1 \input texinfo @c -*- texinfo -*-
3 @setfilename qemu-doc.info
6 @documentencoding UTF-8
8 @settitle QEMU Emulator User Documentation
15 * QEMU: (qemu-doc). The QEMU Emulator User Documentation.
22 @center @titlefont{QEMU Emulator}
24 @center @titlefont{User Documentation}
36 * QEMU PC System emulator::
37 * QEMU System emulator for non PC targets::
38 * QEMU User space emulator::
39 * compilation:: Compilation from the sources
51 * intro_features:: Features
57 QEMU is a FAST! processor emulator using dynamic translation to
58 achieve good emulation speed.
60 QEMU has two operating modes:
63 @cindex operating modes
66 @cindex system emulation
67 Full system emulation. In this mode, QEMU emulates a full system (for
68 example a PC), including one or several processors and various
69 peripherals. It can be used to launch different Operating Systems
70 without rebooting the PC or to debug system code.
73 @cindex user mode emulation
74 User mode emulation. In this mode, QEMU can launch
75 processes compiled for one CPU on another CPU. It can be used to
76 launch the Wine Windows API emulator (@url{http://www.winehq.org}) or
77 to ease cross-compilation and cross-debugging.
81 QEMU can run without a host kernel driver and yet gives acceptable
84 For system emulation, the following hardware targets are supported:
86 @cindex emulated target systems
87 @cindex supported target systems
88 @item PC (x86 or x86_64 processor)
89 @item ISA PC (old style PC without PCI bus)
90 @item PREP (PowerPC processor)
91 @item G3 Beige PowerMac (PowerPC processor)
92 @item Mac99 PowerMac (PowerPC processor, in progress)
93 @item Sun4m/Sun4c/Sun4d (32-bit Sparc processor)
94 @item Sun4u/Sun4v (64-bit Sparc processor, in progress)
95 @item Malta board (32-bit and 64-bit MIPS processors)
96 @item MIPS Magnum (64-bit MIPS processor)
97 @item ARM Integrator/CP (ARM)
98 @item ARM Versatile baseboard (ARM)
99 @item ARM RealView Emulation/Platform baseboard (ARM)
100 @item Spitz, Akita, Borzoi, Terrier and Tosa PDAs (PXA270 processor)
101 @item Luminary Micro LM3S811EVB (ARM Cortex-M3)
102 @item Luminary Micro LM3S6965EVB (ARM Cortex-M3)
103 @item Freescale MCF5208EVB (ColdFire V2).
104 @item Arnewsh MCF5206 evaluation board (ColdFire V2).
105 @item Palm Tungsten|E PDA (OMAP310 processor)
106 @item N800 and N810 tablets (OMAP2420 processor)
107 @item MusicPal (MV88W8618 ARM processor)
108 @item Gumstix "Connex" and "Verdex" motherboards (PXA255/270).
109 @item Siemens SX1 smartphone (OMAP310 processor)
110 @item AXIS-Devboard88 (CRISv32 ETRAX-FS).
111 @item Petalogix Spartan 3aDSP1800 MMU ref design (MicroBlaze).
112 @item Avnet LX60/LX110/LX200 boards (Xtensa)
115 @cindex supported user mode targets
116 For user emulation, x86 (32 and 64 bit), PowerPC (32 and 64 bit),
117 ARM, MIPS (32 bit only), Sparc (32 and 64 bit),
118 Alpha, ColdFire(m68k), CRISv32 and MicroBlaze CPUs are supported.
121 @chapter Installation
123 If you want to compile QEMU yourself, see @ref{compilation}.
126 * install_linux:: Linux
127 * install_windows:: Windows
128 * install_mac:: Macintosh
133 @cindex installation (Linux)
135 If a precompiled package is available for your distribution - you just
136 have to install it. Otherwise, see @ref{compilation}.
138 @node install_windows
140 @cindex installation (Windows)
142 Download the experimental binary installer at
143 @url{http://www.free.oszoo.org/@/download.html}.
144 TODO (no longer available)
149 Download the experimental binary installer at
150 @url{http://www.free.oszoo.org/@/download.html}.
151 TODO (no longer available)
153 @node QEMU PC System emulator
154 @chapter QEMU PC System emulator
155 @cindex system emulation (PC)
158 * pcsys_introduction:: Introduction
159 * pcsys_quickstart:: Quick Start
160 * sec_invocation:: Invocation
162 * pcsys_monitor:: QEMU Monitor
163 * disk_images:: Disk Images
164 * pcsys_network:: Network emulation
165 * pcsys_other_devs:: Other Devices
166 * direct_linux_boot:: Direct Linux Boot
167 * pcsys_usb:: USB emulation
168 * vnc_security:: VNC security
169 * gdb_usage:: GDB usage
170 * pcsys_os_specific:: Target OS specific information
173 @node pcsys_introduction
174 @section Introduction
176 @c man begin DESCRIPTION
178 The QEMU PC System emulator simulates the
179 following peripherals:
183 i440FX host PCI bridge and PIIX3 PCI to ISA bridge
185 Cirrus CLGD 5446 PCI VGA card or dummy VGA card with Bochs VESA
186 extensions (hardware level, including all non standard modes).
188 PS/2 mouse and keyboard
190 2 PCI IDE interfaces with hard disk and CD-ROM support
194 PCI and ISA network adapters
198 Creative SoundBlaster 16 sound card
200 ENSONIQ AudioPCI ES1370 sound card
202 Intel 82801AA AC97 Audio compatible sound card
204 Intel HD Audio Controller and HDA codec
206 Adlib (OPL2) - Yamaha YM3812 compatible chip
208 Gravis Ultrasound GF1 sound card
210 CS4231A compatible sound card
212 PCI UHCI USB controller and a virtual USB hub.
215 SMP is supported with up to 255 CPUs.
217 QEMU uses the PC BIOS from the Seabios project and the Plex86/Bochs LGPL
220 QEMU uses YM3812 emulation by Tatsuyuki Satoh.
222 QEMU uses GUS emulation (GUSEMU32 @url{http://www.deinmeister.de/gusemu/})
223 by Tibor "TS" Schütz.
225 Note that, by default, GUS shares IRQ(7) with parallel ports and so
226 QEMU must be told to not have parallel ports to have working GUS.
229 qemu-system-i386 dos.img -soundhw gus -parallel none
234 qemu-system-i386 dos.img -device gus,irq=5
237 Or some other unclaimed IRQ.
239 CS4231A is the chip used in Windows Sound System and GUSMAX products
243 @node pcsys_quickstart
247 Download and uncompress the linux image (@file{linux.img}) and type:
250 qemu-system-i386 linux.img
253 Linux should boot and give you a prompt.
259 @c man begin SYNOPSIS
260 usage: qemu-system-i386 [options] [@var{disk_image}]
265 @var{disk_image} is a raw hard disk image for IDE hard disk 0. Some
266 targets do not need a disk image.
268 @include qemu-options.texi
277 During the graphical emulation, you can use special key combinations to change
278 modes. The default key mappings are shown below, but if you use @code{-alt-grab}
279 then the modifier is Ctrl-Alt-Shift (instead of Ctrl-Alt) and if you use
280 @code{-ctrl-grab} then the modifier is the right Ctrl key (instead of Ctrl-Alt):
297 Restore the screen's un-scaled dimensions
301 Switch to virtual console 'n'. Standard console mappings are:
304 Target system display
313 Toggle mouse and keyboard grab.
319 @kindex Ctrl-PageDown
320 In the virtual consoles, you can use @key{Ctrl-Up}, @key{Ctrl-Down},
321 @key{Ctrl-PageUp} and @key{Ctrl-PageDown} to move in the back log.
324 During emulation, if you are using the @option{-nographic} option, use
325 @key{Ctrl-a h} to get terminal commands:
338 Save disk data back to file (if -snapshot)
341 Toggle console timestamps
344 Send break (magic sysrq in Linux)
347 Switch between console and monitor
357 The HTML documentation of QEMU for more precise information and Linux
358 user mode emulator invocation.
368 @section QEMU Monitor
371 The QEMU monitor is used to give complex commands to the QEMU
372 emulator. You can use it to:
377 Remove or insert removable media images
378 (such as CD-ROM or floppies).
381 Freeze/unfreeze the Virtual Machine (VM) and save or restore its state
384 @item Inspect the VM state without an external debugger.
390 The following commands are available:
392 @include qemu-monitor.texi
394 @include qemu-monitor-info.texi
396 @subsection Integer expressions
398 The monitor understands integers expressions for every integer
399 argument. You can use register names to get the value of specifics
400 CPU registers by prefixing them with @emph{$}.
405 Since version 0.6.1, QEMU supports many disk image formats, including
406 growable disk images (their size increase as non empty sectors are
407 written), compressed and encrypted disk images. Version 0.8.3 added
408 the new qcow2 disk image format which is essential to support VM
412 * disk_images_quickstart:: Quick start for disk image creation
413 * disk_images_snapshot_mode:: Snapshot mode
414 * vm_snapshots:: VM snapshots
415 * qemu_img_invocation:: qemu-img Invocation
416 * qemu_nbd_invocation:: qemu-nbd Invocation
417 * qemu_ga_invocation:: qemu-ga Invocation
418 * disk_images_formats:: Disk image file formats
419 * host_drives:: Using host drives
420 * disk_images_fat_images:: Virtual FAT disk images
421 * disk_images_nbd:: NBD access
422 * disk_images_sheepdog:: Sheepdog disk images
423 * disk_images_iscsi:: iSCSI LUNs
424 * disk_images_gluster:: GlusterFS disk images
425 * disk_images_ssh:: Secure Shell (ssh) disk images
428 @node disk_images_quickstart
429 @subsection Quick start for disk image creation
431 You can create a disk image with the command:
433 qemu-img create myimage.img mysize
435 where @var{myimage.img} is the disk image filename and @var{mysize} is its
436 size in kilobytes. You can add an @code{M} suffix to give the size in
437 megabytes and a @code{G} suffix for gigabytes.
439 See @ref{qemu_img_invocation} for more information.
441 @node disk_images_snapshot_mode
442 @subsection Snapshot mode
444 If you use the option @option{-snapshot}, all disk images are
445 considered as read only. When sectors in written, they are written in
446 a temporary file created in @file{/tmp}. You can however force the
447 write back to the raw disk images by using the @code{commit} monitor
448 command (or @key{C-a s} in the serial console).
451 @subsection VM snapshots
453 VM snapshots are snapshots of the complete virtual machine including
454 CPU state, RAM, device state and the content of all the writable
455 disks. In order to use VM snapshots, you must have at least one non
456 removable and writable block device using the @code{qcow2} disk image
457 format. Normally this device is the first virtual hard drive.
459 Use the monitor command @code{savevm} to create a new VM snapshot or
460 replace an existing one. A human readable name can be assigned to each
461 snapshot in addition to its numerical ID.
463 Use @code{loadvm} to restore a VM snapshot and @code{delvm} to remove
464 a VM snapshot. @code{info snapshots} lists the available snapshots
465 with their associated information:
468 (qemu) info snapshots
469 Snapshot devices: hda
470 Snapshot list (from hda):
471 ID TAG VM SIZE DATE VM CLOCK
472 1 start 41M 2006-08-06 12:38:02 00:00:14.954
473 2 40M 2006-08-06 12:43:29 00:00:18.633
474 3 msys 40M 2006-08-06 12:44:04 00:00:23.514
477 A VM snapshot is made of a VM state info (its size is shown in
478 @code{info snapshots}) and a snapshot of every writable disk image.
479 The VM state info is stored in the first @code{qcow2} non removable
480 and writable block device. The disk image snapshots are stored in
481 every disk image. The size of a snapshot in a disk image is difficult
482 to evaluate and is not shown by @code{info snapshots} because the
483 associated disk sectors are shared among all the snapshots to save
484 disk space (otherwise each snapshot would need a full copy of all the
487 When using the (unrelated) @code{-snapshot} option
488 (@ref{disk_images_snapshot_mode}), you can always make VM snapshots,
489 but they are deleted as soon as you exit QEMU.
491 VM snapshots currently have the following known limitations:
494 They cannot cope with removable devices if they are removed or
495 inserted after a snapshot is done.
497 A few device drivers still have incomplete snapshot support so their
498 state is not saved or restored properly (in particular USB).
501 @node qemu_img_invocation
502 @subsection @code{qemu-img} Invocation
504 @include qemu-img.texi
506 @node qemu_nbd_invocation
507 @subsection @code{qemu-nbd} Invocation
509 @include qemu-nbd.texi
511 @node qemu_ga_invocation
512 @subsection @code{qemu-ga} Invocation
514 @include qemu-ga.texi
516 @node disk_images_formats
517 @subsection Disk image file formats
519 QEMU supports many image file formats that can be used with VMs as well as with
520 any of the tools (like @code{qemu-img}). This includes the preferred formats
521 raw and qcow2 as well as formats that are supported for compatibility with
522 older QEMU versions or other hypervisors.
524 Depending on the image format, different options can be passed to
525 @code{qemu-img create} and @code{qemu-img convert} using the @code{-o} option.
526 This section describes each format and the options that are supported for it.
531 Raw disk image format. This format has the advantage of
532 being simple and easily exportable to all other emulators. If your
533 file system supports @emph{holes} (for example in ext2 or ext3 on
534 Linux or NTFS on Windows), then only the written sectors will reserve
535 space. Use @code{qemu-img info} to know the real size used by the
536 image or @code{ls -ls} on Unix/Linux.
541 Preallocation mode (allowed values: @code{off}, @code{falloc}, @code{full}).
542 @code{falloc} mode preallocates space for image by calling posix_fallocate().
543 @code{full} mode preallocates space for image by writing zeros to underlying
548 QEMU image format, the most versatile format. Use it to have smaller
549 images (useful if your filesystem does not supports holes, for example
550 on Windows), zlib based compression and support of multiple VM
556 Determines the qcow2 version to use. @code{compat=0.10} uses the
557 traditional image format that can be read by any QEMU since 0.10.
558 @code{compat=1.1} enables image format extensions that only QEMU 1.1 and
559 newer understand (this is the default). Amongst others, this includes
560 zero clusters, which allow efficient copy-on-read for sparse images.
563 File name of a base image (see @option{create} subcommand)
565 Image format of the base image
567 If this option is set to @code{on}, the image is encrypted with 128-bit AES-CBC.
569 The use of encryption in qcow and qcow2 images is considered to be flawed by
570 modern cryptography standards, suffering from a number of design problems:
573 @item The AES-CBC cipher is used with predictable initialization vectors based
574 on the sector number. This makes it vulnerable to chosen plaintext attacks
575 which can reveal the existence of encrypted data.
576 @item The user passphrase is directly used as the encryption key. A poorly
577 chosen or short passphrase will compromise the security of the encryption.
578 @item In the event of the passphrase being compromised there is no way to
579 change the passphrase to protect data in any qcow images. The files must
580 be cloned, using a different encryption passphrase in the new file. The
581 original file must then be securely erased using a program like shred,
582 though even this is ineffective with many modern storage technologies.
585 Use of qcow / qcow2 encryption with QEMU is deprecated, and support for
586 it will go away in a future release. Users are recommended to use an
587 alternative encryption technology such as the Linux dm-crypt / LUKS
591 Changes the qcow2 cluster size (must be between 512 and 2M). Smaller cluster
592 sizes can improve the image file size whereas larger cluster sizes generally
593 provide better performance.
596 Preallocation mode (allowed values: @code{off}, @code{metadata}, @code{falloc},
597 @code{full}). An image with preallocated metadata is initially larger but can
598 improve performance when the image needs to grow. @code{falloc} and @code{full}
599 preallocations are like the same options of @code{raw} format, but sets up
603 If this option is set to @code{on}, reference count updates are postponed with
604 the goal of avoiding metadata I/O and improving performance. This is
605 particularly interesting with @option{cache=writethrough} which doesn't batch
606 metadata updates. The tradeoff is that after a host crash, the reference count
607 tables must be rebuilt, i.e. on the next open an (automatic) @code{qemu-img
608 check -r all} is required, which may take some time.
610 This option can only be enabled if @code{compat=1.1} is specified.
613 If this option is set to @code{on}, it will turn off COW of the file. It's only
614 valid on btrfs, no effect on other file systems.
616 Btrfs has low performance when hosting a VM image file, even more when the guest
617 on the VM also using btrfs as file system. Turning off COW is a way to mitigate
618 this bad performance. Generally there are two ways to turn off COW on btrfs:
619 a) Disable it by mounting with nodatacow, then all newly created files will be
620 NOCOW. b) For an empty file, add the NOCOW file attribute. That's what this option
623 Note: this option is only valid to new or empty files. If there is an existing
624 file which is COW and has data blocks already, it couldn't be changed to NOCOW
625 by setting @code{nocow=on}. One can issue @code{lsattr filename} to check if
626 the NOCOW flag is set or not (Capital 'C' is NOCOW flag).
631 Old QEMU image format with support for backing files and compact image files
632 (when your filesystem or transport medium does not support holes).
634 When converting QED images to qcow2, you might want to consider using the
635 @code{lazy_refcounts=on} option to get a more QED-like behaviour.
640 File name of a base image (see @option{create} subcommand).
642 Image file format of backing file (optional). Useful if the format cannot be
643 autodetected because it has no header, like some vhd/vpc files.
645 Changes the cluster size (must be power-of-2 between 4K and 64K). Smaller
646 cluster sizes can improve the image file size whereas larger cluster sizes
647 generally provide better performance.
649 Changes the number of clusters per L1/L2 table (must be power-of-2 between 1
650 and 16). There is normally no need to change this value but this option can be
651 used for performance benchmarking.
655 Old QEMU image format with support for backing files, compact image files,
656 encryption and compression.
661 File name of a base image (see @option{create} subcommand)
663 If this option is set to @code{on}, the image is encrypted.
667 VirtualBox 1.1 compatible image format.
671 If this option is set to @code{on}, the image is created with metadata
676 VMware 3 and 4 compatible image format.
681 File name of a base image (see @option{create} subcommand).
683 Create a VMDK version 6 image (instead of version 4)
685 Specifies which VMDK subformat to use. Valid options are
686 @code{monolithicSparse} (default),
687 @code{monolithicFlat},
688 @code{twoGbMaxExtentSparse},
689 @code{twoGbMaxExtentFlat} and
690 @code{streamOptimized}.
694 VirtualPC compatible image format (VHD).
698 Specifies which VHD subformat to use. Valid options are
699 @code{dynamic} (default) and @code{fixed}.
703 Hyper-V compatible image format (VHDX).
707 Specifies which VHDX subformat to use. Valid options are
708 @code{dynamic} (default) and @code{fixed}.
709 @item block_state_zero
710 Force use of payload blocks of type 'ZERO'. Can be set to @code{on} (default)
711 or @code{off}. When set to @code{off}, new blocks will be created as
712 @code{PAYLOAD_BLOCK_NOT_PRESENT}, which means parsers are free to return
713 arbitrary data for those blocks. Do not set to @code{off} when using
714 @code{qemu-img convert} with @code{subformat=dynamic}.
716 Block size; min 1 MB, max 256 MB. 0 means auto-calculate based on image size.
722 @subsubsection Read-only formats
723 More disk image file formats are supported in a read-only mode.
726 Bochs images of @code{growing} type.
728 Linux Compressed Loop image, useful only to reuse directly compressed
729 CD-ROM images present for example in the Knoppix CD-ROMs.
733 Parallels disk image format.
738 @subsection Using host drives
740 In addition to disk image files, QEMU can directly access host
741 devices. We describe here the usage for QEMU version >= 0.8.3.
745 On Linux, you can directly use the host device filename instead of a
746 disk image filename provided you have enough privileges to access
747 it. For example, use @file{/dev/cdrom} to access to the CDROM.
751 You can specify a CDROM device even if no CDROM is loaded. QEMU has
752 specific code to detect CDROM insertion or removal. CDROM ejection by
753 the guest OS is supported. Currently only data CDs are supported.
755 You can specify a floppy device even if no floppy is loaded. Floppy
756 removal is currently not detected accurately (if you change floppy
757 without doing floppy access while the floppy is not loaded, the guest
758 OS will think that the same floppy is loaded).
759 Use of the host's floppy device is deprecated, and support for it will
760 be removed in a future release.
762 Hard disks can be used. Normally you must specify the whole disk
763 (@file{/dev/hdb} instead of @file{/dev/hdb1}) so that the guest OS can
764 see it as a partitioned disk. WARNING: unless you know what you do, it
765 is better to only make READ-ONLY accesses to the hard disk otherwise
766 you may corrupt your host data (use the @option{-snapshot} command
767 line option or modify the device permissions accordingly).
770 @subsubsection Windows
774 The preferred syntax is the drive letter (e.g. @file{d:}). The
775 alternate syntax @file{\\.\d:} is supported. @file{/dev/cdrom} is
776 supported as an alias to the first CDROM drive.
778 Currently there is no specific code to handle removable media, so it
779 is better to use the @code{change} or @code{eject} monitor commands to
780 change or eject media.
782 Hard disks can be used with the syntax: @file{\\.\PhysicalDrive@var{N}}
783 where @var{N} is the drive number (0 is the first hard disk).
785 WARNING: unless you know what you do, it is better to only make
786 READ-ONLY accesses to the hard disk otherwise you may corrupt your
787 host data (use the @option{-snapshot} command line so that the
788 modifications are written in a temporary file).
792 @subsubsection Mac OS X
794 @file{/dev/cdrom} is an alias to the first CDROM.
796 Currently there is no specific code to handle removable media, so it
797 is better to use the @code{change} or @code{eject} monitor commands to
798 change or eject media.
800 @node disk_images_fat_images
801 @subsection Virtual FAT disk images
803 QEMU can automatically create a virtual FAT disk image from a
804 directory tree. In order to use it, just type:
807 qemu-system-i386 linux.img -hdb fat:/my_directory
810 Then you access access to all the files in the @file{/my_directory}
811 directory without having to copy them in a disk image or to export
812 them via SAMBA or NFS. The default access is @emph{read-only}.
814 Floppies can be emulated with the @code{:floppy:} option:
817 qemu-system-i386 linux.img -fda fat:floppy:/my_directory
820 A read/write support is available for testing (beta stage) with the
824 qemu-system-i386 linux.img -fda fat:floppy:rw:/my_directory
827 What you should @emph{never} do:
829 @item use non-ASCII filenames ;
830 @item use "-snapshot" together with ":rw:" ;
831 @item expect it to work when loadvm'ing ;
832 @item write to the FAT directory on the host system while accessing it with the guest system.
835 @node disk_images_nbd
836 @subsection NBD access
838 QEMU can access directly to block device exported using the Network Block Device
842 qemu-system-i386 linux.img -hdb nbd://my_nbd_server.mydomain.org:1024/
845 If the NBD server is located on the same host, you can use an unix socket instead
849 qemu-system-i386 linux.img -hdb nbd+unix://?socket=/tmp/my_socket
852 In this case, the block device must be exported using qemu-nbd:
855 qemu-nbd --socket=/tmp/my_socket my_disk.qcow2
858 The use of qemu-nbd allows sharing of a disk between several guests:
860 qemu-nbd --socket=/tmp/my_socket --share=2 my_disk.qcow2
864 and then you can use it with two guests:
866 qemu-system-i386 linux1.img -hdb nbd+unix://?socket=/tmp/my_socket
867 qemu-system-i386 linux2.img -hdb nbd+unix://?socket=/tmp/my_socket
870 If the nbd-server uses named exports (supported since NBD 2.9.18, or with QEMU's
871 own embedded NBD server), you must specify an export name in the URI:
873 qemu-system-i386 -cdrom nbd://localhost/debian-500-ppc-netinst
874 qemu-system-i386 -cdrom nbd://localhost/openSUSE-11.1-ppc-netinst
877 The URI syntax for NBD is supported since QEMU 1.3. An alternative syntax is
878 also available. Here are some example of the older syntax:
880 qemu-system-i386 linux.img -hdb nbd:my_nbd_server.mydomain.org:1024
881 qemu-system-i386 linux2.img -hdb nbd:unix:/tmp/my_socket
882 qemu-system-i386 -cdrom nbd:localhost:10809:exportname=debian-500-ppc-netinst
885 @node disk_images_sheepdog
886 @subsection Sheepdog disk images
888 Sheepdog is a distributed storage system for QEMU. It provides highly
889 available block level storage volumes that can be attached to
890 QEMU-based virtual machines.
892 You can create a Sheepdog disk image with the command:
894 qemu-img create sheepdog:///@var{image} @var{size}
896 where @var{image} is the Sheepdog image name and @var{size} is its
899 To import the existing @var{filename} to Sheepdog, you can use a
902 qemu-img convert @var{filename} sheepdog:///@var{image}
905 You can boot from the Sheepdog disk image with the command:
907 qemu-system-i386 sheepdog:///@var{image}
910 You can also create a snapshot of the Sheepdog image like qcow2.
912 qemu-img snapshot -c @var{tag} sheepdog:///@var{image}
914 where @var{tag} is a tag name of the newly created snapshot.
916 To boot from the Sheepdog snapshot, specify the tag name of the
919 qemu-system-i386 sheepdog:///@var{image}#@var{tag}
922 You can create a cloned image from the existing snapshot.
924 qemu-img create -b sheepdog:///@var{base}#@var{tag} sheepdog:///@var{image}
926 where @var{base} is a image name of the source snapshot and @var{tag}
929 You can use an unix socket instead of an inet socket:
932 qemu-system-i386 sheepdog+unix:///@var{image}?socket=@var{path}
935 If the Sheepdog daemon doesn't run on the local host, you need to
936 specify one of the Sheepdog servers to connect to.
938 qemu-img create sheepdog://@var{hostname}:@var{port}/@var{image} @var{size}
939 qemu-system-i386 sheepdog://@var{hostname}:@var{port}/@var{image}
942 @node disk_images_iscsi
943 @subsection iSCSI LUNs
945 iSCSI is a popular protocol used to access SCSI devices across a computer
948 There are two different ways iSCSI devices can be used by QEMU.
950 The first method is to mount the iSCSI LUN on the host, and make it appear as
951 any other ordinary SCSI device on the host and then to access this device as a
952 /dev/sd device from QEMU. How to do this differs between host OSes.
954 The second method involves using the iSCSI initiator that is built into
955 QEMU. This provides a mechanism that works the same way regardless of which
956 host OS you are running QEMU on. This section will describe this second method
957 of using iSCSI together with QEMU.
959 In QEMU, iSCSI devices are described using special iSCSI URLs
963 iscsi://[<username>[%<password>]@@]<host>[:<port>]/<target-iqn-name>/<lun>
966 Username and password are optional and only used if your target is set up
967 using CHAP authentication for access control.
968 Alternatively the username and password can also be set via environment
969 variables to have these not show up in the process list
972 export LIBISCSI_CHAP_USERNAME=<username>
973 export LIBISCSI_CHAP_PASSWORD=<password>
974 iscsi://<host>/<target-iqn-name>/<lun>
977 Various session related parameters can be set via special options, either
978 in a configuration file provided via '-readconfig' or directly on the
981 If the initiator-name is not specified qemu will use a default name
982 of 'iqn.2008-11.org.linux-kvm[:<name>'] where <name> is the name of the
987 Setting a specific initiator name to use when logging in to the target
988 -iscsi initiator-name=iqn.qemu.test:my-initiator
992 Controlling which type of header digest to negotiate with the target
993 -iscsi header-digest=CRC32C|CRC32C-NONE|NONE-CRC32C|NONE
996 These can also be set via a configuration file
999 user = "CHAP username"
1000 password = "CHAP password"
1001 initiator-name = "iqn.qemu.test:my-initiator"
1002 # header digest is one of CRC32C|CRC32C-NONE|NONE-CRC32C|NONE
1003 header-digest = "CRC32C"
1007 Setting the target name allows different options for different targets
1009 [iscsi "iqn.target.name"]
1010 user = "CHAP username"
1011 password = "CHAP password"
1012 initiator-name = "iqn.qemu.test:my-initiator"
1013 # header digest is one of CRC32C|CRC32C-NONE|NONE-CRC32C|NONE
1014 header-digest = "CRC32C"
1018 Howto use a configuration file to set iSCSI configuration options:
1020 cat >iscsi.conf <<EOF
1023 password = "my password"
1024 initiator-name = "iqn.qemu.test:my-initiator"
1025 header-digest = "CRC32C"
1028 qemu-system-i386 -drive file=iscsi://127.0.0.1/iqn.qemu.test/1 \
1029 -readconfig iscsi.conf
1033 Howto set up a simple iSCSI target on loopback and accessing it via QEMU:
1035 This example shows how to set up an iSCSI target with one CDROM and one DISK
1036 using the Linux STGT software target. This target is available on Red Hat based
1037 systems as the package 'scsi-target-utils'.
1039 tgtd --iscsi portal=127.0.0.1:3260
1040 tgtadm --lld iscsi --op new --mode target --tid 1 -T iqn.qemu.test
1041 tgtadm --lld iscsi --mode logicalunit --op new --tid 1 --lun 1 \
1042 -b /IMAGES/disk.img --device-type=disk
1043 tgtadm --lld iscsi --mode logicalunit --op new --tid 1 --lun 2 \
1044 -b /IMAGES/cd.iso --device-type=cd
1045 tgtadm --lld iscsi --op bind --mode target --tid 1 -I ALL
1047 qemu-system-i386 -iscsi initiator-name=iqn.qemu.test:my-initiator \
1048 -boot d -drive file=iscsi://127.0.0.1/iqn.qemu.test/1 \
1049 -cdrom iscsi://127.0.0.1/iqn.qemu.test/2
1052 @node disk_images_gluster
1053 @subsection GlusterFS disk images
1055 GlusterFS is an user space distributed file system.
1057 You can boot from the GlusterFS disk image with the command:
1059 qemu-system-x86_64 -drive file=gluster[+@var{transport}]://[@var{server}[:@var{port}]]/@var{volname}/@var{image}[?socket=...]
1062 @var{gluster} is the protocol.
1064 @var{transport} specifies the transport type used to connect to gluster
1065 management daemon (glusterd). Valid transport types are
1066 tcp, unix and rdma. If a transport type isn't specified, then tcp
1069 @var{server} specifies the server where the volume file specification for
1070 the given volume resides. This can be either hostname, ipv4 address
1071 or ipv6 address. ipv6 address needs to be within square brackets [ ].
1072 If transport type is unix, then @var{server} field should not be specified.
1073 Instead @var{socket} field needs to be populated with the path to unix domain
1076 @var{port} is the port number on which glusterd is listening. This is optional
1077 and if not specified, QEMU will send 0 which will make gluster to use the
1078 default port. If the transport type is unix, then @var{port} should not be
1081 @var{volname} is the name of the gluster volume which contains the disk image.
1083 @var{image} is the path to the actual disk image that resides on gluster volume.
1085 You can create a GlusterFS disk image with the command:
1087 qemu-img create gluster://@var{server}/@var{volname}/@var{image} @var{size}
1092 qemu-system-x86_64 -drive file=gluster://1.2.3.4/testvol/a.img
1093 qemu-system-x86_64 -drive file=gluster+tcp://1.2.3.4/testvol/a.img
1094 qemu-system-x86_64 -drive file=gluster+tcp://1.2.3.4:24007/testvol/dir/a.img
1095 qemu-system-x86_64 -drive file=gluster+tcp://[1:2:3:4:5:6:7:8]/testvol/dir/a.img
1096 qemu-system-x86_64 -drive file=gluster+tcp://[1:2:3:4:5:6:7:8]:24007/testvol/dir/a.img
1097 qemu-system-x86_64 -drive file=gluster+tcp://server.domain.com:24007/testvol/dir/a.img
1098 qemu-system-x86_64 -drive file=gluster+unix:///testvol/dir/a.img?socket=/tmp/glusterd.socket
1099 qemu-system-x86_64 -drive file=gluster+rdma://1.2.3.4:24007/testvol/a.img
1102 @node disk_images_ssh
1103 @subsection Secure Shell (ssh) disk images
1105 You can access disk images located on a remote ssh server
1106 by using the ssh protocol:
1109 qemu-system-x86_64 -drive file=ssh://[@var{user}@@]@var{server}[:@var{port}]/@var{path}[?host_key_check=@var{host_key_check}]
1112 Alternative syntax using properties:
1115 qemu-system-x86_64 -drive file.driver=ssh[,file.user=@var{user}],file.host=@var{server}[,file.port=@var{port}],file.path=@var{path}[,file.host_key_check=@var{host_key_check}]
1118 @var{ssh} is the protocol.
1120 @var{user} is the remote user. If not specified, then the local
1123 @var{server} specifies the remote ssh server. Any ssh server can be
1124 used, but it must implement the sftp-server protocol. Most Unix/Linux
1125 systems should work without requiring any extra configuration.
1127 @var{port} is the port number on which sshd is listening. By default
1128 the standard ssh port (22) is used.
1130 @var{path} is the path to the disk image.
1132 The optional @var{host_key_check} parameter controls how the remote
1133 host's key is checked. The default is @code{yes} which means to use
1134 the local @file{.ssh/known_hosts} file. Setting this to @code{no}
1135 turns off known-hosts checking. Or you can check that the host key
1136 matches a specific fingerprint:
1137 @code{host_key_check=md5:78:45:8e:14:57:4f:d5:45:83:0a:0e:f3:49:82:c9:c8}
1138 (@code{sha1:} can also be used as a prefix, but note that OpenSSH
1139 tools only use MD5 to print fingerprints).
1141 Currently authentication must be done using ssh-agent. Other
1142 authentication methods may be supported in future.
1144 Note: Many ssh servers do not support an @code{fsync}-style operation.
1145 The ssh driver cannot guarantee that disk flush requests are
1146 obeyed, and this causes a risk of disk corruption if the remote
1147 server or network goes down during writes. The driver will
1148 print a warning when @code{fsync} is not supported:
1150 warning: ssh server @code{ssh.example.com:22} does not support fsync
1152 With sufficiently new versions of libssh2 and OpenSSH, @code{fsync} is
1156 @section Network emulation
1158 QEMU can simulate several network cards (PCI or ISA cards on the PC
1159 target) and can connect them to an arbitrary number of Virtual Local
1160 Area Networks (VLANs). Host TAP devices can be connected to any QEMU
1161 VLAN. VLAN can be connected between separate instances of QEMU to
1162 simulate large networks. For simpler usage, a non privileged user mode
1163 network stack can replace the TAP device to have a basic network
1168 QEMU simulates several VLANs. A VLAN can be symbolised as a virtual
1169 connection between several network devices. These devices can be for
1170 example QEMU virtual Ethernet cards or virtual Host ethernet devices
1173 @subsection Using TAP network interfaces
1175 This is the standard way to connect QEMU to a real network. QEMU adds
1176 a virtual network device on your host (called @code{tapN}), and you
1177 can then configure it as if it was a real ethernet card.
1179 @subsubsection Linux host
1181 As an example, you can download the @file{linux-test-xxx.tar.gz}
1182 archive and copy the script @file{qemu-ifup} in @file{/etc} and
1183 configure properly @code{sudo} so that the command @code{ifconfig}
1184 contained in @file{qemu-ifup} can be executed as root. You must verify
1185 that your host kernel supports the TAP network interfaces: the
1186 device @file{/dev/net/tun} must be present.
1188 See @ref{sec_invocation} to have examples of command lines using the
1189 TAP network interfaces.
1191 @subsubsection Windows host
1193 There is a virtual ethernet driver for Windows 2000/XP systems, called
1194 TAP-Win32. But it is not included in standard QEMU for Windows,
1195 so you will need to get it separately. It is part of OpenVPN package,
1196 so download OpenVPN from : @url{http://openvpn.net/}.
1198 @subsection Using the user mode network stack
1200 By using the option @option{-net user} (default configuration if no
1201 @option{-net} option is specified), QEMU uses a completely user mode
1202 network stack (you don't need root privilege to use the virtual
1203 network). The virtual network configuration is the following:
1207 QEMU VLAN <------> Firewall/DHCP server <-----> Internet
1210 ----> DNS server (10.0.2.3)
1212 ----> SMB server (10.0.2.4)
1215 The QEMU VM behaves as if it was behind a firewall which blocks all
1216 incoming connections. You can use a DHCP client to automatically
1217 configure the network in the QEMU VM. The DHCP server assign addresses
1218 to the hosts starting from 10.0.2.15.
1220 In order to check that the user mode network is working, you can ping
1221 the address 10.0.2.2 and verify that you got an address in the range
1222 10.0.2.x from the QEMU virtual DHCP server.
1224 Note that ICMP traffic in general does not work with user mode networking.
1225 @code{ping}, aka. ICMP echo, to the local router (10.0.2.2) shall work,
1226 however. If you're using QEMU on Linux >= 3.0, it can use unprivileged ICMP
1227 ping sockets to allow @code{ping} to the Internet. The host admin has to set
1228 the ping_group_range in order to grant access to those sockets. To allow ping
1229 for GID 100 (usually users group):
1232 echo 100 100 > /proc/sys/net/ipv4/ping_group_range
1235 When using the built-in TFTP server, the router is also the TFTP
1238 When using the @option{-redir} option, TCP or UDP connections can be
1239 redirected from the host to the guest. It allows for example to
1240 redirect X11, telnet or SSH connections.
1242 @subsection Connecting VLANs between QEMU instances
1244 Using the @option{-net socket} option, it is possible to make VLANs
1245 that span several QEMU instances. See @ref{sec_invocation} to have a
1248 @node pcsys_other_devs
1249 @section Other Devices
1251 @subsection Inter-VM Shared Memory device
1253 With KVM enabled on a Linux host, a shared memory device is available. Guests
1254 map a POSIX shared memory region into the guest as a PCI device that enables
1255 zero-copy communication to the application level of the guests. The basic
1259 qemu-system-i386 -device ivshmem,size=<size in format accepted by -m>[,shm=<shm name>]
1262 If desired, interrupts can be sent between guest VMs accessing the same shared
1263 memory region. Interrupt support requires using a shared memory server and
1264 using a chardev socket to connect to it. The code for the shared memory server
1265 is qemu.git/contrib/ivshmem-server. An example syntax when using the shared
1269 # First start the ivshmem server once and for all
1270 ivshmem-server -p <pidfile> -S <path> -m <shm name> -l <shm size> -n <vectors n>
1272 # Then start your qemu instances with matching arguments
1273 qemu-system-i386 -device ivshmem,size=<shm size>,vectors=<vectors n>,chardev=<id>
1274 [,msi=on][,ioeventfd=on][,role=peer|master]
1275 -chardev socket,path=<path>,id=<id>
1278 When using the server, the guest will be assigned a VM ID (>=0) that allows guests
1279 using the same server to communicate via interrupts. Guests can read their
1280 VM ID from a device register (see example code). Since receiving the shared
1281 memory region from the server is asynchronous, there is a (small) chance the
1282 guest may boot before the shared memory is attached. To allow an application
1283 to ensure shared memory is attached, the VM ID register will return -1 (an
1284 invalid VM ID) until the memory is attached. Once the shared memory is
1285 attached, the VM ID will return the guest's valid VM ID. With these semantics,
1286 the guest application can check to ensure the shared memory is attached to the
1287 guest before proceeding.
1289 The @option{role} argument can be set to either master or peer and will affect
1290 how the shared memory is migrated. With @option{role=master}, the guest will
1291 copy the shared memory on migration to the destination host. With
1292 @option{role=peer}, the guest will not be able to migrate with the device attached.
1293 With the @option{peer} case, the device should be detached and then reattached
1294 after migration using the PCI hotplug support.
1296 @node direct_linux_boot
1297 @section Direct Linux Boot
1299 This section explains how to launch a Linux kernel inside QEMU without
1300 having to make a full bootable image. It is very useful for fast Linux
1305 qemu-system-i386 -kernel arch/i386/boot/bzImage -hda root-2.4.20.img -append "root=/dev/hda"
1308 Use @option{-kernel} to provide the Linux kernel image and
1309 @option{-append} to give the kernel command line arguments. The
1310 @option{-initrd} option can be used to provide an INITRD image.
1312 When using the direct Linux boot, a disk image for the first hard disk
1313 @file{hda} is required because its boot sector is used to launch the
1316 If you do not need graphical output, you can disable it and redirect
1317 the virtual serial port and the QEMU monitor to the console with the
1318 @option{-nographic} option. The typical command line is:
1320 qemu-system-i386 -kernel arch/i386/boot/bzImage -hda root-2.4.20.img \
1321 -append "root=/dev/hda console=ttyS0" -nographic
1324 Use @key{Ctrl-a c} to switch between the serial console and the
1325 monitor (@pxref{pcsys_keys}).
1328 @section USB emulation
1330 QEMU emulates a PCI UHCI USB controller. You can virtually plug
1331 virtual USB devices or real host USB devices (experimental, works only
1332 on Linux hosts). QEMU will automatically create and connect virtual USB hubs
1333 as necessary to connect multiple USB devices.
1337 * host_usb_devices::
1340 @subsection Connecting USB devices
1342 USB devices can be connected with the @option{-usbdevice} commandline option
1343 or the @code{usb_add} monitor command. Available devices are:
1347 Virtual Mouse. This will override the PS/2 mouse emulation when activated.
1349 Pointer device that uses absolute coordinates (like a touchscreen).
1350 This means QEMU is able to report the mouse position without having
1351 to grab the mouse. Also overrides the PS/2 mouse emulation when activated.
1352 @item disk:@var{file}
1353 Mass storage device based on @var{file} (@pxref{disk_images})
1354 @item host:@var{bus.addr}
1355 Pass through the host device identified by @var{bus.addr}
1357 @item host:@var{vendor_id:product_id}
1358 Pass through the host device identified by @var{vendor_id:product_id}
1361 Virtual Wacom PenPartner tablet. This device is similar to the @code{tablet}
1362 above but it can be used with the tslib library because in addition to touch
1363 coordinates it reports touch pressure.
1365 Standard USB keyboard. Will override the PS/2 keyboard (if present).
1366 @item serial:[vendorid=@var{vendor_id}][,product_id=@var{product_id}]:@var{dev}
1367 Serial converter. This emulates an FTDI FT232BM chip connected to host character
1368 device @var{dev}. The available character devices are the same as for the
1369 @code{-serial} option. The @code{vendorid} and @code{productid} options can be
1370 used to override the default 0403:6001. For instance,
1372 usb_add serial:productid=FA00:tcp:192.168.0.2:4444
1374 will connect to tcp port 4444 of ip 192.168.0.2, and plug that to the virtual
1375 serial converter, faking a Matrix Orbital LCD Display (USB ID 0403:FA00).
1377 Braille device. This will use BrlAPI to display the braille output on a real
1379 @item net:@var{options}
1380 Network adapter that supports CDC ethernet and RNDIS protocols. @var{options}
1381 specifies NIC options as with @code{-net nic,}@var{options} (see description).
1382 For instance, user-mode networking can be used with
1384 qemu-system-i386 [...OPTIONS...] -net user,vlan=0 -usbdevice net:vlan=0
1386 Currently this cannot be used in machines that support PCI NICs.
1387 @item bt[:@var{hci-type}]
1388 Bluetooth dongle whose type is specified in the same format as with
1389 the @option{-bt hci} option, @pxref{bt-hcis,,allowed HCI types}. If
1390 no type is given, the HCI logic corresponds to @code{-bt hci,vlan=0}.
1391 This USB device implements the USB Transport Layer of HCI. Example
1394 qemu-system-i386 [...OPTIONS...] -usbdevice bt:hci,vlan=3 -bt device:keyboard,vlan=3
1398 @node host_usb_devices
1399 @subsection Using host USB devices on a Linux host
1401 WARNING: this is an experimental feature. QEMU will slow down when
1402 using it. USB devices requiring real time streaming (i.e. USB Video
1403 Cameras) are not supported yet.
1406 @item If you use an early Linux 2.4 kernel, verify that no Linux driver
1407 is actually using the USB device. A simple way to do that is simply to
1408 disable the corresponding kernel module by renaming it from @file{mydriver.o}
1409 to @file{mydriver.o.disabled}.
1411 @item Verify that @file{/proc/bus/usb} is working (most Linux distributions should enable it by default). You should see something like that:
1417 @item Since only root can access to the USB devices directly, you can either launch QEMU as root or change the permissions of the USB devices you want to use. For testing, the following suffices:
1419 chown -R myuid /proc/bus/usb
1422 @item Launch QEMU and do in the monitor:
1425 Device 1.2, speed 480 Mb/s
1426 Class 00: USB device 1234:5678, USB DISK
1428 You should see the list of the devices you can use (Never try to use
1429 hubs, it won't work).
1431 @item Add the device in QEMU by using:
1433 usb_add host:1234:5678
1436 Normally the guest OS should report that a new USB device is
1437 plugged. You can use the option @option{-usbdevice} to do the same.
1439 @item Now you can try to use the host USB device in QEMU.
1443 When relaunching QEMU, you may have to unplug and plug again the USB
1444 device to make it work again (this is a bug).
1447 @section VNC security
1449 The VNC server capability provides access to the graphical console
1450 of the guest VM across the network. This has a number of security
1451 considerations depending on the deployment scenarios.
1455 * vnc_sec_password::
1456 * vnc_sec_certificate::
1457 * vnc_sec_certificate_verify::
1458 * vnc_sec_certificate_pw::
1460 * vnc_sec_certificate_sasl::
1461 * vnc_generate_cert::
1465 @subsection Without passwords
1467 The simplest VNC server setup does not include any form of authentication.
1468 For this setup it is recommended to restrict it to listen on a UNIX domain
1469 socket only. For example
1472 qemu-system-i386 [...OPTIONS...] -vnc unix:/home/joebloggs/.qemu-myvm-vnc
1475 This ensures that only users on local box with read/write access to that
1476 path can access the VNC server. To securely access the VNC server from a
1477 remote machine, a combination of netcat+ssh can be used to provide a secure
1480 @node vnc_sec_password
1481 @subsection With passwords
1483 The VNC protocol has limited support for password based authentication. Since
1484 the protocol limits passwords to 8 characters it should not be considered
1485 to provide high security. The password can be fairly easily brute-forced by
1486 a client making repeat connections. For this reason, a VNC server using password
1487 authentication should be restricted to only listen on the loopback interface
1488 or UNIX domain sockets. Password authentication is not supported when operating
1489 in FIPS 140-2 compliance mode as it requires the use of the DES cipher. Password
1490 authentication is requested with the @code{password} option, and then once QEMU
1491 is running the password is set with the monitor. Until the monitor is used to
1492 set the password all clients will be rejected.
1495 qemu-system-i386 [...OPTIONS...] -vnc :1,password -monitor stdio
1496 (qemu) change vnc password
1501 @node vnc_sec_certificate
1502 @subsection With x509 certificates
1504 The QEMU VNC server also implements the VeNCrypt extension allowing use of
1505 TLS for encryption of the session, and x509 certificates for authentication.
1506 The use of x509 certificates is strongly recommended, because TLS on its
1507 own is susceptible to man-in-the-middle attacks. Basic x509 certificate
1508 support provides a secure session, but no authentication. This allows any
1509 client to connect, and provides an encrypted session.
1512 qemu-system-i386 [...OPTIONS...] -vnc :1,tls,x509=/etc/pki/qemu -monitor stdio
1515 In the above example @code{/etc/pki/qemu} should contain at least three files,
1516 @code{ca-cert.pem}, @code{server-cert.pem} and @code{server-key.pem}. Unprivileged
1517 users will want to use a private directory, for example @code{$HOME/.pki/qemu}.
1518 NB the @code{server-key.pem} file should be protected with file mode 0600 to
1519 only be readable by the user owning it.
1521 @node vnc_sec_certificate_verify
1522 @subsection With x509 certificates and client verification
1524 Certificates can also provide a means to authenticate the client connecting.
1525 The server will request that the client provide a certificate, which it will
1526 then validate against the CA certificate. This is a good choice if deploying
1527 in an environment with a private internal certificate authority.
1530 qemu-system-i386 [...OPTIONS...] -vnc :1,tls,x509verify=/etc/pki/qemu -monitor stdio
1534 @node vnc_sec_certificate_pw
1535 @subsection With x509 certificates, client verification and passwords
1537 Finally, the previous method can be combined with VNC password authentication
1538 to provide two layers of authentication for clients.
1541 qemu-system-i386 [...OPTIONS...] -vnc :1,password,tls,x509verify=/etc/pki/qemu -monitor stdio
1542 (qemu) change vnc password
1549 @subsection With SASL authentication
1551 The SASL authentication method is a VNC extension, that provides an
1552 easily extendable, pluggable authentication method. This allows for
1553 integration with a wide range of authentication mechanisms, such as
1554 PAM, GSSAPI/Kerberos, LDAP, SQL databases, one-time keys and more.
1555 The strength of the authentication depends on the exact mechanism
1556 configured. If the chosen mechanism also provides a SSF layer, then
1557 it will encrypt the datastream as well.
1559 Refer to the later docs on how to choose the exact SASL mechanism
1560 used for authentication, but assuming use of one supporting SSF,
1561 then QEMU can be launched with:
1564 qemu-system-i386 [...OPTIONS...] -vnc :1,sasl -monitor stdio
1567 @node vnc_sec_certificate_sasl
1568 @subsection With x509 certificates and SASL authentication
1570 If the desired SASL authentication mechanism does not supported
1571 SSF layers, then it is strongly advised to run it in combination
1572 with TLS and x509 certificates. This provides securely encrypted
1573 data stream, avoiding risk of compromising of the security
1574 credentials. This can be enabled, by combining the 'sasl' option
1575 with the aforementioned TLS + x509 options:
1578 qemu-system-i386 [...OPTIONS...] -vnc :1,tls,x509,sasl -monitor stdio
1582 @node vnc_generate_cert
1583 @subsection Generating certificates for VNC
1585 The GNU TLS packages provides a command called @code{certtool} which can
1586 be used to generate certificates and keys in PEM format. At a minimum it
1587 is necessary to setup a certificate authority, and issue certificates to
1588 each server. If using certificates for authentication, then each client
1589 will also need to be issued a certificate. The recommendation is for the
1590 server to keep its certificates in either @code{/etc/pki/qemu} or for
1591 unprivileged users in @code{$HOME/.pki/qemu}.
1595 * vnc_generate_server::
1596 * vnc_generate_client::
1598 @node vnc_generate_ca
1599 @subsubsection Setup the Certificate Authority
1601 This step only needs to be performed once per organization / organizational
1602 unit. First the CA needs a private key. This key must be kept VERY secret
1603 and secure. If this key is compromised the entire trust chain of the certificates
1604 issued with it is lost.
1607 # certtool --generate-privkey > ca-key.pem
1610 A CA needs to have a public certificate. For simplicity it can be a self-signed
1611 certificate, or one issue by a commercial certificate issuing authority. To
1612 generate a self-signed certificate requires one core piece of information, the
1613 name of the organization.
1616 # cat > ca.info <<EOF
1617 cn = Name of your organization
1621 # certtool --generate-self-signed \
1622 --load-privkey ca-key.pem
1623 --template ca.info \
1624 --outfile ca-cert.pem
1627 The @code{ca-cert.pem} file should be copied to all servers and clients wishing to utilize
1628 TLS support in the VNC server. The @code{ca-key.pem} must not be disclosed/copied at all.
1630 @node vnc_generate_server
1631 @subsubsection Issuing server certificates
1633 Each server (or host) needs to be issued with a key and certificate. When connecting
1634 the certificate is sent to the client which validates it against the CA certificate.
1635 The core piece of information for a server certificate is the hostname. This should
1636 be the fully qualified hostname that the client will connect with, since the client
1637 will typically also verify the hostname in the certificate. On the host holding the
1638 secure CA private key:
1641 # cat > server.info <<EOF
1642 organization = Name of your organization
1643 cn = server.foo.example.com
1648 # certtool --generate-privkey > server-key.pem
1649 # certtool --generate-certificate \
1650 --load-ca-certificate ca-cert.pem \
1651 --load-ca-privkey ca-key.pem \
1652 --load-privkey server-key.pem \
1653 --template server.info \
1654 --outfile server-cert.pem
1657 The @code{server-key.pem} and @code{server-cert.pem} files should now be securely copied
1658 to the server for which they were generated. The @code{server-key.pem} is security
1659 sensitive and should be kept protected with file mode 0600 to prevent disclosure.
1661 @node vnc_generate_client
1662 @subsubsection Issuing client certificates
1664 If the QEMU VNC server is to use the @code{x509verify} option to validate client
1665 certificates as its authentication mechanism, each client also needs to be issued
1666 a certificate. The client certificate contains enough metadata to uniquely identify
1667 the client, typically organization, state, city, building, etc. On the host holding
1668 the secure CA private key:
1671 # cat > client.info <<EOF
1675 organization = Name of your organization
1676 cn = client.foo.example.com
1681 # certtool --generate-privkey > client-key.pem
1682 # certtool --generate-certificate \
1683 --load-ca-certificate ca-cert.pem \
1684 --load-ca-privkey ca-key.pem \
1685 --load-privkey client-key.pem \
1686 --template client.info \
1687 --outfile client-cert.pem
1690 The @code{client-key.pem} and @code{client-cert.pem} files should now be securely
1691 copied to the client for which they were generated.
1694 @node vnc_setup_sasl
1696 @subsection Configuring SASL mechanisms
1698 The following documentation assumes use of the Cyrus SASL implementation on a
1699 Linux host, but the principals should apply to any other SASL impl. When SASL
1700 is enabled, the mechanism configuration will be loaded from system default
1701 SASL service config /etc/sasl2/qemu.conf. If running QEMU as an
1702 unprivileged user, an environment variable SASL_CONF_PATH can be used
1703 to make it search alternate locations for the service config.
1705 The default configuration might contain
1708 mech_list: digest-md5
1709 sasldb_path: /etc/qemu/passwd.db
1712 This says to use the 'Digest MD5' mechanism, which is similar to the HTTP
1713 Digest-MD5 mechanism. The list of valid usernames & passwords is maintained
1714 in the /etc/qemu/passwd.db file, and can be updated using the saslpasswd2
1715 command. While this mechanism is easy to configure and use, it is not
1716 considered secure by modern standards, so only suitable for developers /
1719 A more serious deployment might use Kerberos, which is done with the 'gssapi'
1724 keytab: /etc/qemu/krb5.tab
1727 For this to work the administrator of your KDC must generate a Kerberos
1728 principal for the server, with a name of 'qemu/somehost.example.com@@EXAMPLE.COM'
1729 replacing 'somehost.example.com' with the fully qualified host name of the
1730 machine running QEMU, and 'EXAMPLE.COM' with the Kerberos Realm.
1732 Other configurations will be left as an exercise for the reader. It should
1733 be noted that only Digest-MD5 and GSSAPI provides a SSF layer for data
1734 encryption. For all other mechanisms, VNC should always be configured to
1735 use TLS and x509 certificates to protect security credentials from snooping.
1740 QEMU has a primitive support to work with gdb, so that you can do
1741 'Ctrl-C' while the virtual machine is running and inspect its state.
1743 In order to use gdb, launch QEMU with the '-s' option. It will wait for a
1746 qemu-system-i386 -s -kernel arch/i386/boot/bzImage -hda root-2.4.20.img \
1747 -append "root=/dev/hda"
1748 Connected to host network interface: tun0
1749 Waiting gdb connection on port 1234
1752 Then launch gdb on the 'vmlinux' executable:
1757 In gdb, connect to QEMU:
1759 (gdb) target remote localhost:1234
1762 Then you can use gdb normally. For example, type 'c' to launch the kernel:
1767 Here are some useful tips in order to use gdb on system code:
1771 Use @code{info reg} to display all the CPU registers.
1773 Use @code{x/10i $eip} to display the code at the PC position.
1775 Use @code{set architecture i8086} to dump 16 bit code. Then use
1776 @code{x/10i $cs*16+$eip} to dump the code at the PC position.
1779 Advanced debugging options:
1781 The default single stepping behavior is step with the IRQs and timer service routines off. It is set this way because when gdb executes a single step it expects to advance beyond the current instruction. With the IRQs and timer service routines on, a single step might jump into the one of the interrupt or exception vectors instead of executing the current instruction. This means you may hit the same breakpoint a number of times before executing the instruction gdb wants to have executed. Because there are rare circumstances where you want to single step into an interrupt vector the behavior can be controlled from GDB. There are three commands you can query and set the single step behavior:
1783 @item maintenance packet qqemu.sstepbits
1785 This will display the MASK bits used to control the single stepping IE:
1787 (gdb) maintenance packet qqemu.sstepbits
1788 sending: "qqemu.sstepbits"
1789 received: "ENABLE=1,NOIRQ=2,NOTIMER=4"
1791 @item maintenance packet qqemu.sstep
1793 This will display the current value of the mask used when single stepping IE:
1795 (gdb) maintenance packet qqemu.sstep
1796 sending: "qqemu.sstep"
1799 @item maintenance packet Qqemu.sstep=HEX_VALUE
1801 This will change the single step mask, so if wanted to enable IRQs on the single step, but not timers, you would use:
1803 (gdb) maintenance packet Qqemu.sstep=0x5
1804 sending: "qemu.sstep=0x5"
1809 @node pcsys_os_specific
1810 @section Target OS specific information
1814 To have access to SVGA graphic modes under X11, use the @code{vesa} or
1815 the @code{cirrus} X11 driver. For optimal performances, use 16 bit
1816 color depth in the guest and the host OS.
1818 When using a 2.6 guest Linux kernel, you should add the option
1819 @code{clock=pit} on the kernel command line because the 2.6 Linux
1820 kernels make very strict real time clock checks by default that QEMU
1821 cannot simulate exactly.
1823 When using a 2.6 guest Linux kernel, verify that the 4G/4G patch is
1824 not activated because QEMU is slower with this patch. The QEMU
1825 Accelerator Module is also much slower in this case. Earlier Fedora
1826 Core 3 Linux kernel (< 2.6.9-1.724_FC3) were known to incorporate this
1827 patch by default. Newer kernels don't have it.
1831 If you have a slow host, using Windows 95 is better as it gives the
1832 best speed. Windows 2000 is also a good choice.
1834 @subsubsection SVGA graphic modes support
1836 QEMU emulates a Cirrus Logic GD5446 Video
1837 card. All Windows versions starting from Windows 95 should recognize
1838 and use this graphic card. For optimal performances, use 16 bit color
1839 depth in the guest and the host OS.
1841 If you are using Windows XP as guest OS and if you want to use high
1842 resolution modes which the Cirrus Logic BIOS does not support (i.e. >=
1843 1280x1024x16), then you should use the VESA VBE virtual graphic card
1844 (option @option{-std-vga}).
1846 @subsubsection CPU usage reduction
1848 Windows 9x does not correctly use the CPU HLT
1849 instruction. The result is that it takes host CPU cycles even when
1850 idle. You can install the utility from
1851 @url{http://www.user.cityline.ru/~maxamn/amnhltm.zip} to solve this
1852 problem. Note that no such tool is needed for NT, 2000 or XP.
1854 @subsubsection Windows 2000 disk full problem
1856 Windows 2000 has a bug which gives a disk full problem during its
1857 installation. When installing it, use the @option{-win2k-hack} QEMU
1858 option to enable a specific workaround. After Windows 2000 is
1859 installed, you no longer need this option (this option slows down the
1862 @subsubsection Windows 2000 shutdown
1864 Windows 2000 cannot automatically shutdown in QEMU although Windows 98
1865 can. It comes from the fact that Windows 2000 does not automatically
1866 use the APM driver provided by the BIOS.
1868 In order to correct that, do the following (thanks to Struan
1869 Bartlett): go to the Control Panel => Add/Remove Hardware & Next =>
1870 Add/Troubleshoot a device => Add a new device & Next => No, select the
1871 hardware from a list & Next => NT Apm/Legacy Support & Next => Next
1872 (again) a few times. Now the driver is installed and Windows 2000 now
1873 correctly instructs QEMU to shutdown at the appropriate moment.
1875 @subsubsection Share a directory between Unix and Windows
1877 See @ref{sec_invocation} about the help of the option @option{-smb}.
1879 @subsubsection Windows XP security problem
1881 Some releases of Windows XP install correctly but give a security
1884 A problem is preventing Windows from accurately checking the
1885 license for this computer. Error code: 0x800703e6.
1888 The workaround is to install a service pack for XP after a boot in safe
1889 mode. Then reboot, and the problem should go away. Since there is no
1890 network while in safe mode, its recommended to download the full
1891 installation of SP1 or SP2 and transfer that via an ISO or using the
1892 vvfat block device ("-hdb fat:directory_which_holds_the_SP").
1894 @subsection MS-DOS and FreeDOS
1896 @subsubsection CPU usage reduction
1898 DOS does not correctly use the CPU HLT instruction. The result is that
1899 it takes host CPU cycles even when idle. You can install the utility
1900 from @url{http://www.vmware.com/software/dosidle210.zip} to solve this
1903 @node QEMU System emulator for non PC targets
1904 @chapter QEMU System emulator for non PC targets
1906 QEMU is a generic emulator and it emulates many non PC
1907 machines. Most of the options are similar to the PC emulator. The
1908 differences are mentioned in the following sections.
1911 * PowerPC System emulator::
1912 * Sparc32 System emulator::
1913 * Sparc64 System emulator::
1914 * MIPS System emulator::
1915 * ARM System emulator::
1916 * ColdFire System emulator::
1917 * Cris System emulator::
1918 * Microblaze System emulator::
1919 * SH4 System emulator::
1920 * Xtensa System emulator::
1923 @node PowerPC System emulator
1924 @section PowerPC System emulator
1925 @cindex system emulation (PowerPC)
1927 Use the executable @file{qemu-system-ppc} to simulate a complete PREP
1928 or PowerMac PowerPC system.
1930 QEMU emulates the following PowerMac peripherals:
1934 UniNorth or Grackle PCI Bridge
1936 PCI VGA compatible card with VESA Bochs Extensions
1938 2 PMAC IDE interfaces with hard disk and CD-ROM support
1944 VIA-CUDA with ADB keyboard and mouse.
1947 QEMU emulates the following PREP peripherals:
1953 PCI VGA compatible card with VESA Bochs Extensions
1955 2 IDE interfaces with hard disk and CD-ROM support
1959 NE2000 network adapters
1963 PREP Non Volatile RAM
1965 PC compatible keyboard and mouse.
1968 QEMU uses the Open Hack'Ware Open Firmware Compatible BIOS available at
1969 @url{http://perso.magic.fr/l_indien/OpenHackWare/index.htm}.
1971 Since version 0.9.1, QEMU uses OpenBIOS @url{http://www.openbios.org/}
1972 for the g3beige and mac99 PowerMac machines. OpenBIOS is a free (GPL
1973 v2) portable firmware implementation. The goal is to implement a 100%
1974 IEEE 1275-1994 (referred to as Open Firmware) compliant firmware.
1976 @c man begin OPTIONS
1978 The following options are specific to the PowerPC emulation:
1982 @item -g @var{W}x@var{H}[x@var{DEPTH}]
1984 Set the initial VGA graphic mode. The default is 800x600x32.
1986 @item -prom-env @var{string}
1988 Set OpenBIOS variables in NVRAM, for example:
1991 qemu-system-ppc -prom-env 'auto-boot?=false' \
1992 -prom-env 'boot-device=hd:2,\yaboot' \
1993 -prom-env 'boot-args=conf=hd:2,\yaboot.conf'
1996 These variables are not used by Open Hack'Ware.
2003 More information is available at
2004 @url{http://perso.magic.fr/l_indien/qemu-ppc/}.
2006 @node Sparc32 System emulator
2007 @section Sparc32 System emulator
2008 @cindex system emulation (Sparc32)
2010 Use the executable @file{qemu-system-sparc} to simulate the following
2011 Sun4m architecture machines:
2026 SPARCstation Voyager
2033 The emulation is somewhat complete. SMP up to 16 CPUs is supported,
2034 but Linux limits the number of usable CPUs to 4.
2036 QEMU emulates the following sun4m peripherals:
2042 TCX or cgthree Frame buffer
2044 Lance (Am7990) Ethernet
2046 Non Volatile RAM M48T02/M48T08
2048 Slave I/O: timers, interrupt controllers, Zilog serial ports, keyboard
2049 and power/reset logic
2051 ESP SCSI controller with hard disk and CD-ROM support
2053 Floppy drive (not on SS-600MP)
2055 CS4231 sound device (only on SS-5, not working yet)
2058 The number of peripherals is fixed in the architecture. Maximum
2059 memory size depends on the machine type, for SS-5 it is 256MB and for
2062 Since version 0.8.2, QEMU uses OpenBIOS
2063 @url{http://www.openbios.org/}. OpenBIOS is a free (GPL v2) portable
2064 firmware implementation. The goal is to implement a 100% IEEE
2065 1275-1994 (referred to as Open Firmware) compliant firmware.
2067 A sample Linux 2.6 series kernel and ram disk image are available on
2068 the QEMU web site. There are still issues with NetBSD and OpenBSD, but
2069 most kernel versions work. Please note that currently older Solaris kernels
2070 don't work probably due to interface issues between OpenBIOS and
2073 @c man begin OPTIONS
2075 The following options are specific to the Sparc32 emulation:
2079 @item -g @var{W}x@var{H}x[x@var{DEPTH}]
2081 Set the initial graphics mode. For TCX, the default is 1024x768x8 with the
2082 option of 1024x768x24. For cgthree, the default is 1024x768x8 with the option
2083 of 1152x900x8 for people who wish to use OBP.
2085 @item -prom-env @var{string}
2087 Set OpenBIOS variables in NVRAM, for example:
2090 qemu-system-sparc -prom-env 'auto-boot?=false' \
2091 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single'
2094 @item -M [SS-4|SS-5|SS-10|SS-20|SS-600MP|LX|Voyager|SPARCClassic] [|SPARCbook]
2096 Set the emulated machine type. Default is SS-5.
2102 @node Sparc64 System emulator
2103 @section Sparc64 System emulator
2104 @cindex system emulation (Sparc64)
2106 Use the executable @file{qemu-system-sparc64} to simulate a Sun4u
2107 (UltraSPARC PC-like machine), Sun4v (T1 PC-like machine), or generic
2108 Niagara (T1) machine. The Sun4u emulator is mostly complete, being
2109 able to run Linux, NetBSD and OpenBSD in headless (-nographic) mode. The
2110 Sun4v and Niagara emulators are still a work in progress.
2112 QEMU emulates the following peripherals:
2116 UltraSparc IIi APB PCI Bridge
2118 PCI VGA compatible card with VESA Bochs Extensions
2120 PS/2 mouse and keyboard
2122 Non Volatile RAM M48T59
2124 PC-compatible serial ports
2126 2 PCI IDE interfaces with hard disk and CD-ROM support
2131 @c man begin OPTIONS
2133 The following options are specific to the Sparc64 emulation:
2137 @item -prom-env @var{string}
2139 Set OpenBIOS variables in NVRAM, for example:
2142 qemu-system-sparc64 -prom-env 'auto-boot?=false'
2145 @item -M [sun4u|sun4v|Niagara]
2147 Set the emulated machine type. The default is sun4u.
2153 @node MIPS System emulator
2154 @section MIPS System emulator
2155 @cindex system emulation (MIPS)
2157 Four executables cover simulation of 32 and 64-bit MIPS systems in
2158 both endian options, @file{qemu-system-mips}, @file{qemu-system-mipsel}
2159 @file{qemu-system-mips64} and @file{qemu-system-mips64el}.
2160 Five different machine types are emulated:
2164 A generic ISA PC-like machine "mips"
2166 The MIPS Malta prototype board "malta"
2168 An ACER Pica "pica61". This machine needs the 64-bit emulator.
2170 MIPS emulator pseudo board "mipssim"
2172 A MIPS Magnum R4000 machine "magnum". This machine needs the 64-bit emulator.
2175 The generic emulation is supported by Debian 'Etch' and is able to
2176 install Debian into a virtual disk image. The following devices are
2181 A range of MIPS CPUs, default is the 24Kf
2183 PC style serial port
2190 The Malta emulation supports the following devices:
2194 Core board with MIPS 24Kf CPU and Galileo system controller
2196 PIIX4 PCI/USB/SMbus controller
2198 The Multi-I/O chip's serial device
2200 PCI network cards (PCnet32 and others)
2202 Malta FPGA serial device
2204 Cirrus (default) or any other PCI VGA graphics card
2207 The ACER Pica emulation supports:
2213 PC-style IRQ and DMA controllers
2220 The mipssim pseudo board emulation provides an environment similar
2221 to what the proprietary MIPS emulator uses for running Linux.
2226 A range of MIPS CPUs, default is the 24Kf
2228 PC style serial port
2230 MIPSnet network emulation
2233 The MIPS Magnum R4000 emulation supports:
2239 PC-style IRQ controller
2249 @node ARM System emulator
2250 @section ARM System emulator
2251 @cindex system emulation (ARM)
2253 Use the executable @file{qemu-system-arm} to simulate a ARM
2254 machine. The ARM Integrator/CP board is emulated with the following
2259 ARM926E, ARM1026E, ARM946E, ARM1136 or Cortex-A8 CPU
2263 SMC 91c111 Ethernet adapter
2265 PL110 LCD controller
2267 PL050 KMI with PS/2 keyboard and mouse.
2269 PL181 MultiMedia Card Interface with SD card.
2272 The ARM Versatile baseboard is emulated with the following devices:
2276 ARM926E, ARM1136 or Cortex-A8 CPU
2278 PL190 Vectored Interrupt Controller
2282 SMC 91c111 Ethernet adapter
2284 PL110 LCD controller
2286 PL050 KMI with PS/2 keyboard and mouse.
2288 PCI host bridge. Note the emulated PCI bridge only provides access to
2289 PCI memory space. It does not provide access to PCI IO space.
2290 This means some devices (eg. ne2k_pci NIC) are not usable, and others
2291 (eg. rtl8139 NIC) are only usable when the guest drivers use the memory
2292 mapped control registers.
2294 PCI OHCI USB controller.
2296 LSI53C895A PCI SCSI Host Bus Adapter with hard disk and CD-ROM devices.
2298 PL181 MultiMedia Card Interface with SD card.
2301 Several variants of the ARM RealView baseboard are emulated,
2302 including the EB, PB-A8 and PBX-A9. Due to interactions with the
2303 bootloader, only certain Linux kernel configurations work out
2304 of the box on these boards.
2306 Kernels for the PB-A8 board should have CONFIG_REALVIEW_HIGH_PHYS_OFFSET
2307 enabled in the kernel, and expect 512M RAM. Kernels for The PBX-A9 board
2308 should have CONFIG_SPARSEMEM enabled, CONFIG_REALVIEW_HIGH_PHYS_OFFSET
2309 disabled and expect 1024M RAM.
2311 The following devices are emulated:
2315 ARM926E, ARM1136, ARM11MPCore, Cortex-A8 or Cortex-A9 MPCore CPU
2317 ARM AMBA Generic/Distributed Interrupt Controller
2321 SMC 91c111 or SMSC LAN9118 Ethernet adapter
2323 PL110 LCD controller
2325 PL050 KMI with PS/2 keyboard and mouse
2329 PCI OHCI USB controller
2331 LSI53C895A PCI SCSI Host Bus Adapter with hard disk and CD-ROM devices
2333 PL181 MultiMedia Card Interface with SD card.
2336 The XScale-based clamshell PDA models ("Spitz", "Akita", "Borzoi"
2337 and "Terrier") emulation includes the following peripherals:
2341 Intel PXA270 System-on-chip (ARM V5TE core)
2345 IBM/Hitachi DSCM microdrive in a PXA PCMCIA slot - not in "Akita"
2347 On-chip OHCI USB controller
2349 On-chip LCD controller
2351 On-chip Real Time Clock
2353 TI ADS7846 touchscreen controller on SSP bus
2355 Maxim MAX1111 analog-digital converter on I@math{^2}C bus
2357 GPIO-connected keyboard controller and LEDs
2359 Secure Digital card connected to PXA MMC/SD host
2363 WM8750 audio CODEC on I@math{^2}C and I@math{^2}S busses
2366 The Palm Tungsten|E PDA (codename "Cheetah") emulation includes the
2371 Texas Instruments OMAP310 System-on-chip (ARM 925T core)
2373 ROM and RAM memories (ROM firmware image can be loaded with -option-rom)
2375 On-chip LCD controller
2377 On-chip Real Time Clock
2379 TI TSC2102i touchscreen controller / analog-digital converter / Audio
2380 CODEC, connected through MicroWire and I@math{^2}S busses
2382 GPIO-connected matrix keypad
2384 Secure Digital card connected to OMAP MMC/SD host
2389 Nokia N800 and N810 internet tablets (known also as RX-34 and RX-44 / 48)
2390 emulation supports the following elements:
2394 Texas Instruments OMAP2420 System-on-chip (ARM 1136 core)
2396 RAM and non-volatile OneNAND Flash memories
2398 Display connected to EPSON remote framebuffer chip and OMAP on-chip
2399 display controller and a LS041y3 MIPI DBI-C controller
2401 TI TSC2301 (in N800) and TI TSC2005 (in N810) touchscreen controllers
2402 driven through SPI bus
2404 National Semiconductor LM8323-controlled qwerty keyboard driven
2405 through I@math{^2}C bus
2407 Secure Digital card connected to OMAP MMC/SD host
2409 Three OMAP on-chip UARTs and on-chip STI debugging console
2411 A Bluetooth(R) transceiver and HCI connected to an UART
2413 Mentor Graphics "Inventra" dual-role USB controller embedded in a TI
2414 TUSB6010 chip - only USB host mode is supported
2416 TI TMP105 temperature sensor driven through I@math{^2}C bus
2418 TI TWL92230C power management companion with an RTC on I@math{^2}C bus
2420 Nokia RETU and TAHVO multi-purpose chips with an RTC, connected
2424 The Luminary Micro Stellaris LM3S811EVB emulation includes the following
2431 64k Flash and 8k SRAM.
2433 Timers, UARTs, ADC and I@math{^2}C interface.
2435 OSRAM Pictiva 96x16 OLED with SSD0303 controller on I@math{^2}C bus.
2438 The Luminary Micro Stellaris LM3S6965EVB emulation includes the following
2445 256k Flash and 64k SRAM.
2447 Timers, UARTs, ADC, I@math{^2}C and SSI interfaces.
2449 OSRAM Pictiva 128x64 OLED with SSD0323 controller connected via SSI.
2452 The Freecom MusicPal internet radio emulation includes the following
2457 Marvell MV88W8618 ARM core.
2459 32 MB RAM, 256 KB SRAM, 8 MB flash.
2463 MV88W8xx8 Ethernet controller
2465 MV88W8618 audio controller, WM8750 CODEC and mixer
2467 128×64 display with brightness control
2469 2 buttons, 2 navigation wheels with button function
2472 The Siemens SX1 models v1 and v2 (default) basic emulation.
2473 The emulation includes the following elements:
2477 Texas Instruments OMAP310 System-on-chip (ARM 925T core)
2479 ROM and RAM memories (ROM firmware image can be loaded with -pflash)
2481 1 Flash of 16MB and 1 Flash of 8MB
2485 On-chip LCD controller
2487 On-chip Real Time Clock
2489 Secure Digital card connected to OMAP MMC/SD host
2494 A Linux 2.6 test image is available on the QEMU web site. More
2495 information is available in the QEMU mailing-list archive.
2497 @c man begin OPTIONS
2499 The following options are specific to the ARM emulation:
2504 Enable semihosting syscall emulation.
2506 On ARM this implements the "Angel" interface.
2508 Note that this allows guest direct access to the host filesystem,
2509 so should only be used with trusted guest OS.
2513 @node ColdFire System emulator
2514 @section ColdFire System emulator
2515 @cindex system emulation (ColdFire)
2516 @cindex system emulation (M68K)
2518 Use the executable @file{qemu-system-m68k} to simulate a ColdFire machine.
2519 The emulator is able to boot a uClinux kernel.
2521 The M5208EVB emulation includes the following devices:
2525 MCF5208 ColdFire V2 Microprocessor (ISA A+ with EMAC).
2527 Three Two on-chip UARTs.
2529 Fast Ethernet Controller (FEC)
2532 The AN5206 emulation includes the following devices:
2536 MCF5206 ColdFire V2 Microprocessor.
2541 @c man begin OPTIONS
2543 The following options are specific to the ColdFire emulation:
2548 Enable semihosting syscall emulation.
2550 On M68K this implements the "ColdFire GDB" interface used by libgloss.
2552 Note that this allows guest direct access to the host filesystem,
2553 so should only be used with trusted guest OS.
2557 @node Cris System emulator
2558 @section Cris System emulator
2559 @cindex system emulation (Cris)
2563 @node Microblaze System emulator
2564 @section Microblaze System emulator
2565 @cindex system emulation (Microblaze)
2569 @node SH4 System emulator
2570 @section SH4 System emulator
2571 @cindex system emulation (SH4)
2575 @node Xtensa System emulator
2576 @section Xtensa System emulator
2577 @cindex system emulation (Xtensa)
2579 Two executables cover simulation of both Xtensa endian options,
2580 @file{qemu-system-xtensa} and @file{qemu-system-xtensaeb}.
2581 Two different machine types are emulated:
2585 Xtensa emulator pseudo board "sim"
2587 Avnet LX60/LX110/LX200 board
2590 The sim pseudo board emulation provides an environment similar
2591 to one provided by the proprietary Tensilica ISS.
2596 A range of Xtensa CPUs, default is the DC232B
2598 Console and filesystem access via semihosting calls
2601 The Avnet LX60/LX110/LX200 emulation supports:
2605 A range of Xtensa CPUs, default is the DC232B
2609 OpenCores 10/100 Mbps Ethernet MAC
2612 @c man begin OPTIONS
2614 The following options are specific to the Xtensa emulation:
2619 Enable semihosting syscall emulation.
2621 Xtensa semihosting provides basic file IO calls, such as open/read/write/seek/select.
2622 Tensilica baremetal libc for ISS and linux platform "sim" use this interface.
2624 Note that this allows guest direct access to the host filesystem,
2625 so should only be used with trusted guest OS.
2628 @node QEMU User space emulator
2629 @chapter QEMU User space emulator
2632 * Supported Operating Systems ::
2633 * Linux User space emulator::
2634 * BSD User space emulator ::
2637 @node Supported Operating Systems
2638 @section Supported Operating Systems
2640 The following OS are supported in user space emulation:
2644 Linux (referred as qemu-linux-user)
2646 BSD (referred as qemu-bsd-user)
2649 @node Linux User space emulator
2650 @section Linux User space emulator
2655 * Command line options::
2660 @subsection Quick Start
2662 In order to launch a Linux process, QEMU needs the process executable
2663 itself and all the target (x86) dynamic libraries used by it.
2667 @item On x86, you can just try to launch any process by using the native
2671 qemu-i386 -L / /bin/ls
2674 @code{-L /} tells that the x86 dynamic linker must be searched with a
2677 @item Since QEMU is also a linux process, you can launch QEMU with
2678 QEMU (NOTE: you can only do that if you compiled QEMU from the sources):
2681 qemu-i386 -L / qemu-i386 -L / /bin/ls
2684 @item On non x86 CPUs, you need first to download at least an x86 glibc
2685 (@file{qemu-runtime-i386-XXX-.tar.gz} on the QEMU web page). Ensure that
2686 @code{LD_LIBRARY_PATH} is not set:
2689 unset LD_LIBRARY_PATH
2692 Then you can launch the precompiled @file{ls} x86 executable:
2695 qemu-i386 tests/i386/ls
2697 You can look at @file{scripts/qemu-binfmt-conf.sh} so that
2698 QEMU is automatically launched by the Linux kernel when you try to
2699 launch x86 executables. It requires the @code{binfmt_misc} module in the
2702 @item The x86 version of QEMU is also included. You can try weird things such as:
2704 qemu-i386 /usr/local/qemu-i386/bin/qemu-i386 \
2705 /usr/local/qemu-i386/bin/ls-i386
2711 @subsection Wine launch
2715 @item Ensure that you have a working QEMU with the x86 glibc
2716 distribution (see previous section). In order to verify it, you must be
2720 qemu-i386 /usr/local/qemu-i386/bin/ls-i386
2723 @item Download the binary x86 Wine install
2724 (@file{qemu-XXX-i386-wine.tar.gz} on the QEMU web page).
2726 @item Configure Wine on your account. Look at the provided script
2727 @file{/usr/local/qemu-i386/@/bin/wine-conf.sh}. Your previous
2728 @code{$@{HOME@}/.wine} directory is saved to @code{$@{HOME@}/.wine.org}.
2730 @item Then you can try the example @file{putty.exe}:
2733 qemu-i386 /usr/local/qemu-i386/wine/bin/wine \
2734 /usr/local/qemu-i386/wine/c/Program\ Files/putty.exe
2739 @node Command line options
2740 @subsection Command line options
2743 usage: qemu-i386 [-h] [-d] [-L path] [-s size] [-cpu model] [-g port] [-B offset] [-R size] program [arguments...]
2750 Set the x86 elf interpreter prefix (default=/usr/local/qemu-i386)
2752 Set the x86 stack size in bytes (default=524288)
2754 Select CPU model (-cpu help for list and additional feature selection)
2755 @item -E @var{var}=@var{value}
2756 Set environment @var{var} to @var{value}.
2758 Remove @var{var} from the environment.
2760 Offset guest address by the specified number of bytes. This is useful when
2761 the address region required by guest applications is reserved on the host.
2762 This option is currently only supported on some hosts.
2764 Pre-allocate a guest virtual address space of the given size (in bytes).
2765 "G", "M", and "k" suffixes may be used when specifying the size.
2772 Activate logging of the specified items (use '-d help' for a list of log items)
2774 Act as if the host page size was 'pagesize' bytes
2776 Wait gdb connection to port
2778 Run the emulation in single step mode.
2781 Environment variables:
2785 Print system calls and arguments similar to the 'strace' program
2786 (NOTE: the actual 'strace' program will not work because the user
2787 space emulator hasn't implemented ptrace). At the moment this is
2788 incomplete. All system calls that don't have a specific argument
2789 format are printed with information for six arguments. Many
2790 flag-style arguments don't have decoders and will show up as numbers.
2793 @node Other binaries
2794 @subsection Other binaries
2796 @cindex user mode (Alpha)
2797 @command{qemu-alpha} TODO.
2799 @cindex user mode (ARM)
2800 @command{qemu-armeb} TODO.
2802 @cindex user mode (ARM)
2803 @command{qemu-arm} is also capable of running ARM "Angel" semihosted ELF
2804 binaries (as implemented by the arm-elf and arm-eabi Newlib/GDB
2805 configurations), and arm-uclinux bFLT format binaries.
2807 @cindex user mode (ColdFire)
2808 @cindex user mode (M68K)
2809 @command{qemu-m68k} is capable of running semihosted binaries using the BDM
2810 (m5xxx-ram-hosted.ld) or m68k-sim (sim.ld) syscall interfaces, and
2811 coldfire uClinux bFLT format binaries.
2813 The binary format is detected automatically.
2815 @cindex user mode (Cris)
2816 @command{qemu-cris} TODO.
2818 @cindex user mode (i386)
2819 @command{qemu-i386} TODO.
2820 @command{qemu-x86_64} TODO.
2822 @cindex user mode (Microblaze)
2823 @command{qemu-microblaze} TODO.
2825 @cindex user mode (MIPS)
2826 @command{qemu-mips} TODO.
2827 @command{qemu-mipsel} TODO.
2829 @cindex user mode (PowerPC)
2830 @command{qemu-ppc64abi32} TODO.
2831 @command{qemu-ppc64} TODO.
2832 @command{qemu-ppc} TODO.
2834 @cindex user mode (SH4)
2835 @command{qemu-sh4eb} TODO.
2836 @command{qemu-sh4} TODO.
2838 @cindex user mode (SPARC)
2839 @command{qemu-sparc} can execute Sparc32 binaries (Sparc32 CPU, 32 bit ABI).
2841 @command{qemu-sparc32plus} can execute Sparc32 and SPARC32PLUS binaries
2842 (Sparc64 CPU, 32 bit ABI).
2844 @command{qemu-sparc64} can execute some Sparc64 (Sparc64 CPU, 64 bit ABI) and
2845 SPARC32PLUS binaries (Sparc64 CPU, 32 bit ABI).
2847 @node BSD User space emulator
2848 @section BSD User space emulator
2853 * BSD Command line options::
2857 @subsection BSD Status
2861 target Sparc64 on Sparc64: Some trivial programs work.
2864 @node BSD Quick Start
2865 @subsection Quick Start
2867 In order to launch a BSD process, QEMU needs the process executable
2868 itself and all the target dynamic libraries used by it.
2872 @item On Sparc64, you can just try to launch any process by using the native
2876 qemu-sparc64 /bin/ls
2881 @node BSD Command line options
2882 @subsection Command line options
2885 usage: qemu-sparc64 [-h] [-d] [-L path] [-s size] [-bsd type] program [arguments...]
2892 Set the library root path (default=/)
2894 Set the stack size in bytes (default=524288)
2895 @item -ignore-environment
2896 Start with an empty environment. Without this option,
2897 the initial environment is a copy of the caller's environment.
2898 @item -E @var{var}=@var{value}
2899 Set environment @var{var} to @var{value}.
2901 Remove @var{var} from the environment.
2903 Set the type of the emulated BSD Operating system. Valid values are
2904 FreeBSD, NetBSD and OpenBSD (default).
2911 Activate logging of the specified items (use '-d help' for a list of log items)
2913 Act as if the host page size was 'pagesize' bytes
2915 Run the emulation in single step mode.
2919 @chapter Compilation from the sources
2924 * Cross compilation for Windows with Linux::
2932 @subsection Compilation
2934 First you must decompress the sources:
2937 tar zxvf qemu-x.y.z.tar.gz
2941 Then you configure QEMU and build it (usually no options are needed):
2947 Then type as root user:
2951 to install QEMU in @file{/usr/local}.
2957 @item Install the current versions of MSYS and MinGW from
2958 @url{http://www.mingw.org/}. You can find detailed installation
2959 instructions in the download section and the FAQ.
2962 the MinGW development library of SDL 1.2.x
2963 (@file{SDL-devel-1.2.x-@/mingw32.tar.gz}) from
2964 @url{http://www.libsdl.org}. Unpack it in a temporary place and
2965 edit the @file{sdl-config} script so that it gives the
2966 correct SDL directory when invoked.
2968 @item Install the MinGW version of zlib and make sure
2969 @file{zlib.h} and @file{libz.dll.a} are in
2970 MinGW's default header and linker search paths.
2972 @item Extract the current version of QEMU.
2974 @item Start the MSYS shell (file @file{msys.bat}).
2976 @item Change to the QEMU directory. Launch @file{./configure} and
2977 @file{make}. If you have problems using SDL, verify that
2978 @file{sdl-config} can be launched from the MSYS command line.
2980 @item You can install QEMU in @file{Program Files/QEMU} by typing
2981 @file{make install}. Don't forget to copy @file{SDL.dll} in
2982 @file{Program Files/QEMU}.
2986 @node Cross compilation for Windows with Linux
2987 @section Cross compilation for Windows with Linux
2991 Install the MinGW cross compilation tools available at
2992 @url{http://www.mingw.org/}.
2995 the MinGW development library of SDL 1.2.x
2996 (@file{SDL-devel-1.2.x-@/mingw32.tar.gz}) from
2997 @url{http://www.libsdl.org}. Unpack it in a temporary place and
2998 edit the @file{sdl-config} script so that it gives the
2999 correct SDL directory when invoked. Set up the @code{PATH} environment
3000 variable so that @file{sdl-config} can be launched by
3001 the QEMU configuration script.
3003 @item Install the MinGW version of zlib and make sure
3004 @file{zlib.h} and @file{libz.dll.a} are in
3005 MinGW's default header and linker search paths.
3008 Configure QEMU for Windows cross compilation:
3010 PATH=/usr/i686-pc-mingw32/sys-root/mingw/bin:$PATH ./configure --cross-prefix='i686-pc-mingw32-'
3012 The example assumes @file{sdl-config} is installed under @file{/usr/i686-pc-mingw32/sys-root/mingw/bin} and
3013 MinGW cross compilation tools have names like @file{i686-pc-mingw32-gcc} and @file{i686-pc-mingw32-strip}.
3014 We set the @code{PATH} environment variable to ensure the MinGW version of @file{sdl-config} is used and
3015 use --cross-prefix to specify the name of the cross compiler.
3016 You can also use --prefix to set the Win32 install path which defaults to @file{c:/Program Files/QEMU}.
3018 Under Fedora Linux, you can run:
3020 yum -y install mingw32-gcc mingw32-SDL mingw32-zlib
3022 to get a suitable cross compilation environment.
3024 @item You can install QEMU in the installation directory by typing
3025 @code{make install}. Don't forget to copy @file{SDL.dll} and @file{zlib1.dll} into the
3026 installation directory.
3030 Wine can be used to launch the resulting qemu-system-i386.exe
3031 and all other qemu-system-@var{target}.exe compiled for Win32.
3036 System Requirements:
3038 @item Mac OS 10.5 or higher
3039 @item The clang compiler shipped with Xcode 4.2 or higher,
3040 or GCC 4.3 or higher
3043 Additional Requirements (install in order):
3045 @item libffi: @uref{https://sourceware.org/libffi/}
3046 @item gettext: @uref{http://www.gnu.org/software/gettext/}
3047 @item glib: @uref{http://ftp.gnome.org/pub/GNOME/sources/glib/}
3048 @item pkg-config: @uref{http://www.freedesktop.org/wiki/Software/pkg-config/}
3049 @item autoconf: @uref{http://www.gnu.org/software/autoconf/autoconf.html}
3050 @item automake: @uref{http://www.gnu.org/software/automake/}
3051 @item libtool: @uref{http://www.gnu.org/software/libtool/}
3052 @item pixman: @uref{http://www.pixman.org/}
3055 * You may find it easiest to get these from a third-party packager
3056 such as Homebrew, Macports, or Fink.
3058 After downloading the QEMU source code, double-click it to expand it.
3060 Then configure and make QEMU:
3066 If you have a recent version of Mac OS X (OSX 10.7 or better
3067 with Xcode 4.2 or better) we recommend building QEMU with the
3068 default compiler provided by Apple, for your version of Mac OS X
3069 (which will be 'clang'). The configure script will
3070 automatically pick this.
3072 Note: If after the configure step you see a message like this:
3074 ERROR: Your compiler does not support the __thread specifier for
3075 Thread-Local Storage (TLS). Please upgrade to a version that does.
3077 you may have to build your own version of gcc from source. Expect that to take
3078 several hours. More information can be found here:
3079 @uref{https://gcc.gnu.org/install/} @*
3081 These are some of the third party binaries of gcc available for download:
3083 @item Homebrew: @uref{http://brew.sh/}
3084 @item @uref{https://www.litebeam.net/gcc/gcc_472.pkg}
3085 @item @uref{http://www.macports.org/ports.php?by=name&substr=gcc}
3088 You can have several versions of GCC on your system. To specify a certain version,
3089 use the --cc and --cxx options.
3091 ./configure --cxx=<path of your c++ compiler> --cc=<path of your c compiler> <other options>
3095 @section Make targets
3101 Make everything which is typically needed.
3110 Remove most files which were built during make.
3112 @item make distclean
3113 Remove everything which was built during make.
3119 Create documentation in dvi, html, info or pdf format.
3124 @item make defconfig
3125 (Re-)create some build configuration files.
3126 User made changes will be overwritten.
3137 QEMU is a trademark of Fabrice Bellard.
3139 QEMU is released under the GNU General Public License (TODO: add link).
3140 Parts of QEMU have specific licenses, see file LICENSE.
3142 TODO (refer to file LICENSE, include it, include the GPL?)
3156 @section Concept Index
3157 This is the main index. Should we combine all keywords in one index? TODO
3160 @node Function Index
3161 @section Function Index
3162 This index could be used for command line options and monitor functions.
3165 @node Keystroke Index
3166 @section Keystroke Index
3168 This is a list of all keystrokes which have a special function
3169 in system emulation.
3174 @section Program Index
3177 @node Data Type Index
3178 @section Data Type Index
3180 This index could be used for qdev device names and options.
3184 @node Variable Index
3185 @section Variable Index