] | 144 | which may be identified by the id provided in the -device entry. |
Jonathan Cameron | 03b39fc | 2022-06-08 15:54:33 +0100 | [diff] [blame] | 145 | Multiple entries are needed to specify all the targets when |
| 146 | the fixed memory window represents interleaved memory. X is the |
| 147 | target index from 0. |
| 148 | |
| 149 | ``size=size`` sets the size of the CFMW. This must be a multiple of |
| 150 | 256MiB. The region will be aligned to 256MiB but the location is |
| 151 | platform and configuration dependent. |
| 152 | |
| 153 | ``interleave-granularity=granularity`` sets the granularity of |
| 154 | interleave. Default 256KiB. Only 256KiB, 512KiB, 1024KiB, 2048KiB |
| 155 | 4096KiB, 8192KiB and 16384KiB granularities supported. |
| 156 | |
| 157 | Example: |
| 158 | |
| 159 | :: |
| 160 | |
| 161 | -machine cxl-fmw.0.targets.0=cxl.0,cxl-fmw.0.targets.1=cxl.1,cxl-fmw.0.size=128G,cxl-fmw.0.interleave-granularity=512k |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 162 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 163 | |
Sean Christopherson | dfce81f | 2021-09-28 10:40:58 +0200 | [diff] [blame] | 164 | DEF("M", HAS_ARG, QEMU_OPTION_M, |
Yang Zhong | 1105812 | 2021-11-01 12:20:05 -0400 | [diff] [blame] | 165 | " sgx-epc.0.memdev=memid,sgx-epc.0.node=numaid\n", |
Sean Christopherson | dfce81f | 2021-09-28 10:40:58 +0200 | [diff] [blame] | 166 | QEMU_ARCH_ALL) |
| 167 | |
| 168 | SRST |
Yang Zhong | 1105812 | 2021-11-01 12:20:05 -0400 | [diff] [blame] | 169 | ``sgx-epc.0.memdev=@var{memid},sgx-epc.0.node=@var{numaid}`` |
Sean Christopherson | dfce81f | 2021-09-28 10:40:58 +0200 | [diff] [blame] | 170 | Define an SGX EPC section. |
| 171 | ERST |
Jan Kiszka | 80f52a6 | 2011-07-23 12:39:46 +0200 | [diff] [blame] | 172 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 173 | DEF("cpu", HAS_ARG, QEMU_OPTION_cpu, |
Peter Maydell | 585f603 | 2012-10-04 16:22:01 +0100 | [diff] [blame] | 174 | "-cpu cpu select CPU ('-cpu help' for list)\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 175 | SRST |
| 176 | ``-cpu model`` |
| 177 | Select CPU model (``-cpu help`` for list and additional feature |
| 178 | selection) |
| 179 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 180 | |
KONRAD Frederic | 8d4e914 | 2017-02-23 18:29:08 +0000 | [diff] [blame] | 181 | DEF("accel", HAS_ARG, QEMU_OPTION_accel, |
Paolo Bonzini | fe17413 | 2019-11-13 15:16:44 +0100 | [diff] [blame] | 182 | "-accel [accel=]accelerator[,prop[=value][,...]]\n" |
Philippe Mathieu-Daudé | b91b0fc | 2023-06-24 00:52:29 +0200 | [diff] [blame] | 183 | " select accelerator (kvm, xen, hvf, nvmm, whpx or tcg; use 'help' for a list)\n" |
Paolo Bonzini | 46472d8 | 2019-11-13 10:56:53 +0100 | [diff] [blame] | 184 | " igd-passthru=on|off (enable Xen integrated Intel graphics passthrough, default=off)\n" |
Paolo Bonzini | 11bc4a1 | 2019-11-13 10:56:53 +0100 | [diff] [blame] | 185 | " kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n" |
Paolo Bonzini | 23b0898 | 2019-11-13 10:56:53 +0100 | [diff] [blame] | 186 | " kvm-shadow-mem=size of KVM shadow MMU in bytes\n" |
Peter Maydell | 3cfb045 | 2023-04-17 17:40:32 +0100 | [diff] [blame] | 187 | " one-insn-per-tb=on|off (one guest instruction per TCG translation block)\n" |
Richard Henderson | a35b3e1 | 2020-10-28 20:50:29 -0700 | [diff] [blame] | 188 | " split-wx=on|off (enable TCG split w^x mapping)\n" |
Paolo Bonzini | fe17413 | 2019-11-13 15:16:44 +0100 | [diff] [blame] | 189 | " tb-size=n (TCG translation block cache size)\n" |
Peter Xu | 2ea5cb0 | 2021-05-06 12:05:47 -0400 | [diff] [blame] | 190 | " dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n" |
Shameer Kolothum | c8f2eb5 | 2023-09-05 10:12:46 +0100 | [diff] [blame] | 191 | " eager-split-size=n (KVM Eager Page Split chunk size, default 0, disabled. ARM only)\n" |
Chenyi Qiang | e2e69f6 | 2022-09-29 15:20:14 +0800 | [diff] [blame] | 192 | " notify-vmexit=run|internal-error|disable,notify-window=n (enable notify VM exit and set notify window, x86 only)\n" |
Daan De Meyer | aef158b | 2023-10-21 15:40:15 +0200 | [diff] [blame] | 193 | " thread=single|multi (enable multi-threaded TCG)\n" |
| 194 | " device=path (KVM device path, default /dev/kvm)\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 195 | SRST |
| 196 | ``-accel name[,prop=value[,...]]`` |
| 197 | This is used to enable an accelerator. Depending on the target |
Philippe Mathieu-Daudé | b91b0fc | 2023-06-24 00:52:29 +0200 | [diff] [blame] | 198 | architecture, kvm, xen, hvf, nvmm, whpx or tcg can be available. By |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 199 | default, tcg is used. If there is more than one accelerator |
| 200 | specified, the next one is used if the previous one fails to |
| 201 | initialize. |
| 202 | |
| 203 | ``igd-passthru=on|off`` |
| 204 | When Xen is in use, this option controls whether Intel |
| 205 | integrated graphics devices can be passed through to the guest |
| 206 | (default=off) |
| 207 | |
| 208 | ``kernel-irqchip=on|off|split`` |
| 209 | Controls KVM in-kernel irqchip support. The default is full |
| 210 | acceleration of the interrupt controllers. On x86, split irqchip |
| 211 | reduces the kernel attack surface, at a performance cost for |
| 212 | non-MSI interrupts. Disabling the in-kernel irqchip completely |
| 213 | is not recommended except for debugging purposes. |
| 214 | |
| 215 | ``kvm-shadow-mem=size`` |
| 216 | Defines the size of the KVM shadow MMU. |
| 217 | |
Peter Maydell | 3cfb045 | 2023-04-17 17:40:32 +0100 | [diff] [blame] | 218 | ``one-insn-per-tb=on|off`` |
| 219 | Makes the TCG accelerator put only one guest instruction into |
| 220 | each translation block. This slows down emulation a lot, but |
| 221 | can be useful in some situations, such as when trying to analyse |
| 222 | the logs produced by the ``-d`` option. |
| 223 | |
Richard Henderson | a35b3e1 | 2020-10-28 20:50:29 -0700 | [diff] [blame] | 224 | ``split-wx=on|off`` |
| 225 | Controls the use of split w^x mapping for the TCG code generation |
| 226 | buffer. Some operating systems require this to be enabled, and in |
| 227 | such a case this will default on. On other operating systems, this |
| 228 | will default off, but one may enable this for testing or debugging. |
| 229 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 230 | ``tb-size=n`` |
| 231 | Controls the size (in MiB) of the TCG translation block cache. |
| 232 | |
| 233 | ``thread=single|multi`` |
| 234 | Controls number of TCG threads. When the TCG is multi-threaded |
Michael Tokarev | cba42d6 | 2021-03-09 14:15:10 +0300 | [diff] [blame] | 235 | there will be one thread per vCPU therefore taking advantage of |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 236 | additional host cores. The default is to enable multi-threading |
| 237 | where both the back-end and front-ends support it and no |
| 238 | incompatible TCG features have been enabled (e.g. |
| 239 | icount/replay). |
Peter Xu | 2ea5cb0 | 2021-05-06 12:05:47 -0400 | [diff] [blame] | 240 | |
| 241 | ``dirty-ring-size=n`` |
| 242 | When the KVM accelerator is used, it controls the size of the per-vCPU |
| 243 | dirty page ring buffer (number of entries for each vCPU). It should |
| 244 | be a value that is power of two, and it should be 1024 or bigger (but |
| 245 | still less than the maximum value that the kernel supports). 4096 |
| 246 | could be a good initial value if you have no idea which is the best. |
| 247 | Set this value to 0 to disable the feature. By default, this feature |
| 248 | is disabled (dirty-ring-size=0). When enabled, KVM will instead |
| 249 | record dirty pages in a bitmap. |
| 250 | |
Shameer Kolothum | c8f2eb5 | 2023-09-05 10:12:46 +0100 | [diff] [blame] | 251 | ``eager-split-size=n`` |
| 252 | KVM implements dirty page logging at the PAGE_SIZE granularity and |
| 253 | enabling dirty-logging on a huge-page requires breaking it into |
| 254 | PAGE_SIZE pages in the first place. KVM on ARM does this splitting |
| 255 | lazily by default. There are performance benefits in doing huge-page |
| 256 | split eagerly, especially in situations where TLBI costs associated |
| 257 | with break-before-make sequences are considerable and also if guest |
| 258 | workloads are read intensive. The size here specifies how many pages |
| 259 | to break at a time and needs to be a valid block size which is |
| 260 | 1GB/2MB/4KB, 32MB/16KB and 512MB/64KB for 4KB/16KB/64KB PAGE_SIZE |
| 261 | respectively. Be wary of specifying a higher size as it will have an |
| 262 | impact on the memory. By default, this feature is disabled |
| 263 | (eager-split-size=0). |
| 264 | |
Chenyi Qiang | e2e69f6 | 2022-09-29 15:20:14 +0800 | [diff] [blame] | 265 | ``notify-vmexit=run|internal-error|disable,notify-window=n`` |
| 266 | Enables or disables notify VM exit support on x86 host and specify |
| 267 | the corresponding notify window to trigger the VM exit if enabled. |
| 268 | ``run`` option enables the feature. It does nothing and continue |
| 269 | if the exit happens. ``internal-error`` option enables the feature. |
| 270 | It raises a internal error. ``disable`` option doesn't enable the feature. |
| 271 | This feature can mitigate the CPU stuck issue due to event windows don't |
| 272 | open up for a specified of time (i.e. notify-window). |
| 273 | Default: notify-vmexit=run,notify-window=0. |
| 274 | |
Daan De Meyer | aef158b | 2023-10-21 15:40:15 +0200 | [diff] [blame] | 275 | ``device=path`` |
| 276 | Sets the path to the KVM device node. Defaults to ``/dev/kvm``. This |
| 277 | option can be used to pass the KVM device to use via a file descriptor |
| 278 | by setting the value to ``/dev/fdset/NN``. |
| 279 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 280 | ERST |
KONRAD Frederic | 8d4e914 | 2017-02-23 18:29:08 +0000 | [diff] [blame] | 281 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 282 | DEF("smp", HAS_ARG, QEMU_OPTION_smp, |
Pierre Morel | 5de1aff | 2023-10-16 20:39:06 +0200 | [diff] [blame] | 283 | "-smp [[cpus=]n][,maxcpus=maxcpus][,drawers=drawers][,books=books][,sockets=sockets]\n" |
| 284 | " [,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]\n" |
Yanan Wang | 0d87178 | 2021-12-28 17:22:08 +0800 | [diff] [blame] | 285 | " set the number of initial CPUs to 'n' [default=1]\n" |
Daniel P. Berrangé | ce8ee7c | 2021-06-22 15:30:43 +0100 | [diff] [blame] | 286 | " maxcpus= maximum number of total CPUs, including\n" |
Bruce Rogers | ca1a8a0 | 2010-01-06 12:33:57 -0700 | [diff] [blame] | 287 | " offline CPUs for hotplug, etc\n" |
Pierre Morel | 5de1aff | 2023-10-16 20:39:06 +0200 | [diff] [blame] | 288 | " drawers= number of drawers on the machine board\n" |
| 289 | " books= number of books in one drawer\n" |
| 290 | " sockets= number of sockets in one book\n" |
Yanan Wang | 0d87178 | 2021-12-28 17:22:08 +0800 | [diff] [blame] | 291 | " dies= number of dies in one socket\n" |
Yanan Wang | 864c3b5 | 2021-12-28 17:22:09 +0800 | [diff] [blame] | 292 | " clusters= number of clusters in one die\n" |
| 293 | " cores= number of cores in one cluster\n" |
Yanan Wang | 0d87178 | 2021-12-28 17:22:08 +0800 | [diff] [blame] | 294 | " threads= number of threads in one core\n" |
| 295 | "Note: Different machines may have different subsets of the CPU topology\n" |
| 296 | " parameters supported, so the actual meaning of the supported parameters\n" |
| 297 | " will vary accordingly. For example, for a machine type that supports a\n" |
| 298 | " three-level CPU hierarchy of sockets/cores/threads, the parameters will\n" |
| 299 | " sequentially mean as below:\n" |
| 300 | " sockets means the number of sockets on the machine board\n" |
| 301 | " cores means the number of cores in one socket\n" |
| 302 | " threads means the number of threads in one core\n" |
| 303 | " For a particular machine type board, an expected CPU topology hierarchy\n" |
| 304 | " can be defined through the supported sub-option. Unsupported parameters\n" |
| 305 | " can also be provided in addition to the sub-option, but their values\n" |
| 306 | " must be set as 1 in the purpose of correct parsing.\n", |
| 307 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 308 | SRST |
Yanan Wang | 864c3b5 | 2021-12-28 17:22:09 +0800 | [diff] [blame] | 309 | ``-smp [[cpus=]n][,maxcpus=maxcpus][,sockets=sockets][,dies=dies][,clusters=clusters][,cores=cores][,threads=threads]`` |
Daniel P. Berrangé | 80d7835 | 2021-06-22 16:17:09 +0100 | [diff] [blame] | 310 | Simulate a SMP system with '\ ``n``\ ' CPUs initially present on |
| 311 | the machine type board. On boards supporting CPU hotplug, the optional |
| 312 | '\ ``maxcpus``\ ' parameter can be set to enable further CPUs to be |
Yanan Wang | 7d8c5a3 | 2021-09-29 10:58:05 +0800 | [diff] [blame] | 313 | added at runtime. When both parameters are omitted, the maximum number |
| 314 | of CPUs will be calculated from the provided topology members and the |
| 315 | initial CPU count will match the maximum number. When only one of them |
| 316 | is given then the omitted one will be set to its counterpart's value. |
| 317 | Both parameters may be specified, but the maximum number of CPUs must |
Yanan Wang | 0d87178 | 2021-12-28 17:22:08 +0800 | [diff] [blame] | 318 | be equal to or greater than the initial CPU count. Product of the |
| 319 | CPU topology hierarchy must be equal to the maximum number of CPUs. |
| 320 | Both parameters are subject to an upper limit that is determined by |
| 321 | the specific machine type chosen. |
Daniel P. Berrangé | 80d7835 | 2021-06-22 16:17:09 +0100 | [diff] [blame] | 322 | |
Yanan Wang | 0d87178 | 2021-12-28 17:22:08 +0800 | [diff] [blame] | 323 | To control reporting of CPU topology information, values of the topology |
| 324 | parameters can be specified. Machines may only support a subset of the |
| 325 | parameters and different machines may have different subsets supported |
| 326 | which vary depending on capacity of the corresponding CPU targets. So |
| 327 | for a particular machine type board, an expected topology hierarchy can |
| 328 | be defined through the supported sub-option. Unsupported parameters can |
| 329 | also be provided in addition to the sub-option, but their values must be |
| 330 | set as 1 in the purpose of correct parsing. |
Daniel P. Berrangé | 80d7835 | 2021-06-22 16:17:09 +0100 | [diff] [blame] | 331 | |
| 332 | Either the initial CPU count, or at least one of the topology parameters |
Yanan Wang | c2511b1 | 2021-09-29 10:58:02 +0800 | [diff] [blame] | 333 | must be specified. The specified parameters must be greater than zero, |
| 334 | explicit configuration like "cpus=0" is not allowed. Values for any |
| 335 | omitted parameters will be computed from those which are given. |
Yanan Wang | 0d87178 | 2021-12-28 17:22:08 +0800 | [diff] [blame] | 336 | |
| 337 | For example, the following sub-option defines a CPU topology hierarchy |
| 338 | (2 sockets totally on the machine, 2 cores per socket, 2 threads per |
| 339 | core) for a machine that only supports sockets/cores/threads. |
| 340 | Some members of the option can be omitted but their values will be |
| 341 | automatically computed: |
| 342 | |
| 343 | :: |
| 344 | |
| 345 | -smp 8,sockets=2,cores=2,threads=2,maxcpus=8 |
| 346 | |
| 347 | The following sub-option defines a CPU topology hierarchy (2 sockets |
| 348 | totally on the machine, 2 dies per socket, 2 cores per die, 2 threads |
| 349 | per core) for PC machines which support sockets/dies/cores/threads. |
| 350 | Some members of the option can be omitted but their values will be |
| 351 | automatically computed: |
| 352 | |
| 353 | :: |
| 354 | |
| 355 | -smp 16,sockets=2,dies=2,cores=2,threads=2,maxcpus=16 |
| 356 | |
Yanan Wang | d55c316 | 2022-01-07 16:32:27 +0800 | [diff] [blame] | 357 | The following sub-option defines a CPU topology hierarchy (2 sockets |
| 358 | totally on the machine, 2 clusters per socket, 2 cores per cluster, |
| 359 | 2 threads per core) for ARM virt machines which support sockets/clusters |
| 360 | /cores/threads. Some members of the option can be omitted but their values |
| 361 | will be automatically computed: |
| 362 | |
| 363 | :: |
| 364 | |
| 365 | -smp 16,sockets=2,clusters=2,cores=2,threads=2,maxcpus=16 |
| 366 | |
Yanan Wang | c2511b1 | 2021-09-29 10:58:02 +0800 | [diff] [blame] | 367 | Historically preference was given to the coarsest topology parameters |
| 368 | when computing missing values (ie sockets preferred over cores, which |
| 369 | were preferred over threads), however, this behaviour is considered |
Yanan Wang | 4a0af29 | 2021-09-29 10:58:09 +0800 | [diff] [blame] | 370 | liable to change. Prior to 6.2 the preference was sockets over cores |
| 371 | over threads. Since 6.2 the preference is cores over sockets over threads. |
Yanan Wang | 0d87178 | 2021-12-28 17:22:08 +0800 | [diff] [blame] | 372 | |
| 373 | For example, the following option defines a machine board with 2 sockets |
| 374 | of 1 core before 6.2 and 1 socket of 2 cores after 6.2: |
| 375 | |
| 376 | :: |
| 377 | |
| 378 | -smp 2 |
Yicong Yang | 97f4eff | 2022-12-29 14:55:09 +0800 | [diff] [blame] | 379 | |
| 380 | Note: The cluster topology will only be generated in ACPI and exposed |
| 381 | to guest if it's explicitly specified in -smp. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 382 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 383 | |
aliguori | 268a362 | 2009-04-21 22:30:27 +0000 | [diff] [blame] | 384 | DEF("numa", HAS_ARG, QEMU_OPTION_numa, |
Tao Xu | 244b3f4 | 2019-12-13 09:19:22 +0800 | [diff] [blame] | 385 | "-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" |
| 386 | "-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=node]\n" |
Igor Mammedov | 2d19c65 | 2017-11-28 15:53:58 +0100 | [diff] [blame] | 387 | "-numa dist,src=source,dst=destination,val=distance\n" |
Liu Jingqi | 9b12dfa | 2019-12-13 09:19:23 +0800 | [diff] [blame] | 388 | "-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]\n" |
Liu Jingqi | c412a48 | 2019-12-13 09:19:24 +0800 | [diff] [blame] | 389 | "-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" |
| 390 | "-numa hmat-cache,node-id=node,size=size,level=level[,associativity=none|direct|complex][,policy=none|write-back|write-through][,line=size]\n", |
Igor Mammedov | 2d19c65 | 2017-11-28 15:53:58 +0100 | [diff] [blame] | 391 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 392 | SRST |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 393 | ``-numa node[,mem=size][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` |
| 394 | \ |
| 395 | ``-numa node[,memdev=id][,cpus=firstcpu[-lastcpu]][,nodeid=node][,initiator=initiator]`` |
| 396 | \ |
| 397 | ``-numa dist,src=source,dst=destination,val=distance`` |
| 398 | \ |
| 399 | ``-numa cpu,node-id=node[,socket-id=x][,core-id=y][,thread-id=z]`` |
| 400 | \ |
Stefan Weil | 2cb40d4 | 2022-11-10 20:08:25 +0100 | [diff] [blame] | 401 | ``-numa hmat-lb,initiator=node,target=node,hierarchy=hierarchy,data-type=type[,latency=lat][,bandwidth=bw]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 402 | \ |
| 403 | ``-numa hmat-cache,node-id=node,size=size,level=level[,associativity=str][,policy=str][,line=size]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 404 | Define a NUMA node and assign RAM and VCPUs to it. Set the NUMA |
| 405 | distance from a source node to a destination node. Set the ACPI |
| 406 | Heterogeneous Memory Attributes for the given nodes. |
| 407 | |
| 408 | Legacy VCPU assignment uses '\ ``cpus``\ ' option where firstcpu and |
| 409 | lastcpu are CPU indexes. Each '\ ``cpus``\ ' option represent a |
| 410 | contiguous range of CPU indexes (or a single VCPU if lastcpu is |
| 411 | omitted). A non-contiguous set of VCPUs can be represented by |
| 412 | providing multiple '\ ``cpus``\ ' options. If '\ ``cpus``\ ' is |
| 413 | omitted on all nodes, VCPUs are automatically split between them. |
| 414 | |
| 415 | For example, the following option assigns VCPUs 0, 1, 2 and 5 to a |
| 416 | NUMA node: |
| 417 | |
| 418 | :: |
| 419 | |
| 420 | -numa node,cpus=0-2,cpus=5 |
| 421 | |
| 422 | '\ ``cpu``\ ' option is a new alternative to '\ ``cpus``\ ' option |
| 423 | which uses '\ ``socket-id|core-id|thread-id``\ ' properties to |
| 424 | assign CPU objects to a node using topology layout properties of |
| 425 | CPU. The set of properties is machine specific, and depends on used |
| 426 | machine type/'\ ``smp``\ ' options. It could be queried with |
| 427 | '\ ``hotpluggable-cpus``\ ' monitor command. '\ ``node-id``\ ' |
| 428 | property specifies node to which CPU object will be assigned, it's |
| 429 | required for node to be declared with '\ ``node``\ ' option before |
| 430 | it's used with '\ ``cpu``\ ' option. |
| 431 | |
| 432 | For example: |
| 433 | |
| 434 | :: |
| 435 | |
| 436 | -M pc \ |
| 437 | -smp 1,sockets=2,maxcpus=2 \ |
| 438 | -numa node,nodeid=0 -numa node,nodeid=1 \ |
| 439 | -numa cpu,node-id=0,socket-id=0 -numa cpu,node-id=1,socket-id=1 |
| 440 | |
Yohei Kojima | 4f51398 | 2023-04-24 10:22:45 +0100 | [diff] [blame] | 441 | '\ ``memdev``\ ' option assigns RAM from a given memory backend |
| 442 | device to a node. It is recommended to use '\ ``memdev``\ ' option |
| 443 | over legacy '\ ``mem``\ ' option. This is because '\ ``memdev``\ ' |
| 444 | option provides better performance and more control over the |
| 445 | backend's RAM (e.g. '\ ``prealloc``\ ' parameter of |
| 446 | '\ ``-memory-backend-ram``\ ' allows memory preallocation). |
Igor Mammedov | 32a354d | 2020-06-09 09:56:35 -0400 | [diff] [blame] | 447 | |
Yohei Kojima | 4f51398 | 2023-04-24 10:22:45 +0100 | [diff] [blame] | 448 | For compatibility reasons, legacy '\ ``mem``\ ' option is |
| 449 | supported in 5.0 and older machine types. Note that '\ ``mem``\ ' |
| 450 | and '\ ``memdev``\ ' are mutually exclusive. If one node uses |
| 451 | '\ ``memdev``\ ', the rest nodes have to use '\ ``memdev``\ ' |
| 452 | option, and vice versa. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 453 | |
Yohei Kojima | 4f51398 | 2023-04-24 10:22:45 +0100 | [diff] [blame] | 454 | Users must specify memory for all NUMA nodes by '\ ``memdev``\ ' |
| 455 | (or legacy '\ ``mem``\ ' if available). In QEMU 5.2, the support |
| 456 | for '\ ``-numa node``\ ' without memory specified was removed. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 457 | |
| 458 | '\ ``initiator``\ ' is an additional option that points to an |
| 459 | initiator NUMA node that has best performance (the lowest latency or |
| 460 | largest bandwidth) to this NUMA node. Note that this option can be |
| 461 | set only when the machine property 'hmat' is set to 'on'. |
| 462 | |
| 463 | Following example creates a machine with 2 NUMA nodes, node 0 has |
| 464 | CPU. node 1 has only memory, and its initiator is node 0. Note that |
| 465 | because node 0 has CPU, by default the initiator of node 0 is itself |
| 466 | and must be itself. |
| 467 | |
| 468 | :: |
| 469 | |
| 470 | -machine hmat=on \ |
| 471 | -m 2G,slots=2,maxmem=4G \ |
| 472 | -object memory-backend-ram,size=1G,id=m0 \ |
| 473 | -object memory-backend-ram,size=1G,id=m1 \ |
| 474 | -numa node,nodeid=0,memdev=m0 \ |
| 475 | -numa node,nodeid=1,memdev=m1,initiator=0 \ |
| 476 | -smp 2,sockets=2,maxcpus=2 \ |
| 477 | -numa cpu,node-id=0,socket-id=0 \ |
| 478 | -numa cpu,node-id=0,socket-id=1 |
| 479 | |
| 480 | source and destination are NUMA node IDs. distance is the NUMA |
| 481 | distance from source to destination. The distance from a node to |
| 482 | itself is always 10. If any pair of nodes is given a distance, then |
| 483 | all pairs must be given distances. Although, when distances are only |
| 484 | given in one direction for each pair of nodes, then the distances in |
| 485 | the opposite directions are assumed to be the same. If, however, an |
| 486 | asymmetrical pair of distances is given for even one node pair, then |
| 487 | all node pairs must be provided distance values for both directions, |
| 488 | even when they are symmetrical. When a node is unreachable from |
| 489 | another node, set the pair's distance to 255. |
| 490 | |
| 491 | Note that the -``numa`` option doesn't allocate any of the specified |
| 492 | resources, it just assigns existing resources to NUMA nodes. This |
| 493 | means that one still has to use the ``-m``, ``-smp`` options to |
| 494 | allocate RAM and VCPUs respectively. |
| 495 | |
| 496 | Use '\ ``hmat-lb``\ ' to set System Locality Latency and Bandwidth |
| 497 | Information between initiator and target NUMA nodes in ACPI |
| 498 | Heterogeneous Attribute Memory Table (HMAT). Initiator NUMA node can |
| 499 | create memory requests, usually it has one or more processors. |
| 500 | Target NUMA node contains addressable memory. |
| 501 | |
| 502 | In '\ ``hmat-lb``\ ' option, node are NUMA node IDs. hierarchy is |
| 503 | the memory hierarchy of the target NUMA node: if hierarchy is |
| 504 | 'memory', the structure represents the memory performance; if |
| 505 | hierarchy is 'first-level\|second-level\|third-level', this |
| 506 | structure represents aggregated performance of memory side caches |
| 507 | for each domain. type of 'data-type' is type of data represented by |
| 508 | this structure instance: if 'hierarchy' is 'memory', 'data-type' is |
| 509 | 'access\|read\|write' latency or 'access\|read\|write' bandwidth of |
| 510 | the target memory; if 'hierarchy' is |
| 511 | 'first-level\|second-level\|third-level', 'data-type' is |
| 512 | 'access\|read\|write' hit latency or 'access\|read\|write' hit |
| 513 | bandwidth of the target memory side cache. |
| 514 | |
| 515 | lat is latency value in nanoseconds. bw is bandwidth value, the |
| 516 | possible value and units are NUM[M\|G\|T], mean that the bandwidth |
| 517 | value are NUM byte per second (or MB/s, GB/s or TB/s depending on |
| 518 | used suffix). Note that if latency or bandwidth value is 0, means |
| 519 | the corresponding latency or bandwidth information is not provided. |
| 520 | |
| 521 | In '\ ``hmat-cache``\ ' option, node-id is the NUMA-id of the memory |
| 522 | belongs. size is the size of memory side cache in bytes. level is |
| 523 | the cache level described in this structure, note that the cache |
| 524 | level 0 should not be used with '\ ``hmat-cache``\ ' option. |
| 525 | associativity is the cache associativity, the possible value is |
| 526 | 'none/direct(direct-mapped)/complex(complex cache indexing)'. policy |
| 527 | is the write policy. line is the cache Line size in bytes. |
| 528 | |
| 529 | For example, the following options describe 2 NUMA nodes. Node 0 has |
| 530 | 2 cpus and a ram, node 1 has only a ram. The processors in node 0 |
| 531 | access memory in node 0 with access-latency 5 nanoseconds, |
| 532 | access-bandwidth is 200 MB/s; The processors in NUMA node 0 access |
| 533 | memory in NUMA node 1 with access-latency 10 nanoseconds, |
| 534 | access-bandwidth is 100 MB/s. And for memory side cache information, |
| 535 | NUMA node 0 and 1 both have 1 level memory cache, size is 10KB, |
| 536 | policy is write-back, the cache Line size is 8 bytes: |
| 537 | |
| 538 | :: |
| 539 | |
| 540 | -machine hmat=on \ |
| 541 | -m 2G \ |
| 542 | -object memory-backend-ram,size=1G,id=m0 \ |
| 543 | -object memory-backend-ram,size=1G,id=m1 \ |
Yanan Wang | 848dd26 | 2021-09-28 20:11:34 +0800 | [diff] [blame] | 544 | -smp 2,sockets=2,maxcpus=2 \ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 545 | -numa node,nodeid=0,memdev=m0 \ |
| 546 | -numa node,nodeid=1,memdev=m1,initiator=0 \ |
| 547 | -numa cpu,node-id=0,socket-id=0 \ |
| 548 | -numa cpu,node-id=0,socket-id=1 \ |
| 549 | -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-latency,latency=5 \ |
| 550 | -numa hmat-lb,initiator=0,target=0,hierarchy=memory,data-type=access-bandwidth,bandwidth=200M \ |
| 551 | -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-latency,latency=10 \ |
| 552 | -numa hmat-lb,initiator=0,target=1,hierarchy=memory,data-type=access-bandwidth,bandwidth=100M \ |
| 553 | -numa hmat-cache,node-id=0,size=10K,level=1,associativity=direct,policy=write-back,line=8 \ |
| 554 | -numa hmat-cache,node-id=1,size=10K,level=1,associativity=direct,policy=write-back,line=8 |
| 555 | ERST |
aliguori | 268a362 | 2009-04-21 22:30:27 +0000 | [diff] [blame] | 556 | |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 557 | DEF("add-fd", HAS_ARG, QEMU_OPTION_add_fd, |
| 558 | "-add-fd fd=fd,set=set[,opaque=opaque]\n" |
| 559 | " Add 'fd' to fd 'set'\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 560 | SRST |
| 561 | ``-add-fd fd=fd,set=set[,opaque=opaque]`` |
| 562 | Add a file descriptor to an fd set. Valid options are: |
| 563 | |
| 564 | ``fd=fd`` |
| 565 | This option defines the file descriptor of which a duplicate is |
| 566 | added to fd set. The file descriptor cannot be stdin, stdout, or |
| 567 | stderr. |
| 568 | |
| 569 | ``set=set`` |
| 570 | This option defines the ID of the fd set to add the file |
| 571 | descriptor to. |
| 572 | |
| 573 | ``opaque=opaque`` |
| 574 | This option defines a free-form string that can be used to |
| 575 | describe fd. |
| 576 | |
| 577 | You can open an image using pre-opened file descriptors from an fd |
| 578 | set: |
| 579 | |
| 580 | .. parsed-literal:: |
| 581 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 582 | |qemu_system| \\ |
| 583 | -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ |
| 584 | -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 585 | -drive file=/dev/fdset/2,index=0,media=disk |
| 586 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 587 | |
| 588 | DEF("set", HAS_ARG, QEMU_OPTION_set, |
| 589 | "-set group.id.arg=value\n" |
| 590 | " set <arg> parameter for item <id> of type <group>\n" |
| 591 | " i.e. -set drive.$id.file=/path/to/image\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 592 | SRST |
| 593 | ``-set group.id.arg=value`` |
| 594 | Set parameter arg for item id of type group |
| 595 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 596 | |
| 597 | DEF("global", HAS_ARG, QEMU_OPTION_global, |
Paolo Bonzini | 3751d7c | 2015-04-09 14:16:19 +0200 | [diff] [blame] | 598 | "-global driver.property=value\n" |
| 599 | "-global driver=driver,property=property,value=value\n" |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 600 | " set a global default for a driver property\n", |
| 601 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 602 | SRST |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 603 | ``-global driver.prop=value`` |
| 604 | \ |
| 605 | ``-global driver=driver,property=property,value=value`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 606 | Set default value of driver's property prop to value, e.g.: |
| 607 | |
| 608 | .. parsed-literal:: |
| 609 | |
| 610 | |qemu_system_x86| -global ide-hd.physical_block_size=4096 disk-image.img |
| 611 | |
| 612 | In particular, you can use this to set driver properties for devices |
| 613 | which are created automatically by the machine model. To create a |
| 614 | device which is not created automatically and set properties on it, |
| 615 | use -``device``. |
| 616 | |
| 617 | -global driver.prop=value is shorthand for -global |
| 618 | driver=driver,property=prop,value=value. The longhand syntax works |
| 619 | even when driver contains a dot. |
| 620 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 621 | |
| 622 | DEF("boot", HAS_ARG, QEMU_OPTION_boot, |
| 623 | "-boot [order=drives][,once=drives][,menu=on|off]\n" |
Amos Kong | c8a6ae8 | 2013-03-19 14:23:27 +0800 | [diff] [blame] | 624 | " [,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_time][,strict=on|off]\n" |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 625 | " 'drives': floppy (a), hard disk (c), CD-ROM (d), network (n)\n" |
| 626 | " 'sp_name': the file's name that would be passed to bios as logo picture, if menu=on\n" |
| 627 | " 'sp_time': the period that splash picture last if menu=on, unit is ms\n" |
| 628 | " 'rb_timeout': the timeout before guest reboot when boot failed, unit is ms\n", |
| 629 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 630 | SRST |
| 631 | ``-boot [order=drives][,once=drives][,menu=on|off][,splash=sp_name][,splash-time=sp_time][,reboot-timeout=rb_timeout][,strict=on|off]`` |
| 632 | Specify boot order drives as a string of drive letters. Valid drive |
| 633 | letters depend on the target architecture. The x86 PC uses: a, b |
| 634 | (floppy 1 and 2), c (first hard disk), d (first CD-ROM), n-p |
| 635 | (Etherboot from network adapter 1-4), hard disk boot is the default. |
| 636 | To apply a particular boot order only on the first startup, specify |
| 637 | it via ``once``. Note that the ``order`` or ``once`` parameter |
| 638 | should not be used together with the ``bootindex`` property of |
| 639 | devices, since the firmware implementations normally do not support |
| 640 | both at the same time. |
| 641 | |
| 642 | Interactive boot menus/prompts can be enabled via ``menu=on`` as far |
| 643 | as firmware/BIOS supports them. The default is non-interactive boot. |
| 644 | |
| 645 | A splash picture could be passed to bios, enabling user to show it |
| 646 | as logo, when option splash=sp\_name is given and menu=on, If |
| 647 | firmware/BIOS supports them. Currently Seabios for X86 system |
| 648 | support it. limitation: The splash file could be a jpeg file or a |
| 649 | BMP file in 24 BPP format(true color). The resolution should be |
| 650 | supported by the SVGA mode, so the recommended is 320x240, 640x480, |
| 651 | 800x640. |
| 652 | |
| 653 | A timeout could be passed to bios, guest will pause for rb\_timeout |
| 654 | ms when boot failed, then reboot. If rb\_timeout is '-1', guest will |
| 655 | not reboot, qemu passes '-1' to bios by default. Currently Seabios |
| 656 | for X86 system support it. |
| 657 | |
| 658 | Do strict boot via ``strict=on`` as far as firmware/BIOS supports |
| 659 | it. This only effects when boot priority is changed by bootindex |
| 660 | options. The default is non-strict boot. |
| 661 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 662 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 663 | |
| 664 | # try to boot from network first, then from hard disk |
| 665 | |qemu_system_x86| -boot order=nc |
| 666 | # boot from CD-ROM first, switch back to default order after reboot |
| 667 | |qemu_system_x86| -boot once=d |
| 668 | # boot with a splash picture for 5 seconds. |
| 669 | |qemu_system_x86| -boot menu=on,splash=/root/boot.bmp,splash-time=5000 |
| 670 | |
| 671 | Note: The legacy format '-boot drives' is still supported but its |
| 672 | use is discouraged as it may be removed from future versions. |
| 673 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 674 | |
| 675 | DEF("m", HAS_ARG, QEMU_OPTION_m, |
Michael Tokarev | 89f3ea2 | 2016-11-10 17:51:32 +0300 | [diff] [blame] | 676 | "-m [size=]megs[,slots=n,maxmem=size]\n" |
Igor Mammedov | 6e1d3c1 | 2013-11-27 01:27:35 +0100 | [diff] [blame] | 677 | " configure guest RAM\n" |
Alexander Graf | 0daba1f | 2015-06-05 11:05:03 +0200 | [diff] [blame] | 678 | " size: initial amount of guest memory\n" |
Igor Mammedov | c270fb9 | 2014-06-02 15:25:02 +0200 | [diff] [blame] | 679 | " slots: number of hotplug slots (default: none)\n" |
Matthew Rosato | b6fe012 | 2014-08-28 11:25:33 -0400 | [diff] [blame] | 680 | " maxmem: maximum amount of guest memory (default: none)\n" |
Thomas Huth | a635bcf | 2023-07-03 09:56:46 +0200 | [diff] [blame] | 681 | " Note: Some architectures might enforce a specific granularity\n", |
Igor Mammedov | 6e1d3c1 | 2013-11-27 01:27:35 +0100 | [diff] [blame] | 682 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 683 | SRST |
| 684 | ``-m [size=]megs[,slots=n,maxmem=size]`` |
| 685 | Sets guest startup RAM size to megs megabytes. Default is 128 MiB. |
| 686 | Optionally, a suffix of "M" or "G" can be used to signify a value in |
| 687 | megabytes or gigabytes respectively. Optional pair slots, maxmem |
| 688 | could be used to set amount of hotpluggable memory slots and maximum |
| 689 | amount of memory. Note that maxmem must be aligned to the page size. |
| 690 | |
| 691 | For example, the following command-line sets the guest startup RAM |
| 692 | size to 1GB, creates 3 slots to hotplug additional memory and sets |
| 693 | the maximum memory the guest can reach to 4GB: |
| 694 | |
| 695 | .. parsed-literal:: |
| 696 | |
| 697 | |qemu_system| -m 1G,slots=3,maxmem=4G |
| 698 | |
| 699 | If slots and maxmem are not specified, memory hotplug won't be |
| 700 | enabled and the guest startup RAM will never increase. |
| 701 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 702 | |
| 703 | DEF("mem-path", HAS_ARG, QEMU_OPTION_mempath, |
| 704 | "-mem-path FILE provide backing storage for guest RAM\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 705 | SRST |
| 706 | ``-mem-path path`` |
| 707 | Allocate guest RAM from a temporarily created file in path. |
| 708 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 709 | |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 710 | DEF("mem-prealloc", 0, QEMU_OPTION_mem_prealloc, |
| 711 | "-mem-prealloc preallocate guest memory (use with -mem-path)\n", |
| 712 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 713 | SRST |
| 714 | ``-mem-prealloc`` |
| 715 | Preallocate memory when using -mem-path. |
| 716 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 717 | |
| 718 | DEF("k", HAS_ARG, QEMU_OPTION_k, |
| 719 | "-k language use keyboard layout (for example 'fr' for French)\n", |
| 720 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 721 | SRST |
| 722 | ``-k language`` |
| 723 | Use keyboard layout language (for example ``fr`` for French). This |
| 724 | option is only needed where it is not easy to get raw PC keycodes |
| 725 | (e.g. on Macs, with some X11 servers or with a VNC or curses |
| 726 | display). You don't normally need to use it on PC/Linux or |
| 727 | PC/Windows hosts. |
| 728 | |
| 729 | The available layouts are: |
| 730 | |
| 731 | :: |
| 732 | |
| 733 | ar de-ch es fo fr-ca hu ja mk no pt-br sv |
| 734 | da en-gb et fr fr-ch is lt nl pl ru th |
| 735 | de en-us fi fr-be hr it lv nl-be pt sl tr |
| 736 | |
| 737 | The default is ``en-us``. |
| 738 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 739 | |
| 740 | |
Paolo Bonzini | 039a683 | 2022-04-27 12:27:46 +0200 | [diff] [blame] | 741 | DEF("audio", HAS_ARG, QEMU_OPTION_audio, |
Paolo Bonzini | 1ebdbff | 2023-09-21 10:23:58 +0200 | [diff] [blame] | 742 | "-audio [driver=]driver[,prop[=value][,...]]\n" |
| 743 | " specifies default audio backend when `audiodev` is not\n" |
| 744 | " used to create a machine or sound device;" |
| 745 | " options are the same as for -audiodev\n" |
Paolo Bonzini | 039a683 | 2022-04-27 12:27:46 +0200 | [diff] [blame] | 746 | "-audio [driver=]driver,model=value[,prop[=value][,...]]\n" |
| 747 | " specifies the audio backend and device to use;\n" |
| 748 | " apart from 'model', options are the same as for -audiodev.\n" |
| 749 | " use '-audio model=help' to show possible devices.\n", |
| 750 | QEMU_ARCH_ALL) |
| 751 | SRST |
Paolo Bonzini | 1ebdbff | 2023-09-21 10:23:58 +0200 | [diff] [blame] | 752 | ``-audio [driver=]driver[,model=value][,prop[=value][,...]]`` |
| 753 | If the ``model`` option is specified, ``-audio`` is a shortcut |
| 754 | for configuring both the guest audio hardware and the host audio |
| 755 | backend in one go. The guest hardware model can be set with |
| 756 | ``model=modelname``. Use ``model=help`` to list the available |
| 757 | device types. |
Paolo Bonzini | 039a683 | 2022-04-27 12:27:46 +0200 | [diff] [blame] | 758 | |
| 759 | The following two example do exactly the same, to show how ``-audio`` |
| 760 | can be used to shorten the command line length: |
| 761 | |
| 762 | .. parsed-literal:: |
| 763 | |
| 764 | |qemu_system| -audiodev pa,id=pa -device sb16,audiodev=pa |
| 765 | |qemu_system| -audio pa,model=sb16 |
Paolo Bonzini | 1ebdbff | 2023-09-21 10:23:58 +0200 | [diff] [blame] | 766 | |
| 767 | If the ``model`` option is not specified, ``-audio`` is used to |
| 768 | configure a default audio backend that will be used whenever the |
| 769 | ``audiodev`` property is not set on a device or machine. In |
| 770 | particular, ``-audio none`` ensures that no audio is produced even |
| 771 | for machines that have embedded sound hardware. |
| 772 | |
| 773 | In both cases, the driver option is the same as with the corresponding |
| 774 | ``-audiodev`` option below. Use ``driver=help`` to list the available |
| 775 | drivers. |
| 776 | |
Paolo Bonzini | 039a683 | 2022-04-27 12:27:46 +0200 | [diff] [blame] | 777 | ERST |
| 778 | |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 779 | DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, |
| 780 | "-audiodev [driver=]driver,id=id[,prop[=value][,...]]\n" |
| 781 | " specifies the audio backend to use\n" |
Claudio Fontana | 5e03b6d | 2022-09-08 10:14:41 +0200 | [diff] [blame] | 782 | " Use ``-audiodev help`` to list the available drivers\n" |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 783 | " id= identifier of the backend\n" |
| 784 | " timer-period= timer period in microseconds\n" |
Kővágó, Zoltán | 8efac07 | 2019-10-13 21:57:58 +0200 | [diff] [blame] | 785 | " in|out.mixing-engine= use mixing engine to mix streams inside QEMU\n" |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 786 | " in|out.fixed-settings= use fixed settings for host audio\n" |
| 787 | " in|out.frequency= frequency to use with fixed settings\n" |
| 788 | " in|out.channels= number of channels to use with fixed settings\n" |
| 789 | " in|out.format= sample format to use with fixed settings\n" |
Volker Rümelin | 49f77e6 | 2020-03-08 20:33:21 +0100 | [diff] [blame] | 790 | " valid values: s8, s16, s32, u8, u16, u32, f32\n" |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 791 | " in|out.voices= number of voices to use\n" |
Stefan Hajnoczi | 8624725 | 2019-09-18 10:53:33 +0100 | [diff] [blame] | 792 | " in|out.buffer-length= length of buffer in microseconds\n" |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 793 | "-audiodev none,id=id,[,prop[=value][,...]]\n" |
| 794 | " dummy driver that discards all output\n" |
| 795 | #ifdef CONFIG_AUDIO_ALSA |
| 796 | "-audiodev alsa,id=id[,prop[=value][,...]]\n" |
| 797 | " in|out.dev= name of the audio device to use\n" |
Stefan Hajnoczi | dfc5434 | 2019-09-18 10:53:35 +0100 | [diff] [blame] | 798 | " in|out.period-length= length of period in microseconds\n" |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 799 | " in|out.try-poll= attempt to use poll mode\n" |
| 800 | " threshold= threshold (in microseconds) when playback starts\n" |
| 801 | #endif |
| 802 | #ifdef CONFIG_AUDIO_COREAUDIO |
| 803 | "-audiodev coreaudio,id=id[,prop[=value][,...]]\n" |
| 804 | " in|out.buffer-count= number of buffers\n" |
| 805 | #endif |
| 806 | #ifdef CONFIG_AUDIO_DSOUND |
| 807 | "-audiodev dsound,id=id[,prop[=value][,...]]\n" |
| 808 | " latency= add extra latency to playback in microseconds\n" |
| 809 | #endif |
| 810 | #ifdef CONFIG_AUDIO_OSS |
| 811 | "-audiodev oss,id=id[,prop[=value][,...]]\n" |
| 812 | " in|out.dev= path of the audio device to use\n" |
| 813 | " in|out.buffer-count= number of buffers\n" |
| 814 | " in|out.try-poll= attempt to use poll mode\n" |
| 815 | " try-mmap= try using memory mapped access\n" |
| 816 | " exclusive= open device in exclusive mode\n" |
| 817 | " dsp-policy= set timing policy (0..10), -1 to use fragment mode\n" |
| 818 | #endif |
| 819 | #ifdef CONFIG_AUDIO_PA |
| 820 | "-audiodev pa,id=id[,prop[=value][,...]]\n" |
| 821 | " server= PulseAudio server address\n" |
| 822 | " in|out.name= source/sink device name\n" |
Stefan Hajnoczi | 14d4f01 | 2019-10-04 13:56:41 +0100 | [diff] [blame] | 823 | " in|out.latency= desired latency in microseconds\n" |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 824 | #endif |
Dorinda Bassey | c2d3d1c | 2023-04-17 12:56:54 +0200 | [diff] [blame] | 825 | #ifdef CONFIG_AUDIO_PIPEWIRE |
| 826 | "-audiodev pipewire,id=id[,prop[=value][,...]]\n" |
| 827 | " in|out.name= source/sink device name\n" |
| 828 | " in|out.stream-name= name of pipewire stream\n" |
| 829 | " in|out.latency= desired latency in microseconds\n" |
| 830 | #endif |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 831 | #ifdef CONFIG_AUDIO_SDL |
| 832 | "-audiodev sdl,id=id[,prop[=value][,...]]\n" |
Volker Rümelin | 5a0926c | 2021-01-10 11:02:19 +0100 | [diff] [blame] | 833 | " in|out.buffer-count= number of buffers\n" |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 834 | #endif |
Alexandre Ratchov | 663df1c | 2022-09-07 15:23:42 +0200 | [diff] [blame] | 835 | #ifdef CONFIG_AUDIO_SNDIO |
| 836 | "-audiodev sndio,id=id[,prop[=value][,...]]\n" |
| 837 | #endif |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 838 | #ifdef CONFIG_SPICE |
| 839 | "-audiodev spice,id=id[,prop[=value][,...]]\n" |
| 840 | #endif |
Marc-André Lureau | 739362d | 2021-03-09 17:15:28 +0400 | [diff] [blame] | 841 | #ifdef CONFIG_DBUS_DISPLAY |
| 842 | "-audiodev dbus,id=id[,prop[=value][,...]]\n" |
| 843 | #endif |
Kővágó, Zoltán | f0b3d81 | 2019-03-08 23:34:14 +0100 | [diff] [blame] | 844 | "-audiodev wav,id=id[,prop[=value][,...]]\n" |
| 845 | " path= path of wav file to record\n", |
| 846 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 847 | SRST |
| 848 | ``-audiodev [driver=]driver,id=id[,prop[=value][,...]]`` |
| 849 | Adds a new audio backend driver identified by id. There are global |
| 850 | and driver specific properties. Some values can be set differently |
| 851 | for input and output, they're marked with ``in|out.``. You can set |
| 852 | the input's property with ``in.prop`` and the output's property with |
| 853 | ``out.prop``. For example: |
| 854 | |
| 855 | :: |
| 856 | |
| 857 | -audiodev alsa,id=example,in.frequency=44110,out.frequency=8000 |
| 858 | -audiodev alsa,id=example,out.channels=1 # leaves in.channels unspecified |
| 859 | |
| 860 | NOTE: parameter validation is known to be incomplete, in many cases |
| 861 | specifying an invalid option causes QEMU to print an error message |
| 862 | and continue emulation without sound. |
| 863 | |
| 864 | Valid global options are: |
| 865 | |
| 866 | ``id=identifier`` |
| 867 | Identifies the audio backend. |
| 868 | |
| 869 | ``timer-period=period`` |
| 870 | Sets the timer period used by the audio subsystem in |
| 871 | microseconds. Default is 10000 (10 ms). |
| 872 | |
| 873 | ``in|out.mixing-engine=on|off`` |
| 874 | Use QEMU's mixing engine to mix all streams inside QEMU and |
| 875 | convert audio formats when not supported by the backend. When |
| 876 | off, fixed-settings must be off too. Note that disabling this |
| 877 | option means that the selected backend must support multiple |
| 878 | streams and the audio formats used by the virtual cards, |
| 879 | otherwise you'll get no sound. It's not recommended to disable |
| 880 | this option unless you want to use 5.1 or 7.1 audio, as mixing |
| 881 | engine only supports mono and stereo audio. Default is on. |
| 882 | |
| 883 | ``in|out.fixed-settings=on|off`` |
| 884 | Use fixed settings for host audio. When off, it will change |
| 885 | based on how the guest opens the sound card. In this case you |
| 886 | must not specify frequency, channels or format. Default is on. |
| 887 | |
| 888 | ``in|out.frequency=frequency`` |
| 889 | Specify the frequency to use when using fixed-settings. Default |
| 890 | is 44100Hz. |
| 891 | |
| 892 | ``in|out.channels=channels`` |
| 893 | Specify the number of channels to use when using fixed-settings. |
| 894 | Default is 2 (stereo). |
| 895 | |
| 896 | ``in|out.format=format`` |
| 897 | Specify the sample format to use when using fixed-settings. |
| 898 | Valid values are: ``s8``, ``s16``, ``s32``, ``u8``, ``u16``, |
Volker Rümelin | 49f77e6 | 2020-03-08 20:33:21 +0100 | [diff] [blame] | 899 | ``u32``, ``f32``. Default is ``s16``. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 900 | |
| 901 | ``in|out.voices=voices`` |
| 902 | Specify the number of voices to use. Default is 1. |
| 903 | |
| 904 | ``in|out.buffer-length=usecs`` |
| 905 | Sets the size of the buffer in microseconds. |
| 906 | |
| 907 | ``-audiodev none,id=id[,prop[=value][,...]]`` |
| 908 | Creates a dummy backend that discards all outputs. This backend has |
| 909 | no backend specific properties. |
| 910 | |
| 911 | ``-audiodev alsa,id=id[,prop[=value][,...]]`` |
| 912 | Creates backend using the ALSA. This backend is only available on |
| 913 | Linux. |
| 914 | |
| 915 | ALSA specific options are: |
| 916 | |
| 917 | ``in|out.dev=device`` |
| 918 | Specify the ALSA device to use for input and/or output. Default |
| 919 | is ``default``. |
| 920 | |
| 921 | ``in|out.period-length=usecs`` |
| 922 | Sets the period length in microseconds. |
| 923 | |
| 924 | ``in|out.try-poll=on|off`` |
| 925 | Attempt to use poll mode with the device. Default is on. |
| 926 | |
| 927 | ``threshold=threshold`` |
| 928 | Threshold (in microseconds) when playback starts. Default is 0. |
| 929 | |
| 930 | ``-audiodev coreaudio,id=id[,prop[=value][,...]]`` |
| 931 | Creates a backend using Apple's Core Audio. This backend is only |
| 932 | available on Mac OS and only supports playback. |
| 933 | |
| 934 | Core Audio specific options are: |
| 935 | |
| 936 | ``in|out.buffer-count=count`` |
| 937 | Sets the count of the buffers. |
| 938 | |
| 939 | ``-audiodev dsound,id=id[,prop[=value][,...]]`` |
| 940 | Creates a backend using Microsoft's DirectSound. This backend is |
| 941 | only available on Windows and only supports playback. |
| 942 | |
| 943 | DirectSound specific options are: |
| 944 | |
| 945 | ``latency=usecs`` |
| 946 | Add extra usecs microseconds latency to playback. Default is |
| 947 | 10000 (10 ms). |
| 948 | |
| 949 | ``-audiodev oss,id=id[,prop[=value][,...]]`` |
| 950 | Creates a backend using OSS. This backend is available on most |
| 951 | Unix-like systems. |
| 952 | |
| 953 | OSS specific options are: |
| 954 | |
| 955 | ``in|out.dev=device`` |
| 956 | Specify the file name of the OSS device to use. Default is |
| 957 | ``/dev/dsp``. |
| 958 | |
| 959 | ``in|out.buffer-count=count`` |
| 960 | Sets the count of the buffers. |
| 961 | |
| 962 | ``in|out.try-poll=on|of`` |
| 963 | Attempt to use poll mode with the device. Default is on. |
| 964 | |
| 965 | ``try-mmap=on|off`` |
| 966 | Try using memory mapped device access. Default is off. |
| 967 | |
| 968 | ``exclusive=on|off`` |
| 969 | Open the device in exclusive mode (vmix won't work in this |
| 970 | case). Default is off. |
| 971 | |
| 972 | ``dsp-policy=policy`` |
| 973 | Sets the timing policy (between 0 and 10, where smaller number |
| 974 | means smaller latency but higher CPU usage). Use -1 to use |
| 975 | buffer sizes specified by ``buffer`` and ``buffer-count``. This |
| 976 | option is ignored if you do not have OSS 4. Default is 5. |
| 977 | |
| 978 | ``-audiodev pa,id=id[,prop[=value][,...]]`` |
| 979 | Creates a backend using PulseAudio. This backend is available on |
| 980 | most systems. |
| 981 | |
| 982 | PulseAudio specific options are: |
| 983 | |
| 984 | ``server=server`` |
| 985 | Sets the PulseAudio server to connect to. |
| 986 | |
| 987 | ``in|out.name=sink`` |
| 988 | Use the specified source/sink for recording/playback. |
| 989 | |
| 990 | ``in|out.latency=usecs`` |
| 991 | Desired latency in microseconds. The PulseAudio server will try |
| 992 | to honor this value but actual latencies may be lower or higher. |
| 993 | |
Dorinda Bassey | c2d3d1c | 2023-04-17 12:56:54 +0200 | [diff] [blame] | 994 | ``-audiodev pipewire,id=id[,prop[=value][,...]]`` |
Marc-André Lureau | 20c5124 | 2023-05-06 20:37:26 +0400 | [diff] [blame] | 995 | Creates a backend using PipeWire. This backend is available on |
Dorinda Bassey | c2d3d1c | 2023-04-17 12:56:54 +0200 | [diff] [blame] | 996 | most systems. |
| 997 | |
Marc-André Lureau | 20c5124 | 2023-05-06 20:37:26 +0400 | [diff] [blame] | 998 | PipeWire specific options are: |
Dorinda Bassey | c2d3d1c | 2023-04-17 12:56:54 +0200 | [diff] [blame] | 999 | |
| 1000 | ``in|out.latency=usecs`` |
| 1001 | Desired latency in microseconds. |
| 1002 | |
| 1003 | ``in|out.name=sink`` |
| 1004 | Use the specified source/sink for recording/playback. |
| 1005 | |
| 1006 | ``in|out.stream-name`` |
| 1007 | Specify the name of pipewire stream. |
| 1008 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1009 | ``-audiodev sdl,id=id[,prop[=value][,...]]`` |
| 1010 | Creates a backend using SDL. This backend is available on most |
| 1011 | systems, but you should use your platform's native backend if |
Volker Rümelin | 5a0926c | 2021-01-10 11:02:19 +0100 | [diff] [blame] | 1012 | possible. |
| 1013 | |
| 1014 | SDL specific options are: |
| 1015 | |
| 1016 | ``in|out.buffer-count=count`` |
| 1017 | Sets the count of the buffers. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1018 | |
Alexandre Ratchov | 663df1c | 2022-09-07 15:23:42 +0200 | [diff] [blame] | 1019 | ``-audiodev sndio,id=id[,prop[=value][,...]]`` |
| 1020 | Creates a backend using SNDIO. This backend is available on |
| 1021 | OpenBSD and most other Unix-like systems. |
| 1022 | |
| 1023 | Sndio specific options are: |
| 1024 | |
| 1025 | ``in|out.dev=device`` |
| 1026 | Specify the sndio device to use for input and/or output. Default |
| 1027 | is ``default``. |
| 1028 | |
| 1029 | ``in|out.latency=usecs`` |
| 1030 | Sets the desired period length in microseconds. |
| 1031 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1032 | ``-audiodev spice,id=id[,prop[=value][,...]]`` |
| 1033 | Creates a backend that sends audio through SPICE. This backend |
| 1034 | requires ``-spice`` and automatically selected in that case, so |
| 1035 | usually you can ignore this option. This backend has no backend |
| 1036 | specific properties. |
| 1037 | |
| 1038 | ``-audiodev wav,id=id[,prop[=value][,...]]`` |
| 1039 | Creates a backend that writes audio to a WAV file. |
| 1040 | |
| 1041 | Backend specific options are: |
| 1042 | |
| 1043 | ``path=path`` |
| 1044 | Write recorded audio into the specified file. Default is |
| 1045 | ``qemu.wav``. |
| 1046 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1047 | |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1048 | DEF("device", HAS_ARG, QEMU_OPTION_device, |
| 1049 | "-device driver[,prop[=value][,...]]\n" |
| 1050 | " add device (based on driver)\n" |
| 1051 | " prop=value,... sets driver properties\n" |
| 1052 | " use '-device help' to print all possible drivers\n" |
| 1053 | " use '-device driver,help' to print all possible properties\n", |
| 1054 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1055 | SRST |
| 1056 | ``-device driver[,prop[=value][,...]]`` |
| 1057 | Add device driver. prop=value sets driver properties. Valid |
| 1058 | properties depend on the driver. To get help on possible drivers and |
| 1059 | properties, use ``-device help`` and ``-device driver,help``. |
| 1060 | |
| 1061 | Some drivers are: |
| 1062 | |
Corey Minyard | 789101b | 2020-07-17 11:37:02 -0500 | [diff] [blame] | 1063 | ``-device ipmi-bmc-sim,id=id[,prop[=value][,...]]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1064 | Add an IPMI BMC. This is a simulation of a hardware management |
| 1065 | interface processor that normally sits on a system. It provides a |
| 1066 | watchdog and the ability to reset and power control the system. You |
| 1067 | need to connect this to an IPMI interface to make it useful |
| 1068 | |
| 1069 | The IPMI slave address to use for the BMC. The default is 0x20. This |
| 1070 | address is the BMC's address on the I2C network of management |
| 1071 | controllers. If you don't know what this means, it is safe to ignore |
| 1072 | it. |
| 1073 | |
| 1074 | ``id=id`` |
| 1075 | The BMC id for interfaces to use this device. |
| 1076 | |
| 1077 | ``slave_addr=val`` |
| 1078 | Define slave address to use for the BMC. The default is 0x20. |
| 1079 | |
| 1080 | ``sdrfile=file`` |
| 1081 | file containing raw Sensor Data Records (SDR) data. The default |
| 1082 | is none. |
| 1083 | |
| 1084 | ``fruareasize=val`` |
| 1085 | size of a Field Replaceable Unit (FRU) area. The default is |
| 1086 | 1024. |
| 1087 | |
| 1088 | ``frudatafile=file`` |
| 1089 | file containing raw Field Replaceable Unit (FRU) inventory data. |
| 1090 | The default is none. |
| 1091 | |
| 1092 | ``guid=uuid`` |
| 1093 | value for the GUID for the BMC, in standard UUID format. If this |
| 1094 | is set, get "Get GUID" command to the BMC will return it. |
| 1095 | Otherwise "Get GUID" will return an error. |
| 1096 | |
| 1097 | ``-device ipmi-bmc-extern,id=id,chardev=id[,slave_addr=val]`` |
| 1098 | Add a connection to an external IPMI BMC simulator. Instead of |
| 1099 | locally emulating the BMC like the above item, instead connect to an |
| 1100 | external entity that provides the IPMI services. |
| 1101 | |
| 1102 | A connection is made to an external BMC simulator. If you do this, |
| 1103 | it is strongly recommended that you use the "reconnect=" chardev |
| 1104 | option to reconnect to the simulator if the connection is lost. Note |
| 1105 | that if this is not used carefully, it can be a security issue, as |
| 1106 | the interface has the ability to send resets, NMIs, and power off |
| 1107 | the VM. It's best if QEMU makes a connection to an external |
| 1108 | simulator running on a secure port on localhost, so neither the |
| 1109 | simulator nor QEMU is exposed to any outside network. |
| 1110 | |
| 1111 | See the "lanserv/README.vm" file in the OpenIPMI library for more |
| 1112 | details on the external interface. |
| 1113 | |
| 1114 | ``-device isa-ipmi-kcs,bmc=id[,ioport=val][,irq=val]`` |
Stefan Weil | 1e458f1 | 2022-10-30 11:59:44 +0100 | [diff] [blame] | 1115 | Add a KCS IPMI interface on the ISA bus. This also adds a |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1116 | corresponding ACPI and SMBIOS entries, if appropriate. |
| 1117 | |
| 1118 | ``bmc=id`` |
| 1119 | The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern |
| 1120 | above. |
| 1121 | |
| 1122 | ``ioport=val`` |
| 1123 | Define the I/O address of the interface. The default is 0xca0 |
| 1124 | for KCS. |
| 1125 | |
| 1126 | ``irq=val`` |
| 1127 | Define the interrupt to use. The default is 5. To disable |
| 1128 | interrupts, set this to 0. |
| 1129 | |
| 1130 | ``-device isa-ipmi-bt,bmc=id[,ioport=val][,irq=val]`` |
| 1131 | Like the KCS interface, but defines a BT interface. The default port |
| 1132 | is 0xe4 and the default interrupt is 5. |
Corey Minyard | 323679d | 2019-09-23 13:50:33 -0500 | [diff] [blame] | 1133 | |
| 1134 | ``-device pci-ipmi-kcs,bmc=id`` |
Stefan Weil | 1e458f1 | 2022-10-30 11:59:44 +0100 | [diff] [blame] | 1135 | Add a KCS IPMI interface on the PCI bus. |
Corey Minyard | 323679d | 2019-09-23 13:50:33 -0500 | [diff] [blame] | 1136 | |
| 1137 | ``bmc=id`` |
| 1138 | The BMC to connect to, one of ipmi-bmc-sim or ipmi-bmc-extern above. |
| 1139 | |
| 1140 | ``-device pci-ipmi-bt,bmc=id`` |
| 1141 | Like the KCS interface, but defines a BT interface on the PCI bus. |
Peter Xu | 7395b3e | 2021-07-07 11:41:14 -0400 | [diff] [blame] | 1142 | |
| 1143 | ``-device intel-iommu[,option=...]`` |
| 1144 | This is only supported by ``-machine q35``, which will enable Intel VT-d |
| 1145 | emulation within the guest. It supports below options: |
| 1146 | |
| 1147 | ``intremap=on|off`` (default: auto) |
| 1148 | This enables interrupt remapping feature. It's required to enable |
| 1149 | complete x2apic. Currently it only supports kvm kernel-irqchip modes |
| 1150 | ``off`` or ``split``, while full kernel-irqchip is not yet supported. |
| 1151 | The default value is "auto", which will be decided by the mode of |
| 1152 | kernel-irqchip. |
| 1153 | |
| 1154 | ``caching-mode=on|off`` (default: off) |
| 1155 | This enables caching mode for the VT-d emulated device. When |
| 1156 | caching-mode is enabled, each guest DMA buffer mapping will generate an |
| 1157 | IOTLB invalidation from the guest IOMMU driver to the vIOMMU device in |
| 1158 | a synchronous way. It is required for ``-device vfio-pci`` to work |
| 1159 | with the VT-d device, because host assigned devices requires to setup |
| 1160 | the DMA mapping on the host before guest DMA starts. |
| 1161 | |
| 1162 | ``device-iotlb=on|off`` (default: off) |
| 1163 | This enables device-iotlb capability for the emulated VT-d device. So |
| 1164 | far virtio/vhost should be the only real user for this parameter, |
| 1165 | paired with ats=on configured for the device. |
| 1166 | |
| 1167 | ``aw-bits=39|48`` (default: 39) |
| 1168 | This decides the address width of IOVA address space. The address |
| 1169 | space has 39 bits width for 3-level IOMMU page tables, and 48 bits for |
| 1170 | 4-level IOMMU page tables. |
| 1171 | |
| 1172 | Please also refer to the wiki page for general scenarios of VT-d |
| 1173 | emulation in QEMU: https://wiki.qemu.org/Features/VT-d. |
| 1174 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1175 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1176 | |
| 1177 | DEF("name", HAS_ARG, QEMU_OPTION_name, |
Dr. David Alan Gilbert | 8f480de | 2014-01-30 10:20:31 +0000 | [diff] [blame] | 1178 | "-name string1[,process=string2][,debug-threads=on|off]\n" |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1179 | " set the name of the guest\n" |
Roman Bolshakov | 479a574 | 2018-12-17 23:26:01 +0300 | [diff] [blame] | 1180 | " string1 sets the window title and string2 the process name\n" |
| 1181 | " When debug-threads is enabled, individual threads are given a separate name\n" |
Dr. David Alan Gilbert | 8f480de | 2014-01-30 10:20:31 +0000 | [diff] [blame] | 1182 | " NOTE: The thread names are for debugging and not a stable API.\n", |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1183 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1184 | SRST |
| 1185 | ``-name name`` |
| 1186 | Sets the name of the guest. This name will be displayed in the SDL |
| 1187 | window caption. The name will also be used for the VNC server. Also |
| 1188 | optionally set the top visible process name in Linux. Naming of |
| 1189 | individual threads can also be enabled on Linux to aid debugging. |
| 1190 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1191 | |
| 1192 | DEF("uuid", HAS_ARG, QEMU_OPTION_uuid, |
| 1193 | "-uuid %08x-%04x-%04x-%04x-%012x\n" |
| 1194 | " specify machine UUID\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1195 | SRST |
| 1196 | ``-uuid uuid`` |
| 1197 | Set system UUID. |
| 1198 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1199 | |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1200 | DEFHEADING() |
| 1201 | |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 1202 | DEFHEADING(Block device options:) |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 1203 | |
Alex Bennée | 5af2b0f | 2022-08-22 17:56:08 +0100 | [diff] [blame] | 1204 | SRST |
| 1205 | The QEMU block device handling options have a long history and |
| 1206 | have gone through several iterations as the feature set and complexity |
| 1207 | of the block layer have grown. Many online guides to QEMU often |
| 1208 | reference older and deprecated options, which can lead to confusion. |
| 1209 | |
Alex Bennée | c1654c3 | 2023-04-24 10:22:37 +0100 | [diff] [blame] | 1210 | The most explicit way to describe disks is to use a combination of |
Alex Bennée | 5af2b0f | 2022-08-22 17:56:08 +0100 | [diff] [blame] | 1211 | ``-device`` to specify the hardware device and ``-blockdev`` to |
| 1212 | describe the backend. The device defines what the guest sees and the |
Alex Bennée | c1654c3 | 2023-04-24 10:22:37 +0100 | [diff] [blame] | 1213 | backend describes how QEMU handles the data. It is the only guaranteed |
| 1214 | stable interface for describing block devices and as such is |
| 1215 | recommended for management tools and scripting. |
| 1216 | |
| 1217 | The ``-drive`` option combines the device and backend into a single |
| 1218 | command line option which is a more human friendly. There is however no |
| 1219 | interface stability guarantee although some older board models still |
| 1220 | need updating to work with the modern blockdev forms. |
| 1221 | |
| 1222 | Older options like ``-hda`` are essentially macros which expand into |
| 1223 | ``-drive`` options for various drive interfaces. The original forms |
| 1224 | bake in a lot of assumptions from the days when QEMU was emulating a |
| 1225 | legacy PC, they are not recommended for modern configurations. |
Alex Bennée | 5af2b0f | 2022-08-22 17:56:08 +0100 | [diff] [blame] | 1226 | |
| 1227 | ERST |
| 1228 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1229 | DEF("fda", HAS_ARG, QEMU_OPTION_fda, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 1230 | "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL) |
| 1231 | DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1232 | SRST |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1233 | ``-fda file`` |
| 1234 | \ |
| 1235 | ``-fdb file`` |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 1236 | Use file as floppy disk 0/1 image (see the :ref:`disk images` chapter in |
| 1237 | the System Emulation Users Guide). |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1238 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1239 | |
| 1240 | DEF("hda", HAS_ARG, QEMU_OPTION_hda, |
Thomas Huth | bcd8e24 | 2023-08-29 15:29:48 +0200 | [diff] [blame] | 1241 | "-hda/-hdb file use 'file' as hard disk 0/1 image\n", QEMU_ARCH_ALL) |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 1242 | DEF("hdb", HAS_ARG, QEMU_OPTION_hdb, "", QEMU_ARCH_ALL) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1243 | DEF("hdc", HAS_ARG, QEMU_OPTION_hdc, |
Thomas Huth | bcd8e24 | 2023-08-29 15:29:48 +0200 | [diff] [blame] | 1244 | "-hdc/-hdd file use 'file' as hard disk 2/3 image\n", QEMU_ARCH_ALL) |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 1245 | DEF("hdd", HAS_ARG, QEMU_OPTION_hdd, "", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1246 | SRST |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1247 | ``-hda file`` |
| 1248 | \ |
| 1249 | ``-hdb file`` |
| 1250 | \ |
| 1251 | ``-hdc file`` |
| 1252 | \ |
| 1253 | ``-hdd file`` |
Thomas Huth | bcd8e24 | 2023-08-29 15:29:48 +0200 | [diff] [blame] | 1254 | Use file as hard disk 0, 1, 2 or 3 image on the default bus of the |
| 1255 | emulated machine (this is for example the IDE bus on most x86 machines, |
| 1256 | but it can also be SCSI, virtio or something else on other target |
| 1257 | architectures). See also the :ref:`disk images` chapter in the System |
| 1258 | Emulation Users Guide. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1259 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1260 | |
| 1261 | DEF("cdrom", HAS_ARG, QEMU_OPTION_cdrom, |
Thomas Huth | bcd8e24 | 2023-08-29 15:29:48 +0200 | [diff] [blame] | 1262 | "-cdrom file use 'file' as CD-ROM image\n", |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 1263 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1264 | SRST |
| 1265 | ``-cdrom file`` |
Thomas Huth | bcd8e24 | 2023-08-29 15:29:48 +0200 | [diff] [blame] | 1266 | Use file as CD-ROM image on the default bus of the emulated machine |
| 1267 | (which is IDE1 master on x86, so you cannot use ``-hdc`` and ``-cdrom`` |
| 1268 | at the same time there). On systems that support it, you can use the |
| 1269 | host CD-ROM by using ``/dev/cdrom`` as filename. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1270 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1271 | |
Markus Armbruster | 42e5f39 | 2017-02-28 22:27:07 +0100 | [diff] [blame] | 1272 | DEF("blockdev", HAS_ARG, QEMU_OPTION_blockdev, |
| 1273 | "-blockdev [driver=]driver[,node-name=N][,discard=ignore|unmap]\n" |
| 1274 | " [,cache.direct=on|off][,cache.no-flush=on|off]\n" |
Kevin Wolf | c9b749d | 2019-10-15 12:29:58 +0200 | [diff] [blame] | 1275 | " [,read-only=on|off][,auto-read-only=on|off]\n" |
| 1276 | " [,force-share=on|off][,detect-zeroes=on|off|unmap]\n" |
Markus Armbruster | 42e5f39 | 2017-02-28 22:27:07 +0100 | [diff] [blame] | 1277 | " [,driver specific parameters...]\n" |
| 1278 | " configure a block backend\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1279 | SRST |
| 1280 | ``-blockdev option[,option[,option[,...]]]`` |
| 1281 | Define a new block driver node. Some of the options apply to all |
| 1282 | block drivers, other options are only accepted for a specific block |
| 1283 | driver. See below for a list of generic options and options for the |
| 1284 | most common block drivers. |
| 1285 | |
| 1286 | Options that expect a reference to another node (e.g. ``file``) can |
| 1287 | be given in two ways. Either you specify the node name of an already |
| 1288 | existing node (file=node-name), or you define a new node inline, |
| 1289 | adding options for the referenced node after a dot |
| 1290 | (file.filename=path,file.aio=native). |
| 1291 | |
| 1292 | A block driver node created with ``-blockdev`` can be used for a |
| 1293 | guest device by specifying its node name for the ``drive`` property |
| 1294 | in a ``-device`` argument that defines a block device. |
| 1295 | |
| 1296 | ``Valid options for any block driver node:`` |
| 1297 | ``driver`` |
| 1298 | Specifies the block driver to use for the given node. |
| 1299 | |
| 1300 | ``node-name`` |
| 1301 | This defines the name of the block driver node by which it |
| 1302 | will be referenced later. The name must be unique, i.e. it |
| 1303 | must not match the name of a different block driver node, or |
| 1304 | (if you use ``-drive`` as well) the ID of a drive. |
| 1305 | |
| 1306 | If no node name is specified, it is automatically generated. |
| 1307 | The generated node name is not intended to be predictable |
| 1308 | and changes between QEMU invocations. For the top level, an |
| 1309 | explicit node name must be specified. |
| 1310 | |
| 1311 | ``read-only`` |
| 1312 | Open the node read-only. Guest write attempts will fail. |
| 1313 | |
| 1314 | Note that some block drivers support only read-only access, |
| 1315 | either generally or in certain configurations. In this case, |
| 1316 | the default value ``read-only=off`` does not work and the |
| 1317 | option must be specified explicitly. |
| 1318 | |
| 1319 | ``auto-read-only`` |
| 1320 | If ``auto-read-only=on`` is set, QEMU may fall back to |
| 1321 | read-only usage even when ``read-only=off`` is requested, or |
| 1322 | even switch between modes as needed, e.g. depending on |
| 1323 | whether the image file is writable or whether a writing user |
| 1324 | is attached to the node. |
| 1325 | |
| 1326 | ``force-share`` |
| 1327 | Override the image locking system of QEMU by forcing the |
| 1328 | node to utilize weaker shared access for permissions where |
| 1329 | it would normally request exclusive access. When there is |
| 1330 | the potential for multiple instances to have the same file |
| 1331 | open (whether this invocation of QEMU is the first or the |
| 1332 | second instance), both instances must permit shared access |
| 1333 | for the second instance to succeed at opening the file. |
| 1334 | |
| 1335 | Enabling ``force-share=on`` requires ``read-only=on``. |
| 1336 | |
| 1337 | ``cache.direct`` |
| 1338 | The host page cache can be avoided with ``cache.direct=on``. |
| 1339 | This will attempt to do disk IO directly to the guest's |
| 1340 | memory. QEMU may still perform an internal copy of the data. |
| 1341 | |
| 1342 | ``cache.no-flush`` |
| 1343 | In case you don't care about data integrity over host |
| 1344 | failures, you can use ``cache.no-flush=on``. This option |
| 1345 | tells QEMU that it never needs to write any data to the disk |
| 1346 | but can instead keep things in cache. If anything goes |
| 1347 | wrong, like your host losing power, the disk storage getting |
| 1348 | disconnected accidentally, etc. your image will most |
| 1349 | probably be rendered unusable. |
| 1350 | |
| 1351 | ``discard=discard`` |
| 1352 | discard is one of "ignore" (or "off") or "unmap" (or "on") |
| 1353 | and controls whether ``discard`` (also known as ``trim`` or |
| 1354 | ``unmap``) requests are ignored or passed to the filesystem. |
| 1355 | Some machine types may not support discard requests. |
| 1356 | |
| 1357 | ``detect-zeroes=detect-zeroes`` |
| 1358 | detect-zeroes is "off", "on" or "unmap" and enables the |
| 1359 | automatic conversion of plain zero writes by the OS to |
| 1360 | driver specific optimized zero write commands. You may even |
| 1361 | choose "unmap" if discard is set to "unmap" to allow a zero |
| 1362 | write to be converted to an ``unmap`` operation. |
| 1363 | |
| 1364 | ``Driver-specific options for file`` |
| 1365 | This is the protocol-level block driver for accessing regular |
| 1366 | files. |
| 1367 | |
| 1368 | ``filename`` |
| 1369 | The path to the image file in the local filesystem |
| 1370 | |
| 1371 | ``aio`` |
Stefano Garzarella | ad1e691 | 2020-09-24 17:15:11 +0200 | [diff] [blame] | 1372 | Specifies the AIO backend (threads/native/io_uring, |
| 1373 | default: threads) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1374 | |
| 1375 | ``locking`` |
| 1376 | Specifies whether the image file is protected with Linux OFD |
| 1377 | / POSIX locks. The default is to use the Linux Open File |
| 1378 | Descriptor API if available, otherwise no lock is applied. |
| 1379 | (auto/on/off, default: auto) |
| 1380 | |
| 1381 | Example: |
| 1382 | |
| 1383 | :: |
| 1384 | |
| 1385 | -blockdev driver=file,node-name=disk,filename=disk.img |
| 1386 | |
| 1387 | ``Driver-specific options for raw`` |
| 1388 | This is the image format block driver for raw images. It is |
| 1389 | usually stacked on top of a protocol level block driver such as |
| 1390 | ``file``. |
| 1391 | |
| 1392 | ``file`` |
| 1393 | Reference to or definition of the data source block driver |
| 1394 | node (e.g. a ``file`` driver node) |
| 1395 | |
| 1396 | Example 1: |
| 1397 | |
| 1398 | :: |
| 1399 | |
| 1400 | -blockdev driver=file,node-name=disk_file,filename=disk.img |
| 1401 | -blockdev driver=raw,node-name=disk,file=disk_file |
| 1402 | |
| 1403 | Example 2: |
| 1404 | |
| 1405 | :: |
| 1406 | |
| 1407 | -blockdev driver=raw,node-name=disk,file.driver=file,file.filename=disk.img |
| 1408 | |
| 1409 | ``Driver-specific options for qcow2`` |
| 1410 | This is the image format block driver for qcow2 images. It is |
| 1411 | usually stacked on top of a protocol level block driver such as |
| 1412 | ``file``. |
| 1413 | |
| 1414 | ``file`` |
| 1415 | Reference to or definition of the data source block driver |
| 1416 | node (e.g. a ``file`` driver node) |
| 1417 | |
| 1418 | ``backing`` |
| 1419 | Reference to or definition of the backing file block device |
| 1420 | (default is taken from the image file). It is allowed to |
| 1421 | pass ``null`` here in order to disable the default backing |
| 1422 | file. |
| 1423 | |
| 1424 | ``lazy-refcounts`` |
| 1425 | Whether to enable the lazy refcounts feature (on/off; |
| 1426 | default is taken from the image file) |
| 1427 | |
| 1428 | ``cache-size`` |
| 1429 | The maximum total size of the L2 table and refcount block |
| 1430 | caches in bytes (default: the sum of l2-cache-size and |
| 1431 | refcount-cache-size) |
| 1432 | |
| 1433 | ``l2-cache-size`` |
| 1434 | The maximum size of the L2 table cache in bytes (default: if |
| 1435 | cache-size is not specified - 32M on Linux platforms, and 8M |
| 1436 | on non-Linux platforms; otherwise, as large as possible |
| 1437 | within the cache-size, while permitting the requested or the |
| 1438 | minimal refcount cache size) |
| 1439 | |
| 1440 | ``refcount-cache-size`` |
| 1441 | The maximum size of the refcount block cache in bytes |
| 1442 | (default: 4 times the cluster size; or if cache-size is |
| 1443 | specified, the part of it which is not used for the L2 |
| 1444 | cache) |
| 1445 | |
| 1446 | ``cache-clean-interval`` |
| 1447 | Clean unused entries in the L2 and refcount caches. The |
| 1448 | interval is in seconds. The default value is 600 on |
| 1449 | supporting platforms, and 0 on other platforms. Setting it |
| 1450 | to 0 disables this feature. |
| 1451 | |
| 1452 | ``pass-discard-request`` |
| 1453 | Whether discard requests to the qcow2 device should be |
| 1454 | forwarded to the data source (on/off; default: on if |
| 1455 | discard=unmap is specified, off otherwise) |
| 1456 | |
| 1457 | ``pass-discard-snapshot`` |
| 1458 | Whether discard requests for the data source should be |
| 1459 | issued when a snapshot operation (e.g. deleting a snapshot) |
| 1460 | frees clusters in the qcow2 file (on/off; default: on) |
| 1461 | |
| 1462 | ``pass-discard-other`` |
| 1463 | Whether discard requests for the data source should be |
| 1464 | issued on other occasions where a cluster gets freed |
| 1465 | (on/off; default: off) |
| 1466 | |
Jean-Louis Dupond | 42a2890 | 2023-06-05 10:45:24 +0200 | [diff] [blame] | 1467 | ``discard-no-unref`` |
Jean-Louis Dupond | b2b1090 | 2023-10-03 14:52:37 +0200 | [diff] [blame] | 1468 | When enabled, data clusters will remain preallocated when they are |
| 1469 | no longer used, e.g. because they are discarded or converted to |
| 1470 | zero clusters. As usual, whether the old data is discarded or kept |
| 1471 | on the protocol level (i.e. in the image file) depends on the |
| 1472 | setting of the pass-discard-request option. Keeping the clusters |
| 1473 | preallocated prevents qcow2 fragmentation that would otherwise be |
| 1474 | caused by freeing and re-allocating them later. Besides potential |
Jean-Louis Dupond | 42a2890 | 2023-06-05 10:45:24 +0200 | [diff] [blame] | 1475 | performance degradation, such fragmentation can lead to increased |
| 1476 | allocation of clusters past the end of the image file, |
| 1477 | resulting in image files whose file length can grow much larger |
| 1478 | than their guest disk size would suggest. |
| 1479 | If image file length is of concern (e.g. when storing qcow2 |
| 1480 | images directly on block devices), you should consider enabling |
| 1481 | this option. |
| 1482 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1483 | ``overlap-check`` |
| 1484 | Which overlap checks to perform for writes to the image |
| 1485 | (none/constant/cached/all; default: cached). For details or |
| 1486 | finer granularity control refer to the QAPI documentation of |
| 1487 | ``blockdev-add``. |
| 1488 | |
| 1489 | Example 1: |
| 1490 | |
| 1491 | :: |
| 1492 | |
| 1493 | -blockdev driver=file,node-name=my_file,filename=/tmp/disk.qcow2 |
| 1494 | -blockdev driver=qcow2,node-name=hda,file=my_file,overlap-check=none,cache-size=16777216 |
| 1495 | |
| 1496 | Example 2: |
| 1497 | |
| 1498 | :: |
| 1499 | |
| 1500 | -blockdev driver=qcow2,node-name=disk,file.driver=http,file.filename=http://example.com/image.qcow2 |
| 1501 | |
| 1502 | ``Driver-specific options for other drivers`` |
| 1503 | Please refer to the QAPI documentation of the ``blockdev-add`` |
| 1504 | QMP command. |
| 1505 | ERST |
Markus Armbruster | 42e5f39 | 2017-02-28 22:27:07 +0100 | [diff] [blame] | 1506 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1507 | DEF("drive", HAS_ARG, QEMU_OPTION_drive, |
| 1508 | "-drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i]\n" |
Stefan Hajnoczi | 92196b2 | 2011-08-04 12:26:52 +0100 | [diff] [blame] | 1509 | " [,cache=writethrough|writeback|none|directsync|unsafe][,format=f]\n" |
Kevin Wolf | 572023f | 2018-06-13 11:01:30 +0200 | [diff] [blame] | 1510 | " [,snapshot=on|off][,rerror=ignore|stop|report]\n" |
Stefano Garzarella | ad1e691 | 2020-09-24 17:15:11 +0200 | [diff] [blame] | 1511 | " [,werror=ignore|stop|report|enospc][,id=name]\n" |
| 1512 | " [,aio=threads|native|io_uring]\n" |
Stefan Hajnoczi | fb0490f | 2011-11-17 13:40:32 +0000 | [diff] [blame] | 1513 | " [,readonly=on|off][,copy-on-read=on|off]\n" |
Peter Lieven | 2f7133b | 2014-07-28 21:53:02 +0200 | [diff] [blame] | 1514 | " [,discard=ignore|unmap][,detect-zeroes=on|off|unmap]\n" |
Benoît Canet | 3e9fab6 | 2013-09-02 14:14:40 +0200 | [diff] [blame] | 1515 | " [[,bps=b]|[[,bps_rd=r][,bps_wr=w]]]\n" |
| 1516 | " [[,iops=i]|[[,iops_rd=r][,iops_wr=w]]]\n" |
| 1517 | " [[,bps_max=bm]|[[,bps_rd_max=rm][,bps_wr_max=wm]]]\n" |
| 1518 | " [[,iops_max=im]|[[,iops_rd_max=irm][,iops_wr_max=iwm]]]\n" |
Benoît Canet | 2024c1d | 2013-09-02 14:14:41 +0200 | [diff] [blame] | 1519 | " [[,iops_size=is]]\n" |
Alberto Garcia | 76f4afb | 2015-06-08 18:17:44 +0200 | [diff] [blame] | 1520 | " [[,group=g]]\n" |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 1521 | " use 'file' as a drive image\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1522 | SRST |
| 1523 | ``-drive option[,option[,option[,...]]]`` |
| 1524 | Define a new drive. This includes creating a block driver node (the |
| 1525 | backend) as well as a guest device, and is mostly a shortcut for |
| 1526 | defining the corresponding ``-blockdev`` and ``-device`` options. |
| 1527 | |
| 1528 | ``-drive`` accepts all options that are accepted by ``-blockdev``. |
| 1529 | In addition, it knows the following options: |
| 1530 | |
| 1531 | ``file=file`` |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 1532 | This option defines which disk image (see the :ref:`disk images` |
| 1533 | chapter in the System Emulation Users Guide) to use with this drive. |
| 1534 | If the filename contains comma, you must double it (for instance, |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1535 | "file=my,,file" to use file "my,file"). |
| 1536 | |
| 1537 | Special files such as iSCSI devices can be specified using |
| 1538 | protocol specific URLs. See the section for "Device URL Syntax" |
| 1539 | for more information. |
| 1540 | |
| 1541 | ``if=interface`` |
| 1542 | This option defines on which type on interface the drive is |
| 1543 | connected. Available types are: ide, scsi, sd, mtd, floppy, |
| 1544 | pflash, virtio, none. |
| 1545 | |
| 1546 | ``bus=bus,unit=unit`` |
| 1547 | These options define where is connected the drive by defining |
| 1548 | the bus number and the unit id. |
| 1549 | |
| 1550 | ``index=index`` |
Laurent Vivier | 35aab30 | 2022-02-02 15:34:22 +0100 | [diff] [blame] | 1551 | This option defines where the drive is connected by using an |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1552 | index in the list of available connectors of a given interface |
| 1553 | type. |
| 1554 | |
| 1555 | ``media=media`` |
| 1556 | This option defines the type of the media: disk or cdrom. |
| 1557 | |
| 1558 | ``snapshot=snapshot`` |
| 1559 | snapshot is "on" or "off" and controls snapshot mode for the |
| 1560 | given drive (see ``-snapshot``). |
| 1561 | |
| 1562 | ``cache=cache`` |
| 1563 | cache is "none", "writeback", "unsafe", "directsync" or |
| 1564 | "writethrough" and controls how the host cache is used to access |
| 1565 | block data. This is a shortcut that sets the ``cache.direct`` |
| 1566 | and ``cache.no-flush`` options (as in ``-blockdev``), and |
| 1567 | additionally ``cache.writeback``, which provides a default for |
| 1568 | the ``write-cache`` option of block guest devices (as in |
| 1569 | ``-device``). The modes correspond to the following settings: |
| 1570 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1571 | ============= =============== ============ ============== |
| 1572 | \ cache.writeback cache.direct cache.no-flush |
| 1573 | ============= =============== ============ ============== |
| 1574 | writeback on off off |
| 1575 | none on on off |
| 1576 | writethrough off off off |
| 1577 | directsync off on off |
| 1578 | unsafe on off on |
| 1579 | ============= =============== ============ ============== |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1580 | |
| 1581 | The default mode is ``cache=writeback``. |
| 1582 | |
| 1583 | ``aio=aio`` |
Stefano Garzarella | ad1e691 | 2020-09-24 17:15:11 +0200 | [diff] [blame] | 1584 | aio is "threads", "native", or "io_uring" and selects between pthread |
| 1585 | based disk I/O, native Linux AIO, or Linux io_uring API. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1586 | |
| 1587 | ``format=format`` |
| 1588 | Specify which disk format will be used rather than detecting the |
| 1589 | format. Can be used to specify format=raw to avoid interpreting |
| 1590 | an untrusted format header. |
| 1591 | |
| 1592 | ``werror=action,rerror=action`` |
| 1593 | Specify which action to take on write and read errors. Valid |
| 1594 | actions are: "ignore" (ignore the error and try to continue), |
| 1595 | "stop" (pause QEMU), "report" (report the error to the guest), |
| 1596 | "enospc" (pause QEMU only if the host disk is full; report the |
| 1597 | error to the guest otherwise). The default setting is |
| 1598 | ``werror=enospc`` and ``rerror=report``. |
| 1599 | |
| 1600 | ``copy-on-read=copy-on-read`` |
| 1601 | copy-on-read is "on" or "off" and enables whether to copy read |
| 1602 | backing file sectors into the image file. |
| 1603 | |
| 1604 | ``bps=b,bps_rd=r,bps_wr=w`` |
| 1605 | Specify bandwidth throttling limits in bytes per second, either |
| 1606 | for all request types or for reads or writes only. Small values |
| 1607 | can lead to timeouts or hangs inside the guest. A safe minimum |
| 1608 | for disks is 2 MB/s. |
| 1609 | |
| 1610 | ``bps_max=bm,bps_rd_max=rm,bps_wr_max=wm`` |
| 1611 | Specify bursts in bytes per second, either for all request types |
| 1612 | or for reads or writes only. Bursts allow the guest I/O to spike |
| 1613 | above the limit temporarily. |
| 1614 | |
| 1615 | ``iops=i,iops_rd=r,iops_wr=w`` |
| 1616 | Specify request rate limits in requests per second, either for |
| 1617 | all request types or for reads or writes only. |
| 1618 | |
| 1619 | ``iops_max=bm,iops_rd_max=rm,iops_wr_max=wm`` |
| 1620 | Specify bursts in requests per second, either for all request |
| 1621 | types or for reads or writes only. Bursts allow the guest I/O to |
| 1622 | spike above the limit temporarily. |
| 1623 | |
| 1624 | ``iops_size=is`` |
| 1625 | Let every is bytes of a request count as a new request for iops |
| 1626 | throttling purposes. Use this option to prevent guests from |
| 1627 | circumventing iops limits by sending fewer but larger requests. |
| 1628 | |
| 1629 | ``group=g`` |
| 1630 | Join a throttling quota group with given name g. All drives that |
| 1631 | are members of the same group are accounted for together. Use |
| 1632 | this option to prevent guests from circumventing throttling |
| 1633 | limits by using many small disks instead of a single larger |
| 1634 | disk. |
| 1635 | |
| 1636 | By default, the ``cache.writeback=on`` mode is used. It will report |
| 1637 | data writes as completed as soon as the data is present in the host |
| 1638 | page cache. This is safe as long as your guest OS makes sure to |
| 1639 | correctly flush disk caches where needed. If your guest OS does not |
| 1640 | handle volatile disk write caches correctly and your host crashes or |
| 1641 | loses power, then the guest may experience data corruption. |
| 1642 | |
| 1643 | For such guests, you should consider using ``cache.writeback=off``. |
| 1644 | This means that the host page cache will be used to read and write |
| 1645 | data, but write notification will be sent to the guest only after |
| 1646 | QEMU has made sure to flush each write to the disk. Be aware that |
| 1647 | this has a major impact on performance. |
| 1648 | |
| 1649 | When using the ``-snapshot`` option, unsafe caching is always used. |
| 1650 | |
| 1651 | Copy-on-read avoids accessing the same backing file sectors |
| 1652 | repeatedly and is useful when the backing file is over a slow |
| 1653 | network. By default copy-on-read is off. |
| 1654 | |
| 1655 | Instead of ``-cdrom`` you can use: |
| 1656 | |
| 1657 | .. parsed-literal:: |
| 1658 | |
| 1659 | |qemu_system| -drive file=file,index=2,media=cdrom |
| 1660 | |
| 1661 | Instead of ``-hda``, ``-hdb``, ``-hdc``, ``-hdd``, you can use: |
| 1662 | |
| 1663 | .. parsed-literal:: |
| 1664 | |
| 1665 | |qemu_system| -drive file=file,index=0,media=disk |
| 1666 | |qemu_system| -drive file=file,index=1,media=disk |
| 1667 | |qemu_system| -drive file=file,index=2,media=disk |
| 1668 | |qemu_system| -drive file=file,index=3,media=disk |
| 1669 | |
| 1670 | You can open an image using pre-opened file descriptors from an fd |
| 1671 | set: |
| 1672 | |
| 1673 | .. parsed-literal:: |
| 1674 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 1675 | |qemu_system| \\ |
| 1676 | -add-fd fd=3,set=2,opaque="rdwr:/path/to/file" \\ |
| 1677 | -add-fd fd=4,set=2,opaque="rdonly:/path/to/file" \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1678 | -drive file=/dev/fdset/2,index=0,media=disk |
| 1679 | |
| 1680 | You can connect a CDROM to the slave of ide0: |
| 1681 | |
| 1682 | .. parsed-literal:: |
| 1683 | |
| 1684 | |qemu_system_x86| -drive file=file,if=ide,index=1,media=cdrom |
| 1685 | |
| 1686 | If you don't specify the "file=" argument, you define an empty |
| 1687 | drive: |
| 1688 | |
| 1689 | .. parsed-literal:: |
| 1690 | |
| 1691 | |qemu_system_x86| -drive if=ide,index=1,media=cdrom |
| 1692 | |
| 1693 | Instead of ``-fda``, ``-fdb``, you can use: |
| 1694 | |
| 1695 | .. parsed-literal:: |
| 1696 | |
| 1697 | |qemu_system_x86| -drive file=file,index=0,if=floppy |
| 1698 | |qemu_system_x86| -drive file=file,index=1,if=floppy |
| 1699 | |
| 1700 | By default, interface is "ide" and index is automatically |
| 1701 | incremented: |
| 1702 | |
| 1703 | .. parsed-literal:: |
| 1704 | |
John Snow | a234ec3 | 2023-02-02 17:31:21 -0500 | [diff] [blame] | 1705 | |qemu_system_x86| -drive file=a -drive file=b |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1706 | |
| 1707 | is interpreted like: |
| 1708 | |
| 1709 | .. parsed-literal:: |
| 1710 | |
| 1711 | |qemu_system_x86| -hda a -hdb b |
| 1712 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1713 | |
| 1714 | DEF("mtdblock", HAS_ARG, QEMU_OPTION_mtdblock, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 1715 | "-mtdblock file use 'file' as on-board Flash memory image\n", |
| 1716 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1717 | SRST |
| 1718 | ``-mtdblock file`` |
| 1719 | Use file as on-board Flash memory image. |
| 1720 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1721 | |
| 1722 | DEF("sd", HAS_ARG, QEMU_OPTION_sd, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 1723 | "-sd file use 'file' as SecureDigital card image\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1724 | SRST |
| 1725 | ``-sd file`` |
| 1726 | Use file as SecureDigital card image. |
| 1727 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1728 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1729 | DEF("snapshot", 0, QEMU_OPTION_snapshot, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 1730 | "-snapshot write to temporary files instead of disk image files\n", |
| 1731 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1732 | SRST |
| 1733 | ``-snapshot`` |
| 1734 | Write to temporary files instead of disk image files. In this case, |
| 1735 | the raw disk image you use is not written back. You can however |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 1736 | force the write back by pressing C-a s (see the :ref:`disk images` |
| 1737 | chapter in the System Emulation Users Guide). |
Alex Bennée | c1654c3 | 2023-04-24 10:22:37 +0100 | [diff] [blame] | 1738 | |
| 1739 | .. warning:: |
| 1740 | snapshot is incompatible with ``-blockdev`` (instead use qemu-img |
| 1741 | to manually create snapshot images to attach to your blockdev). |
| 1742 | If you have mixed ``-blockdev`` and ``-drive`` declarations you |
| 1743 | can use the 'snapshot' property on your drive declarations |
| 1744 | instead of this global option. |
| 1745 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1746 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 1747 | |
Gautham R Shenoy | 74db920 | 2010-04-29 17:44:43 +0530 | [diff] [blame] | 1748 | DEF("fsdev", HAS_ARG, QEMU_OPTION_fsdev, |
Greg Kurz | b44a6b0 | 2019-05-17 17:34:49 +0200 | [diff] [blame] | 1749 | "-fsdev local,id=id,path=path,security_model=mapped-xattr|mapped-file|passthrough|none\n" |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1750 | " [,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode]\n" |
Pradeep Jagadeesh | b8bbdb8 | 2017-02-28 10:31:46 +0100 | [diff] [blame] | 1751 | " [[,throttling.bps-total=b]|[[,throttling.bps-read=r][,throttling.bps-write=w]]]\n" |
| 1752 | " [[,throttling.iops-total=i]|[[,throttling.iops-read=r][,throttling.iops-write=w]]]\n" |
| 1753 | " [[,throttling.bps-total-max=bm]|[[,throttling.bps-read-max=rm][,throttling.bps-write-max=wm]]]\n" |
| 1754 | " [[,throttling.iops-total-max=im]|[[,throttling.iops-read-max=irm][,throttling.iops-write-max=iwm]]]\n" |
Greg Kurz | b44a6b0 | 2019-05-17 17:34:49 +0200 | [diff] [blame] | 1755 | " [[,throttling.iops-size=is]]\n" |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1756 | "-fsdev proxy,id=id,socket=socket[,writeout=immediate][,readonly=on]\n" |
| 1757 | "-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=immediate][,readonly=on]\n" |
Greg Kurz | b44a6b0 | 2019-05-17 17:34:49 +0200 | [diff] [blame] | 1758 | "-fsdev synth,id=id\n", |
Gautham R Shenoy | 74db920 | 2010-04-29 17:44:43 +0530 | [diff] [blame] | 1759 | QEMU_ARCH_ALL) |
| 1760 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1761 | SRST |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1762 | ``-fsdev local,id=id,path=path,security_model=security_model [,writeout=writeout][,readonly=on][,fmode=fmode][,dmode=dmode] [,throttling.option=value[,throttling.option=value[,...]]]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1763 | \ |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1764 | ``-fsdev proxy,id=id,socket=socket[,writeout=writeout][,readonly=on]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1765 | \ |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1766 | ``-fsdev proxy,id=id,sock_fd=sock_fd[,writeout=writeout][,readonly=on]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1767 | \ |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1768 | ``-fsdev synth,id=id[,readonly=on]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1769 | Define a new file system device. Valid options are: |
| 1770 | |
| 1771 | ``local`` |
| 1772 | Accesses to the filesystem are done by QEMU. |
| 1773 | |
| 1774 | ``proxy`` |
Christian Schoenebeck | 71d72ec | 2023-06-26 13:49:06 +0200 | [diff] [blame] | 1775 | Accesses to the filesystem are done by virtfs-proxy-helper(1). This |
| 1776 | option is deprecated (since QEMU 8.1) and will be removed in a future |
| 1777 | version of QEMU. Use ``local`` instead. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1778 | |
| 1779 | ``synth`` |
| 1780 | Synthetic filesystem, only used by QTests. |
| 1781 | |
| 1782 | ``id=id`` |
| 1783 | Specifies identifier for this device. |
| 1784 | |
| 1785 | ``path=path`` |
| 1786 | Specifies the export path for the file system device. Files |
| 1787 | under this path will be available to the 9p client on the guest. |
| 1788 | |
| 1789 | ``security_model=security_model`` |
| 1790 | Specifies the security model to be used for this export path. |
| 1791 | Supported security models are "passthrough", "mapped-xattr", |
| 1792 | "mapped-file" and "none". In "passthrough" security model, files |
| 1793 | are stored using the same credentials as they are created on the |
| 1794 | guest. This requires QEMU to run as root. In "mapped-xattr" |
| 1795 | security model, some of the file attributes like uid, gid, mode |
| 1796 | bits and link target are stored as file attributes. For |
| 1797 | "mapped-file" these attributes are stored in the hidden |
| 1798 | .virtfs\_metadata directory. Directories exported by this |
| 1799 | security model cannot interact with other unix tools. "none" |
| 1800 | security model is same as passthrough except the sever won't |
| 1801 | report failures if it fails to set file attributes like |
| 1802 | ownership. Security model is mandatory only for local fsdriver. |
| 1803 | Other fsdrivers (like proxy) don't take security model as a |
| 1804 | parameter. |
| 1805 | |
| 1806 | ``writeout=writeout`` |
| 1807 | This is an optional argument. The only supported value is |
| 1808 | "immediate". This means that host page cache will be used to |
| 1809 | read and write data but write notification will be sent to the |
| 1810 | guest only when the data has been reported as written by the |
| 1811 | storage subsystem. |
| 1812 | |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1813 | ``readonly=on`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1814 | Enables exporting 9p share as a readonly mount for guests. By |
| 1815 | default read-write access is given. |
| 1816 | |
| 1817 | ``socket=socket`` |
| 1818 | Enables proxy filesystem driver to use passed socket file for |
| 1819 | communicating with virtfs-proxy-helper(1). |
| 1820 | |
| 1821 | ``sock_fd=sock_fd`` |
| 1822 | Enables proxy filesystem driver to use passed socket descriptor |
| 1823 | for communicating with virtfs-proxy-helper(1). Usually a helper |
| 1824 | like libvirt will create socketpair and pass one of the fds as |
| 1825 | sock\_fd. |
| 1826 | |
| 1827 | ``fmode=fmode`` |
| 1828 | Specifies the default mode for newly created files on the host. |
| 1829 | Works only with security models "mapped-xattr" and |
| 1830 | "mapped-file". |
| 1831 | |
| 1832 | ``dmode=dmode`` |
| 1833 | Specifies the default mode for newly created directories on the |
| 1834 | host. Works only with security models "mapped-xattr" and |
| 1835 | "mapped-file". |
| 1836 | |
| 1837 | ``throttling.bps-total=b,throttling.bps-read=r,throttling.bps-write=w`` |
| 1838 | Specify bandwidth throttling limits in bytes per second, either |
| 1839 | for all request types or for reads or writes only. |
| 1840 | |
| 1841 | ``throttling.bps-total-max=bm,bps-read-max=rm,bps-write-max=wm`` |
| 1842 | Specify bursts in bytes per second, either for all request types |
| 1843 | or for reads or writes only. Bursts allow the guest I/O to spike |
| 1844 | above the limit temporarily. |
| 1845 | |
| 1846 | ``throttling.iops-total=i,throttling.iops-read=r, throttling.iops-write=w`` |
| 1847 | Specify request rate limits in requests per second, either for |
| 1848 | all request types or for reads or writes only. |
| 1849 | |
| 1850 | ``throttling.iops-total-max=im,throttling.iops-read-max=irm, throttling.iops-write-max=iwm`` |
| 1851 | Specify bursts in requests per second, either for all request |
| 1852 | types or for reads or writes only. Bursts allow the guest I/O to |
| 1853 | spike above the limit temporarily. |
| 1854 | |
| 1855 | ``throttling.iops-size=is`` |
| 1856 | Let every is bytes of a request count as a new request for iops |
| 1857 | throttling purposes. |
| 1858 | |
| 1859 | -fsdev option is used along with -device driver "virtio-9p-...". |
| 1860 | |
| 1861 | ``-device virtio-9p-type,fsdev=id,mount_tag=mount_tag`` |
| 1862 | Options for virtio-9p-... driver are: |
| 1863 | |
| 1864 | ``type`` |
| 1865 | Specifies the variant to be used. Supported values are "pci", |
| 1866 | "ccw" or "device", depending on the machine type. |
| 1867 | |
| 1868 | ``fsdev=id`` |
| 1869 | Specifies the id value specified along with -fsdev option. |
| 1870 | |
| 1871 | ``mount_tag=mount_tag`` |
| 1872 | Specifies the tag name to be used by the guest to mount this |
| 1873 | export point. |
| 1874 | ERST |
Gautham R Shenoy | 74db920 | 2010-04-29 17:44:43 +0530 | [diff] [blame] | 1875 | |
Gautham R Shenoy | 3d54abc | 2010-04-29 17:45:03 +0530 | [diff] [blame] | 1876 | DEF("virtfs", HAS_ARG, QEMU_OPTION_virtfs, |
Greg Kurz | b44a6b0 | 2019-05-17 17:34:49 +0200 | [diff] [blame] | 1877 | "-virtfs local,path=path,mount_tag=tag,security_model=mapped-xattr|mapped-file|passthrough|none\n" |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1878 | " [,id=id][,writeout=immediate][,readonly=on][,fmode=fmode][,dmode=dmode][,multidevs=remap|forbid|warn]\n" |
| 1879 | "-virtfs proxy,mount_tag=tag,socket=socket[,id=id][,writeout=immediate][,readonly=on]\n" |
| 1880 | "-virtfs proxy,mount_tag=tag,sock_fd=sock_fd[,id=id][,writeout=immediate][,readonly=on]\n" |
| 1881 | "-virtfs synth,mount_tag=tag[,id=id][,readonly=on]\n", |
Gautham R Shenoy | 3d54abc | 2010-04-29 17:45:03 +0530 | [diff] [blame] | 1882 | QEMU_ARCH_ALL) |
| 1883 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1884 | SRST |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1885 | ``-virtfs local,path=path,mount_tag=mount_tag ,security_model=security_model[,writeout=writeout][,readonly=on] [,fmode=fmode][,dmode=dmode][,multidevs=multidevs]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1886 | \ |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1887 | ``-virtfs proxy,socket=socket,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1888 | \ |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1889 | ``-virtfs proxy,sock_fd=sock_fd,mount_tag=mount_tag [,writeout=writeout][,readonly=on]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 1890 | \ |
| 1891 | ``-virtfs synth,mount_tag=mount_tag`` |
Christian Schoenebeck | 65abaa0 | 2020-05-14 08:06:43 +0200 | [diff] [blame] | 1892 | Define a new virtual filesystem device and expose it to the guest using |
| 1893 | a virtio-9p-device (a.k.a. 9pfs), which essentially means that a certain |
| 1894 | directory on host is made directly accessible by guest as a pass-through |
| 1895 | file system by using the 9P network protocol for communication between |
| 1896 | host and guests, if desired even accessible, shared by several guests |
Stefan Weil | 2cb40d4 | 2022-11-10 20:08:25 +0100 | [diff] [blame] | 1897 | simultaneously. |
Christian Schoenebeck | 65abaa0 | 2020-05-14 08:06:43 +0200 | [diff] [blame] | 1898 | |
| 1899 | Note that ``-virtfs`` is actually just a convenience shortcut for its |
| 1900 | generalized form ``-fsdev -device virtio-9p-pci``. |
| 1901 | |
| 1902 | The general form of pass-through file system options are: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1903 | |
| 1904 | ``local`` |
| 1905 | Accesses to the filesystem are done by QEMU. |
| 1906 | |
| 1907 | ``proxy`` |
| 1908 | Accesses to the filesystem are done by virtfs-proxy-helper(1). |
Christian Schoenebeck | 71d72ec | 2023-06-26 13:49:06 +0200 | [diff] [blame] | 1909 | This option is deprecated (since QEMU 8.1) and will be removed in a |
| 1910 | future version of QEMU. Use ``local`` instead. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1911 | |
| 1912 | ``synth`` |
| 1913 | Synthetic filesystem, only used by QTests. |
| 1914 | |
| 1915 | ``id=id`` |
| 1916 | Specifies identifier for the filesystem device |
| 1917 | |
| 1918 | ``path=path`` |
| 1919 | Specifies the export path for the file system device. Files |
| 1920 | under this path will be available to the 9p client on the guest. |
| 1921 | |
| 1922 | ``security_model=security_model`` |
| 1923 | Specifies the security model to be used for this export path. |
| 1924 | Supported security models are "passthrough", "mapped-xattr", |
| 1925 | "mapped-file" and "none". In "passthrough" security model, files |
| 1926 | are stored using the same credentials as they are created on the |
| 1927 | guest. This requires QEMU to run as root. In "mapped-xattr" |
| 1928 | security model, some of the file attributes like uid, gid, mode |
| 1929 | bits and link target are stored as file attributes. For |
| 1930 | "mapped-file" these attributes are stored in the hidden |
| 1931 | .virtfs\_metadata directory. Directories exported by this |
| 1932 | security model cannot interact with other unix tools. "none" |
| 1933 | security model is same as passthrough except the sever won't |
| 1934 | report failures if it fails to set file attributes like |
| 1935 | ownership. Security model is mandatory only for local fsdriver. |
| 1936 | Other fsdrivers (like proxy) don't take security model as a |
| 1937 | parameter. |
| 1938 | |
| 1939 | ``writeout=writeout`` |
| 1940 | This is an optional argument. The only supported value is |
| 1941 | "immediate". This means that host page cache will be used to |
| 1942 | read and write data but write notification will be sent to the |
| 1943 | guest only when the data has been reported as written by the |
| 1944 | storage subsystem. |
| 1945 | |
Paolo Bonzini | 991c180 | 2020-11-13 03:10:52 -0500 | [diff] [blame] | 1946 | ``readonly=on`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 1947 | Enables exporting 9p share as a readonly mount for guests. By |
| 1948 | default read-write access is given. |
| 1949 | |
| 1950 | ``socket=socket`` |
| 1951 | Enables proxy filesystem driver to use passed socket file for |
| 1952 | communicating with virtfs-proxy-helper(1). Usually a helper like |
| 1953 | libvirt will create socketpair and pass one of the fds as |
| 1954 | sock\_fd. |
| 1955 | |
| 1956 | ``sock_fd`` |
| 1957 | Enables proxy filesystem driver to use passed 'sock\_fd' as the |
| 1958 | socket descriptor for interfacing with virtfs-proxy-helper(1). |
| 1959 | |
| 1960 | ``fmode=fmode`` |
| 1961 | Specifies the default mode for newly created files on the host. |
| 1962 | Works only with security models "mapped-xattr" and |
| 1963 | "mapped-file". |
| 1964 | |
| 1965 | ``dmode=dmode`` |
| 1966 | Specifies the default mode for newly created directories on the |
| 1967 | host. Works only with security models "mapped-xattr" and |
| 1968 | "mapped-file". |
| 1969 | |
| 1970 | ``mount_tag=mount_tag`` |
| 1971 | Specifies the tag name to be used by the guest to mount this |
| 1972 | export point. |
| 1973 | |
| 1974 | ``multidevs=multidevs`` |
| 1975 | Specifies how to deal with multiple devices being shared with a |
| 1976 | 9p export. Supported behaviours are either "remap", "forbid" or |
| 1977 | "warn". The latter is the default behaviour on which virtfs 9p |
| 1978 | expects only one device to be shared with the same export, and |
| 1979 | if more than one device is shared and accessed via the same 9p |
| 1980 | export then only a warning message is logged (once) by qemu on |
| 1981 | host side. In order to avoid file ID collisions on guest you |
| 1982 | should either create a separate virtfs export for each device to |
| 1983 | be shared with guests (recommended way) or you might use "remap" |
| 1984 | instead which allows you to share multiple devices with only one |
| 1985 | export instead, which is achieved by remapping the original |
| 1986 | inode numbers from host to guest in a way that would prevent |
| 1987 | such collisions. Remapping inodes in such use cases is required |
| 1988 | because the original device IDs from host are never passed and |
| 1989 | exposed on guest. Instead all files of an export shared with |
| 1990 | virtfs always share the same device id on guest. So two files |
| 1991 | with identical inode numbers but from actually different devices |
| 1992 | on host would otherwise cause a file ID collision and hence |
| 1993 | potential misbehaviours on guest. "forbid" on the other hand |
| 1994 | assumes like "warn" that only one device is shared by the same |
| 1995 | export, however it will not only log a warning message but also |
| 1996 | deny access to additional devices on guest. Note though that |
| 1997 | "forbid" does currently not block all possible file access |
| 1998 | operations (e.g. readdir() would still return entries from other |
| 1999 | devices). |
| 2000 | ERST |
Gautham R Shenoy | 3d54abc | 2010-04-29 17:45:03 +0530 | [diff] [blame] | 2001 | |
Markus Armbruster | 61d7048 | 2017-10-02 16:03:03 +0200 | [diff] [blame] | 2002 | DEF("iscsi", HAS_ARG, QEMU_OPTION_iscsi, |
Daniel P. Berrangé | c3b3a6c | 2022-12-01 04:25:05 -0500 | [diff] [blame] | 2003 | "-iscsi [user=user][,password=password][,password-secret=secret-id]\n" |
| 2004 | " [,header-digest=CRC32C|CR32C-NONE|NONE-CRC32C|NONE]\n" |
Markus Armbruster | 61d7048 | 2017-10-02 16:03:03 +0200 | [diff] [blame] | 2005 | " [,initiator-name=initiator-iqn][,id=target-iqn]\n" |
| 2006 | " [,timeout=timeout]\n" |
| 2007 | " iSCSI session parameters\n", QEMU_ARCH_ALL) |
| 2008 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2009 | SRST |
| 2010 | ``-iscsi`` |
| 2011 | Configure iSCSI session parameters. |
| 2012 | ERST |
Markus Armbruster | 4474314 | 2017-10-02 16:03:04 +0200 | [diff] [blame] | 2013 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2014 | DEFHEADING() |
| 2015 | |
Thomas Huth | c2a34ab | 2021-03-10 18:33:23 +0100 | [diff] [blame] | 2016 | DEFHEADING(USB convenience options:) |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 2017 | |
| 2018 | DEF("usb", 0, QEMU_OPTION_usb, |
Stefan Hajnoczi | 73f46fe | 2019-08-15 15:14:28 +0100 | [diff] [blame] | 2019 | "-usb enable on-board USB host controller (if not enabled by default)\n", |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 2020 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2021 | SRST |
| 2022 | ``-usb`` |
| 2023 | Enable USB emulation on machine types with an on-board USB host |
| 2024 | controller (if not enabled by default). Note that on-board USB host |
| 2025 | controllers may not support USB 3.0. In this case |
| 2026 | ``-device qemu-xhci`` can be used instead on machines with PCI. |
| 2027 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 2028 | |
| 2029 | DEF("usbdevice", HAS_ARG, QEMU_OPTION_usbdevice, |
| 2030 | "-usbdevice name add the host or guest USB device 'name'\n", |
| 2031 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2032 | SRST |
| 2033 | ``-usbdevice devname`` |
Thomas Huth | c2a34ab | 2021-03-10 18:33:23 +0100 | [diff] [blame] | 2034 | Add the USB device devname, and enable an on-board USB controller |
| 2035 | if possible and necessary (just like it can be done via |
| 2036 | ``-machine usb=on``). Note that this option is mainly intended for |
| 2037 | the user's convenience only. More fine-grained control can be |
| 2038 | achieved by selecting a USB host controller (if necessary) and the |
| 2039 | desired USB device via the ``-device`` option instead. For example, |
| 2040 | instead of using ``-usbdevice mouse`` it is possible to use |
| 2041 | ``-device qemu-xhci -device usb-mouse`` to connect the USB mouse |
| 2042 | to a USB 3.0 controller instead (at least on machines that support |
| 2043 | PCI and do not have an USB controller enabled by default yet). |
| 2044 | For more details, see the chapter about |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 2045 | :ref:`Connecting USB devices` in the System Emulation Users Guide. |
Thomas Huth | c2a34ab | 2021-03-10 18:33:23 +0100 | [diff] [blame] | 2046 | Possible devices for devname are: |
| 2047 | |
| 2048 | ``braille`` |
| 2049 | Braille device. This will use BrlAPI to display the braille |
| 2050 | output on a real or fake device (i.e. it also creates a |
| 2051 | corresponding ``braille`` chardev automatically beside the |
| 2052 | ``usb-braille`` USB device). |
| 2053 | |
Thomas Huth | c2a34ab | 2021-03-10 18:33:23 +0100 | [diff] [blame] | 2054 | ``keyboard`` |
| 2055 | Standard USB keyboard. Will override the PS/2 keyboard (if present). |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2056 | |
| 2057 | ``mouse`` |
| 2058 | Virtual Mouse. This will override the PS/2 mouse emulation when |
| 2059 | activated. |
| 2060 | |
| 2061 | ``tablet`` |
| 2062 | Pointer device that uses absolute coordinates (like a |
| 2063 | touchscreen). This means QEMU is able to report the mouse |
| 2064 | position without having to grab the mouse. Also overrides the |
| 2065 | PS/2 mouse emulation when activated. |
| 2066 | |
Thomas Huth | c2a34ab | 2021-03-10 18:33:23 +0100 | [diff] [blame] | 2067 | ``wacom-tablet`` |
| 2068 | Wacom PenPartner USB tablet. |
| 2069 | |
| 2070 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2071 | ERST |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 2072 | |
Markus Armbruster | 10adb8b | 2013-02-13 19:49:43 +0100 | [diff] [blame] | 2073 | DEFHEADING() |
| 2074 | |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 2075 | DEFHEADING(Display options:) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2076 | |
Jes Sorensen | 1472a95 | 2011-03-16 13:33:31 +0100 | [diff] [blame] | 2077 | DEF("display", HAS_ARG, QEMU_OPTION_display, |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2078 | #if defined(CONFIG_SPICE) |
Marc-André Lureau | d8aec9d | 2019-02-21 12:07:03 +0100 | [diff] [blame] | 2079 | "-display spice-app[,gl=on|off]\n" |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2080 | #endif |
| 2081 | #if defined(CONFIG_SDL) |
Thomas Huth | a743d60 | 2022-05-19 17:56:23 +0200 | [diff] [blame] | 2082 | "-display sdl[,gl=on|core|es|off][,grab-mod=<mod>][,show-cursor=on|off]\n" |
| 2083 | " [,window-close=on|off]\n" |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2084 | #endif |
Robert Ho | f04ec5a | 2016-07-26 18:17:11 +0800 | [diff] [blame] | 2085 | #if defined(CONFIG_GTK) |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2086 | "-display gtk[,full-screen=on|off][,gl=on|off][,grab-on-hover=on|off]\n" |
Felix xq Queißner | c34a933 | 2022-07-12 15:37:53 +0200 | [diff] [blame] | 2087 | " [,show-tabs=on|off][,show-cursor=on|off][,window-close=on|off]\n" |
BALATON Zoltan | e26c940 | 2024-02-09 01:05:06 +0100 | [diff] [blame] | 2088 | " [,show-menubar=on|off][,zoom-to-fit=on|off]\n" |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2089 | #endif |
| 2090 | #if defined(CONFIG_VNC) |
| 2091 | "-display vnc=<display>[,<optargs>]\n" |
| 2092 | #endif |
| 2093 | #if defined(CONFIG_CURSES) |
| 2094 | "-display curses[,charset=<encoding>]\n" |
| 2095 | #endif |
Gustavo Noronha Silva | f844cdb | 2022-03-06 21:11:18 +0900 | [diff] [blame] | 2096 | #if defined(CONFIG_COCOA) |
Gustavo Noronha Silva | 4797adc | 2022-03-06 21:11:19 +0900 | [diff] [blame] | 2097 | "-display cocoa[,full-grab=on|off][,swap-opt-cmd=on|off]\n" |
Akihiko Odaki | 9ab8715 | 2023-12-14 15:31:35 +0900 | [diff] [blame] | 2098 | " [,show-cursor=on|off][,left-command-key=on|off]\n" |
Akihiko Odaki | d502dfc | 2023-12-14 15:31:36 +0900 | [diff] [blame] | 2099 | " [,full-screen=on|off][,zoom-to-fit=on|off]\n" |
Gustavo Noronha Silva | f844cdb | 2022-03-06 21:11:18 +0900 | [diff] [blame] | 2100 | #endif |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2101 | #if defined(CONFIG_OPENGL) |
| 2102 | "-display egl-headless[,rendernode=<file>]\n" |
| 2103 | #endif |
Marc-André Lureau | 142ca62 | 2021-07-15 11:53:53 +0400 | [diff] [blame] | 2104 | #if defined(CONFIG_DBUS_DISPLAY) |
| 2105 | "-display dbus[,addr=<dbusaddr>]\n" |
| 2106 | " [,gl=on|core|es|off][,rendernode=<file>]\n" |
| 2107 | #endif |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2108 | "-display none\n" |
| 2109 | " select display backend type\n" |
| 2110 | " The default display is equivalent to\n " |
| 2111 | #if defined(CONFIG_GTK) |
| 2112 | "\"-display gtk\"\n" |
Robert Ho | f04ec5a | 2016-07-26 18:17:11 +0800 | [diff] [blame] | 2113 | #elif defined(CONFIG_SDL) |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2114 | "\"-display sdl\"\n" |
Robert Ho | f04ec5a | 2016-07-26 18:17:11 +0800 | [diff] [blame] | 2115 | #elif defined(CONFIG_COCOA) |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2116 | "\"-display cocoa\"\n" |
Robert Ho | f04ec5a | 2016-07-26 18:17:11 +0800 | [diff] [blame] | 2117 | #elif defined(CONFIG_VNC) |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2118 | "\"-vnc localhost:0,to=99,id=default\"\n" |
Robert Ho | f04ec5a | 2016-07-26 18:17:11 +0800 | [diff] [blame] | 2119 | #else |
Thomas Huth | 88b40c6 | 2019-10-23 14:01:28 +0200 | [diff] [blame] | 2120 | "\"-display none\"\n" |
Robert Ho | f04ec5a | 2016-07-26 18:17:11 +0800 | [diff] [blame] | 2121 | #endif |
| 2122 | , QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2123 | SRST |
| 2124 | ``-display type`` |
Thomas Huth | 707d93d | 2022-05-19 17:56:25 +0200 | [diff] [blame] | 2125 | Select type of display to use. Use ``-display help`` to list the available |
| 2126 | display types. Valid values for type are |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2127 | |
Ahmed Abouzied | ddc7175 | 2021-06-01 19:41:18 +0200 | [diff] [blame] | 2128 | ``spice-app[,gl=on|off]`` |
| 2129 | Start QEMU as a Spice server and launch the default Spice client |
| 2130 | application. The Spice server will redirect the serial consoles |
| 2131 | and QEMU monitors. (Since 4.0) |
| 2132 | |
Marc-André Lureau | 142ca62 | 2021-07-15 11:53:53 +0400 | [diff] [blame] | 2133 | ``dbus`` |
| 2134 | Export the display over D-Bus interfaces. (Since 7.0) |
| 2135 | |
| 2136 | The connection is registered with the "org.qemu" name (and queued when |
| 2137 | already owned). |
| 2138 | |
| 2139 | ``addr=<dbusaddr>`` : D-Bus bus address to connect to. |
| 2140 | |
Marc-André Lureau | 9999782 | 2021-10-10 00:16:57 +0400 | [diff] [blame] | 2141 | ``p2p=yes|no`` : Use peer-to-peer connection, accepted via QMP ``add_client``. |
| 2142 | |
| 2143 | ``gl=on|off|core|es`` : Use OpenGL for rendering (the D-Bus interface |
| 2144 | will share framebuffers with DMABUF file descriptors). |
Marc-André Lureau | 142ca62 | 2021-07-15 11:53:53 +0400 | [diff] [blame] | 2145 | |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2146 | ``sdl`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2147 | Display video output via SDL (usually in a separate graphics |
| 2148 | window; see the SDL documentation for other possibilities). |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2149 | Valid parameters are: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2150 | |
Thomas Huth | 8e8e844 | 2021-08-25 11:20:21 +0200 | [diff] [blame] | 2151 | ``grab-mod=<mods>`` : Used to select the modifier keys for toggling |
John Snow | 450e0f2 | 2021-10-04 17:52:36 -0400 | [diff] [blame] | 2152 | the mouse grabbing in conjunction with the "g" key. ``<mods>`` can be |
| 2153 | either ``lshift-lctrl-lalt`` or ``rctrl``. |
Thomas Huth | 8e8e844 | 2021-08-25 11:20:21 +0200 | [diff] [blame] | 2154 | |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2155 | ``gl=on|off|core|es`` : Use OpenGL for displaying |
| 2156 | |
| 2157 | ``show-cursor=on|off`` : Force showing the mouse cursor |
| 2158 | |
| 2159 | ``window-close=on|off`` : Allow to quit qemu with window close button |
| 2160 | |
| 2161 | ``gtk`` |
Ahmed Abouzied | ddc7175 | 2021-06-01 19:41:18 +0200 | [diff] [blame] | 2162 | Display video output in a GTK window. This interface provides |
| 2163 | drop-down menus and other UI elements to configure and control |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2164 | the VM during runtime. Valid parameters are: |
Ahmed Abouzied | ddc7175 | 2021-06-01 19:41:18 +0200 | [diff] [blame] | 2165 | |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2166 | ``full-screen=on|off`` : Start in fullscreen mode |
| 2167 | |
| 2168 | ``gl=on|off`` : Use OpenGL for displaying |
| 2169 | |
| 2170 | ``grab-on-hover=on|off`` : Grab keyboard input on mouse hover |
| 2171 | |
Felix xq Queißner | c34a933 | 2022-07-12 15:37:53 +0200 | [diff] [blame] | 2172 | ``show-tabs=on|off`` : Display the tab bar for switching between the |
| 2173 | various graphical interfaces (e.g. VGA and |
| 2174 | virtual console character devices) by default. |
| 2175 | |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2176 | ``show-cursor=on|off`` : Force showing the mouse cursor |
| 2177 | |
| 2178 | ``window-close=on|off`` : Allow to quit qemu with window close button |
| 2179 | |
Bryce Mills | dbccb1a | 2022-10-11 13:58:21 +0000 | [diff] [blame] | 2180 | ``show-menubar=on|off`` : Display the main window menubar, defaults to "on" |
| 2181 | |
Jan Kratochvil | c35d937 | 2023-06-28 18:23:36 +0800 | [diff] [blame] | 2182 | ``zoom-to-fit=on|off`` : Expand video output to the window size, |
| 2183 | defaults to "off" |
| 2184 | |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2185 | ``curses[,charset=<encoding>]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2186 | Display video output via curses. For graphics device models |
| 2187 | which support a text mode, QEMU can display this output using a |
| 2188 | curses/ncurses interface. Nothing is displayed when the graphics |
| 2189 | device is in graphical mode or if the graphics device does not |
| 2190 | support a text mode. Generally only the VGA device models |
| 2191 | support text mode. The font charset used by the guest can be |
| 2192 | specified with the ``charset`` option, for example |
| 2193 | ``charset=CP850`` for IBM CP850 encoding. The default is |
| 2194 | ``CP437``. |
| 2195 | |
Carwyn Ellis | 48941a5 | 2022-01-02 17:41:52 +0000 | [diff] [blame] | 2196 | ``cocoa`` |
| 2197 | Display video output in a Cocoa window. Mac only. This interface |
| 2198 | provides drop-down menus and other UI elements to configure and |
| 2199 | control the VM during runtime. Valid parameters are: |
| 2200 | |
Akihiko Odaki | d502dfc | 2023-12-14 15:31:36 +0900 | [diff] [blame] | 2201 | ``full-grab=on|off`` : Capture all key presses, including system combos. |
| 2202 | This requires accessibility permissions, since it |
| 2203 | performs a global grab on key events. |
| 2204 | (default: off) See |
| 2205 | https://support.apple.com/en-in/guide/mac-help/mh32356/mac |
| 2206 | |
| 2207 | ``swap-opt-cmd=on|off`` : Swap the Option and Command keys so that their |
| 2208 | key codes match their position on non-Mac |
| 2209 | keyboards and you can use Meta/Super and Alt |
| 2210 | where you expect them. (default: off) |
| 2211 | |
Carwyn Ellis | 48941a5 | 2022-01-02 17:41:52 +0000 | [diff] [blame] | 2212 | ``show-cursor=on|off`` : Force showing the mouse cursor |
| 2213 | |
| 2214 | ``left-command-key=on|off`` : Disable forwarding left command key to host |
| 2215 | |
Akihiko Odaki | d502dfc | 2023-12-14 15:31:36 +0900 | [diff] [blame] | 2216 | ``full-screen=on|off`` : Start in fullscreen mode |
| 2217 | |
| 2218 | ``zoom-to-fit=on|off`` : Expand video output to the window size, |
| 2219 | defaults to "off" |
| 2220 | |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2221 | ``egl-headless[,rendernode=<file>]`` |
Ahmed Abouzied | ddc7175 | 2021-06-01 19:41:18 +0200 | [diff] [blame] | 2222 | Offload all OpenGL operations to a local DRI device. For any |
| 2223 | graphical display, this display needs to be paired with either |
| 2224 | VNC or SPICE displays. |
| 2225 | |
Thomas Huth | 95f439b | 2021-06-30 18:32:31 +0200 | [diff] [blame] | 2226 | ``vnc=<display>`` |
| 2227 | Start a VNC server on display <display> |
| 2228 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2229 | ``none`` |
| 2230 | Do not display video output. The guest will still see an |
| 2231 | emulated graphics card, but its output will not be displayed to |
| 2232 | the QEMU user. This option differs from the -nographic option in |
| 2233 | that it only affects what is done with video output; -nographic |
| 2234 | also changes the destination of the serial and parallel port |
| 2235 | data. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2236 | ERST |
Jes Sorensen | 1472a95 | 2011-03-16 13:33:31 +0100 | [diff] [blame] | 2237 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2238 | DEF("nographic", 0, QEMU_OPTION_nographic, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 2239 | "-nographic disable graphical output and redirect serial I/Os to console\n", |
| 2240 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2241 | SRST |
| 2242 | ``-nographic`` |
| 2243 | Normally, if QEMU is compiled with graphical window support, it |
| 2244 | displays output such as guest graphics, guest console, and the QEMU |
| 2245 | monitor in a window. With this option, you can totally disable |
| 2246 | graphical output so that QEMU is a simple command line application. |
| 2247 | The emulated serial port is redirected on the console and muxed with |
| 2248 | the monitor (unless redirected elsewhere explicitly). Therefore, you |
| 2249 | can still use QEMU to debug a Linux kernel with a serial console. |
| 2250 | Use C-a h for help on switching between the console and monitor. |
| 2251 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2252 | |
Marc-André Lureau | 5324e3e | 2021-09-09 12:44:11 +0400 | [diff] [blame] | 2253 | #ifdef CONFIG_SPICE |
Gerd Hoffmann | 29b0040 | 2010-03-11 11:13:27 -0300 | [diff] [blame] | 2254 | DEF("spice", HAS_ARG, QEMU_OPTION_spice, |
Yonit Halperin | 27af778 | 2012-08-21 13:54:20 +0300 | [diff] [blame] | 2255 | "-spice [port=port][,tls-port=secured-port][,x509-dir=<dir>]\n" |
| 2256 | " [,x509-key-file=<file>][,x509-key-password=<file>]\n" |
| 2257 | " [,x509-cert-file=<file>][,x509-cacert-file=<file>]\n" |
Daniel P. Berrangé | a9daa36 | 2021-02-16 19:10:20 +0000 | [diff] [blame] | 2258 | " [,x509-dh-key-file=<file>][,addr=addr]\n" |
| 2259 | " [,ipv4=on|off][,ipv6=on|off][,unix=on|off]\n" |
Yonit Halperin | 27af778 | 2012-08-21 13:54:20 +0300 | [diff] [blame] | 2260 | " [,tls-ciphers=<list>]\n" |
| 2261 | " [,tls-channel=[main|display|cursor|inputs|record|playback]]\n" |
| 2262 | " [,plaintext-channel=[main|display|cursor|inputs|record|playback]]\n" |
Daniel P. Berrangé | 99522f6 | 2021-03-11 11:43:42 +0000 | [diff] [blame] | 2263 | " [,sasl=on|off][,disable-ticketing=on|off]\n" |
Daniel P. Berrangé | 36debaf | 2022-12-01 04:22:11 -0500 | [diff] [blame] | 2264 | " [,password-secret=<secret-id>]\n" |
Yonit Halperin | 27af778 | 2012-08-21 13:54:20 +0300 | [diff] [blame] | 2265 | " [,image-compression=[auto_glz|auto_lz|quic|glz|lz|off]]\n" |
| 2266 | " [,jpeg-wan-compression=[auto|never|always]]\n" |
| 2267 | " [,zlib-glz-wan-compression=[auto|never|always]]\n" |
Daniel P. Berrangé | a9daa36 | 2021-02-16 19:10:20 +0000 | [diff] [blame] | 2268 | " [,streaming-video=[off|all|filter]][,disable-copy-paste=on|off]\n" |
| 2269 | " [,disable-agent-file-xfer=on|off][,agent-mouse=[on|off]]\n" |
Hans de Goede | 5ad24e5 | 2013-06-08 15:37:27 +0200 | [diff] [blame] | 2270 | " [,playback-compression=[on|off]][,seamless-migration=[on|off]]\n" |
Marc-André Lureau | 7b52550 | 2017-02-12 15:21:18 +0400 | [diff] [blame] | 2271 | " [,gl=[on|off]][,rendernode=<file>]\n" |
Thomas Huth | a635bcf | 2023-07-03 09:56:46 +0200 | [diff] [blame] | 2272 | " enable spice\n" |
| 2273 | " at least one of {port, tls-port} is mandatory\n", |
Yonit Halperin | 27af778 | 2012-08-21 13:54:20 +0300 | [diff] [blame] | 2274 | QEMU_ARCH_ALL) |
Marc-André Lureau | 5324e3e | 2021-09-09 12:44:11 +0400 | [diff] [blame] | 2275 | #endif |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2276 | SRST |
| 2277 | ``-spice option[,option[,...]]`` |
| 2278 | Enable the spice remote desktop protocol. Valid options are |
| 2279 | |
| 2280 | ``port=<nr>`` |
| 2281 | Set the TCP port spice is listening on for plaintext channels. |
| 2282 | |
| 2283 | ``addr=<addr>`` |
| 2284 | Set the IP address spice is listening on. Default is any |
| 2285 | address. |
| 2286 | |
Daniel P. Berrangé | a9daa36 | 2021-02-16 19:10:20 +0000 | [diff] [blame] | 2287 | ``ipv4=on|off``; \ ``ipv6=on|off``; \ ``unix=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2288 | Force using the specified IP version. |
| 2289 | |
Daniel P. Berrangé | 99522f6 | 2021-03-11 11:43:42 +0000 | [diff] [blame] | 2290 | ``password-secret=<secret-id>`` |
| 2291 | Set the ID of the ``secret`` object containing the password |
| 2292 | you need to authenticate. |
| 2293 | |
Daniel P. Berrangé | a9daa36 | 2021-02-16 19:10:20 +0000 | [diff] [blame] | 2294 | ``sasl=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2295 | Require that the client use SASL to authenticate with the spice. |
| 2296 | The exact choice of authentication method used is controlled |
| 2297 | from the system / user's SASL configuration file for the 'qemu' |
| 2298 | service. This is typically found in /etc/sasl2/qemu.conf. If |
| 2299 | running QEMU as an unprivileged user, an environment variable |
| 2300 | SASL\_CONF\_PATH can be used to make it search alternate |
| 2301 | locations for the service config. While some SASL auth methods |
| 2302 | can also provide data encryption (eg GSSAPI), it is recommended |
| 2303 | that SASL always be combined with the 'tls' and 'x509' settings |
| 2304 | to enable use of SSL and server certificates. This ensures a |
| 2305 | data encryption preventing compromise of authentication |
| 2306 | credentials. |
| 2307 | |
Daniel P. Berrangé | a9daa36 | 2021-02-16 19:10:20 +0000 | [diff] [blame] | 2308 | ``disable-ticketing=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2309 | Allow client connects without authentication. |
| 2310 | |
Daniel P. Berrangé | a9daa36 | 2021-02-16 19:10:20 +0000 | [diff] [blame] | 2311 | ``disable-copy-paste=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2312 | Disable copy paste between the client and the guest. |
| 2313 | |
Daniel P. Berrangé | a9daa36 | 2021-02-16 19:10:20 +0000 | [diff] [blame] | 2314 | ``disable-agent-file-xfer=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2315 | Disable spice-vdagent based file-xfer between the client and the |
| 2316 | guest. |
| 2317 | |
| 2318 | ``tls-port=<nr>`` |
| 2319 | Set the TCP port spice is listening on for encrypted channels. |
| 2320 | |
| 2321 | ``x509-dir=<dir>`` |
| 2322 | Set the x509 file directory. Expects same filenames as -vnc |
| 2323 | $display,x509=$dir |
| 2324 | |
| 2325 | ``x509-key-file=<file>``; \ ``x509-key-password=<file>``; \ ``x509-cert-file=<file>``; \ ``x509-cacert-file=<file>``; \ ``x509-dh-key-file=<file>`` |
| 2326 | The x509 file names can also be configured individually. |
| 2327 | |
| 2328 | ``tls-ciphers=<list>`` |
| 2329 | Specify which ciphers to use. |
| 2330 | |
| 2331 | ``tls-channel=[main|display|cursor|inputs|record|playback]``; \ ``plaintext-channel=[main|display|cursor|inputs|record|playback]`` |
| 2332 | Force specific channel to be used with or without TLS |
| 2333 | encryption. The options can be specified multiple times to |
| 2334 | configure multiple channels. The special name "default" can be |
| 2335 | used to set the default mode. For channels which are not |
| 2336 | explicitly forced into one mode the spice client is allowed to |
| 2337 | pick tls/plaintext as he pleases. |
| 2338 | |
| 2339 | ``image-compression=[auto_glz|auto_lz|quic|glz|lz|off]`` |
| 2340 | Configure image compression (lossless). Default is auto\_glz. |
| 2341 | |
| 2342 | ``jpeg-wan-compression=[auto|never|always]``; \ ``zlib-glz-wan-compression=[auto|never|always]`` |
| 2343 | Configure wan image compression (lossy for slow links). Default |
| 2344 | is auto. |
| 2345 | |
| 2346 | ``streaming-video=[off|all|filter]`` |
| 2347 | Configure video stream detection. Default is off. |
| 2348 | |
| 2349 | ``agent-mouse=[on|off]`` |
| 2350 | Enable/disable passing mouse events via vdagent. Default is on. |
| 2351 | |
| 2352 | ``playback-compression=[on|off]`` |
| 2353 | Enable/disable audio stream compression (using celt 0.5.1). |
| 2354 | Default is on. |
| 2355 | |
| 2356 | ``seamless-migration=[on|off]`` |
| 2357 | Enable/disable spice seamless migration. Default is off. |
| 2358 | |
| 2359 | ``gl=[on|off]`` |
| 2360 | Enable/disable OpenGL context. Default is off. |
| 2361 | |
| 2362 | ``rendernode=<file>`` |
| 2363 | DRM render node for OpenGL rendering. If not specified, it will |
| 2364 | pick the first available. (Since 2.9) |
| 2365 | ERST |
Gerd Hoffmann | 29b0040 | 2010-03-11 11:13:27 -0300 | [diff] [blame] | 2366 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2367 | DEF("portrait", 0, QEMU_OPTION_portrait, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 2368 | "-portrait rotate graphical output 90 deg left (only PXA LCD)\n", |
| 2369 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2370 | SRST |
| 2371 | ``-portrait`` |
| 2372 | Rotate graphical output 90 deg left (only PXA LCD). |
| 2373 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2374 | |
Vasily Khoruzhick | 9312805 | 2011-06-17 13:04:36 +0300 | [diff] [blame] | 2375 | DEF("rotate", HAS_ARG, QEMU_OPTION_rotate, |
| 2376 | "-rotate <deg> rotate graphical output some deg left (only PXA LCD)\n", |
| 2377 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2378 | SRST |
| 2379 | ``-rotate deg`` |
| 2380 | Rotate graphical output some deg left (only PXA LCD). |
| 2381 | ERST |
Vasily Khoruzhick | 9312805 | 2011-06-17 13:04:36 +0300 | [diff] [blame] | 2382 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2383 | DEF("vga", HAS_ARG, QEMU_OPTION_vga, |
Gerd Hoffmann | a94f0c5 | 2014-09-10 14:28:48 +0200 | [diff] [blame] | 2384 | "-vga [std|cirrus|vmware|qxl|xenfb|tcx|cg3|virtio|none]\n" |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 2385 | " select video card type\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2386 | SRST |
| 2387 | ``-vga type`` |
| 2388 | Select type of VGA card to emulate. Valid values for type are |
| 2389 | |
| 2390 | ``cirrus`` |
| 2391 | Cirrus Logic GD5446 Video card. All Windows versions starting |
| 2392 | from Windows 95 should recognize and use this graphic card. For |
| 2393 | optimal performances, use 16 bit color depth in the guest and |
| 2394 | the host OS. (This card was the default before QEMU 2.2) |
| 2395 | |
| 2396 | ``std`` |
| 2397 | Standard VGA card with Bochs VBE extensions. If your guest OS |
| 2398 | supports the VESA 2.0 VBE extensions (e.g. Windows XP) and if |
| 2399 | you want to use high resolution modes (>= 1280x1024x16) then you |
| 2400 | should use this option. (This card is the default since QEMU |
| 2401 | 2.2) |
| 2402 | |
| 2403 | ``vmware`` |
| 2404 | VMWare SVGA-II compatible adapter. Use it if you have |
| 2405 | sufficiently recent XFree86/XOrg server or Windows guest with a |
| 2406 | driver for this card. |
| 2407 | |
| 2408 | ``qxl`` |
| 2409 | QXL paravirtual graphic card. It is VGA compatible (including |
| 2410 | VESA 2.0 VBE support). Works best with qxl guest drivers |
| 2411 | installed though. Recommended choice when using the spice |
| 2412 | protocol. |
| 2413 | |
| 2414 | ``tcx`` |
| 2415 | (sun4m only) Sun TCX framebuffer. This is the default |
| 2416 | framebuffer for sun4m machines and offers both 8-bit and 24-bit |
| 2417 | colour depths at a fixed resolution of 1024x768. |
| 2418 | |
| 2419 | ``cg3`` |
| 2420 | (sun4m only) Sun cgthree framebuffer. This is a simple 8-bit |
| 2421 | framebuffer for sun4m machines available in both 1024x768 |
| 2422 | (OpenBIOS) and 1152x900 (OBP) resolutions aimed at people |
| 2423 | wishing to run older Solaris versions. |
| 2424 | |
| 2425 | ``virtio`` |
| 2426 | Virtio VGA card. |
| 2427 | |
| 2428 | ``none`` |
| 2429 | Disable VGA card. |
| 2430 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2431 | |
| 2432 | DEF("full-screen", 0, QEMU_OPTION_full_screen, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 2433 | "-full-screen start in full screen\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2434 | SRST |
| 2435 | ``-full-screen`` |
| 2436 | Start in full screen. |
| 2437 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2438 | |
John Snow | 60f9a4e | 2020-02-04 11:56:38 -0500 | [diff] [blame] | 2439 | DEF("g", HAS_ARG, QEMU_OPTION_g , |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 2440 | "-g WxH[xDEPTH] Set the initial graphical resolution and depth\n", |
Laurent Vivier | 8ac919a | 2019-10-26 18:45:43 +0200 | [diff] [blame] | 2441 | QEMU_ARCH_PPC | QEMU_ARCH_SPARC | QEMU_ARCH_M68K) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2442 | SRST |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 2443 | ``-g`` *width*\ ``x``\ *height*\ ``[x``\ *depth*\ ``]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2444 | Set the initial graphical resolution and depth (PPC, SPARC only). |
| 2445 | |
| 2446 | For PPC the default is 800x600x32. |
| 2447 | |
| 2448 | For SPARC with the TCX graphics device, the default is 1024x768x8 |
| 2449 | with the option of 1024x768x24. For cgthree, the default is |
| 2450 | 1024x768x8 with the option of 1152x900x8 for people who wish to use |
| 2451 | OBP. |
| 2452 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2453 | |
Marc-André Lureau | 6261164 | 2023-10-25 17:05:08 +0400 | [diff] [blame] | 2454 | #ifdef CONFIG_VNC |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2455 | DEF("vnc", HAS_ARG, QEMU_OPTION_vnc , |
Robert Ho | f04ec5a | 2016-07-26 18:17:11 +0800 | [diff] [blame] | 2456 | "-vnc <display> shorthand for -display vnc=<display>\n", QEMU_ARCH_ALL) |
Marc-André Lureau | 6261164 | 2023-10-25 17:05:08 +0400 | [diff] [blame] | 2457 | #endif |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2458 | SRST |
| 2459 | ``-vnc display[,option[,option[,...]]]`` |
| 2460 | Normally, if QEMU is compiled with graphical window support, it |
| 2461 | displays output such as guest graphics, guest console, and the QEMU |
| 2462 | monitor in a window. With this option, you can have QEMU listen on |
| 2463 | VNC display display and redirect the VGA display over the VNC |
| 2464 | session. It is very useful to enable the usb tablet device when |
| 2465 | using this option (option ``-device usb-tablet``). When using the |
| 2466 | VNC display, you must use the ``-k`` parameter to set the keyboard |
| 2467 | layout if you are not using en-us. Valid syntax for the display is |
| 2468 | |
| 2469 | ``to=L`` |
| 2470 | With this option, QEMU will try next available VNC displays, |
Manos Pitsidianakis | 835f364 | 2024-02-20 10:52:23 +0200 | [diff] [blame] | 2471 | until the number L, if the originally defined "-vnc display" is |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2472 | not available, e.g. port 5900+display is already used by another |
| 2473 | application. By default, to=0. |
| 2474 | |
| 2475 | ``host:d`` |
| 2476 | TCP connections will only be allowed from host on display d. By |
| 2477 | convention the TCP port is 5900+d. Optionally, host can be |
| 2478 | omitted in which case the server will accept connections from |
| 2479 | any host. |
| 2480 | |
| 2481 | ``unix:path`` |
| 2482 | Connections will be allowed over UNIX domain sockets where path |
| 2483 | is the location of a unix socket to listen for connections on. |
| 2484 | |
| 2485 | ``none`` |
| 2486 | VNC is initialized but not started. The monitor ``change`` |
| 2487 | command can be used to later start the VNC server. |
| 2488 | |
| 2489 | Following the display value there may be one or more option flags |
| 2490 | separated by commas. Valid options are |
| 2491 | |
Daniel P. Berrangé | 82a17d1 | 2021-02-16 19:10:23 +0000 | [diff] [blame] | 2492 | ``reverse=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2493 | Connect to a listening VNC client via a "reverse" connection. |
| 2494 | The client is specified by the display. For reverse network |
| 2495 | connections (host:d,``reverse``), the d argument is a TCP port |
| 2496 | number, not a display number. |
| 2497 | |
Daniel P. Berrangé | 82a17d1 | 2021-02-16 19:10:23 +0000 | [diff] [blame] | 2498 | ``websocket=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2499 | Opens an additional TCP listening port dedicated to VNC |
| 2500 | Websocket connections. If a bare websocket option is given, the |
| 2501 | Websocket port is 5700+display. An alternative port can be |
| 2502 | specified with the syntax ``websocket``\ =port. |
| 2503 | |
| 2504 | If host is specified connections will only be allowed from this |
| 2505 | host. It is possible to control the websocket listen address |
| 2506 | independently, using the syntax ``websocket``\ =host:port. |
| 2507 | |
| 2508 | If no TLS credentials are provided, the websocket connection |
| 2509 | runs in unencrypted mode. If TLS credentials are provided, the |
| 2510 | websocket connection requires encrypted client connections. |
| 2511 | |
Daniel P. Berrangé | 82a17d1 | 2021-02-16 19:10:23 +0000 | [diff] [blame] | 2512 | ``password=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2513 | Require that password based authentication is used for client |
| 2514 | connections. |
| 2515 | |
| 2516 | The password must be set separately using the ``set_password`` |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 2517 | command in the :ref:`QEMU monitor`. The |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2518 | syntax to change your password is: |
| 2519 | ``set_password <protocol> <password>`` where <protocol> could be |
| 2520 | either "vnc" or "spice". |
| 2521 | |
| 2522 | If you would like to change <protocol> password expiration, you |
| 2523 | should use ``expire_password <protocol> <expiration-time>`` |
| 2524 | where expiration time could be one of the following options: |
| 2525 | now, never, +seconds or UNIX time of expiration, e.g. +60 to |
| 2526 | make password expire in 60 seconds, or 1335196800 to make |
| 2527 | password expire on "Mon Apr 23 12:00:00 EDT 2012" (UNIX time for |
| 2528 | this date and time). |
| 2529 | |
| 2530 | You can also use keywords "now" or "never" for the expiration |
| 2531 | time to allow <protocol> password to expire immediately or never |
| 2532 | expire. |
| 2533 | |
Daniel P. Berrangé | 6c6840e | 2021-03-11 11:43:41 +0000 | [diff] [blame] | 2534 | ``password-secret=<secret-id>`` |
| 2535 | Require that password based authentication is used for client |
| 2536 | connections, using the password provided by the ``secret`` |
| 2537 | object identified by ``secret-id``. |
| 2538 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2539 | ``tls-creds=ID`` |
| 2540 | Provides the ID of a set of TLS credentials to use to secure the |
| 2541 | VNC server. They will apply to both the normal VNC server socket |
| 2542 | and the websocket socket (if enabled). Setting TLS credentials |
| 2543 | will cause the VNC server socket to enable the VeNCrypt auth |
| 2544 | mechanism. The credentials should have been previously created |
| 2545 | using the ``-object tls-creds`` argument. |
| 2546 | |
| 2547 | ``tls-authz=ID`` |
| 2548 | Provides the ID of the QAuthZ authorization object against which |
| 2549 | the client's x509 distinguished name will validated. This object |
| 2550 | is only resolved at time of use, so can be deleted and recreated |
| 2551 | on the fly while the VNC server is active. If missing, it will |
| 2552 | default to denying access. |
| 2553 | |
Daniel P. Berrangé | 82a17d1 | 2021-02-16 19:10:23 +0000 | [diff] [blame] | 2554 | ``sasl=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2555 | Require that the client use SASL to authenticate with the VNC |
| 2556 | server. The exact choice of authentication method used is |
| 2557 | controlled from the system / user's SASL configuration file for |
| 2558 | the 'qemu' service. This is typically found in |
| 2559 | /etc/sasl2/qemu.conf. If running QEMU as an unprivileged user, |
| 2560 | an environment variable SASL\_CONF\_PATH can be used to make it |
| 2561 | search alternate locations for the service config. While some |
| 2562 | SASL auth methods can also provide data encryption (eg GSSAPI), |
| 2563 | it is recommended that SASL always be combined with the 'tls' |
| 2564 | and 'x509' settings to enable use of SSL and server |
| 2565 | certificates. This ensures a data encryption preventing |
| 2566 | compromise of authentication credentials. See the |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 2567 | :ref:`VNC security` section in the System Emulation Users Guide |
| 2568 | for details on using SASL authentication. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2569 | |
| 2570 | ``sasl-authz=ID`` |
| 2571 | Provides the ID of the QAuthZ authorization object against which |
| 2572 | the client's SASL username will validated. This object is only |
| 2573 | resolved at time of use, so can be deleted and recreated on the |
| 2574 | fly while the VNC server is active. If missing, it will default |
| 2575 | to denying access. |
| 2576 | |
Daniel P. Berrangé | 82a17d1 | 2021-02-16 19:10:23 +0000 | [diff] [blame] | 2577 | ``acl=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2578 | Legacy method for enabling authorization of clients against the |
| 2579 | x509 distinguished name and SASL username. It results in the |
| 2580 | creation of two ``authz-list`` objects with IDs of |
| 2581 | ``vnc.username`` and ``vnc.x509dname``. The rules for these |
| 2582 | objects must be configured with the HMP ACL commands. |
| 2583 | |
| 2584 | This option is deprecated and should no longer be used. The new |
| 2585 | ``sasl-authz`` and ``tls-authz`` options are a replacement. |
| 2586 | |
Daniel P. Berrangé | 82a17d1 | 2021-02-16 19:10:23 +0000 | [diff] [blame] | 2587 | ``lossy=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2588 | Enable lossy compression methods (gradient, JPEG, ...). If this |
| 2589 | option is set, VNC client may receive lossy framebuffer updates |
| 2590 | depending on its encoding settings. Enabling this option can |
| 2591 | save a lot of bandwidth at the expense of quality. |
| 2592 | |
Daniel P. Berrangé | 82a17d1 | 2021-02-16 19:10:23 +0000 | [diff] [blame] | 2593 | ``non-adaptive=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2594 | Disable adaptive encodings. Adaptive encodings are enabled by |
| 2595 | default. An adaptive encoding will try to detect frequently |
| 2596 | updated screen regions, and send updates in these regions using |
| 2597 | a lossy encoding (like JPEG). This can be really helpful to save |
| 2598 | bandwidth when playing videos. Disabling adaptive encodings |
| 2599 | restores the original static behavior of encodings like Tight. |
| 2600 | |
| 2601 | ``share=[allow-exclusive|force-shared|ignore]`` |
| 2602 | Set display sharing policy. 'allow-exclusive' allows clients to |
| 2603 | ask for exclusive access. As suggested by the rfb spec this is |
| 2604 | implemented by dropping other connections. Connecting multiple |
| 2605 | clients in parallel requires all clients asking for a shared |
| 2606 | session (vncviewer: -shared switch). This is the default. |
| 2607 | 'force-shared' disables exclusive client access. Useful for |
| 2608 | shared desktop sessions, where you don't want someone forgetting |
| 2609 | specify -shared disconnect everybody else. 'ignore' completely |
| 2610 | ignores the shared flag and allows everybody connect |
| 2611 | unconditionally. Doesn't conform to the rfb spec but is |
| 2612 | traditional QEMU behavior. |
| 2613 | |
| 2614 | ``key-delay-ms`` |
| 2615 | Set keyboard delay, for key down and key up events, in |
| 2616 | milliseconds. Default is 10. Keyboards are low-bandwidth |
| 2617 | devices, so this slowdown can help the device and guest to keep |
| 2618 | up and not lose events in case events are arriving in bulk. |
| 2619 | Possible causes for the latter are flaky network connections, or |
| 2620 | scripts for automated testing. |
| 2621 | |
| 2622 | ``audiodev=audiodev`` |
| 2623 | Use the specified audiodev when the VNC client requests audio |
| 2624 | transmission. When not using an -audiodev argument, this option |
| 2625 | must be omitted, otherwise is must be present and specify a |
| 2626 | valid audiodev. |
Daniel P. Berrangé | 7b5fa0b | 2020-12-11 16:08:25 +0000 | [diff] [blame] | 2627 | |
Daniel P. Berrangé | 82a17d1 | 2021-02-16 19:10:23 +0000 | [diff] [blame] | 2628 | ``power-control=on|off`` |
Daniel P. Berrangé | 7b5fa0b | 2020-12-11 16:08:25 +0000 | [diff] [blame] | 2629 | Permit the remote client to issue shutdown, reboot or reset power |
| 2630 | control requests. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2631 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2632 | |
Michael Ellerman | a3adb7a | 2011-12-19 17:19:31 +1100 | [diff] [blame] | 2633 | ARCHHEADING(, QEMU_ARCH_I386) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2634 | |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 2635 | ARCHHEADING(i386 target only:, QEMU_ARCH_I386) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2636 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2637 | DEF("win2k-hack", 0, QEMU_OPTION_win2k_hack, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 2638 | "-win2k-hack use it when installing Windows 2000 to avoid a disk full bug\n", |
| 2639 | QEMU_ARCH_I386) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2640 | SRST |
| 2641 | ``-win2k-hack`` |
| 2642 | Use it when installing Windows 2000 to avoid a disk full bug. After |
| 2643 | Windows 2000 is installed, you no longer need this option (this |
| 2644 | option slows down the IDE transfers). |
| 2645 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2646 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2647 | DEF("no-fd-bootchk", 0, QEMU_OPTION_no_fd_bootchk, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 2648 | "-no-fd-bootchk disable boot signature checking for floppy disks\n", |
| 2649 | QEMU_ARCH_I386) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2650 | SRST |
| 2651 | ``-no-fd-bootchk`` |
| 2652 | Disable boot signature checking for floppy disks in BIOS. May be |
| 2653 | needed to boot from old floppy disks. |
| 2654 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2655 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2656 | DEF("acpitable", HAS_ARG, QEMU_OPTION_acpitable, |
Michael Tokarev | 104bf02 | 2011-05-12 18:44:17 +0400 | [diff] [blame] | 2657 | "-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" |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 2658 | " ACPI table description\n", QEMU_ARCH_I386) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2659 | SRST |
| 2660 | ``-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]...]`` |
| 2661 | Add ACPI table with specified header fields and context from |
| 2662 | specified files. For file=, take whole ACPI table from the specified |
| 2663 | files, including all ACPI headers (possible overridden by other |
| 2664 | options). For data=, only data portion of the table is used, all |
| 2665 | header information is specified in the command line. If a SLIC table |
| 2666 | is supplied to QEMU, then the SLIC's oem\_id and oem\_table\_id |
| 2667 | fields will override the same in the RSDT and the FADT (a.k.a. |
| 2668 | FACP), in order to ensure the field matches required by the |
| 2669 | Microsoft SLIC spec and the ACPI spec. |
| 2670 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2671 | |
aliguori | b6f6e3d | 2009-04-17 18:59:56 +0000 | [diff] [blame] | 2672 | DEF("smbios", HAS_ARG, QEMU_OPTION_smbios, |
| 2673 | "-smbios file=binary\n" |
Bruce Rogers | ca1a8a0 | 2010-01-06 12:33:57 -0700 | [diff] [blame] | 2674 | " load SMBIOS entry from binary file\n" |
Gabriel L. Somlo | b155eb1 | 2015-02-05 11:45:30 -0500 | [diff] [blame] | 2675 | "-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d]\n" |
| 2676 | " [,uefi=on|off]\n" |
Bruce Rogers | ca1a8a0 | 2010-01-06 12:33:57 -0700 | [diff] [blame] | 2677 | " specify SMBIOS type 0 fields\n" |
aliguori | b6f6e3d | 2009-04-17 18:59:56 +0000 | [diff] [blame] | 2678 | "-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str]\n" |
| 2679 | " [,uuid=uuid][,sku=str][,family=str]\n" |
Gabriel L. Somlo | b155eb1 | 2015-02-05 11:45:30 -0500 | [diff] [blame] | 2680 | " specify SMBIOS type 1 fields\n" |
| 2681 | "-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str]\n" |
| 2682 | " [,asset=str][,location=str]\n" |
| 2683 | " specify SMBIOS type 2 fields\n" |
| 2684 | "-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str]\n" |
| 2685 | " [,sku=str]\n" |
| 2686 | " specify SMBIOS type 3 fields\n" |
| 2687 | "-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str]\n" |
Ying Fang | c906e03 | 2020-08-06 11:56:33 +0800 | [diff] [blame] | 2688 | " [,asset=str][,part=str][,max-speed=%d][,current-speed=%d]\n" |
Heinrich Schuchardt | b5831d7 | 2024-01-23 19:42:26 +0100 | [diff] [blame] | 2689 | " [,processor-family=%d,processor-id=%d]\n" |
Gabriel L. Somlo | b155eb1 | 2015-02-05 11:45:30 -0500 | [diff] [blame] | 2690 | " specify SMBIOS type 4 fields\n" |
Hal Martin | fd8caa2 | 2022-08-12 15:51:53 +0200 | [diff] [blame] | 2691 | "-smbios type=8[,external_reference=str][,internal_reference=str][,connector_type=%d][,port_type=%d]\n" |
| 2692 | " specify SMBIOS type 8 fields\n" |
Daniel P. Berrangé | 48a7ff4 | 2020-09-23 14:38:04 +0100 | [diff] [blame] | 2693 | "-smbios type=11[,value=str][,path=filename]\n" |
| 2694 | " specify SMBIOS type 11 fields\n" |
Gabriel L. Somlo | b155eb1 | 2015-02-05 11:45:30 -0500 | [diff] [blame] | 2695 | "-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str]\n" |
Gabriel L. Somlo | 3ebd6cc | 2015-03-11 13:58:01 -0400 | [diff] [blame] | 2696 | " [,asset=str][,part=str][,speed=%d]\n" |
Vincent Bernat | 05dfb44 | 2021-04-01 19:11:38 +0200 | [diff] [blame] | 2697 | " specify SMBIOS type 17 fields\n" |
| 2698 | "-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]\n" |
| 2699 | " specify SMBIOS type 41 fields\n", |
Heinrich Schuchardt | e2ff0de | 2024-01-23 19:42:29 +0100 | [diff] [blame] | 2700 | QEMU_ARCH_I386 | QEMU_ARCH_ARM | QEMU_ARCH_LOONGARCH | QEMU_ARCH_RISCV) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2701 | SRST |
| 2702 | ``-smbios file=binary`` |
| 2703 | Load SMBIOS entry from binary file. |
| 2704 | |
| 2705 | ``-smbios type=0[,vendor=str][,version=str][,date=str][,release=%d.%d][,uefi=on|off]`` |
| 2706 | Specify SMBIOS type 0 fields |
| 2707 | |
| 2708 | ``-smbios type=1[,manufacturer=str][,product=str][,version=str][,serial=str][,uuid=uuid][,sku=str][,family=str]`` |
| 2709 | Specify SMBIOS type 1 fields |
| 2710 | |
| 2711 | ``-smbios type=2[,manufacturer=str][,product=str][,version=str][,serial=str][,asset=str][,location=str]`` |
| 2712 | Specify SMBIOS type 2 fields |
| 2713 | |
| 2714 | ``-smbios type=3[,manufacturer=str][,version=str][,serial=str][,asset=str][,sku=str]`` |
| 2715 | Specify SMBIOS type 3 fields |
| 2716 | |
Heinrich Schuchardt | b5831d7 | 2024-01-23 19:42:26 +0100 | [diff] [blame] | 2717 | ``-smbios type=4[,sock_pfx=str][,manufacturer=str][,version=str][,serial=str][,asset=str][,part=str][,processor-family=%d][,processor-id=%d]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2718 | Specify SMBIOS type 4 fields |
| 2719 | |
Daniel P. Berrangé | 48a7ff4 | 2020-09-23 14:38:04 +0100 | [diff] [blame] | 2720 | ``-smbios type=11[,value=str][,path=filename]`` |
| 2721 | Specify SMBIOS type 11 fields |
| 2722 | |
| 2723 | This argument can be repeated multiple times, and values are added in the order they are parsed. |
| 2724 | Applications intending to use OEM strings data are encouraged to use their application name as |
| 2725 | a prefix for the value string. This facilitates passing information for multiple applications |
| 2726 | concurrently. |
| 2727 | |
| 2728 | The ``value=str`` syntax provides the string data inline, while the ``path=filename`` syntax |
| 2729 | loads data from a file on disk. Note that the file is not permitted to contain any NUL bytes. |
| 2730 | |
| 2731 | Both the ``value`` and ``path`` options can be repeated multiple times and will be added to |
| 2732 | the SMBIOS table in the order in which they appear. |
| 2733 | |
| 2734 | Note that on the x86 architecture, the total size of all SMBIOS tables is limited to 65535 |
| 2735 | bytes. Thus the OEM strings data is not suitable for passing large amounts of data into the |
| 2736 | guest. Instead it should be used as a indicator to inform the guest where to locate the real |
| 2737 | data set, for example, by specifying the serial ID of a block device. |
| 2738 | |
| 2739 | An example passing three strings is |
| 2740 | |
| 2741 | .. parsed-literal:: |
| 2742 | |
| 2743 | -smbios type=11,value=cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/,\\ |
| 2744 | value=anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os,\\ |
| 2745 | path=/some/file/with/oemstringsdata.txt |
| 2746 | |
| 2747 | In the guest OS this is visible with the ``dmidecode`` command |
| 2748 | |
| 2749 | .. parsed-literal:: |
| 2750 | |
| 2751 | $ dmidecode -t 11 |
| 2752 | Handle 0x0E00, DMI type 11, 5 bytes |
| 2753 | OEM Strings |
| 2754 | String 1: cloud-init:ds=nocloud-net;s=http://10.10.0.1:8000/ |
| 2755 | String 2: anaconda:method=http://dl.fedoraproject.org/pub/fedora/linux/releases/25/x86_64/os |
| 2756 | String 3: myapp:some extra data |
| 2757 | |
| 2758 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2759 | ``-smbios type=17[,loc_pfx=str][,bank=str][,manufacturer=str][,serial=str][,asset=str][,part=str][,speed=%d]`` |
| 2760 | Specify SMBIOS type 17 fields |
Vincent Bernat | 05dfb44 | 2021-04-01 19:11:38 +0200 | [diff] [blame] | 2761 | |
| 2762 | ``-smbios type=41[,designation=str][,kind=str][,instance=%d][,pcidev=str]`` |
| 2763 | Specify SMBIOS type 41 fields |
| 2764 | |
| 2765 | This argument can be repeated multiple times. Its main use is to allow network interfaces be created |
| 2766 | as ``enoX`` on Linux, with X being the instance number, instead of the name depending on the interface |
| 2767 | position on the PCI bus. |
| 2768 | |
| 2769 | Here is an example of use: |
| 2770 | |
| 2771 | .. parsed-literal:: |
| 2772 | |
| 2773 | -netdev user,id=internet \\ |
| 2774 | -device virtio-net-pci,mac=50:54:00:00:00:42,netdev=internet,id=internet-dev \\ |
| 2775 | -smbios type=41,designation='Onboard LAN',instance=1,kind=ethernet,pcidev=internet-dev |
| 2776 | |
| 2777 | In the guest OS, the device should then appear as ``eno1``: |
| 2778 | |
| 2779 | ..parsed-literal:: |
| 2780 | |
| 2781 | $ ip -brief l |
| 2782 | lo UNKNOWN 00:00:00:00:00:00 <LOOPBACK,UP,LOWER_UP> |
| 2783 | eno1 UP 50:54:00:00:00:42 <BROADCAST,MULTICAST,UP,LOWER_UP> |
| 2784 | |
| 2785 | Currently, the PCI device has to be attached to the root bus. |
| 2786 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 2787 | ERST |
aliguori | b6f6e3d | 2009-04-17 18:59:56 +0000 | [diff] [blame] | 2788 | |
Markus Armbruster | c70a01e | 2013-02-13 19:49:40 +0100 | [diff] [blame] | 2789 | DEFHEADING() |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2790 | |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 2791 | DEFHEADING(Network options:) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2792 | |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2793 | DEF("netdev", HAS_ARG, QEMU_OPTION_netdev, |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2794 | #ifdef CONFIG_SLIRP |
Daniel P. Berrangé | 8b0dc24 | 2021-02-16 19:10:21 +0000 | [diff] [blame] | 2795 | "-netdev user,id=str[,ipv4=on|off][,net=addr[/mask]][,host=addr]\n" |
| 2796 | " [,ipv6=on|off][,ipv6-net=addr[/int]][,ipv6-host=addr]\n" |
Samuel Thibault | 0b11c03 | 2016-03-20 12:29:54 +0100 | [diff] [blame] | 2797 | " [,restrict=on|off][,hostname=host][,dhcpstart=addr]\n" |
Benjamin Drung | f18d137 | 2018-02-27 17:06:01 +0100 | [diff] [blame] | 2798 | " [,dns=addr][,ipv6-dns=addr][,dnssearch=domain][,domainname=domain]\n" |
Fam Zheng | 0fca92b | 2018-09-14 15:26:16 +0800 | [diff] [blame] | 2799 | " [,tftp=dir][,tftp-server-name=name][,bootfile=f][,hostfwd=rule][,guestfwd=rule]" |
Jan Kiszka | ad196a9 | 2009-06-24 14:42:28 +0200 | [diff] [blame] | 2800 | #ifndef _WIN32 |
Jan Kiszka | c92ef6a | 2009-06-24 14:42:28 +0200 | [diff] [blame] | 2801 | "[,smb=dir[,smbserver=addr]]\n" |
Jan Kiszka | ad196a9 | 2009-06-24 14:42:28 +0200 | [diff] [blame] | 2802 | #endif |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2803 | " configure a user mode network backend with ID 'str',\n" |
| 2804 | " its DHCP server and optional services\n" |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2805 | #endif |
| 2806 | #ifdef _WIN32 |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2807 | "-netdev tap,id=str,ifname=name\n" |
| 2808 | " configure a host TAP network backend with ID 'str'\n" |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2809 | #else |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2810 | "-netdev tap,id=str[,fd=h][,fds=x:y:...:z][,ifname=name][,script=file][,downscript=dfile]\n" |
Alexey Kardashevskiy | 584613e | 2016-09-13 17:11:54 +1000 | [diff] [blame] | 2811 | " [,br=bridge][,helper=helper][,sndbuf=nbytes][,vnet_hdr=on|off][,vhost=on|off]\n" |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2812 | " [,vhostfd=h][,vhostfds=x:y:...:z][,vhostforce=on|off][,queues=n]\n" |
Jason Wang | 69e87b3 | 2016-07-06 09:57:55 +0800 | [diff] [blame] | 2813 | " [,poll-us=n]\n" |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2814 | " configure a host TAP network backend with ID 'str'\n" |
Alexey Kardashevskiy | 584613e | 2016-09-13 17:11:54 +1000 | [diff] [blame] | 2815 | " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" |
Corey Bryant | a7c36ee | 2012-01-26 09:42:27 -0500 | [diff] [blame] | 2816 | " use network scripts 'file' (default=" DEFAULT_NETWORK_SCRIPT ")\n" |
| 2817 | " to configure it and 'dfile' (default=" DEFAULT_NETWORK_DOWN_SCRIPT ")\n" |
| 2818 | " to deconfigure it\n" |
Bruce Rogers | ca1a8a0 | 2010-01-06 12:33:57 -0700 | [diff] [blame] | 2819 | " use '[down]script=no' to disable script execution\n" |
Corey Bryant | a7c36ee | 2012-01-26 09:42:27 -0500 | [diff] [blame] | 2820 | " use network helper 'helper' (default=" DEFAULT_BRIDGE_HELPER ") to\n" |
| 2821 | " configure it\n" |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2822 | " use 'fd=h' to connect to an already opened TAP interface\n" |
Jason Wang | 2ca81ba | 2013-02-20 18:04:01 +0800 | [diff] [blame] | 2823 | " use 'fds=x:y:...:z' to connect to already opened multiqueue capable TAP interfaces\n" |
Bruce Rogers | ca1a8a0 | 2010-01-06 12:33:57 -0700 | [diff] [blame] | 2824 | " use 'sndbuf=nbytes' to limit the size of the send buffer (the\n" |
Michael S. Tsirkin | f157ed2 | 2011-02-01 14:25:40 +0200 | [diff] [blame] | 2825 | " default is disabled 'sndbuf=0' to enable flow control set 'sndbuf=1048576')\n" |
Bruce Rogers | ca1a8a0 | 2010-01-06 12:33:57 -0700 | [diff] [blame] | 2826 | " use vnet_hdr=off to avoid enabling the IFF_VNET_HDR tap flag\n" |
| 2827 | " use vnet_hdr=on to make the lack of IFF_VNET_HDR support an error condition\n" |
Michael S. Tsirkin | 82b0d80 | 2010-03-17 13:08:24 +0200 | [diff] [blame] | 2828 | " use vhost=on to enable experimental in kernel accelerator\n" |
mst@redhat.com | 5430a28 | 2011-02-01 22:13:42 +0200 | [diff] [blame] | 2829 | " (only has effect for virtio guests which use MSIX)\n" |
| 2830 | " use vhostforce=on to force vhost on for non-MSIX virtio guests\n" |
Michael S. Tsirkin | 82b0d80 | 2010-03-17 13:08:24 +0200 | [diff] [blame] | 2831 | " use 'vhostfd=h' to connect to an already opened vhost net device\n" |
Jason Wang | 2ca81ba | 2013-02-20 18:04:01 +0800 | [diff] [blame] | 2832 | " use 'vhostfds=x:y:...:z to connect to multiple already opened vhost net devices\n" |
Jason Wang | ec39601 | 2013-02-22 22:57:52 +0800 | [diff] [blame] | 2833 | " use 'queues=n' to specify the number of queues to be created for multiqueue TAP\n" |
Michael Tokarev | cba42d6 | 2021-03-09 14:15:10 +0300 | [diff] [blame] | 2834 | " use 'poll-us=n' to specify the maximum number of microseconds that could be\n" |
Jason Wang | 69e87b3 | 2016-07-06 09:57:55 +0800 | [diff] [blame] | 2835 | " spent on busy polling for vhost net\n" |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2836 | "-netdev bridge,id=str[,br=bridge][,helper=helper]\n" |
| 2837 | " configure a host TAP network backend with ID 'str' that is\n" |
| 2838 | " connected to a bridge (default=" DEFAULT_BRIDGE_INTERFACE ")\n" |
| 2839 | " using the program 'helper (default=" DEFAULT_BRIDGE_HELPER ")\n" |
Mark McLoughlin | 0df0ff6 | 2009-06-18 18:21:34 +0100 | [diff] [blame] | 2840 | #endif |
Anton Ivanov | 3fb69aa | 2014-06-20 10:34:41 +0100 | [diff] [blame] | 2841 | #ifdef __linux__ |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2842 | "-netdev l2tpv3,id=str,src=srcaddr,dst=dstaddr[,srcport=srcport][,dstport=dstport]\n" |
Daniel P. Berrangé | 8b0dc24 | 2021-02-16 19:10:21 +0000 | [diff] [blame] | 2843 | " [,rxsession=rxsession],txsession=txsession[,ipv6=on|off][,udp=on|off]\n" |
| 2844 | " [,cookie64=on|off][,counter][,pincounter][,txcookie=txcookie]\n" |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2845 | " [,rxcookie=rxcookie][,offset=offset]\n" |
| 2846 | " configure a network backend with ID 'str' connected to\n" |
| 2847 | " an Ethernet over L2TPv3 pseudowire.\n" |
Anton Ivanov | 3fb69aa | 2014-06-20 10:34:41 +0100 | [diff] [blame] | 2848 | " Linux kernel 3.3+ as well as most routers can talk\n" |
Michael Tokarev | 2f47b40 | 2014-07-24 20:10:17 +0400 | [diff] [blame] | 2849 | " L2TPv3. This transport allows connecting a VM to a VM,\n" |
Anton Ivanov | 3fb69aa | 2014-06-20 10:34:41 +0100 | [diff] [blame] | 2850 | " VM to a router and even VM to Host. It is a nearly-universal\n" |
Stefan Hajnoczi | 21843dc | 2020-02-29 11:17:27 +0000 | [diff] [blame] | 2851 | " standard (RFC3931). Note - this implementation uses static\n" |
Anton Ivanov | 3fb69aa | 2014-06-20 10:34:41 +0100 | [diff] [blame] | 2852 | " pre-configured tunnels (same as the Linux kernel).\n" |
| 2853 | " use 'src=' to specify source address\n" |
| 2854 | " use 'dst=' to specify destination address\n" |
| 2855 | " use 'udp=on' to specify udp encapsulation\n" |
Gonglei | 3952651 | 2014-08-14 14:35:48 +0800 | [diff] [blame] | 2856 | " use 'srcport=' to specify source udp port\n" |
Anton Ivanov | 3fb69aa | 2014-06-20 10:34:41 +0100 | [diff] [blame] | 2857 | " use 'dstport=' to specify destination udp port\n" |
| 2858 | " use 'ipv6=on' to force v6\n" |
| 2859 | " L2TPv3 uses cookies to prevent misconfiguration as\n" |
| 2860 | " well as a weak security measure\n" |
| 2861 | " use 'rxcookie=0x012345678' to specify a rxcookie\n" |
| 2862 | " use 'txcookie=0x012345678' to specify a txcookie\n" |
| 2863 | " use 'cookie64=on' to set cookie size to 64 bit, otherwise 32\n" |
| 2864 | " use 'counter=off' to force a 'cut-down' L2TPv3 with no counter\n" |
| 2865 | " use 'pincounter=on' to work around broken counter handling in peer\n" |
| 2866 | " use 'offset=X' to add an extra offset between header and data\n" |
| 2867 | #endif |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2868 | "-netdev socket,id=str[,fd=h][,listen=[host]:port][,connect=host:port]\n" |
| 2869 | " configure a network backend to connect to another network\n" |
| 2870 | " using a socket connection\n" |
| 2871 | "-netdev socket,id=str[,fd=h][,mcast=maddr:port[,localaddr=addr]]\n" |
| 2872 | " configure a network backend to connect to a multicast maddr and port\n" |
Mike Ryan | 3a75e74 | 2010-12-01 11:16:47 -0800 | [diff] [blame] | 2873 | " use 'localaddr=addr' to specify the host address to send packets from\n" |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2874 | "-netdev socket,id=str[,fd=h][,udp=host:port][,localaddr=host:port]\n" |
| 2875 | " configure a network backend to connect to another network\n" |
| 2876 | " using an UDP tunnel\n" |
Laurent Vivier | 148fbf0 | 2023-01-19 11:16:45 +0100 | [diff] [blame] | 2877 | "-netdev stream,id=str[,server=on|off],addr.type=inet,addr.host=host,addr.port=port[,to=maxport][,numeric=on|off][,keep-alive=on|off][,mptcp=on|off][,addr.ipv4=on|off][,addr.ipv6=on|off][,reconnect=seconds]\n" |
| 2878 | "-netdev stream,id=str[,server=on|off],addr.type=unix,addr.path=path[,abstract=on|off][,tight=on|off][,reconnect=seconds]\n" |
| 2879 | "-netdev stream,id=str[,server=on|off],addr.type=fd,addr.str=file-descriptor[,reconnect=seconds]\n" |
Laurent Vivier | 5166fe0 | 2022-10-21 11:09:11 +0200 | [diff] [blame] | 2880 | " configure a network backend to connect to another network\n" |
| 2881 | " using a socket connection in stream mode.\n" |
| 2882 | "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=inet,local.host=addr]\n" |
| 2883 | "-netdev dgram,id=str,remote.type=inet,remote.host=maddr,remote.port=port[,local.type=fd,local.str=file-descriptor]\n" |
| 2884 | " configure a network backend to connect to a multicast maddr and port\n" |
| 2885 | " use ``local.host=addr`` to specify the host address to send packets from\n" |
| 2886 | "-netdev dgram,id=str,local.type=inet,local.host=addr,local.port=port[,remote.type=inet,remote.host=addr,remote.port=port]\n" |
Laurent Vivier | 784e7a2 | 2022-10-21 11:09:17 +0200 | [diff] [blame] | 2887 | "-netdev dgram,id=str,local.type=unix,local.path=path[,remote.type=unix,remote.path=path]\n" |
Laurent Vivier | 5166fe0 | 2022-10-21 11:09:11 +0200 | [diff] [blame] | 2888 | "-netdev dgram,id=str,local.type=fd,local.str=file-descriptor\n" |
| 2889 | " configure a network backend to connect to another network\n" |
| 2890 | " using an UDP tunnel\n" |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2891 | #ifdef CONFIG_VDE |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2892 | "-netdev vde,id=str[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]\n" |
| 2893 | " configure a network backend to connect to port 'n' of a vde switch\n" |
| 2894 | " running on host and listening for incoming connections on 'socketpath'.\n" |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 2895 | " Use group 'groupname' and mode 'octalmode' to change default\n" |
| 2896 | " ownership and permissions for communication port.\n" |
| 2897 | #endif |
Vincenzo Maffione | 5895213 | 2013-11-06 11:44:06 +0100 | [diff] [blame] | 2898 | #ifdef CONFIG_NETMAP |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2899 | "-netdev netmap,id=str,ifname=name[,devname=nmname]\n" |
Vincenzo Maffione | 5895213 | 2013-11-06 11:44:06 +0100 | [diff] [blame] | 2900 | " attach to the existing netmap-enabled network interface 'name', or to a\n" |
| 2901 | " VALE port (created on the fly) called 'name' ('nmname' is name of the \n" |
| 2902 | " netmap device, defaults to '/dev/netmap')\n" |
| 2903 | #endif |
Ilya Maximets | cb039ef | 2023-09-13 20:34:37 +0200 | [diff] [blame] | 2904 | #ifdef CONFIG_AF_XDP |
| 2905 | "-netdev af-xdp,id=str,ifname=name[,mode=native|skb][,force-copy=on|off]\n" |
| 2906 | " [,queues=n][,start-queue=m][,inhibit=on|off][,sock-fds=x:y:...:z]\n" |
| 2907 | " attach to the existing network interface 'name' with AF_XDP socket\n" |
| 2908 | " use 'mode=MODE' to specify an XDP program attach mode\n" |
| 2909 | " use 'force-copy=on|off' to force XDP copy mode even if device supports zero-copy (default: off)\n" |
| 2910 | " use 'inhibit=on|off' to inhibit loading of a default XDP program (default: off)\n" |
| 2911 | " with inhibit=on,\n" |
| 2912 | " use 'sock-fds' to provide file descriptors for already open AF_XDP sockets\n" |
| 2913 | " added to a socket map in XDP program. One socket per queue.\n" |
| 2914 | " use 'queues=n' to specify how many queues of a multiqueue interface should be used\n" |
| 2915 | " use 'start-queue=m' to specify the first queue that should be used\n" |
| 2916 | #endif |
Thomas Huth | 253dc14 | 2018-02-21 11:18:32 +0100 | [diff] [blame] | 2917 | #ifdef CONFIG_POSIX |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2918 | "-netdev vhost-user,id=str,chardev=dev[,vhostforce=on|off]\n" |
| 2919 | " configure a vhost-user network, backed by a chardev 'dev'\n" |
Thomas Huth | 253dc14 | 2018-02-21 11:18:32 +0100 | [diff] [blame] | 2920 | #endif |
Cindy Lu | 108a648 | 2020-07-01 22:55:37 +0800 | [diff] [blame] | 2921 | #ifdef __linux__ |
Si-Wei Liu | 8801ccd | 2022-10-08 00:58:58 -0700 | [diff] [blame] | 2922 | "-netdev vhost-vdpa,id=str[,vhostdev=/path/to/dev][,vhostfd=h]\n" |
Cindy Lu | 108a648 | 2020-07-01 22:55:37 +0800 | [diff] [blame] | 2923 | " configure a vhost-vdpa network,Establish a vhost-vdpa netdev\n" |
Si-Wei Liu | 8801ccd | 2022-10-08 00:58:58 -0700 | [diff] [blame] | 2924 | " use 'vhostdev=/path/to/dev' to open a vhost vdpa device\n" |
| 2925 | " use 'vhostfd=h' to connect to an already opened vhost vdpa device\n" |
Cindy Lu | 108a648 | 2020-07-01 22:55:37 +0800 | [diff] [blame] | 2926 | #endif |
Vladislav Yaroshchuk | b0290db | 2022-03-17 20:28:38 +0300 | [diff] [blame] | 2927 | #ifdef CONFIG_VMNET |
| 2928 | "-netdev vmnet-host,id=str[,isolated=on|off][,net-uuid=uuid]\n" |
| 2929 | " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" |
| 2930 | " configure a vmnet network backend in host mode with ID 'str',\n" |
| 2931 | " isolate this interface from others with 'isolated',\n" |
| 2932 | " configure the address range and choose a subnet mask,\n" |
| 2933 | " specify network UUID 'uuid' to disable DHCP and interact with\n" |
| 2934 | " vmnet-host interfaces within this isolated network\n" |
| 2935 | "-netdev vmnet-shared,id=str[,isolated=on|off][,nat66-prefix=addr]\n" |
| 2936 | " [,start-address=addr,end-address=addr,subnet-mask=mask]\n" |
| 2937 | " configure a vmnet network backend in shared mode with ID 'str',\n" |
| 2938 | " configure the address range and choose a subnet mask,\n" |
| 2939 | " set IPv6 ULA prefix (of length 64) to use for internal network,\n" |
| 2940 | " isolate this interface from others with 'isolated'\n" |
| 2941 | "-netdev vmnet-bridged,id=str,ifname=name[,isolated=on|off]\n" |
| 2942 | " configure a vmnet network backend in bridged mode with ID 'str',\n" |
| 2943 | " use 'ifname=name' to select a physical network interface to be bridged,\n" |
| 2944 | " isolate this interface from others with 'isolated'\n" |
| 2945 | #endif |
Thomas Huth | 18d65d2 | 2018-01-15 20:50:55 +0100 | [diff] [blame] | 2946 | "-netdev hubport,id=str,hubid=n[,netdev=nd]\n" |
Thomas Huth | af1a5c3 | 2018-04-30 20:02:23 +0200 | [diff] [blame] | 2947 | " configure a hub port on the hub with ID 'n'\n", QEMU_ARCH_ALL) |
Thomas Huth | 78cd6f7 | 2018-02-21 11:18:36 +0100 | [diff] [blame] | 2948 | DEF("nic", HAS_ARG, QEMU_OPTION_nic, |
BALATON Zoltan | dfaa7d5 | 2018-07-16 21:12:08 +0200 | [diff] [blame] | 2949 | "-nic [tap|bridge|" |
Thomas Huth | 78cd6f7 | 2018-02-21 11:18:36 +0100 | [diff] [blame] | 2950 | #ifdef CONFIG_SLIRP |
| 2951 | "user|" |
| 2952 | #endif |
| 2953 | #ifdef __linux__ |
| 2954 | "l2tpv3|" |
| 2955 | #endif |
| 2956 | #ifdef CONFIG_VDE |
| 2957 | "vde|" |
| 2958 | #endif |
| 2959 | #ifdef CONFIG_NETMAP |
| 2960 | "netmap|" |
| 2961 | #endif |
Ilya Maximets | cb039ef | 2023-09-13 20:34:37 +0200 | [diff] [blame] | 2962 | #ifdef CONFIG_AF_XDP |
| 2963 | "af-xdp|" |
| 2964 | #endif |
Thomas Huth | 78cd6f7 | 2018-02-21 11:18:36 +0100 | [diff] [blame] | 2965 | #ifdef CONFIG_POSIX |
| 2966 | "vhost-user|" |
| 2967 | #endif |
Vladislav Yaroshchuk | b0290db | 2022-03-17 20:28:38 +0300 | [diff] [blame] | 2968 | #ifdef CONFIG_VMNET |
| 2969 | "vmnet-host|vmnet-shared|vmnet-bridged|" |
| 2970 | #endif |
Thomas Huth | 78cd6f7 | 2018-02-21 11:18:36 +0100 | [diff] [blame] | 2971 | "socket][,option][,...][mac=macaddr]\n" |
| 2972 | " initialize an on-board / default host NIC (using MAC address\n" |
| 2973 | " macaddr) and connect it to the given host network backend\n" |
BALATON Zoltan | dfaa7d5 | 2018-07-16 21:12:08 +0200 | [diff] [blame] | 2974 | "-nic none use it alone to have zero network devices (the default is to\n" |
Thomas Huth | 78cd6f7 | 2018-02-21 11:18:36 +0100 | [diff] [blame] | 2975 | " provided a 'user' network connection)\n", |
| 2976 | QEMU_ARCH_ALL) |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2977 | DEF("net", HAS_ARG, QEMU_OPTION_net, |
Thomas Huth | af1a5c3 | 2018-04-30 20:02:23 +0200 | [diff] [blame] | 2978 | "-net nic[,macaddr=mac][,model=type][,name=str][,addr=str][,vectors=v]\n" |
Thomas Huth | 0e60a82 | 2017-12-19 16:28:55 +0100 | [diff] [blame] | 2979 | " configure or create an on-board (or machine default) NIC and\n" |
Thomas Huth | af1a5c3 | 2018-04-30 20:02:23 +0200 | [diff] [blame] | 2980 | " connect it to hub 0 (please use -nic unless you need a hub)\n" |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 2981 | "-net [" |
Mark McLoughlin | a1ea458 | 2009-10-08 19:58:26 +0100 | [diff] [blame] | 2982 | #ifdef CONFIG_SLIRP |
| 2983 | "user|" |
| 2984 | #endif |
| 2985 | "tap|" |
Corey Bryant | a7c36ee | 2012-01-26 09:42:27 -0500 | [diff] [blame] | 2986 | "bridge|" |
Mark McLoughlin | a1ea458 | 2009-10-08 19:58:26 +0100 | [diff] [blame] | 2987 | #ifdef CONFIG_VDE |
| 2988 | "vde|" |
| 2989 | #endif |
Vincenzo Maffione | 5895213 | 2013-11-06 11:44:06 +0100 | [diff] [blame] | 2990 | #ifdef CONFIG_NETMAP |
| 2991 | "netmap|" |
| 2992 | #endif |
Ilya Maximets | cb039ef | 2023-09-13 20:34:37 +0200 | [diff] [blame] | 2993 | #ifdef CONFIG_AF_XDP |
| 2994 | "af-xdp|" |
| 2995 | #endif |
Vladislav Yaroshchuk | b0290db | 2022-03-17 20:28:38 +0300 | [diff] [blame] | 2996 | #ifdef CONFIG_VMNET |
| 2997 | "vmnet-host|vmnet-shared|vmnet-bridged|" |
| 2998 | #endif |
Thomas Huth | af1a5c3 | 2018-04-30 20:02:23 +0200 | [diff] [blame] | 2999 | "socket][,option][,option][,...]\n" |
Thomas Huth | 6a8b4a5 | 2015-05-15 16:58:24 +0200 | [diff] [blame] | 3000 | " old way to initialize a host network interface\n" |
| 3001 | " (use the -netdev option if possible instead)\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3002 | SRST |
Ilya Maximets | cb039ef | 2023-09-13 20:34:37 +0200 | [diff] [blame] | 3003 | ``-nic [tap|bridge|user|l2tpv3|vde|netmap|af-xdp|vhost-user|socket][,...][,mac=macaddr][,model=mn]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3004 | This option is a shortcut for configuring both the on-board |
| 3005 | (default) guest NIC hardware and the host network backend in one go. |
| 3006 | The host backend options are the same as with the corresponding |
| 3007 | ``-netdev`` options below. The guest NIC model can be set with |
| 3008 | ``model=modelname``. Use ``model=help`` to list the available device |
| 3009 | types. The hardware MAC address can be set with ``mac=macaddr``. |
| 3010 | |
| 3011 | The following two example do exactly the same, to show how ``-nic`` |
| 3012 | can be used to shorten the command line length: |
| 3013 | |
| 3014 | .. parsed-literal:: |
| 3015 | |
| 3016 | |qemu_system| -netdev user,id=n1,ipv6=off -device e1000,netdev=n1,mac=52:54:98:76:54:32 |
| 3017 | |qemu_system| -nic user,ipv6=off,model=e1000,mac=52:54:98:76:54:32 |
| 3018 | |
| 3019 | ``-nic none`` |
| 3020 | Indicate that no network devices should be configured. It is used to |
| 3021 | override the default configuration (default NIC with "user" host |
| 3022 | network backend) which is activated if no other networking options |
| 3023 | are provided. |
| 3024 | |
| 3025 | ``-netdev user,id=id[,option][,option][,...]`` |
| 3026 | Configure user mode host network backend which requires no |
| 3027 | administrator privilege to run. Valid options are: |
| 3028 | |
| 3029 | ``id=id`` |
| 3030 | Assign symbolic name for use in monitor commands. |
| 3031 | |
| 3032 | ``ipv4=on|off and ipv6=on|off`` |
| 3033 | Specify that either IPv4 or IPv6 must be enabled. If neither is |
| 3034 | specified both protocols are enabled. |
| 3035 | |
| 3036 | ``net=addr[/mask]`` |
| 3037 | Set IP network address the guest will see. Optionally specify |
| 3038 | the netmask, either in the form a.b.c.d or as number of valid |
| 3039 | top-most bits. Default is 10.0.2.0/24. |
| 3040 | |
| 3041 | ``host=addr`` |
| 3042 | Specify the guest-visible address of the host. Default is the |
| 3043 | 2nd IP in the guest network, i.e. x.x.x.2. |
| 3044 | |
| 3045 | ``ipv6-net=addr[/int]`` |
| 3046 | Set IPv6 network address the guest will see (default is |
| 3047 | fec0::/64). The network prefix is given in the usual hexadecimal |
| 3048 | IPv6 address notation. The prefix size is optional, and is given |
| 3049 | as the number of valid top-most bits (default is 64). |
| 3050 | |
| 3051 | ``ipv6-host=addr`` |
| 3052 | Specify the guest-visible IPv6 address of the host. Default is |
| 3053 | the 2nd IPv6 in the guest network, i.e. xxxx::2. |
| 3054 | |
| 3055 | ``restrict=on|off`` |
| 3056 | If this option is enabled, the guest will be isolated, i.e. it |
| 3057 | will not be able to contact the host and no guest IP packets |
| 3058 | will be routed over the host to the outside. This option does |
| 3059 | not affect any explicitly set forwarding rules. |
| 3060 | |
| 3061 | ``hostname=name`` |
| 3062 | Specifies the client hostname reported by the built-in DHCP |
| 3063 | server. |
| 3064 | |
| 3065 | ``dhcpstart=addr`` |
| 3066 | Specify the first of the 16 IPs the built-in DHCP server can |
| 3067 | assign. Default is the 15th to 31st IP in the guest network, |
| 3068 | i.e. x.x.x.15 to x.x.x.31. |
| 3069 | |
| 3070 | ``dns=addr`` |
| 3071 | Specify the guest-visible address of the virtual nameserver. The |
| 3072 | address must be different from the host address. Default is the |
| 3073 | 3rd IP in the guest network, i.e. x.x.x.3. |
| 3074 | |
| 3075 | ``ipv6-dns=addr`` |
| 3076 | Specify the guest-visible address of the IPv6 virtual |
| 3077 | nameserver. The address must be different from the host address. |
| 3078 | Default is the 3rd IP in the guest network, i.e. xxxx::3. |
| 3079 | |
| 3080 | ``dnssearch=domain`` |
| 3081 | Provides an entry for the domain-search list sent by the |
| 3082 | built-in DHCP server. More than one domain suffix can be |
| 3083 | transmitted by specifying this option multiple times. If |
| 3084 | supported, this will cause the guest to automatically try to |
| 3085 | append the given domain suffix(es) in case a domain name can not |
| 3086 | be resolved. |
| 3087 | |
| 3088 | Example: |
| 3089 | |
| 3090 | .. parsed-literal:: |
| 3091 | |
| 3092 | |qemu_system| -nic user,dnssearch=mgmt.example.org,dnssearch=example.org |
| 3093 | |
| 3094 | ``domainname=domain`` |
| 3095 | Specifies the client domain name reported by the built-in DHCP |
| 3096 | server. |
| 3097 | |
| 3098 | ``tftp=dir`` |
| 3099 | When using the user mode network stack, activate a built-in TFTP |
| 3100 | server. The files in dir will be exposed as the root of a TFTP |
| 3101 | server. The TFTP client on the guest must be configured in |
| 3102 | binary mode (use the command ``bin`` of the Unix TFTP client). |
Michael Tokarev | b30fa6b | 2024-02-08 09:00:50 +0300 | [diff] [blame] | 3103 | The built-in TFTP server is read-only; it does not implement any |
| 3104 | command for writing files. QEMU will not write to this directory. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3105 | |
| 3106 | ``tftp-server-name=name`` |
| 3107 | In BOOTP reply, broadcast name as the "TFTP server name" |
| 3108 | (RFC2132 option 66). This can be used to advise the guest to |
| 3109 | load boot files or configurations from a different server than |
| 3110 | the host address. |
| 3111 | |
| 3112 | ``bootfile=file`` |
| 3113 | When using the user mode network stack, broadcast file as the |
| 3114 | BOOTP filename. In conjunction with ``tftp``, this can be used |
| 3115 | to network boot a guest from a local directory. |
| 3116 | |
| 3117 | Example (using pxelinux): |
| 3118 | |
| 3119 | .. parsed-literal:: |
| 3120 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3121 | |qemu_system| -hda linux.img -boot n -device e1000,netdev=n1 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3122 | -netdev user,id=n1,tftp=/path/to/tftp/files,bootfile=/pxelinux.0 |
| 3123 | |
| 3124 | ``smb=dir[,smbserver=addr]`` |
| 3125 | When using the user mode network stack, activate a built-in SMB |
| 3126 | server so that Windows OSes can access to the host files in |
| 3127 | ``dir`` transparently. The IP address of the SMB server can be |
| 3128 | set to addr. By default the 4th IP in the guest network is used, |
| 3129 | i.e. x.x.x.4. |
| 3130 | |
| 3131 | In the guest Windows OS, the line: |
| 3132 | |
| 3133 | :: |
| 3134 | |
| 3135 | 10.0.2.4 smbserver |
| 3136 | |
| 3137 | must be added in the file ``C:\WINDOWS\LMHOSTS`` (for windows |
| 3138 | 9x/Me) or ``C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS`` (Windows |
| 3139 | NT/2000). |
| 3140 | |
| 3141 | Then ``dir`` can be accessed in ``\\smbserver\qemu``. |
| 3142 | |
| 3143 | Note that a SAMBA server must be installed on the host OS. |
| 3144 | |
| 3145 | ``hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport`` |
| 3146 | Redirect incoming TCP or UDP connections to the host port |
| 3147 | hostport to the guest IP address guestaddr on guest port |
| 3148 | guestport. If guestaddr is not specified, its value is x.x.x.15 |
| 3149 | (default first address given by the built-in DHCP server). By |
| 3150 | specifying hostaddr, the rule can be bound to a specific host |
| 3151 | interface. If no connection type is set, TCP is used. This |
| 3152 | option can be given multiple times. |
| 3153 | |
| 3154 | For example, to redirect host X11 connection from screen 1 to |
| 3155 | guest screen 0, use the following: |
| 3156 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3157 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3158 | |
| 3159 | # on the host |
| 3160 | |qemu_system| -nic user,hostfwd=tcp:127.0.0.1:6001-:6000 |
| 3161 | # this host xterm should open in the guest X11 server |
| 3162 | xterm -display :1 |
| 3163 | |
| 3164 | To redirect telnet connections from host port 5555 to telnet |
| 3165 | port on the guest, use the following: |
| 3166 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3167 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3168 | |
| 3169 | # on the host |
| 3170 | |qemu_system| -nic user,hostfwd=tcp::5555-:23 |
| 3171 | telnet localhost 5555 |
| 3172 | |
| 3173 | Then when you use on the host ``telnet localhost 5555``, you |
| 3174 | connect to the guest telnet server. |
| 3175 | |
| 3176 | ``guestfwd=[tcp]:server:port-dev``; \ ``guestfwd=[tcp]:server:port-cmd:command`` |
| 3177 | Forward guest TCP connections to the IP address server on port |
| 3178 | port to the character device dev or to a program executed by |
| 3179 | cmd:command which gets spawned for each connection. This option |
| 3180 | can be given multiple times. |
| 3181 | |
| 3182 | You can either use a chardev directly and have that one used |
| 3183 | throughout QEMU's lifetime, like in the following example: |
| 3184 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3185 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3186 | |
| 3187 | # open 10.10.1.1:4321 on bootup, connect 10.0.2.100:1234 to it whenever |
| 3188 | # the guest accesses it |
| 3189 | |qemu_system| -nic user,guestfwd=tcp:10.0.2.100:1234-tcp:10.10.1.1:4321 |
| 3190 | |
| 3191 | Or you can execute a command on every TCP connection established |
| 3192 | by the guest, so that QEMU behaves similar to an inetd process |
| 3193 | for that virtual server: |
| 3194 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3195 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3196 | |
| 3197 | # call "netcat 10.10.1.1 4321" on every TCP connection to 10.0.2.100:1234 |
| 3198 | # and connect the TCP stream to its stdin/stdout |
| 3199 | |qemu_system| -nic 'user,id=n1,guestfwd=tcp:10.0.2.100:1234-cmd:netcat 10.10.1.1 4321' |
| 3200 | |
| 3201 | ``-netdev tap,id=id[,fd=h][,ifname=name][,script=file][,downscript=dfile][,br=bridge][,helper=helper]`` |
| 3202 | Configure a host TAP network backend with ID id. |
| 3203 | |
| 3204 | Use the network script file to configure it and the network script |
| 3205 | dfile to deconfigure it. If name is not provided, the OS |
| 3206 | automatically provides one. The default network configure script is |
| 3207 | ``/etc/qemu-ifup`` and the default network deconfigure script is |
| 3208 | ``/etc/qemu-ifdown``. Use ``script=no`` or ``downscript=no`` to |
| 3209 | disable script execution. |
| 3210 | |
| 3211 | If running QEMU as an unprivileged user, use the network helper |
Tianjia Zhang | 8d73ec8 | 2020-07-27 12:59:25 +0800 | [diff] [blame] | 3212 | to configure the TAP interface and attach it to the bridge. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3213 | The default network helper executable is |
| 3214 | ``/path/to/qemu-bridge-helper`` and the default bridge device is |
| 3215 | ``br0``. |
| 3216 | |
| 3217 | ``fd``\ =h can be used to specify the handle of an already opened |
| 3218 | host TAP interface. |
| 3219 | |
| 3220 | Examples: |
| 3221 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3222 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3223 | |
| 3224 | #launch a QEMU instance with the default network script |
| 3225 | |qemu_system| linux.img -nic tap |
| 3226 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3227 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3228 | |
| 3229 | #launch a QEMU instance with two NICs, each one connected |
| 3230 | #to a TAP device |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3231 | |qemu_system| linux.img \\ |
| 3232 | -netdev tap,id=nd0,ifname=tap0 -device e1000,netdev=nd0 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3233 | -netdev tap,id=nd1,ifname=tap1 -device rtl8139,netdev=nd1 |
| 3234 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3235 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3236 | |
| 3237 | #launch a QEMU instance with the default network helper to |
| 3238 | #connect a TAP device to bridge br0 |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3239 | |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3240 | -netdev tap,id=n1,"helper=/path/to/qemu-bridge-helper" |
| 3241 | |
| 3242 | ``-netdev bridge,id=id[,br=bridge][,helper=helper]`` |
| 3243 | Connect a host TAP network interface to a host bridge device. |
| 3244 | |
| 3245 | Use the network helper helper to configure the TAP interface and |
| 3246 | attach it to the bridge. The default network helper executable is |
| 3247 | ``/path/to/qemu-bridge-helper`` and the default bridge device is |
| 3248 | ``br0``. |
| 3249 | |
| 3250 | Examples: |
| 3251 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3252 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3253 | |
| 3254 | #launch a QEMU instance with the default network helper to |
| 3255 | #connect a TAP device to bridge br0 |
| 3256 | |qemu_system| linux.img -netdev bridge,id=n1 -device virtio-net,netdev=n1 |
| 3257 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3258 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3259 | |
| 3260 | #launch a QEMU instance with the default network helper to |
| 3261 | #connect a TAP device to bridge qemubr0 |
| 3262 | |qemu_system| linux.img -netdev bridge,br=qemubr0,id=n1 -device virtio-net,netdev=n1 |
| 3263 | |
| 3264 | ``-netdev socket,id=id[,fd=h][,listen=[host]:port][,connect=host:port]`` |
| 3265 | This host network backend can be used to connect the guest's network |
| 3266 | to another QEMU virtual machine using a TCP socket connection. If |
| 3267 | ``listen`` is specified, QEMU waits for incoming connections on port |
| 3268 | (host is optional). ``connect`` is used to connect to another QEMU |
| 3269 | instance using the ``listen`` option. ``fd``\ =h specifies an |
| 3270 | already opened TCP socket. |
| 3271 | |
| 3272 | Example: |
| 3273 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3274 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3275 | |
| 3276 | # launch a first QEMU instance |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3277 | |qemu_system| linux.img \\ |
| 3278 | -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3279 | -netdev socket,id=n1,listen=:1234 |
| 3280 | # connect the network of this instance to the network of the first instance |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3281 | |qemu_system| linux.img \\ |
| 3282 | -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3283 | -netdev socket,id=n2,connect=127.0.0.1:1234 |
| 3284 | |
| 3285 | ``-netdev socket,id=id[,fd=h][,mcast=maddr:port[,localaddr=addr]]`` |
| 3286 | Configure a socket host network backend to share the guest's network |
| 3287 | traffic with another QEMU virtual machines using a UDP multicast |
| 3288 | socket, effectively making a bus for every QEMU with same multicast |
| 3289 | address maddr and port. NOTES: |
| 3290 | |
| 3291 | 1. Several QEMU can be running on different hosts and share same bus |
| 3292 | (assuming correct multicast setup for these hosts). |
| 3293 | |
| 3294 | 2. mcast support is compatible with User Mode Linux (argument |
| 3295 | ``ethN=mcast``), see http://user-mode-linux.sf.net. |
| 3296 | |
| 3297 | 3. Use ``fd=h`` to specify an already opened UDP multicast socket. |
| 3298 | |
| 3299 | Example: |
| 3300 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3301 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3302 | |
| 3303 | # launch one QEMU instance |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3304 | |qemu_system| linux.img \\ |
| 3305 | -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3306 | -netdev socket,id=n1,mcast=230.0.0.1:1234 |
| 3307 | # launch another QEMU instance on same "bus" |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3308 | |qemu_system| linux.img \\ |
| 3309 | -device e1000,netdev=n2,mac=52:54:00:12:34:57 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3310 | -netdev socket,id=n2,mcast=230.0.0.1:1234 |
| 3311 | # launch yet another QEMU instance on same "bus" |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3312 | |qemu_system| linux.img \\ |
| 3313 | -device e1000,netdev=n3,mac=52:54:00:12:34:58 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3314 | -netdev socket,id=n3,mcast=230.0.0.1:1234 |
| 3315 | |
| 3316 | Example (User Mode Linux compat.): |
| 3317 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3318 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3319 | |
| 3320 | # launch QEMU instance (note mcast address selected is UML's default) |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3321 | |qemu_system| linux.img \\ |
| 3322 | -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3323 | -netdev socket,id=n1,mcast=239.192.168.1:1102 |
| 3324 | # launch UML |
| 3325 | /path/to/linux ubd0=/path/to/root_fs eth0=mcast |
| 3326 | |
| 3327 | Example (send packets from host's 1.2.3.4): |
| 3328 | |
| 3329 | .. parsed-literal:: |
| 3330 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3331 | |qemu_system| linux.img \\ |
| 3332 | -device e1000,netdev=n1,mac=52:54:00:12:34:56 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3333 | -netdev socket,id=n1,mcast=239.192.168.1:1102,localaddr=1.2.3.4 |
| 3334 | |
Daniel P. Berrangé | 8b0dc24 | 2021-02-16 19:10:21 +0000 | [diff] [blame] | 3335 | ``-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]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3336 | Configure a L2TPv3 pseudowire host network backend. L2TPv3 (RFC3931) |
| 3337 | is a popular protocol to transport Ethernet (and other Layer 2) data |
| 3338 | frames between two systems. It is present in routers, firewalls and |
| 3339 | the Linux kernel (from version 3.3 onwards). |
| 3340 | |
| 3341 | This transport allows a VM to communicate to another VM, router or |
| 3342 | firewall directly. |
| 3343 | |
| 3344 | ``src=srcaddr`` |
| 3345 | source address (mandatory) |
| 3346 | |
| 3347 | ``dst=dstaddr`` |
| 3348 | destination address (mandatory) |
| 3349 | |
| 3350 | ``udp`` |
| 3351 | select udp encapsulation (default is ip). |
| 3352 | |
| 3353 | ``srcport=srcport`` |
| 3354 | source udp port. |
| 3355 | |
| 3356 | ``dstport=dstport`` |
| 3357 | destination udp port. |
| 3358 | |
| 3359 | ``ipv6`` |
| 3360 | force v6, otherwise defaults to v4. |
| 3361 | |
| 3362 | ``rxcookie=rxcookie``; \ ``txcookie=txcookie`` |
| 3363 | Cookies are a weak form of security in the l2tpv3 specification. |
| 3364 | Their function is mostly to prevent misconfiguration. By default |
| 3365 | they are 32 bit. |
| 3366 | |
| 3367 | ``cookie64`` |
| 3368 | Set cookie size to 64 bit instead of the default 32 |
| 3369 | |
| 3370 | ``counter=off`` |
| 3371 | Force a 'cut-down' L2TPv3 with no counter as in |
| 3372 | draft-mkonstan-l2tpext-keyed-ipv6-tunnel-00 |
| 3373 | |
| 3374 | ``pincounter=on`` |
| 3375 | Work around broken counter handling in peer. This may also help |
| 3376 | on networks which have packet reorder. |
| 3377 | |
| 3378 | ``offset=offset`` |
| 3379 | Add an extra offset between header and data |
| 3380 | |
| 3381 | For example, to attach a VM running on host 4.3.2.1 via L2TPv3 to |
| 3382 | the bridge br-lan on the remote Linux host 1.2.3.4: |
| 3383 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3384 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3385 | |
| 3386 | # Setup tunnel on linux host using raw ip as encapsulation |
| 3387 | # on 1.2.3.4 |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3388 | ip l2tp add tunnel remote 4.3.2.1 local 1.2.3.4 tunnel_id 1 peer_tunnel_id 1 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3389 | encap udp udp_sport 16384 udp_dport 16384 |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3390 | ip l2tp add session tunnel_id 1 name vmtunnel0 session_id \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3391 | 0xFFFFFFFF peer_session_id 0xFFFFFFFF |
| 3392 | ifconfig vmtunnel0 mtu 1500 |
| 3393 | ifconfig vmtunnel0 up |
| 3394 | brctl addif br-lan vmtunnel0 |
| 3395 | |
| 3396 | |
| 3397 | # on 4.3.2.1 |
| 3398 | # launch QEMU instance - if your network has reorder or is very lossy add ,pincounter |
| 3399 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 3400 | |qemu_system| linux.img -device e1000,netdev=n1 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3401 | -netdev l2tpv3,id=n1,src=4.2.3.1,dst=1.2.3.4,udp,srcport=16384,dstport=16384,rxsession=0xffffffff,txsession=0xffffffff,counter |
| 3402 | |
| 3403 | ``-netdev vde,id=id[,sock=socketpath][,port=n][,group=groupname][,mode=octalmode]`` |
| 3404 | Configure VDE backend to connect to PORT n of a vde switch running |
| 3405 | on host and listening for incoming connections on socketpath. Use |
| 3406 | GROUP groupname and MODE octalmode to change default ownership and |
| 3407 | permissions for communication port. This option is only available if |
| 3408 | QEMU has been compiled with vde support enabled. |
| 3409 | |
| 3410 | Example: |
| 3411 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3412 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3413 | |
| 3414 | # launch vde switch |
| 3415 | vde_switch -F -sock /tmp/myswitch |
| 3416 | # launch QEMU instance |
| 3417 | |qemu_system| linux.img -nic vde,sock=/tmp/myswitch |
| 3418 | |
Ilya Maximets | cb039ef | 2023-09-13 20:34:37 +0200 | [diff] [blame] | 3419 | ``-netdev af-xdp,id=str,ifname=name[,mode=native|skb][,force-copy=on|off][,queues=n][,start-queue=m][,inhibit=on|off][,sock-fds=x:y:...:z]`` |
| 3420 | Configure AF_XDP backend to connect to a network interface 'name' |
| 3421 | using AF_XDP socket. A specific program attach mode for a default |
| 3422 | XDP program can be forced with 'mode', defaults to best-effort, |
| 3423 | where the likely most performant mode will be in use. Number of queues |
| 3424 | 'n' should generally match the number or queues in the interface, |
| 3425 | defaults to 1. Traffic arriving on non-configured device queues will |
| 3426 | not be delivered to the network backend. |
| 3427 | |
| 3428 | .. parsed-literal:: |
| 3429 | |
| 3430 | # set number of queues to 4 |
| 3431 | ethtool -L eth0 combined 4 |
| 3432 | # launch QEMU instance |
| 3433 | |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ |
| 3434 | -netdev af-xdp,id=n1,ifname=eth0,queues=4 |
| 3435 | |
| 3436 | 'start-queue' option can be specified if a particular range of queues |
| 3437 | [m, m + n] should be in use. For example, this is may be necessary in |
| 3438 | order to use certain NICs in native mode. Kernel allows the driver to |
| 3439 | create a separate set of XDP queues on top of regular ones, and only |
| 3440 | these queues can be used for AF_XDP sockets. NICs that work this way |
| 3441 | may also require an additional traffic redirection with ethtool to these |
| 3442 | special queues. |
| 3443 | |
| 3444 | .. parsed-literal:: |
| 3445 | |
| 3446 | # set number of queues to 1 |
| 3447 | ethtool -L eth0 combined 1 |
| 3448 | # redirect all the traffic to the second queue (id: 1) |
| 3449 | # note: drivers may require non-empty key/mask pair. |
| 3450 | ethtool -N eth0 flow-type ether \\ |
| 3451 | dst 00:00:00:00:00:00 m FF:FF:FF:FF:FF:FE action 1 |
| 3452 | ethtool -N eth0 flow-type ether \\ |
| 3453 | dst 00:00:00:00:00:01 m FF:FF:FF:FF:FF:FE action 1 |
| 3454 | # launch QEMU instance |
| 3455 | |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ |
| 3456 | -netdev af-xdp,id=n1,ifname=eth0,queues=1,start-queue=1 |
| 3457 | |
| 3458 | XDP program can also be loaded externally. In this case 'inhibit' option |
| 3459 | should be set to 'on' and 'sock-fds' provided with file descriptors for |
| 3460 | already open but not bound XDP sockets already added to a socket map for |
| 3461 | corresponding queues. One socket per queue. |
| 3462 | |
| 3463 | .. parsed-literal:: |
| 3464 | |
| 3465 | |qemu_system| linux.img -device virtio-net-pci,netdev=n1 \\ |
| 3466 | -netdev af-xdp,id=n1,ifname=eth0,queues=3,inhibit=on,sock-fds=15:16:17 |
| 3467 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3468 | ``-netdev vhost-user,chardev=id[,vhostforce=on|off][,queues=n]`` |
| 3469 | Establish a vhost-user netdev, backed by a chardev id. The chardev |
| 3470 | should be a unix domain socket backed one. The vhost-user uses a |
| 3471 | specifically defined protocol to pass vhost ioctl replacement |
| 3472 | messages to an application on the other end of the socket. On |
| 3473 | non-MSIX guests, the feature can be forced with vhostforce. Use |
| 3474 | 'queues=n' to specify the number of queues to be created for |
| 3475 | multiqueue vhost-user. |
| 3476 | |
| 3477 | Example: |
| 3478 | |
| 3479 | :: |
| 3480 | |
| 3481 | qemu -m 512 -object memory-backend-file,id=mem,size=512M,mem-path=/hugetlbfs,share=on \ |
| 3482 | -numa node,memdev=mem \ |
| 3483 | -chardev socket,id=chr0,path=/path/to/socket \ |
| 3484 | -netdev type=vhost-user,id=net0,chardev=chr0 \ |
| 3485 | -device virtio-net-pci,netdev=net0 |
| 3486 | |
Si-Wei Liu | 8801ccd | 2022-10-08 00:58:58 -0700 | [diff] [blame] | 3487 | ``-netdev vhost-vdpa[,vhostdev=/path/to/dev][,vhostfd=h]`` |
Cindy Lu | 108a648 | 2020-07-01 22:55:37 +0800 | [diff] [blame] | 3488 | Establish a vhost-vdpa netdev. |
| 3489 | |
| 3490 | vDPA device is a device that uses a datapath which complies with |
| 3491 | the virtio specifications with a vendor specific control path. |
| 3492 | vDPA devices can be both physically located on the hardware or |
| 3493 | emulated by software. |
| 3494 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3495 | ``-netdev hubport,id=id,hubid=hubid[,netdev=nd]`` |
| 3496 | Create a hub port on the emulated hub with ID hubid. |
| 3497 | |
| 3498 | The hubport netdev lets you connect a NIC to a QEMU emulated hub |
| 3499 | instead of a single netdev. Alternatively, you can also connect the |
| 3500 | hubport to another netdev with ID nd by using the ``netdev=nd`` |
| 3501 | option. |
| 3502 | |
| 3503 | ``-net nic[,netdev=nd][,macaddr=mac][,model=type] [,name=name][,addr=addr][,vectors=v]`` |
| 3504 | Legacy option to configure or create an on-board (or machine |
| 3505 | default) Network Interface Card(NIC) and connect it either to the |
| 3506 | emulated hub with ID 0 (i.e. the default hub), or to the netdev nd. |
| 3507 | If model is omitted, then the default NIC model associated with the |
| 3508 | machine type is used. Note that the default NIC model may change in |
| 3509 | future QEMU releases, so it is highly recommended to always specify |
| 3510 | a model. Optionally, the MAC address can be changed to mac, the |
| 3511 | device address set to addr (PCI cards only), and a name can be |
| 3512 | assigned for use in monitor commands. Optionally, for PCI cards, you |
| 3513 | can specify the number v of MSI-X vectors that the card should have; |
| 3514 | this option currently only affects virtio cards; set v = 0 to |
| 3515 | disable MSI-X. If no ``-net`` option is specified, a single NIC is |
| 3516 | created. QEMU can emulate several different models of network card. |
| 3517 | Use ``-net nic,model=help`` for a list of available devices for your |
| 3518 | target. |
| 3519 | |
| 3520 | ``-net user|tap|bridge|socket|l2tpv3|vde[,...][,name=name]`` |
| 3521 | Configure a host network backend (with the options corresponding to |
| 3522 | the same ``-netdev`` option) and connect it to the emulated hub 0 |
| 3523 | (the default hub). Use name to specify the name of the hub port. |
| 3524 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 3525 | |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3526 | DEFHEADING() |
| 3527 | |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 3528 | DEFHEADING(Character device options:) |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3529 | |
| 3530 | DEF("chardev", HAS_ARG, QEMU_OPTION_chardev, |
Lin Ma | 517b3d4 | 2016-08-17 01:13:52 +0800 | [diff] [blame] | 3531 | "-chardev help\n" |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3532 | "-chardev null,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
Markus Armbruster | ba858d1 | 2021-09-28 09:14:49 +0200 | [diff] [blame] | 3533 | "-chardev socket,id=id[,host=host],port=port[,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]\n" |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3534 | " [,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds][,mux=on|off]\n" |
Daniel P. Berrange | fd4a5fd | 2019-03-08 15:21:50 +0000 | [diff] [blame] | 3535 | " [,logfile=PATH][,logappend=on|off][,tls-creds=ID][,tls-authz=ID] (tcp)\n" |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3536 | "-chardev socket,id=id,path=path[,server=on|off][,wait=on|off][,telnet=on|off][,websocket=on|off][,reconnect=seconds]\n" |
xiaoqiang zhao | e339273 | 2020-05-16 11:13:27 +0800 | [diff] [blame] | 3537 | " [,mux=on|off][,logfile=PATH][,logappend=on|off][,abstract=on|off][,tight=on|off] (unix)\n" |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3538 | "-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr]\n" |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3539 | " [,localport=localport][,ipv4=on|off][,ipv6=on|off][,mux=on|off]\n" |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3540 | " [,logfile=PATH][,logappend=on|off]\n" |
| 3541 | "-chardev msmouse,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3542 | "-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]\n" |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3543 | " [,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
| 3544 | "-chardev ringbuf,id=id[,size=size][,logfile=PATH][,logappend=on|off]\n" |
Peter Maydell | 5b18a6b | 2023-04-13 16:07:24 +0100 | [diff] [blame] | 3545 | "-chardev file,id=id,path=path[,input-path=input-file][,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3546 | "-chardev pipe,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3547 | #ifdef _WIN32 |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3548 | "-chardev console,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
| 3549 | "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3550 | #else |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3551 | "-chardev pty,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
| 3552 | "-chardev stdio,id=id[,mux=on|off][,signal=on|off][,logfile=PATH][,logappend=on|off]\n" |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3553 | #endif |
| 3554 | #ifdef CONFIG_BRLAPI |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3555 | "-chardev braille,id=id[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3556 | #endif |
| 3557 | #if defined(__linux__) || defined(__sun__) || defined(__FreeBSD__) \ |
| 3558 | || defined(__NetBSD__) || defined(__OpenBSD__) || defined(__DragonFly__) |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3559 | "-chardev serial,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3560 | #endif |
| 3561 | #if defined(__linux__) || defined(__FreeBSD__) || defined(__DragonFly__) |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3562 | "-chardev parallel,id=id,path=path[,mux=on|off][,logfile=PATH][,logappend=on|off]\n" |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3563 | #endif |
Alon Levy | cbcc633 | 2011-01-19 10:49:50 +0200 | [diff] [blame] | 3564 | #if defined(CONFIG_SPICE) |
Daniel P. Berrange | d0d7708 | 2016-01-11 12:44:41 +0000 | [diff] [blame] | 3565 | "-chardev spicevmc,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" |
| 3566 | "-chardev spiceport,id=id,name=name[,debug=debug][,logfile=PATH][,logappend=on|off]\n" |
Alon Levy | cbcc633 | 2011-01-19 10:49:50 +0200 | [diff] [blame] | 3567 | #endif |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 3568 | , QEMU_ARCH_ALL |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3569 | ) |
| 3570 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3571 | SRST |
| 3572 | The general form of a character device option is: |
| 3573 | |
| 3574 | ``-chardev backend,id=id[,mux=on|off][,options]`` |
| 3575 | Backend is one of: ``null``, ``socket``, ``udp``, ``msmouse``, |
| 3576 | ``vc``, ``ringbuf``, ``file``, ``pipe``, ``console``, ``serial``, |
Paolo Bonzini | 6f9f630 | 2022-12-16 10:56:53 +0100 | [diff] [blame] | 3577 | ``pty``, ``stdio``, ``braille``, ``parallel``, |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3578 | ``spicevmc``, ``spiceport``. The specific backend will determine the |
| 3579 | applicable options. |
| 3580 | |
| 3581 | Use ``-chardev help`` to print all available chardev backend types. |
| 3582 | |
| 3583 | All devices must have an id, which can be any string up to 127 |
| 3584 | characters long. It is used to uniquely identify this device in |
| 3585 | other command line directives. |
| 3586 | |
| 3587 | A character device may be used in multiplexing mode by multiple |
| 3588 | front-ends. Specify ``mux=on`` to enable this mode. A multiplexer is |
| 3589 | a "1:N" device, and here the "1" end is your specified chardev |
| 3590 | backend, and the "N" end is the various parts of QEMU that can talk |
| 3591 | to a chardev. If you create a chardev with ``id=myid`` and |
| 3592 | ``mux=on``, QEMU will create a multiplexer with your specified ID, |
| 3593 | and you can then configure multiple front ends to use that chardev |
| 3594 | ID for their input/output. Up to four different front ends can be |
| 3595 | connected to a single multiplexed chardev. (Without multiplexing |
| 3596 | enabled, a chardev can only be used by a single front end.) For |
| 3597 | instance you could use this to allow a single stdio chardev to be |
| 3598 | used by two serial ports and the QEMU monitor: |
| 3599 | |
| 3600 | :: |
| 3601 | |
| 3602 | -chardev stdio,mux=on,id=char0 \ |
| 3603 | -mon chardev=char0,mode=readline \ |
| 3604 | -serial chardev:char0 \ |
| 3605 | -serial chardev:char0 |
| 3606 | |
| 3607 | You can have more than one multiplexer in a system configuration; |
| 3608 | for instance you could have a TCP port multiplexed between UART 0 |
| 3609 | and UART 1, and stdio multiplexed between the QEMU monitor and a |
| 3610 | parallel port: |
| 3611 | |
| 3612 | :: |
| 3613 | |
| 3614 | -chardev stdio,mux=on,id=char0 \ |
| 3615 | -mon chardev=char0,mode=readline \ |
| 3616 | -parallel chardev:char0 \ |
| 3617 | -chardev tcp,...,mux=on,id=char1 \ |
| 3618 | -serial chardev:char1 \ |
| 3619 | -serial chardev:char1 |
| 3620 | |
| 3621 | When you're using a multiplexed character device, some escape |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 3622 | sequences are interpreted in the input. See the chapter about |
| 3623 | :ref:`keys in the character backend multiplexer` in the |
| 3624 | System Emulation Users Guide for more details. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3625 | |
| 3626 | Note that some other command line options may implicitly create |
| 3627 | multiplexed character backends; for instance ``-serial mon:stdio`` |
| 3628 | creates a multiplexed stdio backend connected to the serial port and |
| 3629 | the QEMU monitor, and ``-nographic`` also multiplexes the console |
| 3630 | and the monitor to stdio. |
| 3631 | |
| 3632 | There is currently no support for multiplexing in the other |
| 3633 | direction (where a single QEMU front end takes input and output from |
| 3634 | multiple chardevs). |
| 3635 | |
| 3636 | Every backend supports the ``logfile`` option, which supplies the |
| 3637 | path to a file to record all data transmitted via the backend. The |
| 3638 | ``logappend`` option controls whether the log file will be truncated |
| 3639 | or appended to when opened. |
| 3640 | |
| 3641 | The available backends are: |
| 3642 | |
| 3643 | ``-chardev null,id=id`` |
| 3644 | A void device. This device will not emit any data, and will drop any |
| 3645 | data it receives. The null backend does not take any options. |
| 3646 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3647 | ``-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]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3648 | Create a two-way stream socket, which can be either a TCP or a unix |
| 3649 | socket. A unix socket will be created if ``path`` is specified. |
| 3650 | Behaviour is undefined if TCP options are specified for a unix |
| 3651 | socket. |
| 3652 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3653 | ``server=on|off`` specifies that the socket shall be a listening socket. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3654 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3655 | ``wait=on|off`` specifies that QEMU should not block waiting for a client |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3656 | to connect to a listening socket. |
| 3657 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3658 | ``telnet=on|off`` specifies that traffic on the socket should interpret |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3659 | telnet escape sequences. |
| 3660 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3661 | ``websocket=on|off`` specifies that the socket uses WebSocket protocol for |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3662 | communication. |
| 3663 | |
| 3664 | ``reconnect`` sets the timeout for reconnecting on non-server |
| 3665 | sockets when the remote end goes away. qemu will delay this many |
| 3666 | seconds and then attempt to reconnect. Zero disables reconnecting, |
| 3667 | and is the default. |
| 3668 | |
| 3669 | ``tls-creds`` requests enablement of the TLS protocol for |
| 3670 | encryption, and specifies the id of the TLS credentials to use for |
| 3671 | the handshake. The credentials must be previously created with the |
| 3672 | ``-object tls-creds`` argument. |
| 3673 | |
| 3674 | ``tls-auth`` provides the ID of the QAuthZ authorization object |
| 3675 | against which the client's x509 distinguished name will be |
| 3676 | validated. This object is only resolved at time of use, so can be |
| 3677 | deleted and recreated on the fly while the chardev server is active. |
| 3678 | If missing, it will default to denying access. |
| 3679 | |
| 3680 | TCP and unix socket options are given below: |
| 3681 | |
Paolo Bonzini | a9b1315 | 2021-02-25 11:47:52 +0100 | [diff] [blame] | 3682 | ``TCP options: port=port[,host=host][,to=to][,ipv4=on|off][,ipv6=on|off][,nodelay=on|off]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3683 | ``host`` for a listening socket specifies the local address to |
| 3684 | be bound. For a connecting socket species the remote host to |
| 3685 | connect to. ``host`` is optional for listening sockets. If not |
| 3686 | specified it defaults to ``0.0.0.0``. |
| 3687 | |
| 3688 | ``port`` for a listening socket specifies the local port to be |
| 3689 | bound. For a connecting socket specifies the port on the remote |
| 3690 | host to connect to. ``port`` can be given as either a port |
| 3691 | number or a service name. ``port`` is required. |
| 3692 | |
| 3693 | ``to`` is only relevant to listening sockets. If it is |
| 3694 | specified, and ``port`` cannot be bound, QEMU will attempt to |
| 3695 | bind to subsequent ports up to and including ``to`` until it |
| 3696 | succeeds. ``to`` must be specified as a port number. |
| 3697 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3698 | ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 |
| 3699 | or IPv6 must be used. If neither is specified the socket may |
| 3700 | use either protocol. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3701 | |
Paolo Bonzini | a9b1315 | 2021-02-25 11:47:52 +0100 | [diff] [blame] | 3702 | ``nodelay=on|off`` disables the Nagle algorithm. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3703 | |
xiaoqiang zhao | e339273 | 2020-05-16 11:13:27 +0800 | [diff] [blame] | 3704 | ``unix options: path=path[,abstract=on|off][,tight=on|off]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3705 | ``path`` specifies the local path of the unix socket. ``path`` |
| 3706 | is required. |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3707 | ``abstract=on|off`` specifies the use of the abstract socket namespace, |
xiaoqiang zhao | e339273 | 2020-05-16 11:13:27 +0800 | [diff] [blame] | 3708 | rather than the filesystem. Optional, defaults to false. |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3709 | ``tight=on|off`` sets the socket length of abstract sockets to their minimum, |
xiaoqiang zhao | e339273 | 2020-05-16 11:13:27 +0800 | [diff] [blame] | 3710 | rather than the full sun_path length. Optional, defaults to true. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3711 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3712 | ``-chardev udp,id=id[,host=host],port=port[,localaddr=localaddr][,localport=localport][,ipv4=on|off][,ipv6=on|off]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3713 | Sends all traffic from the guest to a remote host over UDP. |
| 3714 | |
| 3715 | ``host`` specifies the remote host to connect to. If not specified |
| 3716 | it defaults to ``localhost``. |
| 3717 | |
| 3718 | ``port`` specifies the port on the remote host to connect to. |
| 3719 | ``port`` is required. |
| 3720 | |
| 3721 | ``localaddr`` specifies the local address to bind to. If not |
| 3722 | specified it defaults to ``0.0.0.0``. |
| 3723 | |
| 3724 | ``localport`` specifies the local port to bind to. If not specified |
| 3725 | any available local port will be used. |
| 3726 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 3727 | ``ipv4=on|off`` and ``ipv6=on|off`` specify that either IPv4 or IPv6 must be used. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3728 | If neither is specified the device may use either protocol. |
| 3729 | |
| 3730 | ``-chardev msmouse,id=id`` |
| 3731 | Forward QEMU's emulated msmouse events to the guest. ``msmouse`` |
| 3732 | does not take any options. |
| 3733 | |
| 3734 | ``-chardev vc,id=id[[,width=width][,height=height]][[,cols=cols][,rows=rows]]`` |
| 3735 | Connect to a QEMU text console. ``vc`` may optionally be given a |
| 3736 | specific size. |
| 3737 | |
| 3738 | ``width`` and ``height`` specify the width and height respectively |
| 3739 | of the console, in pixels. |
| 3740 | |
| 3741 | ``cols`` and ``rows`` specify that the console be sized to fit a |
| 3742 | text console with the given dimensions. |
| 3743 | |
| 3744 | ``-chardev ringbuf,id=id[,size=size]`` |
| 3745 | Create a ring buffer with fixed size ``size``. size must be a power |
| 3746 | of two and defaults to ``64K``. |
| 3747 | |
Peter Maydell | 5b18a6b | 2023-04-13 16:07:24 +0100 | [diff] [blame] | 3748 | ``-chardev file,id=id,path=path[,input-path=input-path]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3749 | Log all traffic received from the guest to a file. |
| 3750 | |
| 3751 | ``path`` specifies the path of the file to be opened. This file will |
| 3752 | be created if it does not already exist, and overwritten if it does. |
| 3753 | ``path`` is required. |
| 3754 | |
Peter Maydell | 5b18a6b | 2023-04-13 16:07:24 +0100 | [diff] [blame] | 3755 | If ``input-path`` is specified, this is the path of a second file |
| 3756 | which will be used for input. If ``input-path`` is not specified, |
| 3757 | no input will be available from the chardev. |
| 3758 | |
| 3759 | Note that ``input-path`` is not supported on Windows hosts. |
| 3760 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3761 | ``-chardev pipe,id=id,path=path`` |
| 3762 | Create a two-way connection to the guest. The behaviour differs |
| 3763 | slightly between Windows hosts and other hosts: |
| 3764 | |
| 3765 | On Windows, a single duplex pipe will be created at |
| 3766 | ``\\.pipe\path``. |
| 3767 | |
| 3768 | On other hosts, 2 pipes will be created called ``path.in`` and |
| 3769 | ``path.out``. Data written to ``path.in`` will be received by the |
| 3770 | guest. Data written by the guest can be read from ``path.out``. QEMU |
| 3771 | will not create these fifos, and requires them to be present. |
| 3772 | |
| 3773 | ``path`` forms part of the pipe path as described above. ``path`` is |
| 3774 | required. |
| 3775 | |
| 3776 | ``-chardev console,id=id`` |
| 3777 | Send traffic from the guest to QEMU's standard output. ``console`` |
| 3778 | does not take any options. |
| 3779 | |
| 3780 | ``console`` is only available on Windows hosts. |
| 3781 | |
| 3782 | ``-chardev serial,id=id,path=path`` |
| 3783 | Send traffic from the guest to a serial device on the host. |
| 3784 | |
| 3785 | On Unix hosts serial will actually accept any tty device, not only |
| 3786 | serial lines. |
| 3787 | |
| 3788 | ``path`` specifies the name of the serial device to open. |
| 3789 | |
| 3790 | ``-chardev pty,id=id`` |
| 3791 | Create a new pseudo-terminal on the host and connect to it. ``pty`` |
| 3792 | does not take any options. |
| 3793 | |
| 3794 | ``pty`` is not available on Windows hosts. |
| 3795 | |
| 3796 | ``-chardev stdio,id=id[,signal=on|off]`` |
| 3797 | Connect to standard input and standard output of the QEMU process. |
| 3798 | |
| 3799 | ``signal`` controls if signals are enabled on the terminal, that |
| 3800 | includes exiting QEMU with the key sequence Control-c. This option |
| 3801 | is enabled by default, use ``signal=off`` to disable it. |
| 3802 | |
| 3803 | ``-chardev braille,id=id`` |
| 3804 | Connect to a local BrlAPI server. ``braille`` does not take any |
| 3805 | options. |
| 3806 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 3807 | ``-chardev parallel,id=id,path=path`` |
| 3808 | \ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3809 | ``parallel`` is only available on Linux, FreeBSD and DragonFlyBSD |
| 3810 | hosts. |
| 3811 | |
| 3812 | Connect to a local parallel port. |
| 3813 | |
| 3814 | ``path`` specifies the path to the parallel port device. ``path`` is |
| 3815 | required. |
| 3816 | |
| 3817 | ``-chardev spicevmc,id=id,debug=debug,name=name`` |
| 3818 | ``spicevmc`` is only available when spice support is built in. |
| 3819 | |
| 3820 | ``debug`` debug level for spicevmc |
| 3821 | |
| 3822 | ``name`` name of spice channel to connect to |
| 3823 | |
| 3824 | Connect to a spice virtual machine channel, such as vdiport. |
| 3825 | |
| 3826 | ``-chardev spiceport,id=id,debug=debug,name=name`` |
| 3827 | ``spiceport`` is only available when spice support is built in. |
| 3828 | |
| 3829 | ``debug`` debug level for spicevmc |
| 3830 | |
| 3831 | ``name`` name of spice port to connect to |
| 3832 | |
| 3833 | Connect to a spice port, allowing a Spice client to handle the |
| 3834 | traffic identified by a name (preferably a fqdn). |
| 3835 | ERST |
Matthew Booth | 7273a2d | 2009-10-30 13:41:12 +0000 | [diff] [blame] | 3836 | |
| 3837 | DEFHEADING() |
| 3838 | |
Stefan Berger | d1a0cf7 | 2013-02-27 12:47:49 -0500 | [diff] [blame] | 3839 | #ifdef CONFIG_TPM |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 3840 | DEFHEADING(TPM device options:) |
Stefan Berger | d1a0cf7 | 2013-02-27 12:47:49 -0500 | [diff] [blame] | 3841 | |
| 3842 | DEF("tpmdev", HAS_ARG, QEMU_OPTION_tpmdev, \ |
Stefan Berger | 92dcc23 | 2013-02-27 12:47:54 -0500 | [diff] [blame] | 3843 | "-tpmdev passthrough,id=id[,path=path][,cancel-path=path]\n" |
| 3844 | " use path to provide path to a character device; default is /dev/tpm0\n" |
| 3845 | " use cancel-path to provide path to TPM's cancel sysfs entry; if\n" |
Amarnath Valluri | f4ede81 | 2017-09-29 14:10:20 +0300 | [diff] [blame] | 3846 | " not provided it will be searched for in /sys/class/misc/tpm?/device\n" |
| 3847 | "-tpmdev emulator,id=id,chardev=dev\n" |
| 3848 | " configure the TPM device using chardev backend\n", |
Stefan Berger | d1a0cf7 | 2013-02-27 12:47:49 -0500 | [diff] [blame] | 3849 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3850 | SRST |
| 3851 | The general form of a TPM device option is: |
| 3852 | |
| 3853 | ``-tpmdev backend,id=id[,options]`` |
| 3854 | The specific backend type will determine the applicable options. The |
| 3855 | ``-tpmdev`` option creates the TPM backend and requires a |
| 3856 | ``-device`` option that specifies the TPM frontend interface model. |
| 3857 | |
| 3858 | Use ``-tpmdev help`` to print all available TPM backend types. |
| 3859 | |
| 3860 | The available backends are: |
| 3861 | |
| 3862 | ``-tpmdev passthrough,id=id,path=path,cancel-path=cancel-path`` |
| 3863 | (Linux-host only) Enable access to the host's TPM using the |
| 3864 | passthrough driver. |
| 3865 | |
| 3866 | ``path`` specifies the path to the host's TPM device, i.e., on a |
| 3867 | Linux host this would be ``/dev/tpm0``. ``path`` is optional and by |
| 3868 | default ``/dev/tpm0`` is used. |
| 3869 | |
| 3870 | ``cancel-path`` specifies the path to the host TPM device's sysfs |
| 3871 | entry allowing for cancellation of an ongoing TPM command. |
| 3872 | ``cancel-path`` is optional and by default QEMU will search for the |
| 3873 | sysfs entry to use. |
| 3874 | |
| 3875 | Some notes about using the host's TPM with the passthrough driver: |
| 3876 | |
| 3877 | The TPM device accessed by the passthrough driver must not be used |
| 3878 | by any other application on the host. |
| 3879 | |
| 3880 | Since the host's firmware (BIOS/UEFI) has already initialized the |
| 3881 | TPM, the VM's firmware (BIOS/UEFI) will not be able to initialize |
| 3882 | the TPM again and may therefore not show a TPM-specific menu that |
| 3883 | would otherwise allow the user to configure the TPM, e.g., allow the |
| 3884 | user to enable/disable or activate/deactivate the TPM. Further, if |
| 3885 | TPM ownership is released from within a VM then the host's TPM will |
| 3886 | get disabled and deactivated. To enable and activate the TPM again |
| 3887 | afterwards, the host has to be rebooted and the user is required to |
| 3888 | enter the firmware's menu to enable and activate the TPM. If the TPM |
| 3889 | is left disabled and/or deactivated most TPM commands will fail. |
| 3890 | |
| 3891 | To create a passthrough TPM use the following two options: |
| 3892 | |
| 3893 | :: |
| 3894 | |
| 3895 | -tpmdev passthrough,id=tpm0 -device tpm-tis,tpmdev=tpm0 |
| 3896 | |
| 3897 | Note that the ``-tpmdev`` id is ``tpm0`` and is referenced by |
| 3898 | ``tpmdev=tpm0`` in the device option. |
| 3899 | |
| 3900 | ``-tpmdev emulator,id=id,chardev=dev`` |
| 3901 | (Linux-host only) Enable access to a TPM emulator using Unix domain |
| 3902 | socket based chardev backend. |
| 3903 | |
| 3904 | ``chardev`` specifies the unique ID of a character device backend |
| 3905 | that provides connection to the software TPM server. |
| 3906 | |
| 3907 | To create a TPM emulator backend device with chardev socket backend: |
| 3908 | |
| 3909 | :: |
| 3910 | |
| 3911 | -chardev socket,id=chrtpm,path=/tmp/swtpm-sock -tpmdev emulator,id=tpm0,chardev=chrtpm -device tpm-tis,tpmdev=tpm0 |
| 3912 | ERST |
Stefan Berger | d1a0cf7 | 2013-02-27 12:47:49 -0500 | [diff] [blame] | 3913 | |
| 3914 | DEFHEADING() |
| 3915 | |
| 3916 | #endif |
| 3917 | |
Alex Bennée | 1235cf7 | 2022-07-25 15:05:20 +0100 | [diff] [blame] | 3918 | DEFHEADING(Boot Image or Kernel specific:) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3919 | SRST |
Alex Bennée | 1235cf7 | 2022-07-25 15:05:20 +0100 | [diff] [blame] | 3920 | There are broadly 4 ways you can boot a system with QEMU. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3921 | |
Alex Bennée | 1235cf7 | 2022-07-25 15:05:20 +0100 | [diff] [blame] | 3922 | - specify a firmware and let it control finding a kernel |
| 3923 | - specify a firmware and pass a hint to the kernel to boot |
| 3924 | - direct kernel image boot |
| 3925 | - manually load files into the guest's address space |
| 3926 | |
| 3927 | The third method is useful for quickly testing kernels but as there is |
| 3928 | no firmware to pass configuration information to the kernel the |
| 3929 | hardware must either be probeable, the kernel built for the exact |
| 3930 | configuration or passed some configuration data (e.g. a DTB blob) |
| 3931 | which tells the kernel what drivers it needs. This exact details are |
| 3932 | often hardware specific. |
| 3933 | |
| 3934 | The final method is the most generic way of loading images into the |
| 3935 | guest address space and used mostly for ``bare metal`` type |
| 3936 | development where the reset vectors of the processor are taken into |
| 3937 | account. |
| 3938 | |
| 3939 | ERST |
| 3940 | |
| 3941 | SRST |
| 3942 | |
| 3943 | For x86 machines and some other architectures ``-bios`` will generally |
| 3944 | do the right thing with whatever it is given. For other machines the |
| 3945 | more strict ``-pflash`` option needs an image that is sized for the |
| 3946 | flash device for the given machine type. |
| 3947 | |
| 3948 | Please see the :ref:`system-targets-ref` section of the manual for |
| 3949 | more detailed documentation. |
| 3950 | |
| 3951 | ERST |
| 3952 | |
| 3953 | DEF("bios", HAS_ARG, QEMU_OPTION_bios, \ |
| 3954 | "-bios file set the filename for the BIOS\n", QEMU_ARCH_ALL) |
| 3955 | SRST |
| 3956 | ``-bios file`` |
| 3957 | Set the filename for the BIOS. |
| 3958 | ERST |
| 3959 | |
| 3960 | DEF("pflash", HAS_ARG, QEMU_OPTION_pflash, |
| 3961 | "-pflash file use 'file' as a parallel flash image\n", QEMU_ARCH_ALL) |
| 3962 | SRST |
| 3963 | ``-pflash file`` |
| 3964 | Use file as a parallel flash image. |
| 3965 | ERST |
| 3966 | |
| 3967 | SRST |
| 3968 | |
| 3969 | The kernel options were designed to work with Linux kernels although |
| 3970 | other things (like hypervisors) can be packaged up as a kernel |
| 3971 | executable image. The exact format of a executable image is usually |
| 3972 | architecture specific. |
| 3973 | |
| 3974 | The way in which the kernel is started (what address it is loaded at, |
| 3975 | what if any information is passed to it via CPU registers, the state |
| 3976 | of the hardware when it is started, and so on) is also architecture |
| 3977 | specific. Typically it follows the specification laid down by the |
| 3978 | Linux kernel for how kernels for that architecture must be started. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3979 | |
| 3980 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 3981 | |
| 3982 | DEF("kernel", HAS_ARG, QEMU_OPTION_kernel, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 3983 | "-kernel bzImage use 'bzImage' as kernel image\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3984 | SRST |
| 3985 | ``-kernel bzImage`` |
| 3986 | Use bzImage as kernel image. The kernel can be either a Linux kernel |
| 3987 | or in multiboot format. |
| 3988 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 3989 | |
| 3990 | DEF("append", HAS_ARG, QEMU_OPTION_append, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 3991 | "-append cmdline use 'cmdline' as kernel command line\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 3992 | SRST |
| 3993 | ``-append cmdline`` |
| 3994 | Use cmdline as kernel command line |
| 3995 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 3996 | |
| 3997 | DEF("initrd", HAS_ARG, QEMU_OPTION_initrd, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 3998 | "-initrd file use 'file' as initial ram disk\n", QEMU_ARCH_ALL) |
David Woodhouse | 1eeb432 | 2024-01-30 19:01:43 +0000 | [diff] [blame] | 3999 | SRST(initrd) |
David Woodhouse | cc9d10b | 2023-10-19 15:30:23 +0100 | [diff] [blame] | 4000 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4001 | ``-initrd file`` |
| 4002 | Use file as initial ram disk. |
| 4003 | |
| 4004 | ``-initrd "file1 arg=foo,file2"`` |
| 4005 | This syntax is only available with multiboot. |
| 4006 | |
David Woodhouse | cc9d10b | 2023-10-19 15:30:23 +0100 | [diff] [blame] | 4007 | Use file1 and file2 as modules and pass ``arg=foo`` as parameter to the |
| 4008 | first module. Commas can be provided in module parameters by doubling |
| 4009 | them on the command line to escape them: |
| 4010 | |
| 4011 | ``-initrd "bzImage earlyprintk=xen,,keep root=/dev/xvda1,initrd.img"`` |
| 4012 | Multiboot only. Use bzImage as the first module with |
| 4013 | "``earlyprintk=xen,keep root=/dev/xvda1``" as its command line, |
| 4014 | and initrd.img as the second module. |
| 4015 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4016 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4017 | |
Grant Likely | 412beee | 2012-03-02 11:56:38 +0000 | [diff] [blame] | 4018 | DEF("dtb", HAS_ARG, QEMU_OPTION_dtb, \ |
Peter A. G. Crosthwaite | 379b5c7 | 2012-03-04 21:03:54 +1000 | [diff] [blame] | 4019 | "-dtb file use 'file' as device tree image\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4020 | SRST |
| 4021 | ``-dtb file`` |
| 4022 | Use file as a device tree binary (dtb) image and pass it to the |
| 4023 | kernel on boot. |
| 4024 | ERST |
Grant Likely | 412beee | 2012-03-02 11:56:38 +0000 | [diff] [blame] | 4025 | |
Alex Bennée | 1235cf7 | 2022-07-25 15:05:20 +0100 | [diff] [blame] | 4026 | SRST |
| 4027 | |
| 4028 | Finally you can also manually load images directly into the address |
| 4029 | space of the guest. This is most useful for developers who already |
| 4030 | know the layout of their guest and take care to ensure something sane |
| 4031 | will happen when the reset vector executes. |
| 4032 | |
| 4033 | The generic loader can be invoked by using the loader device: |
| 4034 | |
| 4035 | ``-device loader,addr=<addr>,data=<data>,data-len=<data-len>[,data-be=<data-be>][,cpu-num=<cpu-num>]`` |
| 4036 | |
| 4037 | there is also the guest loader which operates in a similar way but |
| 4038 | tweaks the DTB so a hypervisor loaded via ``-kernel`` can find where |
| 4039 | the guest image is: |
| 4040 | |
| 4041 | ``-device guest-loader,addr=<addr>[,kernel=<path>,[bootargs=<arguments>]][,initrd=<path>]`` |
| 4042 | |
| 4043 | ERST |
| 4044 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4045 | DEFHEADING() |
| 4046 | |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 4047 | DEFHEADING(Debug/Expert options:) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4048 | |
Markus Armbruster | 6dd7547 | 2021-03-18 16:55:10 +0100 | [diff] [blame] | 4049 | DEF("compat", HAS_ARG, QEMU_OPTION_compat, |
Markus Armbruster | dbb675c | 2021-03-18 16:55:19 +0100 | [diff] [blame] | 4050 | "-compat [deprecated-input=accept|reject|crash][,deprecated-output=accept|hide]\n" |
Markus Armbruster | 57df0df | 2021-10-28 12:25:20 +0200 | [diff] [blame] | 4051 | " Policy for handling deprecated management interfaces\n" |
| 4052 | "-compat [unstable-input=accept|reject|crash][,unstable-output=accept|hide]\n" |
| 4053 | " Policy for handling unstable management interfaces\n", |
Markus Armbruster | 6dd7547 | 2021-03-18 16:55:10 +0100 | [diff] [blame] | 4054 | QEMU_ARCH_ALL) |
| 4055 | SRST |
| 4056 | ``-compat [deprecated-input=@var{input-policy}][,deprecated-output=@var{output-policy}]`` |
| 4057 | Set policy for handling deprecated management interfaces (experimental): |
| 4058 | |
| 4059 | ``deprecated-input=accept`` (default) |
| 4060 | Accept deprecated commands and arguments |
| 4061 | ``deprecated-input=reject`` |
| 4062 | Reject deprecated commands and arguments |
Markus Armbruster | dbb675c | 2021-03-18 16:55:19 +0100 | [diff] [blame] | 4063 | ``deprecated-input=crash`` |
| 4064 | Crash on deprecated commands and arguments |
Markus Armbruster | 6dd7547 | 2021-03-18 16:55:10 +0100 | [diff] [blame] | 4065 | ``deprecated-output=accept`` (default) |
| 4066 | Emit deprecated command results and events |
| 4067 | ``deprecated-output=hide`` |
| 4068 | Suppress deprecated command results and events |
| 4069 | |
| 4070 | Limitation: covers only syntactic aspects of QMP. |
Markus Armbruster | 57df0df | 2021-10-28 12:25:20 +0200 | [diff] [blame] | 4071 | |
| 4072 | ``-compat [unstable-input=@var{input-policy}][,unstable-output=@var{output-policy}]`` |
| 4073 | Set policy for handling unstable management interfaces (experimental): |
| 4074 | |
| 4075 | ``unstable-input=accept`` (default) |
| 4076 | Accept unstable commands and arguments |
| 4077 | ``unstable-input=reject`` |
| 4078 | Reject unstable commands and arguments |
| 4079 | ``unstable-input=crash`` |
| 4080 | Crash on unstable commands and arguments |
| 4081 | ``unstable-output=accept`` (default) |
| 4082 | Emit unstable command results and events |
| 4083 | ``unstable-output=hide`` |
| 4084 | Suppress unstable command results and events |
| 4085 | |
| 4086 | Limitation: covers only syntactic aspects of QMP. |
Markus Armbruster | 6dd7547 | 2021-03-18 16:55:10 +0100 | [diff] [blame] | 4087 | ERST |
| 4088 | |
Gabriel L. Somlo | 81b2b81 | 2015-04-29 11:21:53 -0400 | [diff] [blame] | 4089 | DEF("fw_cfg", HAS_ARG, QEMU_OPTION_fwcfg, |
| 4090 | "-fw_cfg [name=]<name>,file=<file>\n" |
Markus Armbruster | 63d3145 | 2016-04-18 18:29:50 +0200 | [diff] [blame] | 4091 | " add named fw_cfg entry with contents from file\n" |
Gabriel L. Somlo | 6407d76 | 2015-09-29 12:29:01 -0400 | [diff] [blame] | 4092 | "-fw_cfg [name=]<name>,string=<str>\n" |
Markus Armbruster | 63d3145 | 2016-04-18 18:29:50 +0200 | [diff] [blame] | 4093 | " add named fw_cfg entry with contents from string\n", |
Gabriel L. Somlo | 81b2b81 | 2015-04-29 11:21:53 -0400 | [diff] [blame] | 4094 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4095 | SRST |
| 4096 | ``-fw_cfg [name=]name,file=file`` |
| 4097 | Add named fw\_cfg entry with contents from file file. |
Yihuan Pan | fd49b21 | 2023-12-13 22:17:07 +0800 | [diff] [blame] | 4098 | If the filename contains comma, you must double it (for instance, |
| 4099 | "file=my,,file" to use file "my,file"). |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4100 | |
| 4101 | ``-fw_cfg [name=]name,string=str`` |
| 4102 | Add named fw\_cfg entry with contents from string str. |
Yihuan Pan | fd49b21 | 2023-12-13 22:17:07 +0800 | [diff] [blame] | 4103 | If the string contains comma, you must double it (for instance, |
| 4104 | "string=my,,string" to use file "my,string"). |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4105 | |
| 4106 | The terminating NUL character of the contents of str will not be |
| 4107 | included as part of the fw\_cfg item data. To insert contents with |
| 4108 | embedded NUL characters, you have to use the file parameter. |
| 4109 | |
| 4110 | The fw\_cfg entries are passed by QEMU through to the guest. |
| 4111 | |
| 4112 | Example: |
| 4113 | |
| 4114 | :: |
| 4115 | |
| 4116 | -fw_cfg name=opt/com.mycompany/blob,file=./my_blob.bin |
| 4117 | |
| 4118 | creates an fw\_cfg entry named opt/com.mycompany/blob with contents |
| 4119 | from ./my\_blob.bin. |
| 4120 | ERST |
Gabriel L. Somlo | 81b2b81 | 2015-04-29 11:21:53 -0400 | [diff] [blame] | 4121 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4122 | DEF("serial", HAS_ARG, QEMU_OPTION_serial, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4123 | "-serial dev redirect the serial port to char device 'dev'\n", |
| 4124 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4125 | SRST |
| 4126 | ``-serial dev`` |
| 4127 | Redirect the virtual serial port to host character device dev. The |
| 4128 | default device is ``vc`` in graphical mode and ``stdio`` in non |
| 4129 | graphical mode. |
| 4130 | |
| 4131 | This option can be used several times to simulate up to 4 serial |
| 4132 | ports. |
| 4133 | |
Peter Maydell | 747bfaf | 2024-01-22 16:36:07 +0000 | [diff] [blame] | 4134 | You can use ``-serial none`` to suppress the creation of default |
| 4135 | serial devices. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4136 | |
| 4137 | Available character devices are: |
| 4138 | |
| 4139 | ``vc[:WxH]`` |
| 4140 | Virtual console. Optionally, a width and height can be given in |
| 4141 | pixel with |
| 4142 | |
| 4143 | :: |
| 4144 | |
| 4145 | vc:800x600 |
| 4146 | |
| 4147 | It is also possible to specify width or height in characters: |
| 4148 | |
| 4149 | :: |
| 4150 | |
| 4151 | vc:80Cx24C |
| 4152 | |
| 4153 | ``pty`` |
| 4154 | [Linux only] Pseudo TTY (a new PTY is automatically allocated) |
| 4155 | |
| 4156 | ``none`` |
Peter Maydell | 747bfaf | 2024-01-22 16:36:07 +0000 | [diff] [blame] | 4157 | No device is allocated. Note that for machine types which |
| 4158 | emulate systems where a serial device is always present in |
| 4159 | real hardware, this may be equivalent to the ``null`` option, |
| 4160 | in that the serial device is still present but all output |
| 4161 | is discarded. For boards where the number of serial ports is |
| 4162 | truly variable, this suppresses the creation of the device. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4163 | |
| 4164 | ``null`` |
Peter Maydell | 747bfaf | 2024-01-22 16:36:07 +0000 | [diff] [blame] | 4165 | A guest will see the UART or serial device as present in the |
| 4166 | machine, but all output is discarded, and there is no input. |
| 4167 | Conceptually equivalent to redirecting the output to ``/dev/null``. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4168 | |
| 4169 | ``chardev:id`` |
| 4170 | Use a named character device defined with the ``-chardev`` |
| 4171 | option. |
| 4172 | |
| 4173 | ``/dev/XXX`` |
| 4174 | [Linux only] Use host tty, e.g. ``/dev/ttyS0``. The host serial |
| 4175 | port parameters are set according to the emulated ones. |
| 4176 | |
| 4177 | ``/dev/parportN`` |
| 4178 | [Linux only, parallel port only] Use host parallel port N. |
| 4179 | Currently SPP and EPP parallel port features can be used. |
| 4180 | |
| 4181 | ``file:filename`` |
| 4182 | Write output to filename. No character can be read. |
| 4183 | |
| 4184 | ``stdio`` |
| 4185 | [Unix only] standard input/output |
| 4186 | |
| 4187 | ``pipe:filename`` |
| 4188 | name pipe filename |
| 4189 | |
| 4190 | ``COMn`` |
| 4191 | [Windows only] Use host serial port n |
| 4192 | |
| 4193 | ``udp:[remote_host]:remote_port[@[src_ip]:src_port]`` |
| 4194 | This implements UDP Net Console. When remote\_host or src\_ip |
| 4195 | are not specified they default to ``0.0.0.0``. When not using a |
| 4196 | specified src\_port a random port is automatically chosen. |
| 4197 | |
| 4198 | If you just want a simple readonly console you can use |
| 4199 | ``netcat`` or ``nc``, by starting QEMU with: |
| 4200 | ``-serial udp::4555`` and nc as: ``nc -u -l -p 4555``. Any time |
| 4201 | QEMU writes something to that port it will appear in the |
| 4202 | netconsole session. |
| 4203 | |
| 4204 | If you plan to send characters back via netconsole or you want |
| 4205 | to stop and start QEMU a lot of times, you should have QEMU use |
| 4206 | the same source port each time by using something like ``-serial |
| 4207 | udp::4555@:4556`` to QEMU. Another approach is to use a patched |
| 4208 | version of netcat which can listen to a TCP port and send and |
| 4209 | receive characters via udp. If you have a patched version of |
| 4210 | netcat which activates telnet remote echo and single char |
| 4211 | transfer, then you can use the following options to set up a |
| 4212 | netcat redirector to allow telnet on port 5555 to access the |
| 4213 | QEMU port. |
| 4214 | |
| 4215 | ``QEMU Options:`` |
| 4216 | -serial udp::4555@:4556 |
| 4217 | |
| 4218 | ``netcat options:`` |
| 4219 | -u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T |
| 4220 | |
| 4221 | ``telnet options:`` |
| 4222 | localhost 5555 |
| 4223 | |
Paolo Bonzini | a9b1315 | 2021-02-25 11:47:52 +0100 | [diff] [blame] | 4224 | ``tcp:[host]:port[,server=on|off][,wait=on|off][,nodelay=on|off][,reconnect=seconds]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4225 | The TCP Net Console has two modes of operation. It can send the |
| 4226 | serial I/O to a location or wait for a connection from a |
| 4227 | location. By default the TCP Net Console is sent to host at the |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 4228 | port. If you use the ``server=on`` option QEMU will wait for a client |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4229 | socket application to connect to the port before continuing, |
Paolo Bonzini | a9b1315 | 2021-02-25 11:47:52 +0100 | [diff] [blame] | 4230 | unless the ``wait=on|off`` option was specified. The ``nodelay=on|off`` |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 4231 | option disables the Nagle buffering algorithm. The ``reconnect=on`` |
| 4232 | option only applies if ``server=no`` is set, if the connection goes |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4233 | down it will attempt to reconnect at the given interval. If host |
| 4234 | is omitted, 0.0.0.0 is assumed. Only one TCP connection at a |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 4235 | time is accepted. You can use ``telnet=on`` to connect to the |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4236 | corresponding character device. |
| 4237 | |
| 4238 | ``Example to send tcp console to 192.168.0.2 port 4444`` |
| 4239 | -serial tcp:192.168.0.2:4444 |
| 4240 | |
| 4241 | ``Example to listen and wait on port 4444 for connection`` |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 4242 | -serial tcp::4444,server=on |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4243 | |
| 4244 | ``Example to not wait and listen on ip 192.168.0.100 port 4444`` |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 4245 | -serial tcp:192.168.0.100:4444,server=on,wait=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4246 | |
Paolo Bonzini | a9b1315 | 2021-02-25 11:47:52 +0100 | [diff] [blame] | 4247 | ``telnet:host:port[,server=on|off][,wait=on|off][,nodelay=on|off]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4248 | The telnet protocol is used instead of raw tcp sockets. The |
| 4249 | options work the same as if you had specified ``-serial tcp``. |
| 4250 | The difference is that the port acts like a telnet server or |
| 4251 | client using telnet option negotiation. This will also allow you |
| 4252 | to send the MAGIC\_SYSRQ sequence if you use a telnet that |
| 4253 | supports sending the break sequence. Typically in unix telnet |
| 4254 | you do it with Control-] and then type "send break" followed by |
| 4255 | pressing the enter key. |
| 4256 | |
Paolo Bonzini | a9b1315 | 2021-02-25 11:47:52 +0100 | [diff] [blame] | 4257 | ``websocket:host:port,server=on[,wait=on|off][,nodelay=on|off]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4258 | The WebSocket protocol is used instead of raw tcp socket. The |
| 4259 | port acts as a WebSocket server. Client mode is not supported. |
| 4260 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 4261 | ``unix:path[,server=on|off][,wait=on|off][,reconnect=seconds]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4262 | A unix domain socket is used instead of a tcp socket. The option |
| 4263 | works the same as if you had specified ``-serial tcp`` except |
| 4264 | the unix domain socket path is used for connections. |
| 4265 | |
| 4266 | ``mon:dev_string`` |
| 4267 | This is a special option to allow the monitor to be multiplexed |
| 4268 | onto another serial port. The monitor is accessed with key |
| 4269 | sequence of Control-a and then pressing c. dev\_string should be |
| 4270 | any one of the serial devices specified above. An example to |
| 4271 | multiplex the monitor onto a telnet server listening on port |
| 4272 | 4444 would be: |
| 4273 | |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 4274 | ``-serial mon:telnet::4444,server=on,wait=off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4275 | |
| 4276 | When the monitor is multiplexed to stdio in this way, Ctrl+C |
| 4277 | will not terminate QEMU any more but will be passed to the guest |
| 4278 | instead. |
| 4279 | |
| 4280 | ``braille`` |
| 4281 | Braille device. This will use BrlAPI to display the braille |
| 4282 | output on a real or fake device. |
| 4283 | |
| 4284 | ``msmouse`` |
| 4285 | Three button serial mouse. Configure the guest to use Microsoft |
| 4286 | protocol. |
| 4287 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4288 | |
| 4289 | DEF("parallel", HAS_ARG, QEMU_OPTION_parallel, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4290 | "-parallel dev redirect the parallel port to char device 'dev'\n", |
| 4291 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4292 | SRST |
| 4293 | ``-parallel dev`` |
| 4294 | Redirect the virtual parallel port to host device dev (same devices |
| 4295 | as the serial port). On Linux hosts, ``/dev/parportN`` can be used |
| 4296 | to use hardware devices connected on the corresponding host parallel |
| 4297 | port. |
| 4298 | |
| 4299 | This option can be used several times to simulate up to 3 parallel |
| 4300 | ports. |
| 4301 | |
| 4302 | Use ``-parallel none`` to disable all parallel ports. |
| 4303 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4304 | |
| 4305 | DEF("monitor", HAS_ARG, QEMU_OPTION_monitor, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4306 | "-monitor dev redirect the monitor to char device 'dev'\n", |
| 4307 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4308 | SRST |
| 4309 | ``-monitor dev`` |
| 4310 | Redirect the monitor to host device dev (same devices as the serial |
| 4311 | port). The default device is ``vc`` in graphical mode and ``stdio`` |
| 4312 | in non graphical mode. Use ``-monitor none`` to disable the default |
| 4313 | monitor. |
| 4314 | ERST |
Gerd Hoffmann | 6ca5582 | 2009-12-08 13:11:52 +0100 | [diff] [blame] | 4315 | DEF("qmp", HAS_ARG, QEMU_OPTION_qmp, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4316 | "-qmp dev like -monitor but opens in 'control' mode\n", |
| 4317 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4318 | SRST |
| 4319 | ``-qmp dev`` |
Peter Maydell | 0ec4468 | 2023-05-15 17:22:45 +0100 | [diff] [blame] | 4320 | Like ``-monitor`` but opens in 'control' mode. For example, to make |
| 4321 | QMP available on localhost port 4444:: |
| 4322 | |
| 4323 | -qmp tcp:localhost:4444,server=on,wait=off |
| 4324 | |
| 4325 | Not all options are configurable via this syntax; for maximum |
| 4326 | flexibility use the ``-mon`` option and an accompanying ``-chardev``. |
| 4327 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4328 | ERST |
Max Reitz | 4821cd4 | 2014-11-17 13:31:04 +0100 | [diff] [blame] | 4329 | DEF("qmp-pretty", HAS_ARG, QEMU_OPTION_qmp_pretty, \ |
| 4330 | "-qmp-pretty dev like -qmp but uses pretty JSON formatting\n", |
| 4331 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4332 | SRST |
| 4333 | ``-qmp-pretty dev`` |
Peter Maydell | 0ec4468 | 2023-05-15 17:22:45 +0100 | [diff] [blame] | 4334 | Like ``-qmp`` but uses pretty JSON formatting. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4335 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4336 | |
Gerd Hoffmann | 22a0e04 | 2009-12-08 13:11:51 +0100 | [diff] [blame] | 4337 | DEF("mon", HAS_ARG, QEMU_OPTION_mon, \ |
Vicente Jimenez Aguilar | ef67072 | 2017-11-14 09:11:27 +0100 | [diff] [blame] | 4338 | "-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4339 | SRST |
| 4340 | ``-mon [chardev=]name[,mode=readline|control][,pretty[=on|off]]`` |
Peter Maydell | 0ec4468 | 2023-05-15 17:22:45 +0100 | [diff] [blame] | 4341 | Set up a monitor connected to the chardev ``name``. |
| 4342 | QEMU supports two monitors: the Human Monitor Protocol |
| 4343 | (HMP; for human interaction), and the QEMU Monitor Protocol |
| 4344 | (QMP; a JSON RPC-style protocol). |
| 4345 | The default is HMP; ``mode=control`` selects QMP instead. |
| 4346 | ``pretty`` is only valid when ``mode=control``, |
Ali Shirvani | 16b3f3b | 2021-05-19 11:41:45 +0430 | [diff] [blame] | 4347 | turning on JSON pretty printing to ease |
Daniel P. Berrangé | 283d845 | 2021-02-19 17:56:13 +0000 | [diff] [blame] | 4348 | human reading and debugging. |
Peter Maydell | 0ec4468 | 2023-05-15 17:22:45 +0100 | [diff] [blame] | 4349 | |
| 4350 | For example:: |
| 4351 | |
| 4352 | -chardev socket,id=mon1,host=localhost,port=4444,server=on,wait=off \ |
| 4353 | -mon chardev=mon1,mode=control,pretty=on |
| 4354 | |
| 4355 | enables the QMP monitor on localhost port 4444 with pretty-printing. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4356 | ERST |
Gerd Hoffmann | 22a0e04 | 2009-12-08 13:11:51 +0100 | [diff] [blame] | 4357 | |
H. Peter Anvin | c9f398e | 2009-12-29 13:51:36 -0800 | [diff] [blame] | 4358 | DEF("debugcon", HAS_ARG, QEMU_OPTION_debugcon, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4359 | "-debugcon dev redirect the debug console to char device 'dev'\n", |
| 4360 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4361 | SRST |
| 4362 | ``-debugcon dev`` |
| 4363 | Redirect the debug console to host device dev (same devices as the |
| 4364 | serial port). The debug console is an I/O port which is typically |
| 4365 | port 0xe9; writing to that I/O port sends output to this device. The |
| 4366 | default device is ``vc`` in graphical mode and ``stdio`` in non |
| 4367 | graphical mode. |
| 4368 | ERST |
H. Peter Anvin | c9f398e | 2009-12-29 13:51:36 -0800 | [diff] [blame] | 4369 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4370 | DEF("pidfile", HAS_ARG, QEMU_OPTION_pidfile, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4371 | "-pidfile file write PID to 'file'\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4372 | SRST |
| 4373 | ``-pidfile file`` |
| 4374 | Store the QEMU process PID in file. It is useful if you launch QEMU |
| 4375 | from a script. |
| 4376 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4377 | |
Igor Mammedov | 047f703 | 2018-05-11 19:24:43 +0200 | [diff] [blame] | 4378 | DEF("preconfig", 0, QEMU_OPTION_preconfig, \ |
Markus Armbruster | 361ac94 | 2018-07-05 11:14:02 +0200 | [diff] [blame] | 4379 | "--preconfig pause QEMU before machine is initialized (experimental)\n", |
Igor Mammedov | 047f703 | 2018-05-11 19:24:43 +0200 | [diff] [blame] | 4380 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4381 | SRST |
| 4382 | ``--preconfig`` |
| 4383 | Pause QEMU for interactive configuration before the machine is |
| 4384 | created, which allows querying and configuring properties that will |
| 4385 | affect machine initialization. Use QMP command 'x-exit-preconfig' to |
| 4386 | exit the preconfig state and move to the next state (i.e. run guest |
| 4387 | if -S isn't used or pause the second time if -S is used). This |
| 4388 | option is experimental. |
| 4389 | ERST |
Igor Mammedov | 047f703 | 2018-05-11 19:24:43 +0200 | [diff] [blame] | 4390 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4391 | DEF("S", 0, QEMU_OPTION_S, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4392 | "-S freeze CPU at startup (use 'c' to start execution)\n", |
| 4393 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4394 | SRST |
| 4395 | ``-S`` |
| 4396 | Do not start CPU at startup (you must type 'c' in the monitor). |
| 4397 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4398 | |
Michael S. Tsirkin | 6f131f1 | 2018-06-22 22:22:05 +0300 | [diff] [blame] | 4399 | DEF("overcommit", HAS_ARG, QEMU_OPTION_overcommit, |
BALATON Zoltan | dfaa7d5 | 2018-07-16 21:12:08 +0200 | [diff] [blame] | 4400 | "-overcommit [mem-lock=on|off][cpu-pm=on|off]\n" |
Michael S. Tsirkin | 6f131f1 | 2018-06-22 22:22:05 +0300 | [diff] [blame] | 4401 | " run qemu with overcommit hints\n" |
| 4402 | " mem-lock=on|off controls memory lock support (default: off)\n" |
| 4403 | " cpu-pm=on|off controls cpu power management (default: off)\n", |
| 4404 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4405 | SRST |
| 4406 | ``-overcommit mem-lock=on|off`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 4407 | \ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4408 | ``-overcommit cpu-pm=on|off`` |
| 4409 | Run qemu with hints about host resource overcommit. The default is |
| 4410 | to assume that host overcommits all resources. |
| 4411 | |
| 4412 | Locking qemu and guest memory can be enabled via ``mem-lock=on`` |
| 4413 | (disabled by default). This works when host memory is not |
Thomas Huth | c8c9dc4 | 2020-12-10 16:58:07 +0100 | [diff] [blame] | 4414 | overcommitted and reduces the worst-case latency for guest. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4415 | |
| 4416 | Guest ability to manage power state of host cpus (increasing latency |
| 4417 | for other processes on the same host cpu, but decreasing latency for |
| 4418 | guest) can be enabled via ``cpu-pm=on`` (disabled by default). This |
| 4419 | works best when host CPU is not overcommitted. When used, host |
| 4420 | estimates of CPU cycle and power utilization will be incorrect, not |
| 4421 | taking into account guest idle time. |
| 4422 | ERST |
Michael S. Tsirkin | 6f131f1 | 2018-06-22 22:22:05 +0300 | [diff] [blame] | 4423 | |
aliguori | 59030a8 | 2009-04-05 18:43:41 +0000 | [diff] [blame] | 4424 | DEF("gdb", HAS_ARG, QEMU_OPTION_gdb, \ |
Peter Maydell | e5910d4 | 2020-04-03 10:40:14 +0100 | [diff] [blame] | 4425 | "-gdb dev accept gdb connection on 'dev'. (QEMU defaults to starting\n" |
| 4426 | " the guest without waiting for gdb to connect; use -S too\n" |
| 4427 | " if you want it to not start execution.)\n", |
| 4428 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4429 | SRST |
| 4430 | ``-gdb dev`` |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 4431 | Accept a gdb connection on device dev (see the :ref:`GDB usage` chapter |
| 4432 | in the System Emulation Users Guide). Note that this option does not pause QEMU |
Peter Maydell | e5910d4 | 2020-04-03 10:40:14 +0100 | [diff] [blame] | 4433 | execution -- if you want QEMU to not start the guest until you |
| 4434 | connect with gdb and issue a ``continue`` command, you will need to |
| 4435 | also pass the ``-S`` option to QEMU. |
| 4436 | |
| 4437 | The most usual configuration is to listen on a local TCP socket:: |
| 4438 | |
| 4439 | -gdb tcp::3117 |
| 4440 | |
| 4441 | but you can specify other backends; UDP, pseudo TTY, or even stdio |
| 4442 | are all reasonable use cases. For example, a stdio connection |
| 4443 | allows you to start QEMU from within gdb and establish the |
| 4444 | connection via a pipe: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4445 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 4446 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4447 | |
| 4448 | (gdb) target remote | exec |qemu_system| -gdb stdio ... |
| 4449 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4450 | |
aliguori | 59030a8 | 2009-04-05 18:43:41 +0000 | [diff] [blame] | 4451 | DEF("s", 0, QEMU_OPTION_s, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4452 | "-s shorthand for -gdb tcp::" DEFAULT_GDBSTUB_PORT "\n", |
| 4453 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4454 | SRST |
| 4455 | ``-s`` |
| 4456 | Shorthand for -gdb tcp::1234, i.e. open a gdbserver on TCP port 1234 |
Thomas Huth | 923e931 | 2020-11-16 15:47:36 +0100 | [diff] [blame] | 4457 | (see the :ref:`GDB usage` chapter in the System Emulation Users Guide). |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4458 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4459 | |
| 4460 | DEF("d", HAS_ARG, QEMU_OPTION_d, \ |
Peter Maydell | 989b697 | 2013-02-26 17:52:40 +0000 | [diff] [blame] | 4461 | "-d item1,... enable logging of specified items (use '-d help' for a list of log items)\n", |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4462 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4463 | SRST |
| 4464 | ``-d item1[,...]`` |
| 4465 | Enable logging of specified items. Use '-d help' for a list of log |
| 4466 | items. |
| 4467 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4468 | |
Matthew Fernandez | c235d73 | 2011-06-07 16:32:40 +0000 | [diff] [blame] | 4469 | DEF("D", HAS_ARG, QEMU_OPTION_D, \ |
Peter Maydell | 989b697 | 2013-02-26 17:52:40 +0000 | [diff] [blame] | 4470 | "-D logfile output log to logfile (default stderr)\n", |
Matthew Fernandez | c235d73 | 2011-06-07 16:32:40 +0000 | [diff] [blame] | 4471 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4472 | SRST |
| 4473 | ``-D logfile`` |
| 4474 | Output log in logfile instead of to stderr |
| 4475 | ERST |
Matthew Fernandez | c235d73 | 2011-06-07 16:32:40 +0000 | [diff] [blame] | 4476 | |
Alex Bennée | 3514552 | 2016-03-15 14:30:20 +0000 | [diff] [blame] | 4477 | DEF("dfilter", HAS_ARG, QEMU_OPTION_DFILTER, \ |
| 4478 | "-dfilter range,.. filter debug output to range of addresses (useful for -d cpu,exec,etc..)\n", |
| 4479 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4480 | SRST |
| 4481 | ``-dfilter range1[,...]`` |
| 4482 | Filter debug output to that relevant to a range of target addresses. |
| 4483 | The filter spec can be either start+size, start-size or start..end |
| 4484 | where start end and size are the addresses and sizes required. For |
| 4485 | example: |
| 4486 | |
| 4487 | :: |
| 4488 | |
| 4489 | -dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,0xffffffc000060000-0x1000 |
| 4490 | |
| 4491 | Will dump output for any code in the 0x1000 sized block starting at |
| 4492 | 0x8000 and the 0x200 sized block starting at 0xffffffc000080000 and |
| 4493 | another 0x1000 sized block starting at 0xffffffc00005f000. |
| 4494 | ERST |
Alex Bennée | 3514552 | 2016-03-15 14:30:20 +0000 | [diff] [blame] | 4495 | |
Richard Henderson | 9c09a25 | 2019-03-14 13:06:29 -0700 | [diff] [blame] | 4496 | DEF("seed", HAS_ARG, QEMU_OPTION_seed, \ |
| 4497 | "-seed number seed the pseudo-random number generator\n", |
| 4498 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4499 | SRST |
| 4500 | ``-seed number`` |
| 4501 | Force the guest to use a deterministic pseudo-random number |
| 4502 | generator, seeded with number. This does not affect crypto routines |
| 4503 | within the host. |
| 4504 | ERST |
Richard Henderson | 9c09a25 | 2019-03-14 13:06:29 -0700 | [diff] [blame] | 4505 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4506 | DEF("L", HAS_ARG, QEMU_OPTION_L, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4507 | "-L path set the directory for the BIOS, VGA BIOS and keymaps\n", |
| 4508 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4509 | SRST |
| 4510 | ``-L path`` |
| 4511 | Set the directory for the BIOS, VGA BIOS and keymaps. |
| 4512 | |
| 4513 | To list all the data directories, use ``-L help``. |
| 4514 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4515 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4516 | DEF("enable-kvm", 0, QEMU_OPTION_enable_kvm, \ |
Thomas Huth | 21abf01 | 2022-04-27 15:49:06 +0200 | [diff] [blame] | 4517 | "-enable-kvm enable KVM full virtualization support\n", |
| 4518 | QEMU_ARCH_ARM | QEMU_ARCH_I386 | QEMU_ARCH_MIPS | QEMU_ARCH_PPC | |
| 4519 | QEMU_ARCH_RISCV | QEMU_ARCH_S390X) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4520 | SRST |
| 4521 | ``-enable-kvm`` |
| 4522 | Enable KVM full virtualization support. This option is only |
| 4523 | available if KVM support is enabled when compiling. |
| 4524 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4525 | |
aliguori | e37630c | 2009-04-22 15:19:10 +0000 | [diff] [blame] | 4526 | DEF("xen-domid", HAS_ARG, QEMU_OPTION_xen_domid, |
Thomas Huth | eeb3647 | 2022-04-27 15:31:56 +0200 | [diff] [blame] | 4527 | "-xen-domid id specify xen guest domain id\n", |
| 4528 | QEMU_ARCH_ARM | QEMU_ARCH_I386) |
aliguori | e37630c | 2009-04-22 15:19:10 +0000 | [diff] [blame] | 4529 | DEF("xen-attach", 0, QEMU_OPTION_xen_attach, |
| 4530 | "-xen-attach attach to existing xen domain\n" |
Anthony PERARD | 1077bca | 2018-09-14 12:18:30 +0100 | [diff] [blame] | 4531 | " libxl will use this when starting QEMU\n", |
Thomas Huth | eeb3647 | 2022-04-27 15:31:56 +0200 | [diff] [blame] | 4532 | QEMU_ARCH_ARM | QEMU_ARCH_I386) |
Paul Durrant | 1c59947 | 2017-03-22 09:39:15 +0000 | [diff] [blame] | 4533 | DEF("xen-domid-restrict", 0, QEMU_OPTION_xen_domid_restrict, |
| 4534 | "-xen-domid-restrict restrict set of available xen operations\n" |
| 4535 | " to specified domain id. (Does not affect\n" |
| 4536 | " xenpv machine type).\n", |
Thomas Huth | eeb3647 | 2022-04-27 15:31:56 +0200 | [diff] [blame] | 4537 | QEMU_ARCH_ARM | QEMU_ARCH_I386) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4538 | SRST |
| 4539 | ``-xen-domid id`` |
| 4540 | Specify xen guest domain id (XEN only). |
| 4541 | |
| 4542 | ``-xen-attach`` |
| 4543 | Attach to existing xen domain. libxl will use this when starting |
| 4544 | QEMU (XEN only). Restrict set of available xen operations to |
| 4545 | specified domain id (XEN only). |
| 4546 | ERST |
aliguori | e37630c | 2009-04-22 15:19:10 +0000 | [diff] [blame] | 4547 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4548 | DEF("no-reboot", 0, QEMU_OPTION_no_reboot, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4549 | "-no-reboot exit instead of rebooting\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4550 | SRST |
| 4551 | ``-no-reboot`` |
| 4552 | Exit instead of rebooting. |
| 4553 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4554 | |
| 4555 | DEF("no-shutdown", 0, QEMU_OPTION_no_shutdown, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4556 | "-no-shutdown stop before shutdown\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4557 | SRST |
| 4558 | ``-no-shutdown`` |
| 4559 | Don't exit QEMU on guest shutdown, but instead only stop the |
| 4560 | emulation. This allows for instance switching to monitor to commit |
| 4561 | changes to the disk image. |
| 4562 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4563 | |
Alejandro Jimenez | 2a5ad60 | 2020-12-11 11:52:41 -0500 | [diff] [blame] | 4564 | DEF("action", HAS_ARG, QEMU_OPTION_action, |
Paolo Bonzini | c27025e | 2021-01-20 14:30:27 +0100 | [diff] [blame] | 4565 | "-action reboot=reset|shutdown\n" |
| 4566 | " action when guest reboots [default=reset]\n" |
Alejandro Jimenez | 2a5ad60 | 2020-12-11 11:52:41 -0500 | [diff] [blame] | 4567 | "-action shutdown=poweroff|pause\n" |
| 4568 | " action when guest shuts down [default=poweroff]\n" |
Ilya Leoshkevich | 0882caf | 2022-07-26 00:37:45 +0200 | [diff] [blame] | 4569 | "-action panic=pause|shutdown|exit-failure|none\n" |
Paolo Bonzini | c27025e | 2021-01-20 14:30:27 +0100 | [diff] [blame] | 4570 | " action when guest panics [default=shutdown]\n" |
Alejandro Jimenez | 2a5ad60 | 2020-12-11 11:52:41 -0500 | [diff] [blame] | 4571 | "-action watchdog=reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" |
| 4572 | " action when watchdog fires [default=reset]\n", |
| 4573 | QEMU_ARCH_ALL) |
| 4574 | SRST |
| 4575 | ``-action event=action`` |
| 4576 | The action parameter serves to modify QEMU's default behavior when |
| 4577 | certain guest events occur. It provides a generic method for specifying the |
| 4578 | same behaviors that are modified by the ``-no-reboot`` and ``-no-shutdown`` |
| 4579 | parameters. |
| 4580 | |
| 4581 | Examples: |
| 4582 | |
Alejandro Jimenez | c753e8e | 2020-12-11 17:31:52 -0500 | [diff] [blame] | 4583 | ``-action panic=none`` |
Alejandro Jimenez | 2a5ad60 | 2020-12-11 11:52:41 -0500 | [diff] [blame] | 4584 | ``-action reboot=shutdown,shutdown=pause`` |
Paolo Bonzini | 5433af7 | 2022-09-10 13:44:47 +0200 | [diff] [blame] | 4585 | ``-device i6300esb -action watchdog=pause`` |
Alejandro Jimenez | 2a5ad60 | 2020-12-11 11:52:41 -0500 | [diff] [blame] | 4586 | |
| 4587 | ERST |
| 4588 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4589 | DEF("loadvm", HAS_ARG, QEMU_OPTION_loadvm, \ |
| 4590 | "-loadvm [tag|id]\n" \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4591 | " start right away with a saved state (loadvm in monitor)\n", |
| 4592 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4593 | SRST |
| 4594 | ``-loadvm file`` |
| 4595 | Start right away with a saved state (``loadvm`` in monitor) |
| 4596 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4597 | |
| 4598 | #ifndef _WIN32 |
| 4599 | DEF("daemonize", 0, QEMU_OPTION_daemonize, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4600 | "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4601 | #endif |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4602 | SRST |
| 4603 | ``-daemonize`` |
| 4604 | Daemonize the QEMU process after initialization. QEMU will not |
| 4605 | detach from standard IO until it is ready to receive connections on |
| 4606 | any of its devices. This option is a useful way for external |
| 4607 | programs to launch QEMU without having to cope with initialization |
| 4608 | race conditions. |
| 4609 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4610 | |
| 4611 | DEF("option-rom", HAS_ARG, QEMU_OPTION_option_rom, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4612 | "-option-rom rom load a file, rom, into the option ROM space\n", |
| 4613 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4614 | SRST |
| 4615 | ``-option-rom file`` |
| 4616 | Load the contents of file as an option ROM. This option is useful to |
| 4617 | load things like EtherBoot. |
| 4618 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4619 | |
Jan Kiszka | 1ed2fc1 | 2009-09-15 13:36:04 +0200 | [diff] [blame] | 4620 | DEF("rtc", HAS_ARG, QEMU_OPTION_rtc, \ |
Artem Pisarenko | 238d124 | 2018-10-18 13:12:52 +0600 | [diff] [blame] | 4621 | "-rtc [base=utc|localtime|<datetime>][,clock=host|rt|vm][,driftfix=none|slew]\n" \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4622 | " set the RTC base and clock, enable drift fix for clock ticks (x86 only)\n", |
| 4623 | QEMU_ARCH_ALL) |
Jan Kiszka | 1ed2fc1 | 2009-09-15 13:36:04 +0200 | [diff] [blame] | 4624 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4625 | SRST |
| 4626 | ``-rtc [base=utc|localtime|datetime][,clock=host|rt|vm][,driftfix=none|slew]`` |
| 4627 | Specify ``base`` as ``utc`` or ``localtime`` to let the RTC start at |
| 4628 | the current UTC or local time, respectively. ``localtime`` is |
| 4629 | required for correct date in MS-DOS or Windows. To start at a |
| 4630 | specific point in time, provide datetime in the format |
| 4631 | ``2006-06-17T16:01:21`` or ``2006-06-17``. The default base is UTC. |
| 4632 | |
| 4633 | By default the RTC is driven by the host system time. This allows |
| 4634 | using of the RTC as accurate reference clock inside the guest, |
| 4635 | specifically if the host time is smoothly following an accurate |
| 4636 | external reference clock, e.g. via NTP. If you want to isolate the |
| 4637 | guest time from the host, you can set ``clock`` to ``rt`` instead, |
| 4638 | which provides a host monotonic clock if host support it. To even |
| 4639 | prevent the RTC from progressing during suspension, you can set |
| 4640 | ``clock`` to ``vm`` (virtual clock). '\ ``clock=vm``\ ' is |
| 4641 | recommended especially in icount mode in order to preserve |
| 4642 | determinism; however, note that in icount mode the speed of the |
| 4643 | virtual clock is variable and can in general differ from the host |
| 4644 | clock. |
| 4645 | |
| 4646 | Enable ``driftfix`` (i386 targets only) if you experience time drift |
| 4647 | problems, specifically with Windows' ACPI HAL. This option will try |
| 4648 | to figure out how many timer interrupts were not processed by the |
| 4649 | Windows guest and will re-inject them. |
| 4650 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4651 | |
| 4652 | DEF("icount", HAS_ARG, QEMU_OPTION_icount, \ |
Peter Maydell | fa64790 | 2020-11-21 21:35:06 +0000 | [diff] [blame] | 4653 | "-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=<filename>[,rrsnapshot=<snapshot>]]\n" \ |
aliguori | bc14ca2 | 2009-04-05 18:43:37 +0000 | [diff] [blame] | 4654 | " enable virtual instruction counter with 2^N clock ticks per\n" \ |
Victor CLEMENT | f1f4b57 | 2015-05-29 17:14:05 +0200 | [diff] [blame] | 4655 | " instruction, enable aligning the host and virtual clocks\n" \ |
Peter Maydell | fa64790 | 2020-11-21 21:35:06 +0000 | [diff] [blame] | 4656 | " or disable real time cpu sleeping, and optionally enable\n" \ |
| 4657 | " record-and-replay mode\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4658 | SRST |
Peter Maydell | fa64790 | 2020-11-21 21:35:06 +0000 | [diff] [blame] | 4659 | ``-icount [shift=N|auto][,align=on|off][,sleep=on|off][,rr=record|replay,rrfile=filename[,rrsnapshot=snapshot]]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4660 | Enable virtual instruction counter. The virtual cpu will execute one |
| 4661 | instruction every 2^N ns of virtual time. If ``auto`` is specified |
| 4662 | then the virtual cpu speed will be automatically adjusted to keep |
| 4663 | virtual time within a few seconds of real time. |
| 4664 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4665 | Note that while this option can give deterministic behavior, it does |
| 4666 | not provide cycle accurate emulation. Modern CPUs contain |
| 4667 | superscalar out of order cores with complex cache hierarchies. The |
| 4668 | number of instructions executed often has little or no correlation |
| 4669 | with actual performance. |
| 4670 | |
Peter Maydell | fa64790 | 2020-11-21 21:35:06 +0000 | [diff] [blame] | 4671 | When the virtual cpu is sleeping, the virtual time will advance at |
| 4672 | default speed unless ``sleep=on`` is specified. With |
| 4673 | ``sleep=on``, the virtual time will jump to the next timer |
| 4674 | deadline instantly whenever the virtual cpu goes to sleep mode and |
| 4675 | will not advance if no timer is enabled. This behavior gives |
| 4676 | deterministic execution times from the guest point of view. |
| 4677 | The default if icount is enabled is ``sleep=off``. |
| 4678 | ``sleep=on`` cannot be used together with either ``shift=auto`` |
| 4679 | or ``align=on``. |
| 4680 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4681 | ``align=on`` will activate the delay algorithm which will try to |
| 4682 | synchronise the host clock and the virtual clock. The goal is to |
| 4683 | have a guest running at the real frequency imposed by the shift |
| 4684 | option. Whenever the guest clock is behind the host clock and if |
| 4685 | ``align=on`` is specified then we print a message to the user to |
| 4686 | inform about the delay. Currently this option does not work when |
| 4687 | ``shift`` is ``auto``. Note: The sync algorithm will work for those |
| 4688 | shift values for which the guest clock runs ahead of the host clock. |
| 4689 | Typically this happens when the shift value is high (how high |
Peter Maydell | fa64790 | 2020-11-21 21:35:06 +0000 | [diff] [blame] | 4690 | depends on the host machine). The default if icount is enabled |
| 4691 | is ``align=off``. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4692 | |
Peter Maydell | fa64790 | 2020-11-21 21:35:06 +0000 | [diff] [blame] | 4693 | When the ``rr`` option is specified deterministic record/replay is |
| 4694 | enabled. The ``rrfile=`` option must also be provided to |
| 4695 | specify the path to the replay log. In record mode data is written |
| 4696 | to this file, and in replay mode it is read back. |
| 4697 | If the ``rrsnapshot`` option is given then it specifies a VM snapshot |
| 4698 | name. In record mode, a new VM snapshot with the given name is created |
| 4699 | at the start of execution recording. In replay mode this option |
| 4700 | specifies the snapshot name used to load the initial VM state. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4701 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4702 | |
Richard W.M. Jones | 9dd986c | 2009-04-25 13:56:19 +0100 | [diff] [blame] | 4703 | DEF("watchdog-action", HAS_ARG, QEMU_OPTION_watchdog_action, \ |
Markus Armbruster | 7ad9270 | 2017-10-02 16:03:07 +0200 | [diff] [blame] | 4704 | "-watchdog-action reset|shutdown|poweroff|inject-nmi|pause|debug|none\n" \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4705 | " action when watchdog fires [default=reset]\n", |
| 4706 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4707 | SRST |
| 4708 | ``-watchdog-action action`` |
| 4709 | The action controls what QEMU will do when the watchdog timer |
| 4710 | expires. The default is ``reset`` (forcefully reset the guest). |
| 4711 | Other possible actions are: ``shutdown`` (attempt to gracefully |
| 4712 | shutdown the guest), ``poweroff`` (forcefully poweroff the guest), |
| 4713 | ``inject-nmi`` (inject a NMI into the guest), ``pause`` (pause the |
| 4714 | guest), ``debug`` (print a debug message and continue), or ``none`` |
| 4715 | (do nothing). |
| 4716 | |
| 4717 | Note that the ``shutdown`` action requires that the guest responds |
| 4718 | to ACPI signals, which it may not be able to do in the sort of |
| 4719 | situations where the watchdog would have expired, and thus |
| 4720 | ``-watchdog-action shutdown`` is not recommended for production use. |
| 4721 | |
| 4722 | Examples: |
| 4723 | |
Paolo Bonzini | 5433af7 | 2022-09-10 13:44:47 +0200 | [diff] [blame] | 4724 | ``-device i6300esb -watchdog-action pause`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4725 | |
| 4726 | ERST |
Richard W.M. Jones | 9dd986c | 2009-04-25 13:56:19 +0100 | [diff] [blame] | 4727 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4728 | DEF("echr", HAS_ARG, QEMU_OPTION_echr, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4729 | "-echr chr set terminal escape character instead of ctrl-a\n", |
| 4730 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4731 | SRST |
| 4732 | ``-echr numeric_ascii_value`` |
| 4733 | Change the escape character used for switching to the monitor when |
| 4734 | using monitor and serial sharing. The default is ``0x01`` when using |
| 4735 | the ``-nographic`` option. ``0x01`` is equal to pressing |
| 4736 | ``Control-a``. You can select a different character from the ascii |
| 4737 | control keys where 1 through 26 map to Control-a through Control-z. |
| 4738 | For instance you could use the either of the following to change the |
| 4739 | escape character to Control-t. |
| 4740 | |
| 4741 | ``-echr 0x14``; \ ``-echr 20`` |
| 4742 | |
| 4743 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4744 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4745 | DEF("incoming", HAS_ARG, QEMU_OPTION_incoming, \ |
Daniel P. Berrangé | bf24095 | 2021-02-16 19:10:22 +0000 | [diff] [blame] | 4746 | "-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]\n" \ |
| 4747 | "-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]\n" \ |
Michael Tokarev | 7c60180 | 2015-02-10 22:40:47 +0300 | [diff] [blame] | 4748 | "-incoming unix:socketpath\n" \ |
| 4749 | " prepare for incoming migration, listen on\n" \ |
| 4750 | " specified protocol and socket address\n" \ |
| 4751 | "-incoming fd:fd\n" \ |
Steve Sistare | 385f510 | 2023-09-08 07:22:11 -0700 | [diff] [blame] | 4752 | "-incoming file:filename[,offset=offset]\n" \ |
Michael Tokarev | 7c60180 | 2015-02-10 22:40:47 +0300 | [diff] [blame] | 4753 | "-incoming exec:cmdline\n" \ |
| 4754 | " accept incoming migration on given file descriptor\n" \ |
Dr. David Alan Gilbert | 1597051 | 2015-05-29 19:52:52 +0100 | [diff] [blame] | 4755 | " or from given external command\n" \ |
| 4756 | "-incoming defer\n" \ |
| 4757 | " wait for the URI to be specified via migrate_incoming\n", |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4758 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4759 | SRST |
Daniel P. Berrangé | bf24095 | 2021-02-16 19:10:22 +0000 | [diff] [blame] | 4760 | ``-incoming tcp:[host]:port[,to=maxport][,ipv4=on|off][,ipv6=on|off]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 4761 | \ |
Daniel P. Berrangé | bf24095 | 2021-02-16 19:10:22 +0000 | [diff] [blame] | 4762 | ``-incoming rdma:host:port[,ipv4=on|off][,ipv6=on|off]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4763 | Prepare for incoming migration, listen on a given tcp port. |
| 4764 | |
| 4765 | ``-incoming unix:socketpath`` |
| 4766 | Prepare for incoming migration, listen on a given unix socket. |
| 4767 | |
| 4768 | ``-incoming fd:fd`` |
Steve Sistare | 2a9e2e5 | 2023-09-08 07:22:10 -0700 | [diff] [blame] | 4769 | Accept incoming migration from a given file descriptor. |
| 4770 | |
Steve Sistare | 385f510 | 2023-09-08 07:22:11 -0700 | [diff] [blame] | 4771 | ``-incoming file:filename[,offset=offset]`` |
| 4772 | Accept incoming migration from a given file starting at offset. |
| 4773 | offset allows the common size suffixes, or a 0x prefix, but not both. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4774 | |
| 4775 | ``-incoming exec:cmdline`` |
| 4776 | Accept incoming migration as an output from specified external |
| 4777 | command. |
| 4778 | |
| 4779 | ``-incoming defer`` |
| 4780 | Wait for the URI to be specified via migrate\_incoming. The monitor |
| 4781 | can be used to change settings (such as migration parameters) prior |
| 4782 | to issuing the migrate\_incoming to allow the migration to begin. |
| 4783 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4784 | |
Ashijeet Acharya | d15c05f | 2017-01-16 17:01:51 +0530 | [diff] [blame] | 4785 | DEF("only-migratable", 0, QEMU_OPTION_only_migratable, \ |
| 4786 | "-only-migratable allow only migratable devices\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4787 | SRST |
| 4788 | ``-only-migratable`` |
| 4789 | Only allow migratable devices. Devices will not be allowed to enter |
| 4790 | an unmigratable state. |
| 4791 | ERST |
Ashijeet Acharya | d15c05f | 2017-01-16 17:01:51 +0530 | [diff] [blame] | 4792 | |
Gerd Hoffmann | d8c208d | 2009-12-08 13:11:46 +0100 | [diff] [blame] | 4793 | DEF("nodefaults", 0, QEMU_OPTION_nodefaults, \ |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4794 | "-nodefaults don't create default devices\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4795 | SRST |
| 4796 | ``-nodefaults`` |
| 4797 | Don't create default devices. Normally, QEMU sets the default |
| 4798 | devices like serial port, parallel port, virtual console, monitor |
| 4799 | device, VGA adapter, floppy and CD-ROM drive and others. The |
| 4800 | ``-nodefaults`` option will disable all those default devices. |
| 4801 | ERST |
Gerd Hoffmann | d8c208d | 2009-12-08 13:11:46 +0100 | [diff] [blame] | 4802 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4803 | #ifndef _WIN32 |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4804 | DEF("runas", HAS_ARG, QEMU_OPTION_runas, \ |
Ian Jackson | 2c42f1e | 2017-09-15 18:10:44 +0100 | [diff] [blame] | 4805 | "-runas user change to user id user just before starting the VM\n" \ |
| 4806 | " user can be numeric uid:gid instead\n", |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4807 | QEMU_ARCH_ALL) |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4808 | #endif |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4809 | SRST |
| 4810 | ``-runas user`` |
| 4811 | Immediately before starting guest execution, drop root privileges, |
| 4812 | switching to the specified user. |
| 4813 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4814 | |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4815 | DEF("prom-env", HAS_ARG, QEMU_OPTION_prom_env, |
| 4816 | "-prom-env variable=value\n" |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4817 | " set OpenBIOS nvram variables\n", |
| 4818 | QEMU_ARCH_PPC | QEMU_ARCH_SPARC) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4819 | SRST |
| 4820 | ``-prom-env variable=value`` |
| 4821 | Set OpenBIOS nvram variable to given value (PPC, SPARC only). |
| 4822 | |
| 4823 | :: |
| 4824 | |
| 4825 | qemu-system-sparc -prom-env 'auto-boot?=false' \ |
| 4826 | -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single' |
| 4827 | |
| 4828 | :: |
| 4829 | |
| 4830 | qemu-system-ppc -prom-env 'auto-boot?=false' \ |
| 4831 | -prom-env 'boot-device=hd:2,\yaboot' \ |
| 4832 | -prom-env 'boot-args=conf=hd:2,\yaboot.conf' |
| 4833 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4834 | DEF("semihosting", 0, QEMU_OPTION_semihosting, |
Michael Walle | f7bbcfb | 2014-04-22 20:18:42 +0200 | [diff] [blame] | 4835 | "-semihosting semihosting mode\n", |
Markus Armbruster | 9d49bcf | 2021-05-03 10:40:33 +0200 | [diff] [blame] | 4836 | QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | |
Keith Packard | a10b9d9 | 2021-01-08 22:42:52 +0000 | [diff] [blame] | 4837 | QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4838 | SRST |
| 4839 | ``-semihosting`` |
Alex Bennée | 2da9d21 | 2023-01-24 18:01:13 +0000 | [diff] [blame] | 4840 | Enable :ref:`Semihosting` mode (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V only). |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4841 | |
Alex Bennée | 2da9d21 | 2023-01-24 18:01:13 +0000 | [diff] [blame] | 4842 | .. warning:: |
| 4843 | Note that this allows guest direct access to the host filesystem, so |
| 4844 | should only be used with a trusted guest OS. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4845 | |
| 4846 | See the -semihosting-config option documentation for further |
| 4847 | information about the facilities this enables. |
| 4848 | ERST |
Liviu Ionescu | a38bb07 | 2014-12-11 12:07:48 +0000 | [diff] [blame] | 4849 | DEF("semihosting-config", HAS_ARG, QEMU_OPTION_semihosting_config, |
Peter Maydell | 5202861 | 2022-08-22 15:12:24 +0100 | [diff] [blame] | 4850 | "-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]\n" \ |
Leon Alrae | a59d31a | 2015-06-19 14:17:45 +0100 | [diff] [blame] | 4851 | " semihosting configuration\n", |
Markus Armbruster | 9d49bcf | 2021-05-03 10:40:33 +0200 | [diff] [blame] | 4852 | QEMU_ARCH_ARM | QEMU_ARCH_M68K | QEMU_ARCH_XTENSA | |
Keith Packard | a10b9d9 | 2021-01-08 22:42:52 +0000 | [diff] [blame] | 4853 | QEMU_ARCH_MIPS | QEMU_ARCH_NIOS2 | QEMU_ARCH_RISCV) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4854 | SRST |
Peter Maydell | 5202861 | 2022-08-22 15:12:24 +0100 | [diff] [blame] | 4855 | ``-semihosting-config [enable=on|off][,target=native|gdb|auto][,chardev=id][,userspace=on|off][,arg=str[,...]]`` |
Alex Bennée | 2da9d21 | 2023-01-24 18:01:13 +0000 | [diff] [blame] | 4856 | Enable and configure :ref:`Semihosting` (ARM, M68K, Xtensa, MIPS, Nios II, RISC-V |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4857 | only). |
| 4858 | |
Alex Bennée | 2da9d21 | 2023-01-24 18:01:13 +0000 | [diff] [blame] | 4859 | .. warning:: |
| 4860 | Note that this allows guest direct access to the host filesystem, so |
| 4861 | should only be used with a trusted guest OS. |
Keith Packard | a10b9d9 | 2021-01-08 22:42:52 +0000 | [diff] [blame] | 4862 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4863 | ``target=native|gdb|auto`` |
| 4864 | Defines where the semihosting calls will be addressed, to QEMU |
| 4865 | (``native``) or to GDB (``gdb``). The default is ``auto``, which |
| 4866 | means ``gdb`` during debug sessions and ``native`` otherwise. |
| 4867 | |
| 4868 | ``chardev=str1`` |
| 4869 | Send the output to a chardev backend output for native or auto |
| 4870 | output when not in gdb |
| 4871 | |
Peter Maydell | 5202861 | 2022-08-22 15:12:24 +0100 | [diff] [blame] | 4872 | ``userspace=on|off`` |
| 4873 | Allows code running in guest userspace to access the semihosting |
| 4874 | interface. The default is that only privileged guest code can |
| 4875 | make semihosting calls. Note that setting ``userspace=on`` should |
| 4876 | only be used if all guest code is trusted (for example, in |
| 4877 | bare-metal test case code). |
| 4878 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4879 | ``arg=str1,arg=str2,...`` |
| 4880 | Allows the user to pass input arguments, and can be used |
| 4881 | multiple times to build up a list. The old-style |
| 4882 | ``-kernel``/``-append`` method of passing a command line is |
| 4883 | still supported for backward compatibility. If both the |
| 4884 | ``--semihosting-config arg`` and the ``-kernel``/``-append`` are |
| 4885 | specified, the former is passed to semihosting as it always |
| 4886 | takes precedence. |
| 4887 | ERST |
blueswir1 | 5824d65 | 2009-03-28 06:44:27 +0000 | [diff] [blame] | 4888 | DEF("old-param", 0, QEMU_OPTION_old_param, |
Blue Swirl | ad96090 | 2010-03-29 19:23:52 +0000 | [diff] [blame] | 4889 | "-old-param old param mode\n", QEMU_ARCH_ARM) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4890 | SRST |
| 4891 | ``-old-param`` |
| 4892 | Old param mode (ARM only). |
| 4893 | ERST |
Stefan Weil | 95d5f08 | 2010-01-20 22:25:27 +0100 | [diff] [blame] | 4894 | |
Eduardo Otubo | 7d76ad4 | 2012-08-14 18:44:08 -0300 | [diff] [blame] | 4895 | DEF("sandbox", HAS_ARG, QEMU_OPTION_sandbox, \ |
Eduardo Otubo | 73a1e64 | 2017-03-13 22:13:27 +0100 | [diff] [blame] | 4896 | "-sandbox on[,obsolete=allow|deny][,elevateprivileges=allow|deny|children]\n" \ |
Eduardo Otubo | 24f8cdc | 2017-03-13 22:18:51 +0100 | [diff] [blame] | 4897 | " [,spawn=allow|deny][,resourcecontrol=allow|deny]\n" \ |
Eduardo Otubo | 2b716fa | 2017-03-01 23:17:29 +0100 | [diff] [blame] | 4898 | " Enable seccomp mode 2 system call filter (default 'off').\n" \ |
| 4899 | " use 'obsolete' to allow obsolete system calls that are provided\n" \ |
| 4900 | " by the kernel, but typically no longer used by modern\n" \ |
Eduardo Otubo | 73a1e64 | 2017-03-13 22:13:27 +0100 | [diff] [blame] | 4901 | " C library implementations.\n" \ |
Philippe Mathieu-Daudé | d42304b | 2021-03-03 19:46:43 +0100 | [diff] [blame] | 4902 | " use 'elevateprivileges' to allow or deny the QEMU process ability\n" \ |
| 4903 | " to elevate privileges using set*uid|gid system calls.\n" \ |
Eduardo Otubo | 73a1e64 | 2017-03-13 22:13:27 +0100 | [diff] [blame] | 4904 | " The value 'children' will deny set*uid|gid system calls for\n" \ |
Eduardo Otubo | 995a226 | 2017-03-13 22:16:01 +0100 | [diff] [blame] | 4905 | " main QEMU process but will allow forks and execves to run unprivileged\n" \ |
| 4906 | " use 'spawn' to avoid QEMU to spawn new threads or processes by\n" \ |
Philippe Mathieu-Daudé | d42304b | 2021-03-03 19:46:43 +0100 | [diff] [blame] | 4907 | " blocking *fork and execve\n" \ |
Eduardo Otubo | 24f8cdc | 2017-03-13 22:18:51 +0100 | [diff] [blame] | 4908 | " use 'resourcecontrol' to disable process affinity and schedular priority\n", |
Eduardo Otubo | 7d76ad4 | 2012-08-14 18:44:08 -0300 | [diff] [blame] | 4909 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4910 | SRST |
| 4911 | ``-sandbox arg[,obsolete=string][,elevateprivileges=string][,spawn=string][,resourcecontrol=string]`` |
| 4912 | Enable Seccomp mode 2 system call filter. 'on' will enable syscall |
| 4913 | filtering and 'off' will disable it. The default is 'off'. |
| 4914 | |
| 4915 | ``obsolete=string`` |
| 4916 | Enable Obsolete system calls |
| 4917 | |
| 4918 | ``elevateprivileges=string`` |
| 4919 | Disable set\*uid\|gid system calls |
| 4920 | |
| 4921 | ``spawn=string`` |
| 4922 | Disable \*fork and execve |
| 4923 | |
| 4924 | ``resourcecontrol=string`` |
| 4925 | Disable process affinity and schedular priority |
| 4926 | ERST |
Eduardo Otubo | 7d76ad4 | 2012-08-14 18:44:08 -0300 | [diff] [blame] | 4927 | |
Gerd Hoffmann | 715a664 | 2009-10-14 10:39:28 +0200 | [diff] [blame] | 4928 | DEF("readconfig", HAS_ARG, QEMU_OPTION_readconfig, |
Paolo Bonzini | e960a7e | 2022-04-14 10:57:21 -0400 | [diff] [blame] | 4929 | "-readconfig <file>\n" |
| 4930 | " read config file\n", QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4931 | SRST |
| 4932 | ``-readconfig file`` |
| 4933 | Read device configuration from file. This approach is useful when |
| 4934 | you want to spawn QEMU process with many command line options but |
| 4935 | you don't want to exceed the command line character limit. |
| 4936 | ERST |
Thomas Huth | 2feac45 | 2018-08-21 12:59:56 +0200 | [diff] [blame] | 4937 | |
Eduardo Habkost | f29a561 | 2012-05-02 13:07:29 -0300 | [diff] [blame] | 4938 | DEF("no-user-config", 0, QEMU_OPTION_nouserconfig, |
| 4939 | "-no-user-config\n" |
Eduardo Habkost | 3478eae | 2017-10-04 00:00:25 -0300 | [diff] [blame] | 4940 | " do not load default user-provided config files at startup\n", |
Eduardo Habkost | f29a561 | 2012-05-02 13:07:29 -0300 | [diff] [blame] | 4941 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4942 | SRST |
| 4943 | ``-no-user-config`` |
| 4944 | The ``-no-user-config`` option makes QEMU not load any of the |
| 4945 | user-provided config files on sysconfdir. |
| 4946 | ERST |
Thomas Huth | 2feac45 | 2018-08-21 12:59:56 +0200 | [diff] [blame] | 4947 | |
Prerna Saxena | ab6540d | 2010-08-09 11:48:32 +0100 | [diff] [blame] | 4948 | DEF("trace", HAS_ARG, QEMU_OPTION_trace, |
Paolo Bonzini | 10578a2 | 2016-01-07 16:55:26 +0300 | [diff] [blame] | 4949 | "-trace [[enable=]<pattern>][,events=<file>][,file=<file>]\n" |
Lluís | 23d15e8 | 2011-08-31 20:31:31 +0200 | [diff] [blame] | 4950 | " specify tracing options\n", |
Prerna Saxena | ab6540d | 2010-08-09 11:48:32 +0100 | [diff] [blame] | 4951 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4952 | SRST |
| 4953 | ``-trace [[enable=]pattern][,events=file][,file=file]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 4954 | .. include:: ../qemu-option-trace.rst.inc |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4955 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4956 | ERST |
Lluís Vilanova | 42229a7 | 2017-07-24 17:28:22 +0300 | [diff] [blame] | 4957 | DEF("plugin", HAS_ARG, QEMU_OPTION_plugin, |
Mahmoud Mandour | 3a445ac | 2021-07-30 15:58:05 +0200 | [diff] [blame] | 4958 | "-plugin [file=]<file>[,<argname>=<argvalue>]\n" |
Lluís Vilanova | 42229a7 | 2017-07-24 17:28:22 +0300 | [diff] [blame] | 4959 | " load a plugin\n", |
| 4960 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4961 | SRST |
Mahmoud Mandour | 3a445ac | 2021-07-30 15:58:05 +0200 | [diff] [blame] | 4962 | ``-plugin file=file[,argname=argvalue]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4963 | Load a plugin. |
| 4964 | |
| 4965 | ``file=file`` |
| 4966 | Load the given plugin from a shared library file. |
| 4967 | |
Mahmoud Mandour | 3a445ac | 2021-07-30 15:58:05 +0200 | [diff] [blame] | 4968 | ``argname=argvalue`` |
| 4969 | Argument passed to the plugin. (Can be given multiple times.) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 4970 | ERST |
Stefan Weil | 3dbf2c7 | 2010-01-16 18:19:44 +0100 | [diff] [blame] | 4971 | |
Markus Armbruster | 31e70d6 | 2013-02-13 19:49:37 +0100 | [diff] [blame] | 4972 | HXCOMM Internal use |
| 4973 | DEF("qtest", HAS_ARG, QEMU_OPTION_qtest, "", QEMU_ARCH_ALL) |
| 4974 | DEF("qtest-log", HAS_ARG, QEMU_OPTION_qtest_log, "", QEMU_ARCH_ALL) |
Anthony Liguori | c7f0f3b | 2012-03-28 15:42:02 +0200 | [diff] [blame] | 4975 | |
Thomas Huth | 9ffcbe2 | 2023-07-03 09:44:47 +0200 | [diff] [blame] | 4976 | #ifdef CONFIG_POSIX |
Claudio Imbrenda | 80bd81c | 2023-05-05 14:00:51 +0200 | [diff] [blame] | 4977 | DEF("run-with", HAS_ARG, QEMU_OPTION_run_with, |
Thomas Huth | 9ffcbe2 | 2023-07-03 09:44:47 +0200 | [diff] [blame] | 4978 | "-run-with [async-teardown=on|off][,chroot=dir]\n" |
| 4979 | " Set miscellaneous QEMU process lifecycle options:\n" |
| 4980 | " async-teardown=on enables asynchronous teardown (Linux only)\n" |
| 4981 | " chroot=dir chroot to dir just before starting the VM\n", |
Claudio Imbrenda | 80bd81c | 2023-05-05 14:00:51 +0200 | [diff] [blame] | 4982 | QEMU_ARCH_ALL) |
| 4983 | SRST |
Thomas Huth | 9ffcbe2 | 2023-07-03 09:44:47 +0200 | [diff] [blame] | 4984 | ``-run-with [async-teardown=on|off][,chroot=dir]`` |
Claudio Imbrenda | 80bd81c | 2023-05-05 14:00:51 +0200 | [diff] [blame] | 4985 | Set QEMU process lifecycle options. |
| 4986 | |
| 4987 | ``async-teardown=on`` enables asynchronous teardown. A new process called |
| 4988 | "cleanup/<QEMU_PID>" will be created at startup sharing the address |
| 4989 | space with the main QEMU process, using clone. It will wait for the |
| 4990 | main QEMU process to terminate completely, and then exit. This allows |
| 4991 | QEMU to terminate very quickly even if the guest was huge, leaving the |
| 4992 | teardown of the address space to the cleanup process. Since the cleanup |
| 4993 | process shares the same cgroups as the main QEMU process, accounting is |
| 4994 | performed correctly. This only works if the cleanup process is not |
| 4995 | forcefully killed with SIGKILL before the main QEMU process has |
| 4996 | terminated completely. |
Thomas Huth | 9ffcbe2 | 2023-07-03 09:44:47 +0200 | [diff] [blame] | 4997 | |
| 4998 | ``chroot=dir`` can be used for doing a chroot to the specified directory |
| 4999 | immediately before starting the guest execution. This is especially useful |
| 5000 | in combination with -runas. |
Claudio Imbrenda | 80bd81c | 2023-05-05 14:00:51 +0200 | [diff] [blame] | 5001 | ERST |
| 5002 | #endif |
Claudio Imbrenda | c891c24 | 2022-08-12 15:34:53 +0200 | [diff] [blame] | 5003 | |
Seiji Aguchi | 5e2ac51 | 2013-07-03 23:02:46 -0400 | [diff] [blame] | 5004 | DEF("msg", HAS_ARG, QEMU_OPTION_msg, |
Mario Smarduch | 2880ffb | 2020-06-26 13:19:00 -0700 | [diff] [blame] | 5005 | "-msg [timestamp[=on|off]][,guest-name=[on|off]]\n" |
Markus Armbruster | deda497 | 2019-10-10 10:15:08 +0200 | [diff] [blame] | 5006 | " control error message format\n" |
Mario Smarduch | 2880ffb | 2020-06-26 13:19:00 -0700 | [diff] [blame] | 5007 | " timestamp=on enables timestamps (default: off)\n" |
| 5008 | " guest-name=on enables guest name prefix but only if\n" |
| 5009 | " -name guest option is set (default: off)\n", |
Seiji Aguchi | 5e2ac51 | 2013-07-03 23:02:46 -0400 | [diff] [blame] | 5010 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5011 | SRST |
Mario Smarduch | 2880ffb | 2020-06-26 13:19:00 -0700 | [diff] [blame] | 5012 | ``-msg [timestamp[=on|off]][,guest-name[=on|off]]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5013 | Control error message format. |
| 5014 | |
| 5015 | ``timestamp=on|off`` |
| 5016 | Prefix messages with a timestamp. Default is off. |
Mario Smarduch | 2880ffb | 2020-06-26 13:19:00 -0700 | [diff] [blame] | 5017 | |
| 5018 | ``guest-name=on|off`` |
| 5019 | Prefix messages with guest name but only if -name guest option is set |
| 5020 | otherwise the option is ignored. Default is off. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5021 | ERST |
Seiji Aguchi | 5e2ac51 | 2013-07-03 23:02:46 -0400 | [diff] [blame] | 5022 | |
Amit Shah | abfd9ce | 2014-06-20 18:56:08 +0530 | [diff] [blame] | 5023 | DEF("dump-vmstate", HAS_ARG, QEMU_OPTION_dump_vmstate, |
| 5024 | "-dump-vmstate <file>\n" |
| 5025 | " Output vmstate information in JSON format to file.\n" |
| 5026 | " Use the scripts/vmstate-static-checker.py file to\n" |
| 5027 | " check for possible regressions in migration code\n" |
Laurent Vivier | 2382053 | 2015-09-04 21:30:04 +0200 | [diff] [blame] | 5028 | " by comparing two such vmstate dumps.\n", |
Amit Shah | abfd9ce | 2014-06-20 18:56:08 +0530 | [diff] [blame] | 5029 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5030 | SRST |
| 5031 | ``-dump-vmstate file`` |
| 5032 | Dump json-encoded vmstate information for current machine type to |
| 5033 | file in file |
| 5034 | ERST |
Amit Shah | abfd9ce | 2014-06-20 18:56:08 +0530 | [diff] [blame] | 5035 | |
Emilio G. Cota | 12df189 | 2018-08-15 11:42:49 -0400 | [diff] [blame] | 5036 | DEF("enable-sync-profile", 0, QEMU_OPTION_enable_sync_profile, |
| 5037 | "-enable-sync-profile\n" |
| 5038 | " enable synchronization profiling\n", |
| 5039 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5040 | SRST |
| 5041 | ``-enable-sync-profile`` |
| 5042 | Enable synchronization profiling. |
| 5043 | ERST |
Emilio G. Cota | 12df189 | 2018-08-15 11:42:49 -0400 | [diff] [blame] | 5044 | |
Ilya Leoshkevich | 5584e2d | 2023-01-12 16:20:13 +0100 | [diff] [blame] | 5045 | #if defined(CONFIG_TCG) && defined(CONFIG_LINUX) |
| 5046 | DEF("perfmap", 0, QEMU_OPTION_perfmap, |
| 5047 | "-perfmap generate a /tmp/perf-${pid}.map file for perf\n", |
| 5048 | QEMU_ARCH_ALL) |
| 5049 | SRST |
| 5050 | ``-perfmap`` |
| 5051 | Generate a map file for Linux perf tools that will allow basic profiling |
| 5052 | information to be broken down into basic blocks. |
| 5053 | ERST |
| 5054 | |
| 5055 | DEF("jitdump", 0, QEMU_OPTION_jitdump, |
| 5056 | "-jitdump generate a jit-${pid}.dump file for perf\n", |
| 5057 | QEMU_ARCH_ALL) |
| 5058 | SRST |
| 5059 | ``-jitdump`` |
| 5060 | Generate a dump file for Linux perf tools that maps basic blocks to symbol |
| 5061 | names, line numbers and JITted code. |
| 5062 | ERST |
| 5063 | #endif |
| 5064 | |
Paolo Bonzini | 43f187a | 2017-01-04 13:50:37 +0100 | [diff] [blame] | 5065 | DEFHEADING() |
Markus Armbruster | de6b4f9 | 2017-10-02 16:03:00 +0200 | [diff] [blame] | 5066 | |
| 5067 | DEFHEADING(Generic object creation:) |
Daniel P. Berrange | b9174d4 | 2015-05-13 17:14:03 +0100 | [diff] [blame] | 5068 | |
| 5069 | DEF("object", HAS_ARG, QEMU_OPTION_object, |
| 5070 | "-object TYPENAME[,PROP1=VALUE1,...]\n" |
| 5071 | " create a new object of type TYPENAME setting properties\n" |
| 5072 | " in the order they are specified. Note that the 'id'\n" |
| 5073 | " property must be set. These objects are placed in the\n" |
| 5074 | " '/objects' path.\n", |
| 5075 | QEMU_ARCH_ALL) |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5076 | SRST |
| 5077 | ``-object typename[,prop1=value1,...]`` |
| 5078 | Create a new object of type typename setting properties in the order |
| 5079 | they are specified. Note that the 'id' property must be set. These |
| 5080 | objects are placed in the '/objects' path. |
| 5081 | |
David Hildenbrand | e92666b | 2023-09-06 14:04:55 +0200 | [diff] [blame] | 5082 | ``-object memory-backend-file,id=id,size=size,mem-path=dir,share=on|off,discard-data=on|off,merge=on|off,dump=on|off,prealloc=on|off,host-nodes=host-nodes,policy=default|preferred|bind|interleave,align=align,offset=offset,readonly=on|off,rom=on|off|auto`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5083 | Creates a memory file backend object, which can be used to back |
| 5084 | the guest RAM with huge pages. |
| 5085 | |
| 5086 | The ``id`` parameter is a unique ID that will be used to |
Robert Hoo | 56c9f00 | 2021-04-22 16:42:02 +0800 | [diff] [blame] | 5087 | reference this memory region in other parameters, e.g. ``-numa``, |
| 5088 | ``-device nvdimm``, etc. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5089 | |
| 5090 | The ``size`` option provides the size of the memory region, and |
Robert Hoo | 56c9f00 | 2021-04-22 16:42:02 +0800 | [diff] [blame] | 5091 | accepts common suffixes, e.g. ``500M``. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5092 | |
| 5093 | The ``mem-path`` provides the path to either a shared memory or |
| 5094 | huge page filesystem mount. |
| 5095 | |
| 5096 | The ``share`` boolean option determines whether the memory |
| 5097 | region is marked as private to QEMU, or shared. The latter |
| 5098 | allows a co-operating external process to access the QEMU memory |
| 5099 | region. |
| 5100 | |
| 5101 | The ``share`` is also required for pvrdma devices due to |
| 5102 | limitations in the RDMA API provided by Linux. |
| 5103 | |
| 5104 | Setting share=on might affect the ability to configure NUMA |
| 5105 | bindings for the memory backend under some circumstances, see |
| 5106 | Documentation/vm/numa\_memory\_policy.txt on the Linux kernel |
| 5107 | source tree for additional details. |
| 5108 | |
| 5109 | Setting the ``discard-data`` boolean option to on indicates that |
| 5110 | file contents can be destroyed when QEMU exits, to avoid |
| 5111 | unnecessarily flushing data to the backing file. Note that |
| 5112 | ``discard-data`` is only an optimization, and QEMU might not |
| 5113 | discard file contents if it aborts unexpectedly or is terminated |
| 5114 | using SIGKILL. |
| 5115 | |
| 5116 | The ``merge`` boolean option enables memory merge, also known as |
| 5117 | MADV\_MERGEABLE, so that Kernel Samepage Merging will consider |
| 5118 | the pages for memory deduplication. |
| 5119 | |
| 5120 | Setting the ``dump`` boolean option to off excludes the memory |
| 5121 | from core dumps. This feature is also known as MADV\_DONTDUMP. |
| 5122 | |
| 5123 | The ``prealloc`` boolean option enables memory preallocation. |
| 5124 | |
| 5125 | The ``host-nodes`` option binds the memory range to a list of |
| 5126 | NUMA host nodes. |
| 5127 | |
| 5128 | The ``policy`` option sets the NUMA policy to one of the |
| 5129 | following values: |
| 5130 | |
| 5131 | ``default`` |
| 5132 | default host policy |
| 5133 | |
| 5134 | ``preferred`` |
| 5135 | prefer the given host node list for allocation |
| 5136 | |
| 5137 | ``bind`` |
| 5138 | restrict memory allocation to the given host node list |
| 5139 | |
| 5140 | ``interleave`` |
| 5141 | interleave memory allocations across the given host node |
| 5142 | list |
| 5143 | |
| 5144 | The ``align`` option specifies the base address alignment when |
| 5145 | QEMU mmap(2) ``mem-path``, and accepts common suffixes, eg |
| 5146 | ``2M``. Some backend store specified by ``mem-path`` requires an |
| 5147 | alignment different than the default one used by QEMU, eg the |
| 5148 | device DAX /dev/dax0.0 requires 2M alignment rather than 4K. In |
| 5149 | such cases, users can specify the required alignment via this |
| 5150 | option. |
| 5151 | |
Alexander Graf | 4b870dc | 2023-04-03 22:14:21 +0000 | [diff] [blame] | 5152 | The ``offset`` option specifies the offset into the target file |
| 5153 | that the region starts at. You can use this parameter to back |
| 5154 | multiple regions with a single file. |
| 5155 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5156 | The ``pmem`` option specifies whether the backing file specified |
| 5157 | by ``mem-path`` is in host persistent memory that can be |
| 5158 | accessed using the SNIA NVM programming model (e.g. Intel |
| 5159 | NVDIMM). If ``pmem`` is set to 'on', QEMU will take necessary |
| 5160 | operations to guarantee the persistence of its own writes to |
| 5161 | ``mem-path`` (e.g. in vNVDIMM label emulation and live |
| 5162 | migration). Also, we will map the backend-file with MAP\_SYNC |
| 5163 | flag, which ensures the file metadata is in sync for |
| 5164 | ``mem-path`` in case of host crash or a power failure. MAP\_SYNC |
| 5165 | requires support from both the host kernel (since Linux kernel |
| 5166 | 4.15) and the filesystem of ``mem-path`` mounted with DAX |
| 5167 | option. |
| 5168 | |
Stefan Hajnoczi | 86635aa | 2021-01-04 17:13:19 +0000 | [diff] [blame] | 5169 | The ``readonly`` option specifies whether the backing file is opened |
| 5170 | read-only or read-write (default). |
| 5171 | |
David Hildenbrand | e92666b | 2023-09-06 14:04:55 +0200 | [diff] [blame] | 5172 | The ``rom`` option specifies whether to create Read Only Memory |
| 5173 | (ROM) that cannot be modified by the VM. Any write attempts to such |
| 5174 | ROM will be denied. Most use cases want proper RAM instead of ROM. |
| 5175 | However, selected use cases, like R/O NVDIMMs, can benefit from |
| 5176 | ROM. If set to ``on``, create ROM; if set to ``off``, create |
| 5177 | writable RAM; if set to ``auto`` (default), the value of the |
| 5178 | ``readonly`` option is used. This option is primarily helpful when |
| 5179 | we want to have writable RAM in configurations that would |
| 5180 | traditionally create ROM before the ``rom`` option was introduced: |
| 5181 | VM templating, where we want to open a file readonly |
| 5182 | (``readonly=on``) and mark the memory to be private for QEMU |
| 5183 | (``share=off``). For this use case, we need writable RAM instead |
| 5184 | of ROM, and want to also set ``rom=off``. |
| 5185 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5186 | ``-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`` |
| 5187 | Creates a memory backend object, which can be used to back the |
| 5188 | guest RAM. Memory backend objects offer more control than the |
| 5189 | ``-m`` option that is traditionally used to define guest RAM. |
| 5190 | Please refer to ``memory-backend-file`` for a description of the |
| 5191 | options. |
| 5192 | |
| 5193 | ``-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`` |
| 5194 | Creates an anonymous memory file backend object, which allows |
| 5195 | QEMU to share the memory with an external process (e.g. when |
| 5196 | using vhost-user). The memory is allocated with memfd and |
| 5197 | optional sealing. (Linux only) |
| 5198 | |
| 5199 | The ``seal`` option creates a sealed-file, that will block |
| 5200 | further resizing the memory ('on' by default). |
| 5201 | |
| 5202 | The ``hugetlb`` option specify the file to be created resides in |
| 5203 | the hugetlbfs filesystem (since Linux 4.14). Used in conjunction |
| 5204 | with the ``hugetlb`` option, the ``hugetlbsize`` option specify |
| 5205 | the hugetlb page size on systems that support multiple hugetlb |
| 5206 | page sizes (it must be a power of 2 value supported by the |
| 5207 | system). |
| 5208 | |
| 5209 | In some versions of Linux, the ``hugetlb`` option is |
| 5210 | incompatible with the ``seal`` option (requires at least Linux |
| 5211 | 4.16). |
| 5212 | |
| 5213 | Please refer to ``memory-backend-file`` for a description of the |
| 5214 | other options. |
| 5215 | |
| 5216 | The ``share`` boolean option is on by default with memfd. |
| 5217 | |
Eric Auger | 6e6d8ac | 2023-11-21 16:44:00 +0800 | [diff] [blame] | 5218 | ``-object iommufd,id=id[,fd=fd]`` |
| 5219 | Creates an iommufd backend which allows control of DMA mapping |
| 5220 | through the ``/dev/iommu`` device. |
| 5221 | |
| 5222 | The ``id`` parameter is a unique ID which frontends (such as |
| 5223 | vfio-pci of vdpa) will use to connect with the iommufd backend. |
| 5224 | |
| 5225 | The ``fd`` parameter is an optional pre-opened file descriptor |
| 5226 | resulting from ``/dev/iommu`` opening. Usually the iommufd is shared |
| 5227 | across all subsystems, bringing the benefit of centralized |
| 5228 | reference counting. |
| 5229 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5230 | ``-object rng-builtin,id=id`` |
| 5231 | Creates a random number generator backend which obtains entropy |
| 5232 | from QEMU builtin functions. The ``id`` parameter is a unique ID |
| 5233 | that will be used to reference this entropy backend from the |
| 5234 | ``virtio-rng`` device. By default, the ``virtio-rng`` device |
| 5235 | uses this RNG backend. |
| 5236 | |
| 5237 | ``-object rng-random,id=id,filename=/dev/random`` |
| 5238 | Creates a random number generator backend which obtains entropy |
| 5239 | from a device on the host. The ``id`` parameter is a unique ID |
| 5240 | that will be used to reference this entropy backend from the |
| 5241 | ``virtio-rng`` device. The ``filename`` parameter specifies |
| 5242 | which file to obtain entropy from and if omitted defaults to |
| 5243 | ``/dev/urandom``. |
| 5244 | |
| 5245 | ``-object rng-egd,id=id,chardev=chardevid`` |
| 5246 | Creates a random number generator backend which obtains entropy |
| 5247 | from an external daemon running on the host. The ``id`` |
| 5248 | parameter is a unique ID that will be used to reference this |
| 5249 | entropy backend from the ``virtio-rng`` device. The ``chardev`` |
| 5250 | parameter is the unique ID of a character device backend that |
| 5251 | provides the connection to the RNG daemon. |
| 5252 | |
| 5253 | ``-object tls-creds-anon,id=id,endpoint=endpoint,dir=/path/to/cred/dir,verify-peer=on|off`` |
| 5254 | Creates a TLS anonymous credentials object, which can be used to |
| 5255 | provide TLS support on network backends. The ``id`` parameter is |
| 5256 | a unique ID which network backends will use to access the |
| 5257 | credentials. The ``endpoint`` is either ``server`` or ``client`` |
| 5258 | depending on whether the QEMU network backend that uses the |
| 5259 | credentials will be acting as a client or as a server. If |
| 5260 | ``verify-peer`` is enabled (the default) then once the handshake |
| 5261 | is completed, the peer credentials will be verified, though this |
| 5262 | is a no-op for anonymous credentials. |
| 5263 | |
| 5264 | The dir parameter tells QEMU where to find the credential files. |
| 5265 | For server endpoints, this directory may contain a file |
| 5266 | dh-params.pem providing diffie-hellman parameters to use for the |
| 5267 | TLS server. If the file is missing, QEMU will generate a set of |
| 5268 | DH parameters at startup. This is a computationally expensive |
| 5269 | operation that consumes random pool entropy, so it is |
| 5270 | recommended that a persistent set of parameters be generated |
| 5271 | upfront and saved. |
| 5272 | |
| 5273 | ``-object tls-creds-psk,id=id,endpoint=endpoint,dir=/path/to/keys/dir[,username=username]`` |
| 5274 | Creates a TLS Pre-Shared Keys (PSK) credentials object, which |
| 5275 | can be used to provide TLS support on network backends. The |
| 5276 | ``id`` parameter is a unique ID which network backends will use |
| 5277 | to access the credentials. The ``endpoint`` is either ``server`` |
| 5278 | or ``client`` depending on whether the QEMU network backend that |
| 5279 | uses the credentials will be acting as a client or as a server. |
| 5280 | For clients only, ``username`` is the username which will be |
| 5281 | sent to the server. If omitted it defaults to "qemu". |
| 5282 | |
| 5283 | The dir parameter tells QEMU where to find the keys file. It is |
| 5284 | called "dir/keys.psk" and contains "username:key" pairs. This |
| 5285 | file can most easily be created using the GnuTLS ``psktool`` |
| 5286 | program. |
| 5287 | |
| 5288 | For server endpoints, dir may also contain a file dh-params.pem |
| 5289 | providing diffie-hellman parameters to use for the TLS server. |
| 5290 | If the file is missing, QEMU will generate a set of DH |
| 5291 | parameters at startup. This is a computationally expensive |
| 5292 | operation that consumes random pool entropy, so it is |
| 5293 | recommended that a persistent set of parameters be generated up |
| 5294 | front and saved. |
| 5295 | |
| 5296 | ``-object tls-creds-x509,id=id,endpoint=endpoint,dir=/path/to/cred/dir,priority=priority,verify-peer=on|off,passwordid=id`` |
| 5297 | Creates a TLS anonymous credentials object, which can be used to |
| 5298 | provide TLS support on network backends. The ``id`` parameter is |
| 5299 | a unique ID which network backends will use to access the |
| 5300 | credentials. The ``endpoint`` is either ``server`` or ``client`` |
| 5301 | depending on whether the QEMU network backend that uses the |
| 5302 | credentials will be acting as a client or as a server. If |
| 5303 | ``verify-peer`` is enabled (the default) then once the handshake |
| 5304 | is completed, the peer credentials will be verified. With x509 |
| 5305 | certificates, this implies that the clients must be provided |
| 5306 | with valid client certificates too. |
| 5307 | |
| 5308 | The dir parameter tells QEMU where to find the credential files. |
| 5309 | For server endpoints, this directory may contain a file |
| 5310 | dh-params.pem providing diffie-hellman parameters to use for the |
| 5311 | TLS server. If the file is missing, QEMU will generate a set of |
| 5312 | DH parameters at startup. This is a computationally expensive |
| 5313 | operation that consumes random pool entropy, so it is |
| 5314 | recommended that a persistent set of parameters be generated |
| 5315 | upfront and saved. |
| 5316 | |
| 5317 | For x509 certificate credentials the directory will contain |
| 5318 | further files providing the x509 certificates. The certificates |
| 5319 | must be stored in PEM format, in filenames ca-cert.pem, |
| 5320 | ca-crl.pem (optional), server-cert.pem (only servers), |
| 5321 | server-key.pem (only servers), client-cert.pem (only clients), |
| 5322 | and client-key.pem (only clients). |
| 5323 | |
| 5324 | For the server-key.pem and client-key.pem files which contain |
| 5325 | sensitive private keys, it is possible to use an encrypted |
| 5326 | version by providing the passwordid parameter. This provides the |
| 5327 | ID of a previously created ``secret`` object containing the |
| 5328 | password for decryption. |
| 5329 | |
| 5330 | The priority parameter allows to override the global default |
| 5331 | priority used by gnutls. This can be useful if the system |
| 5332 | administrator needs to use a weaker set of crypto priorities for |
| 5333 | QEMU without potentially forcing the weakness onto all |
| 5334 | applications. Or conversely if one wants wants a stronger |
| 5335 | default for QEMU than for all other applications, they can do |
| 5336 | this through this parameter. Its format is a gnutls priority |
| 5337 | string as described at |
| 5338 | https://gnutls.org/manual/html_node/Priority-Strings.html. |
| 5339 | |
Philippe Mathieu-Daudé | 993aec2 | 2018-10-11 20:21:11 +0200 | [diff] [blame] | 5340 | ``-object tls-cipher-suites,id=id,priority=priority`` |
| 5341 | Creates a TLS cipher suites object, which can be used to control |
| 5342 | the TLS cipher/protocol algorithms that applications are permitted |
| 5343 | to use. |
| 5344 | |
| 5345 | The ``id`` parameter is a unique ID which frontends will use to |
| 5346 | access the ordered list of permitted TLS cipher suites from the |
| 5347 | host. |
| 5348 | |
| 5349 | The ``priority`` parameter allows to override the global default |
| 5350 | priority used by gnutls. This can be useful if the system |
| 5351 | administrator needs to use a weaker set of crypto priorities for |
| 5352 | QEMU without potentially forcing the weakness onto all |
| 5353 | applications. Or conversely if one wants wants a stronger |
| 5354 | default for QEMU than for all other applications, they can do |
| 5355 | this through this parameter. Its format is a gnutls priority |
| 5356 | string as described at |
| 5357 | https://gnutls.org/manual/html_node/Priority-Strings.html. |
| 5358 | |
Philippe Mathieu-Daudé | 69699f3 | 2020-05-14 15:15:47 +0200 | [diff] [blame] | 5359 | An example of use of this object is to control UEFI HTTPS Boot. |
| 5360 | The tls-cipher-suites object exposes the ordered list of permitted |
| 5361 | TLS cipher suites from the host side to the guest firmware, via |
| 5362 | fw_cfg. The list is represented as an array of IANA_TLS_CIPHER |
| 5363 | objects. The firmware uses the IANA_TLS_CIPHER array for configuring |
| 5364 | guest-side TLS. |
| 5365 | |
| 5366 | In the following example, the priority at which the host-side policy |
| 5367 | is retrieved is given by the ``priority`` property. |
| 5368 | Given that QEMU uses GNUTLS, ``priority=@SYSTEM`` may be used to |
| 5369 | refer to /etc/crypto-policies/back-ends/gnutls.config. |
| 5370 | |
| 5371 | .. parsed-literal:: |
| 5372 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5373 | # |qemu_system| \\ |
| 5374 | -object tls-cipher-suites,id=mysuite0,priority=@SYSTEM \\ |
Philippe Mathieu-Daudé | 69699f3 | 2020-05-14 15:15:47 +0200 | [diff] [blame] | 5375 | -fw_cfg name=etc/edk2/https/ciphers,gen_id=mysuite0 |
| 5376 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5377 | ``-object filter-buffer,id=id,netdev=netdevid,interval=t[,queue=all|rx|tx][,status=on|off][,position=head|tail|id=<id>][,insert=behind|before]`` |
| 5378 | Interval t can't be 0, this filter batches the packet delivery: |
| 5379 | all packets arriving in a given interval on netdev netdevid are |
| 5380 | delayed until the end of the interval. Interval is in |
| 5381 | microseconds. ``status`` is optional that indicate whether the |
| 5382 | netfilter is on (enabled) or off (disabled), the default status |
| 5383 | for netfilter will be 'on'. |
| 5384 | |
| 5385 | queue all\|rx\|tx is an option that can be applied to any |
| 5386 | netfilter. |
| 5387 | |
| 5388 | ``all``: the filter is attached both to the receive and the |
| 5389 | transmit queue of the netdev (default). |
| 5390 | |
| 5391 | ``rx``: the filter is attached to the receive queue of the |
| 5392 | netdev, where it will receive packets sent to the netdev. |
| 5393 | |
| 5394 | ``tx``: the filter is attached to the transmit queue of the |
| 5395 | netdev, where it will receive packets sent by the netdev. |
| 5396 | |
| 5397 | position head\|tail\|id=<id> is an option to specify where the |
| 5398 | filter should be inserted in the filter list. It can be applied |
| 5399 | to any netfilter. |
| 5400 | |
| 5401 | ``head``: the filter is inserted at the head of the filter list, |
| 5402 | before any existing filters. |
| 5403 | |
| 5404 | ``tail``: the filter is inserted at the tail of the filter list, |
| 5405 | behind any existing filters (default). |
| 5406 | |
| 5407 | ``id=<id>``: the filter is inserted before or behind the filter |
| 5408 | specified by <id>, see the insert option below. |
| 5409 | |
| 5410 | insert behind\|before is an option to specify where to insert |
| 5411 | the new filter relative to the one specified with |
| 5412 | position=id=<id>. It can be applied to any netfilter. |
| 5413 | |
| 5414 | ``before``: insert before the specified filter. |
| 5415 | |
| 5416 | ``behind``: insert behind the specified filter (default). |
| 5417 | |
| 5418 | ``-object filter-mirror,id=id,netdev=netdevid,outdev=chardevid,queue=all|rx|tx[,vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` |
| 5419 | filter-mirror on netdev netdevid,mirror net packet to |
| 5420 | chardevchardevid, if it has the vnet\_hdr\_support flag, |
| 5421 | filter-mirror will mirror packet with vnet\_hdr\_len. |
| 5422 | |
| 5423 | ``-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]`` |
| 5424 | filter-redirector on netdev netdevid,redirect filter's net |
| 5425 | packet to chardev chardevid,and redirect indev's packet to |
| 5426 | filter.if it has the vnet\_hdr\_support flag, filter-redirector |
| 5427 | will redirect packet with vnet\_hdr\_len. Create a |
| 5428 | filter-redirector we need to differ outdev id from indev id, id |
| 5429 | can not be the same. we can just use indev or outdev, but at |
| 5430 | least one of indev or outdev need to be specified. |
| 5431 | |
| 5432 | ``-object filter-rewriter,id=id,netdev=netdevid,queue=all|rx|tx,[vnet_hdr_support][,position=head|tail|id=<id>][,insert=behind|before]`` |
| 5433 | Filter-rewriter is a part of COLO project.It will rewrite tcp |
| 5434 | packet to secondary from primary to keep secondary tcp |
| 5435 | connection,and rewrite tcp packet to primary from secondary make |
| 5436 | tcp packet can be handled by client.if it has the |
| 5437 | vnet\_hdr\_support flag, we can parse packet with vnet header. |
| 5438 | |
| 5439 | usage: colo secondary: -object |
| 5440 | filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 -object |
| 5441 | filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 -object |
| 5442 | filter-rewriter,id=rew0,netdev=hn0,queue=all |
| 5443 | |
| 5444 | ``-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=len][,position=head|tail|id=<id>][,insert=behind|before]`` |
| 5445 | Dump the network traffic on netdev dev to the file specified by |
| 5446 | filename. At most len bytes (64k by default) per packet are |
| 5447 | stored. The file format is libpcap, so it can be analyzed with |
| 5448 | tools such as tcpdump or Wireshark. |
| 5449 | |
Zhang Chen | a2e5cb7 | 2020-06-24 09:20:41 +0800 | [diff] [blame] | 5450 | ``-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}]`` |
Zhang Chen | 2b28a7e | 2020-06-24 09:20:42 +0800 | [diff] [blame] | 5451 | Colo-compare gets packet from primary\_in chardevid and |
| 5452 | secondary\_in, then compare whether the payload of primary packet |
| 5453 | and secondary packet are the same. If same, it will output |
| 5454 | primary packet to out\_dev, else it will notify COLO-framework to do |
| 5455 | checkpoint and send primary packet to out\_dev. In order to |
| 5456 | improve efficiency, we need to put the task of comparison in |
| 5457 | another iothread. If it has the vnet\_hdr\_support flag, |
| 5458 | colo compare will send/recv packet with vnet\_hdr\_len. |
| 5459 | The compare\_timeout=@var{ms} determines the maximum time of the |
| 5460 | colo-compare hold the packet. The expired\_scan\_cycle=@var{ms} |
| 5461 | is to set the period of scanning expired primary node network packets. |
| 5462 | The max\_queue\_size=@var{size} is to set the max compare queue |
| 5463 | size depend on user environment. |
| 5464 | If user want to use Xen COLO, need to add the notify\_dev to |
Zhang Chen | 9cc43c9 | 2020-03-18 16:23:19 +0800 | [diff] [blame] | 5465 | notify Xen colo-frame to do checkpoint. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5466 | |
Zhang Chen | 2b28a7e | 2020-06-24 09:20:42 +0800 | [diff] [blame] | 5467 | COLO-compare must be used with the help of filter-mirror, |
| 5468 | filter-redirector and filter-rewriter. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5469 | |
| 5470 | :: |
| 5471 | |
| 5472 | KVM COLO |
| 5473 | |
| 5474 | primary: |
Michael Tokarev | 7aa94e5 | 2024-01-07 14:24:59 +0300 | [diff] [blame] | 5475 | -netdev tap,id=hn0,vhost=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5476 | -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 5477 | -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off |
| 5478 | -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off |
| 5479 | -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5480 | -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 5481 | -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5482 | -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 |
| 5483 | -object iothread,id=iothread1 |
| 5484 | -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 |
| 5485 | -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out |
| 5486 | -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 |
| 5487 | -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,iothread=iothread1 |
| 5488 | |
| 5489 | secondary: |
Michael Tokarev | 7aa94e5 | 2024-01-07 14:24:59 +0300 | [diff] [blame] | 5490 | -netdev tap,id=hn0,vhost=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5491 | -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 |
| 5492 | -chardev socket,id=red0,host=3.3.3.3,port=9003 |
| 5493 | -chardev socket,id=red1,host=3.3.3.3,port=9004 |
| 5494 | -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 |
| 5495 | -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 |
| 5496 | |
| 5497 | |
| 5498 | Xen COLO |
| 5499 | |
| 5500 | primary: |
Michael Tokarev | 7aa94e5 | 2024-01-07 14:24:59 +0300 | [diff] [blame] | 5501 | -netdev tap,id=hn0,vhost=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5502 | -device e1000,id=e0,netdev=hn0,mac=52:a4:00:12:78:66 |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 5503 | -chardev socket,id=mirror0,host=3.3.3.3,port=9003,server=on,wait=off |
| 5504 | -chardev socket,id=compare1,host=3.3.3.3,port=9004,server=on,wait=off |
| 5505 | -chardev socket,id=compare0,host=3.3.3.3,port=9001,server=on,wait=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5506 | -chardev socket,id=compare0-0,host=3.3.3.3,port=9001 |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 5507 | -chardev socket,id=compare_out,host=3.3.3.3,port=9005,server=on,wait=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5508 | -chardev socket,id=compare_out0,host=3.3.3.3,port=9005 |
Daniel P. Berrangé | bfdc126 | 2021-02-16 19:10:19 +0000 | [diff] [blame] | 5509 | -chardev socket,id=notify_way,host=3.3.3.3,port=9009,server=on,wait=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5510 | -object filter-mirror,id=m0,netdev=hn0,queue=tx,outdev=mirror0 |
| 5511 | -object filter-redirector,netdev=hn0,id=redire0,queue=rx,indev=compare_out |
| 5512 | -object filter-redirector,netdev=hn0,id=redire1,queue=rx,outdev=compare0 |
| 5513 | -object iothread,id=iothread1 |
Manos Pitsidianakis | 835f364 | 2024-02-20 10:52:23 +0200 | [diff] [blame] | 5514 | -object colo-compare,id=comp0,primary_in=compare0-0,secondary_in=compare1,outdev=compare_out0,notify_dev=notify_way,iothread=iothread1 |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5515 | |
| 5516 | secondary: |
Michael Tokarev | 7aa94e5 | 2024-01-07 14:24:59 +0300 | [diff] [blame] | 5517 | -netdev tap,id=hn0,vhost=off |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5518 | -device e1000,netdev=hn0,mac=52:a4:00:12:78:66 |
| 5519 | -chardev socket,id=red0,host=3.3.3.3,port=9003 |
| 5520 | -chardev socket,id=red1,host=3.3.3.3,port=9004 |
| 5521 | -object filter-redirector,id=f1,netdev=hn0,queue=tx,indev=red0 |
| 5522 | -object filter-redirector,id=f2,netdev=hn0,queue=rx,outdev=red1 |
| 5523 | |
| 5524 | If you want to know the detail of above command line, you can |
| 5525 | read the colo-compare git log. |
| 5526 | |
| 5527 | ``-object cryptodev-backend-builtin,id=id[,queues=queues]`` |
Stefan Weil | 1e458f1 | 2022-10-30 11:59:44 +0100 | [diff] [blame] | 5528 | Creates a cryptodev backend which executes crypto operations from |
| 5529 | the QEMU cipher APIs. The id parameter is a unique ID that will |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5530 | be used to reference this cryptodev backend from the |
| 5531 | ``virtio-crypto`` device. The queues parameter is optional, |
| 5532 | which specify the queue number of cryptodev backend, the default |
| 5533 | of queues is 1. |
| 5534 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5535 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5536 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5537 | # |qemu_system| \\ |
| 5538 | [...] \\ |
| 5539 | -object cryptodev-backend-builtin,id=cryptodev0 \\ |
| 5540 | -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5541 | [...] |
| 5542 | |
| 5543 | ``-object cryptodev-vhost-user,id=id,chardev=chardevid[,queues=queues]`` |
| 5544 | Creates a vhost-user cryptodev backend, backed by a chardev |
| 5545 | chardevid. The id parameter is a unique ID that will be used to |
| 5546 | reference this cryptodev backend from the ``virtio-crypto`` |
| 5547 | device. The chardev should be a unix domain socket backed one. |
| 5548 | The vhost-user uses a specifically defined protocol to pass |
| 5549 | vhost ioctl replacement messages to an application on the other |
| 5550 | end of the socket. The queues parameter is optional, which |
| 5551 | specify the queue number of cryptodev backend for multiqueue |
| 5552 | vhost-user, the default of queues is 1. |
| 5553 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5554 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5555 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5556 | # |qemu_system| \\ |
| 5557 | [...] \\ |
| 5558 | -chardev socket,id=chardev0,path=/path/to/socket \\ |
| 5559 | -object cryptodev-vhost-user,id=cryptodev0,chardev=chardev0 \\ |
| 5560 | -device virtio-crypto-pci,id=crypto0,cryptodev=cryptodev0 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5561 | [...] |
| 5562 | |
| 5563 | ``-object secret,id=id,data=string,format=raw|base64[,keyid=secretid,iv=string]`` |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5564 | \ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5565 | ``-object secret,id=id,file=filename,format=raw|base64[,keyid=secretid,iv=string]`` |
| 5566 | Defines a secret to store a password, encryption key, or some |
| 5567 | other sensitive data. The sensitive data can either be passed |
| 5568 | directly via the data parameter, or indirectly via the file |
| 5569 | parameter. Using the data parameter is insecure unless the |
| 5570 | sensitive data is encrypted. |
| 5571 | |
| 5572 | The sensitive data can be provided in raw format (the default), |
| 5573 | or base64. When encoded as JSON, the raw format only supports |
| 5574 | valid UTF-8 characters, so base64 is recommended for sending |
| 5575 | binary data. QEMU will convert from which ever format is |
| 5576 | provided to the format it needs internally. eg, an RBD password |
| 5577 | can be provided in raw format, even though it will be base64 |
| 5578 | encoded when passed onto the RBD sever. |
| 5579 | |
| 5580 | For added protection, it is possible to encrypt the data |
| 5581 | associated with a secret using the AES-256-CBC cipher. Use of |
| 5582 | encryption is indicated by providing the keyid and iv |
| 5583 | parameters. The keyid parameter provides the ID of a previously |
| 5584 | defined secret that contains the AES-256 decryption key. This |
| 5585 | key should be 32-bytes long and be base64 encoded. The iv |
| 5586 | parameter provides the random initialization vector used for |
| 5587 | encryption of this particular secret and should be a base64 |
| 5588 | encrypted string of the 16-byte IV. |
| 5589 | |
| 5590 | The simplest (insecure) usage is to provide the secret inline |
| 5591 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5592 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5593 | |
| 5594 | # |qemu_system| -object secret,id=sec0,data=letmein,format=raw |
| 5595 | |
| 5596 | The simplest secure usage is to provide the secret via a file |
| 5597 | |
| 5598 | # printf "letmein" > mypasswd.txt # QEMU\_SYSTEM\_MACRO -object |
| 5599 | secret,id=sec0,file=mypasswd.txt,format=raw |
| 5600 | |
| 5601 | For greater security, AES-256-CBC should be used. To illustrate |
| 5602 | usage, consider the openssl command line tool which can encrypt |
| 5603 | the data. Note that when encrypting, the plaintext must be |
| 5604 | padded to the cipher block size (32 bytes) using the standard |
| 5605 | PKCS#5/6 compatible padding algorithm. |
| 5606 | |
| 5607 | First a master key needs to be created in base64 encoding: |
| 5608 | |
| 5609 | :: |
| 5610 | |
| 5611 | # openssl rand -base64 32 > key.b64 |
| 5612 | # KEY=$(base64 -d key.b64 | hexdump -v -e '/1 "%02X"') |
| 5613 | |
| 5614 | Each secret to be encrypted needs to have a random |
| 5615 | initialization vector generated. These do not need to be kept |
| 5616 | secret |
| 5617 | |
| 5618 | :: |
| 5619 | |
| 5620 | # openssl rand -base64 16 > iv.b64 |
| 5621 | # IV=$(base64 -d iv.b64 | hexdump -v -e '/1 "%02X"') |
| 5622 | |
| 5623 | The secret to be defined can now be encrypted, in this case |
| 5624 | we're telling openssl to base64 encode the result, but it could |
| 5625 | be left as raw bytes if desired. |
| 5626 | |
| 5627 | :: |
| 5628 | |
| 5629 | # SECRET=$(printf "letmein" | |
| 5630 | openssl enc -aes-256-cbc -a -K $KEY -iv $IV) |
| 5631 | |
| 5632 | When launching QEMU, create a master secret pointing to |
| 5633 | ``key.b64`` and specify that to be used to decrypt the user |
| 5634 | password. Pass the contents of ``iv.b64`` to the second secret |
| 5635 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5636 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5637 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5638 | # |qemu_system| \\ |
| 5639 | -object secret,id=secmaster0,format=base64,file=key.b64 \\ |
| 5640 | -object secret,id=sec0,keyid=secmaster0,format=base64,\\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5641 | data=$SECRET,iv=$(<iv.b64) |
| 5642 | |
Dov Murik | 55cdf56 | 2021-11-11 10:00:43 +0000 | [diff] [blame] | 5643 | ``-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]`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5644 | Create a Secure Encrypted Virtualization (SEV) guest object, |
| 5645 | which can be used to provide the guest memory encryption support |
| 5646 | on AMD processors. |
| 5647 | |
| 5648 | When memory encryption is enabled, one of the physical address |
| 5649 | bit (aka the C-bit) is utilized to mark if a memory page is |
| 5650 | protected. The ``cbitpos`` is used to provide the C-bit |
| 5651 | position. The C-bit position is Host family dependent hence user |
| 5652 | must provide this value. On EPYC, the value should be 47. |
| 5653 | |
| 5654 | When memory encryption is enabled, we loose certain bits in |
| 5655 | physical address space. The ``reduced-phys-bits`` is used to |
| 5656 | provide the number of bits we loose in physical address space. |
| 5657 | Similar to C-bit, the value is Host family dependent. On EPYC, |
Tom Lendacky | 326e301 | 2022-09-30 10:14:28 -0500 | [diff] [blame] | 5658 | a guest will lose a maximum of 1 bit, so the value should be 1. |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5659 | |
| 5660 | The ``sev-device`` provides the device file to use for |
| 5661 | communicating with the SEV firmware running inside AMD Secure |
| 5662 | Processor. The default device is '/dev/sev'. If hardware |
| 5663 | supports memory encryption then /dev/sev devices are created by |
| 5664 | CCP driver. |
| 5665 | |
| 5666 | The ``policy`` provides the guest policy to be enforced by the |
| 5667 | SEV firmware and restrict what configuration and operational |
| 5668 | commands can be performed on this guest by the hypervisor. The |
| 5669 | policy should be provided by the guest owner and is bound to the |
| 5670 | guest and cannot be changed throughout the lifetime of the |
| 5671 | guest. The default is 0. |
| 5672 | |
| 5673 | If guest ``policy`` allows sharing the key with another SEV |
| 5674 | guest then ``handle`` can be use to provide handle of the guest |
| 5675 | from which to share the key. |
| 5676 | |
| 5677 | The ``dh-cert-file`` and ``session-file`` provides the guest |
| 5678 | owner's Public Diffie-Hillman key defined in SEV spec. The PDH |
| 5679 | and session parameters are used for establishing a cryptographic |
| 5680 | session with the guest owner to negotiate keys used for |
| 5681 | attestation. The file must be encoded in base64. |
| 5682 | |
Dov Murik | 55cdf56 | 2021-11-11 10:00:43 +0000 | [diff] [blame] | 5683 | The ``kernel-hashes`` adds the hashes of given kernel/initrd/ |
| 5684 | cmdline to a designated guest firmware page for measured Linux |
| 5685 | boot with -kernel. The default is off. (Since 6.2) |
| 5686 | |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5687 | e.g to launch a SEV guest |
| 5688 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5689 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5690 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5691 | # |qemu_system_x86| \\ |
| 5692 | ...... \\ |
Tom Lendacky | 326e301 | 2022-09-30 10:14:28 -0500 | [diff] [blame] | 5693 | -object sev-guest,id=sev0,cbitpos=47,reduced-phys-bits=1 \\ |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5694 | -machine ...,memory-encryption=sev0 \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5695 | ..... |
| 5696 | |
| 5697 | ``-object authz-simple,id=id,identity=string`` |
| 5698 | Create an authorization object that will control access to |
| 5699 | network services. |
| 5700 | |
| 5701 | The ``identity`` parameter is identifies the user and its format |
| 5702 | depends on the network service that authorization object is |
| 5703 | associated with. For authorizing based on TLS x509 certificates, |
| 5704 | the identity must be the x509 distinguished name. Note that care |
| 5705 | must be taken to escape any commas in the distinguished name. |
| 5706 | |
| 5707 | An example authorization object to validate a x509 distinguished |
| 5708 | name would look like: |
| 5709 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5710 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5711 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5712 | # |qemu_system| \\ |
| 5713 | ... \\ |
| 5714 | -object 'authz-simple,id=auth0,identity=CN=laptop.example.com,,O=Example Org,,L=London,,ST=London,,C=GB' \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5715 | ... |
| 5716 | |
| 5717 | Note the use of quotes due to the x509 distinguished name |
| 5718 | containing whitespace, and escaping of ','. |
| 5719 | |
Daniel P. Berrangé | 4d7beea | 2020-11-04 13:57:21 +0000 | [diff] [blame] | 5720 | ``-object authz-listfile,id=id,filename=path,refresh=on|off`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5721 | Create an authorization object that will control access to |
| 5722 | network services. |
| 5723 | |
| 5724 | The ``filename`` parameter is the fully qualified path to a file |
| 5725 | containing the access control list rules in JSON format. |
| 5726 | |
| 5727 | An example set of rules that match against SASL usernames might |
| 5728 | look like: |
| 5729 | |
| 5730 | :: |
| 5731 | |
| 5732 | { |
| 5733 | "rules": [ |
| 5734 | { "match": "fred", "policy": "allow", "format": "exact" }, |
| 5735 | { "match": "bob", "policy": "allow", "format": "exact" }, |
| 5736 | { "match": "danb", "policy": "deny", "format": "glob" }, |
| 5737 | { "match": "dan*", "policy": "allow", "format": "exact" }, |
| 5738 | ], |
| 5739 | "policy": "deny" |
| 5740 | } |
| 5741 | |
| 5742 | When checking access the object will iterate over all the rules |
| 5743 | and the first rule to match will have its ``policy`` value |
| 5744 | returned as the result. If no rules match, then the default |
| 5745 | ``policy`` value is returned. |
| 5746 | |
| 5747 | The rules can either be an exact string match, or they can use |
| 5748 | the simple UNIX glob pattern matching to allow wildcards to be |
| 5749 | used. |
| 5750 | |
| 5751 | If ``refresh`` is set to true the file will be monitored and |
| 5752 | automatically reloaded whenever its content changes. |
| 5753 | |
| 5754 | As with the ``authz-simple`` object, the format of the identity |
| 5755 | strings being matched depends on the network service, but is |
| 5756 | usually a TLS x509 distinguished name, or a SASL username. |
| 5757 | |
| 5758 | An example authorization object to validate a SASL username |
| 5759 | would look like: |
| 5760 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5761 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5762 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5763 | # |qemu_system| \\ |
| 5764 | ... \\ |
Daniel P. Berrangé | 4d7beea | 2020-11-04 13:57:21 +0000 | [diff] [blame] | 5765 | -object authz-simple,id=auth0,filename=/etc/qemu/vnc-sasl.acl,refresh=on \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5766 | ... |
| 5767 | |
| 5768 | ``-object authz-pam,id=id,service=string`` |
| 5769 | Create an authorization object that will control access to |
| 5770 | network services. |
| 5771 | |
| 5772 | The ``service`` parameter provides the name of a PAM service to |
| 5773 | use for authorization. It requires that a file |
| 5774 | ``/etc/pam.d/service`` exist to provide the configuration for |
| 5775 | the ``account`` subsystem. |
| 5776 | |
| 5777 | An example authorization object to validate a TLS x509 |
| 5778 | distinguished name would look like: |
| 5779 | |
Peter Maydell | 09ce5f2 | 2020-02-28 15:36:13 +0000 | [diff] [blame] | 5780 | .. parsed-literal:: |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5781 | |
Laszlo Ersek | 353a06b | 2020-09-08 19:21:11 +0200 | [diff] [blame] | 5782 | # |qemu_system| \\ |
| 5783 | ... \\ |
| 5784 | -object authz-pam,id=auth0,service=qemu-vnc \\ |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5785 | ... |
| 5786 | |
| 5787 | There would then be a corresponding config file for PAM at |
| 5788 | ``/etc/pam.d/qemu-vnc`` that contains: |
| 5789 | |
| 5790 | :: |
| 5791 | |
| 5792 | account requisite pam_listfile.so item=user sense=allow \ |
| 5793 | file=/etc/qemu/vnc.allow |
| 5794 | |
| 5795 | Finally the ``/etc/qemu/vnc.allow`` file would contain the list |
Stefan Weil | 1e458f1 | 2022-10-30 11:59:44 +0100 | [diff] [blame] | 5796 | of x509 distinguished names that are permitted access |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5797 | |
| 5798 | :: |
| 5799 | |
| 5800 | CN=laptop.example.com,O=Example Home,L=London,ST=London,C=GB |
| 5801 | |
Stefano Garzarella | 1793ad0 | 2021-07-21 11:42:10 +0200 | [diff] [blame] | 5802 | ``-object iothread,id=id,poll-max-ns=poll-max-ns,poll-grow=poll-grow,poll-shrink=poll-shrink,aio-max-batch=aio-max-batch`` |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5803 | Creates a dedicated event loop thread that devices can be |
| 5804 | assigned to. This is known as an IOThread. By default device |
| 5805 | emulation happens in vCPU threads or the main event loop thread. |
| 5806 | This can become a scalability bottleneck. IOThreads allow device |
| 5807 | emulation and I/O to run on other host CPUs. |
| 5808 | |
| 5809 | The ``id`` parameter is a unique ID that will be used to |
| 5810 | reference this IOThread from ``-device ...,iothread=id``. |
| 5811 | Multiple devices can be assigned to an IOThread. Note that not |
| 5812 | all devices support an ``iothread`` parameter. |
| 5813 | |
| 5814 | The ``query-iothreads`` QMP command lists IOThreads and reports |
| 5815 | their thread IDs so that the user can configure host CPU |
| 5816 | pinning/affinity. |
| 5817 | |
| 5818 | IOThreads use an adaptive polling algorithm to reduce event loop |
| 5819 | latency. Instead of entering a blocking system call to monitor |
| 5820 | file descriptors and then pay the cost of being woken up when an |
| 5821 | event occurs, the polling algorithm spins waiting for events for |
| 5822 | a short time. The algorithm's default parameters are suitable |
| 5823 | for many cases but can be adjusted based on knowledge of the |
| 5824 | workload and/or host device latency. |
| 5825 | |
| 5826 | The ``poll-max-ns`` parameter is the maximum number of |
| 5827 | nanoseconds to busy wait for events. Polling can be disabled by |
| 5828 | setting this value to 0. |
| 5829 | |
| 5830 | The ``poll-grow`` parameter is the multiplier used to increase |
| 5831 | the polling time when the algorithm detects it is missing events |
| 5832 | due to not polling long enough. |
| 5833 | |
| 5834 | The ``poll-shrink`` parameter is the divisor used to decrease |
| 5835 | the polling time when the algorithm detects it is spending too |
| 5836 | long polling without encountering events. |
| 5837 | |
Stefano Garzarella | 1793ad0 | 2021-07-21 11:42:10 +0200 | [diff] [blame] | 5838 | The ``aio-max-batch`` parameter is the maximum number of requests |
| 5839 | in a batch for the AIO engine, 0 means that the engine will use |
| 5840 | its default. |
| 5841 | |
| 5842 | The IOThread parameters can be modified at run-time using the |
Peter Maydell | e2fcbf4 | 2020-03-06 10:07:14 +0000 | [diff] [blame] | 5843 | ``qom-set`` command (where ``iothread1`` is the IOThread's |
| 5844 | ``id``): |
| 5845 | |
| 5846 | :: |
| 5847 | |
| 5848 | (qemu) qom-set /objects/iothread1 poll-max-ns 100000 |
| 5849 | ERST |
Daniel P. Berrange | b9174d4 | 2015-05-13 17:14:03 +0100 | [diff] [blame] | 5850 | |
| 5851 | |
Stefan Weil | 3dbf2c7 | 2010-01-16 18:19:44 +0100 | [diff] [blame] | 5852 | HXCOMM This is the last statement. Insert new options before this line! |
Paolo Bonzini | fd5fc4b | 2021-05-17 07:34:21 -0400 | [diff] [blame] | 5853 | |
| 5854 | #undef DEF |
| 5855 | #undef DEFHEADING |
| 5856 | #undef ARCHHEADING |