Statistics
| Branch: | Revision:

root / qemu-doc.texi @ 7d85892b

History | View | Annotate | Download (81.5 kB)

1
\input texinfo @c -*- texinfo -*-
2
@c %**start of header
3
@setfilename qemu-doc.info
4
@settitle QEMU Emulator User Documentation
5
@exampleindent 0
6
@paragraphindent 0
7
@c %**end of header
8

    
9
@iftex
10
@titlepage
11
@sp 7
12
@center @titlefont{QEMU Emulator}
13
@sp 1
14
@center @titlefont{User Documentation}
15
@sp 3
16
@end titlepage
17
@end iftex
18

    
19
@ifnottex
20
@node Top
21
@top
22

    
23
@menu
24
* Introduction::
25
* Installation::
26
* QEMU PC System emulator::
27
* QEMU System emulator for non PC targets::
28
* QEMU User space emulator::
29
* compilation:: Compilation from the sources
30
* Index::
31
@end menu
32
@end ifnottex
33

    
34
@contents
35

    
36
@node Introduction
37
@chapter Introduction
38

    
39
@menu
40
* intro_features:: Features
41
@end menu
42

    
43
@node intro_features
44
@section Features
45

    
46
QEMU is a FAST! processor emulator using dynamic translation to
47
achieve good emulation speed.
48

    
49
QEMU has two operating modes:
50

    
51
@itemize @minus
52

    
53
@item
54
Full system emulation. In this mode, QEMU emulates a full system (for
55
example a PC), including one or several processors and various
56
peripherals. It can be used to launch different Operating Systems
57
without rebooting the PC or to debug system code.
58

    
59
@item
60
User mode emulation. In this mode, QEMU can launch
61
processes compiled for one CPU on another CPU. It can be used to
62
launch the Wine Windows API emulator (@url{http://www.winehq.org}) or
63
to ease cross-compilation and cross-debugging.
64

    
65
@end itemize
66

    
67
QEMU can run without an host kernel driver and yet gives acceptable
68
performance.
69

    
70
For system emulation, the following hardware targets are supported:
71
@itemize
72
@item PC (x86 or x86_64 processor)
73
@item ISA PC (old style PC without PCI bus)
74
@item PREP (PowerPC processor)
75
@item G3 BW PowerMac (PowerPC processor)
76
@item Mac99 PowerMac (PowerPC processor, in progress)
77
@item Sun4m/Sun4d (32-bit Sparc processor)
78
@item Sun4u (64-bit Sparc processor, in progress)
79
@item Malta board (32-bit and 64-bit MIPS processors)
80
@item ARM Integrator/CP (ARM)
81
@item ARM Versatile baseboard (ARM)
82
@item ARM RealView Emulation baseboard (ARM)
83
@item Spitz, Akita, Borzoi and Terrier PDAs (PXA270 processor)
84
@item Luminary Micro LM3S811EVB (ARM Cortex-M3)
85
@item Luminary Micro LM3S6965EVB (ARM Cortex-M3)
86
@item Freescale MCF5208EVB (ColdFire V2).
87
@item Arnewsh MCF5206 evaluation board (ColdFire V2).
88
@item Palm Tungsten|E PDA (OMAP310 processor)
89
@end itemize
90

    
91
For user emulation, x86, PowerPC, ARM, 32-bit MIPS, Sparc32/64 and ColdFire(m68k) CPUs are supported.
92

    
93
@node Installation
94
@chapter Installation
95

    
96
If you want to compile QEMU yourself, see @ref{compilation}.
97

    
98
@menu
99
* install_linux::   Linux
100
* install_windows:: Windows
101
* install_mac::     Macintosh
102
@end menu
103

    
104
@node install_linux
105
@section Linux
106

    
107
If a precompiled package is available for your distribution - you just
108
have to install it. Otherwise, see @ref{compilation}.
109

    
110
@node install_windows
111
@section Windows
112

    
113
Download the experimental binary installer at
114
@url{http://www.free.oszoo.org/@/download.html}.
115

    
116
@node install_mac
117
@section Mac OS X
118

    
119
Download the experimental binary installer at
120
@url{http://www.free.oszoo.org/@/download.html}.
121

    
122
@node QEMU PC System emulator
123
@chapter QEMU PC System emulator
124

    
125
@menu
126
* pcsys_introduction:: Introduction
127
* pcsys_quickstart::   Quick Start
128
* sec_invocation::     Invocation
129
* pcsys_keys::         Keys
130
* pcsys_monitor::      QEMU Monitor
131
* disk_images::        Disk Images
132
* pcsys_network::      Network emulation
133
* direct_linux_boot::  Direct Linux Boot
134
* pcsys_usb::          USB emulation
135
* vnc_security::       VNC security
136
* gdb_usage::          GDB usage
137
* pcsys_os_specific::  Target OS specific information
138
@end menu
139

    
140
@node pcsys_introduction
141
@section Introduction
142

    
143
@c man begin DESCRIPTION
144

    
145
The QEMU PC System emulator simulates the
146
following peripherals:
147

    
148
@itemize @minus
149
@item
150
i440FX host PCI bridge and PIIX3 PCI to ISA bridge
151
@item
152
Cirrus CLGD 5446 PCI VGA card or dummy VGA card with Bochs VESA
153
extensions (hardware level, including all non standard modes).
154
@item
155
PS/2 mouse and keyboard
156
@item
157
2 PCI IDE interfaces with hard disk and CD-ROM support
158
@item
159
Floppy disk
160
@item
161
PCI/ISA PCI network adapters
162
@item
163
Serial ports
164
@item
165
Creative SoundBlaster 16 sound card
166
@item
167
ENSONIQ AudioPCI ES1370 sound card
168
@item
169
Adlib(OPL2) - Yamaha YM3812 compatible chip
170
@item
171
PCI UHCI USB controller and a virtual USB hub.
172
@end itemize
173

    
174
SMP is supported with up to 255 CPUs.
175

    
176
Note that adlib is only available when QEMU was configured with
177
-enable-adlib
178

    
179
QEMU uses the PC BIOS from the Bochs project and the Plex86/Bochs LGPL
180
VGA BIOS.
181

    
182
QEMU uses YM3812 emulation by Tatsuyuki Satoh.
183

    
184
@c man end
185

    
186
@node pcsys_quickstart
187
@section Quick Start
188

    
189
Download and uncompress the linux image (@file{linux.img}) and type:
190

    
191
@example
192
qemu linux.img
193
@end example
194

    
195
Linux should boot and give you a prompt.
196

    
197
@node sec_invocation
198
@section Invocation
199

    
200
@example
201
@c man begin SYNOPSIS
202
usage: qemu [options] [@var{disk_image}]
203
@c man end
204
@end example
205

    
206
@c man begin OPTIONS
207
@var{disk_image} is a raw hard disk image for IDE hard disk 0.
208

    
209
General options:
210
@table @option
211
@item -M @var{machine}
212
Select the emulated @var{machine} (@code{-M ?} for list)
213

    
214
@item -fda @var{file}
215
@item -fdb @var{file}
216
Use @var{file} as floppy disk 0/1 image (@pxref{disk_images}). You can
217
use the host floppy by using @file{/dev/fd0} as filename (@pxref{host_drives}).
218

    
219
@item -hda @var{file}
220
@item -hdb @var{file}
221
@item -hdc @var{file}
222
@item -hdd @var{file}
223
Use @var{file} as hard disk 0, 1, 2 or 3 image (@pxref{disk_images}).
224

    
225
@item -cdrom @var{file}
226
Use @var{file} as CD-ROM image (you cannot use @option{-hdc} and
227
@option{-cdrom} at the same time). You can use the host CD-ROM by
228
using @file{/dev/cdrom} as filename (@pxref{host_drives}).
229

    
230
@item -drive @var{option}[,@var{option}[,@var{option}[,...]]]
231

    
232
Define a new drive. Valid options are:
233

    
234
@table @code
235
@item file=@var{file}
236
This option defines which disk image (@pxref{disk_images}) to use with
237
this drive.
238
@item if=@var{interface}
239
This option defines on which type on interface the drive is connected.
240
Available types are: ide, scsi, sd, mtd, floppy, pflash.
241
@item bus=@var{bus},unit=@var{unit}
242
These options define where is connected the drive by defining the bus number and
243
the unit id.
244
@item index=@var{index}
245
This option defines where is connected the drive by using an index in the list
246
of available connectors of a given interface type.
247
@item media=@var{media}
248
This option defines the type of the media: disk or cdrom.
249
@item cyls=@var{c},heads=@var{h},secs=@var{s}[,trans=@var{t}]
250
These options have the same definition as they have in @option{-hdachs}.
251
@item snapshot=@var{snapshot}
252
@var{snapshot} is "on" or "off" and allows to enable snapshot for given drive (see @option{-snapshot}).
253
@item cache=@var{cache}
254
@var{cache} is "on" or "off" and allows to disable host cache to access data.
255
@end table
256

    
257
Instead of @option{-cdrom} you can use:
258
@example
259
qemu -drive file=file,index=2,media=cdrom
260
@end example
261

    
262
Instead of @option{-hda}, @option{-hdb}, @option{-hdc}, @option{-hdd}, you can
263
use:
264
@example
265
qemu -drive file=file,index=0,media=disk
266
qemu -drive file=file,index=1,media=disk
267
qemu -drive file=file,index=2,media=disk
268
qemu -drive file=file,index=3,media=disk
269
@end example
270

    
271
You can connect a CDROM to the slave of ide0:
272
@example
273
qemu -drive file=file,if=ide,index=1,media=cdrom
274
@end example
275

    
276
If you don't specify the "file=" argument, you define an empty drive:
277
@example
278
qemu -drive if=ide,index=1,media=cdrom
279
@end example
280

    
281
You can connect a SCSI disk with unit ID 6 on the bus #0:
282
@example
283
qemu -drive file=file,if=scsi,bus=0,unit=6
284
@end example
285

    
286
Instead of @option{-fda}, @option{-fdb}, you can use:
287
@example
288
qemu -drive file=file,index=0,if=floppy
289
qemu -drive file=file,index=1,if=floppy
290
@end example
291

    
292
By default, @var{interface} is "ide" and @var{index} is automatically
293
incremented:
294
@example
295
qemu -drive file=a -drive file=b"
296
@end example
297
is interpreted like:
298
@example
299
qemu -hda a -hdb b
300
@end example
301

    
302
@item -boot [a|c|d|n]
303
Boot on floppy (a), hard disk (c), CD-ROM (d), or Etherboot (n). Hard disk boot
304
is the default.
305

    
306
@item -snapshot
307
Write to temporary files instead of disk image files. In this case,
308
the raw disk image you use is not written back. You can however force
309
the write back by pressing @key{C-a s} (@pxref{disk_images}).
310

    
311
@item -no-fd-bootchk
312
Disable boot signature checking for floppy disks in Bochs BIOS. It may
313
be needed to boot from old floppy disks.
314

    
315
@item -m @var{megs}
316
Set virtual RAM size to @var{megs} megabytes. Default is 128 MiB.
317

    
318
@item -smp @var{n}
319
Simulate an SMP system with @var{n} CPUs. On the PC target, up to 255
320
CPUs are supported. On Sparc32 target, Linux limits the number of usable CPUs
321
to 4.
322

    
323
@item -audio-help
324

    
325
Will show the audio subsystem help: list of drivers, tunable
326
parameters.
327

    
328
@item -soundhw @var{card1}[,@var{card2},...] or -soundhw all
329

    
330
Enable audio and selected sound hardware. Use ? to print all
331
available sound hardware.
332

    
333
@example
334
qemu -soundhw sb16,adlib hda
335
qemu -soundhw es1370 hda
336
qemu -soundhw all hda
337
qemu -soundhw ?
338
@end example
339

    
340
@item -localtime
341
Set the real time clock to local time (the default is to UTC
342
time). This option is needed to have correct date in MS-DOS or
343
Windows.
344

    
345
@item -startdate @var{date}
346
Set the initial date of the real time clock. Valid format for
347
@var{date} are: @code{now} or @code{2006-06-17T16:01:21} or
348
@code{2006-06-17}. The default value is @code{now}.
349

    
350
@item -pidfile @var{file}
351
Store the QEMU process PID in @var{file}. It is useful if you launch QEMU
352
from a script.
353

    
354
@item -daemonize
355
Daemonize the QEMU process after initialization.  QEMU will not detach from
356
standard IO until it is ready to receive connections on any of its devices.
357
This option is a useful way for external programs to launch QEMU without having
358
to cope with initialization race conditions.
359

    
360
@item -win2k-hack
361
Use it when installing Windows 2000 to avoid a disk full bug. After
362
Windows 2000 is installed, you no longer need this option (this option
363
slows down the IDE transfers).
364

    
365
@item -option-rom @var{file}
366
Load the contents of @var{file} as an option ROM.
367
This option is useful to load things like EtherBoot.
368

    
369
@item -name @var{name}
370
Sets the @var{name} of the guest.
371
This name will be display in the SDL window caption.
372
The @var{name} will also be used for the VNC server.
373

    
374
@end table
375

    
376
Display options:
377
@table @option
378

    
379
@item -nographic
380

    
381
Normally, QEMU uses SDL to display the VGA output. With this option,
382
you can totally disable graphical output so that QEMU is a simple
383
command line application. The emulated serial port is redirected on
384
the console. Therefore, you can still use QEMU to debug a Linux kernel
385
with a serial console.
386

    
387
@item -no-frame
388

    
389
Do not use decorations for SDL windows and start them using the whole
390
available screen space. This makes the using QEMU in a dedicated desktop
391
workspace more convenient.
392

    
393
@item -full-screen
394
Start in full screen.
395

    
396
@item -vnc @var{display}[,@var{option}[,@var{option}[,...]]]
397

    
398
Normally, QEMU uses SDL to display the VGA output.  With this option,
399
you can have QEMU listen on VNC display @var{display} and redirect the VGA
400
display over the VNC session.  It is very useful to enable the usb
401
tablet device when using this option (option @option{-usbdevice
402
tablet}). When using the VNC display, you must use the @option{-k}
403
parameter to set the keyboard layout if you are not using en-us. Valid
404
syntax for the @var{display} is
405

    
406
@table @code
407

    
408
@item @var{interface}:@var{d}
409

    
410
TCP connections will only be allowed from @var{interface} on display @var{d}.
411
By convention the TCP port is 5900+@var{d}. Optionally, @var{interface} can
412
be omitted in which case the server will bind to all interfaces.
413

    
414
@item @var{unix}:@var{path}
415

    
416
Connections will be allowed over UNIX domain sockets where @var{path} is the
417
location of a unix socket to listen for connections on.
418

    
419
@item none
420

    
421
VNC is initialized by not started. The monitor @code{change} command can be used
422
to later start the VNC server.
423

    
424
@end table
425

    
426
Following the @var{display} value there may be one or more @var{option} flags
427
separated by commas. Valid options are
428

    
429
@table @code
430

    
431
@item password
432

    
433
Require that password based authentication is used for client connections.
434
The password must be set separately using the @code{change} command in the
435
@ref{pcsys_monitor}
436

    
437
@item tls
438

    
439
Require that client use TLS when communicating with the VNC server. This
440
uses anonymous TLS credentials so is susceptible to a man-in-the-middle
441
attack. It is recommended that this option be combined with either the
442
@var{x509} or @var{x509verify} options.
443

    
444
@item x509=@var{/path/to/certificate/dir}
445

    
446
Valid if @option{tls} is specified. Require that x509 credentials are used
447
for negotiating the TLS session. The server will send its x509 certificate
448
to the client. It is recommended that a password be set on the VNC server
449
to provide authentication of the client when this is used. The path following
450
this option specifies where the x509 certificates are to be loaded from.
451
See the @ref{vnc_security} section for details on generating certificates.
452

    
453
@item x509verify=@var{/path/to/certificate/dir}
454

    
455
Valid if @option{tls} is specified. Require that x509 credentials are used
456
for negotiating the TLS session. The server will send its x509 certificate
457
to the client, and request that the client send its own x509 certificate.
458
The server will validate the client's certificate against the CA certificate,
459
and reject clients when validation fails. If the certificate authority is
460
trusted, this is a sufficient authentication mechanism. You may still wish
461
to set a password on the VNC server as a second authentication layer. The
462
path following this option specifies where the x509 certificates are to
463
be loaded from. See the @ref{vnc_security} section for details on generating
464
certificates.
465

    
466
@end table
467

    
468
@item -k @var{language}
469

    
470
Use keyboard layout @var{language} (for example @code{fr} for
471
French). This option is only needed where it is not easy to get raw PC
472
keycodes (e.g. on Macs, with some X11 servers or with a VNC
473
display). You don't normally need to use it on PC/Linux or PC/Windows
474
hosts.
475

    
476
The available layouts are:
477
@example
478
ar  de-ch  es  fo     fr-ca  hu  ja  mk     no  pt-br  sv
479
da  en-gb  et  fr     fr-ch  is  lt  nl     pl  ru     th
480
de  en-us  fi  fr-be  hr     it  lv  nl-be  pt  sl     tr
481
@end example
482

    
483
The default is @code{en-us}.
484

    
485
@end table
486

    
487
USB options:
488
@table @option
489

    
490
@item -usb
491
Enable the USB driver (will be the default soon)
492

    
493
@item -usbdevice @var{devname}
494
Add the USB device @var{devname}. @xref{usb_devices}.
495
@end table
496

    
497
Network options:
498

    
499
@table @option
500

    
501
@item -net nic[,vlan=@var{n}][,macaddr=@var{addr}][,model=@var{type}]
502
Create a new Network Interface Card and connect it to VLAN @var{n} (@var{n}
503
= 0 is the default). The NIC is an ne2k_pci by default on the PC
504
target. Optionally, the MAC address can be changed. If no
505
@option{-net} option is specified, a single NIC is created.
506
Qemu can emulate several different models of network card.
507
Valid values for @var{type} are
508
@code{i82551}, @code{i82557b}, @code{i82559er},
509
@code{ne2k_pci}, @code{ne2k_isa}, @code{pcnet}, @code{rtl8139},
510
@code{smc91c111}, @code{lance} and @code{mcf_fec}.
511
Not all devices are supported on all targets.  Use -net nic,model=?
512
for a list of available devices for your target.
513

    
514
@item -net user[,vlan=@var{n}][,hostname=@var{name}]
515
Use the user mode network stack which requires no administrator
516
privilege to run.  @option{hostname=name} can be used to specify the client
517
hostname reported by the builtin DHCP server.
518

    
519
@item -net tap[,vlan=@var{n}][,fd=@var{h}][,ifname=@var{name}][,script=@var{file}]
520
Connect the host TAP network interface @var{name} to VLAN @var{n} and
521
use the network script @var{file} to configure it. The default
522
network script is @file{/etc/qemu-ifup}. Use @option{script=no} to
523
disable script execution. If @var{name} is not
524
provided, the OS automatically provides one. @option{fd}=@var{h} can be
525
used to specify the handle of an already opened host TAP interface. Example:
526

    
527
@example
528
qemu linux.img -net nic -net tap
529
@end example
530

    
531
More complicated example (two NICs, each one connected to a TAP device)
532
@example
533
qemu linux.img -net nic,vlan=0 -net tap,vlan=0,ifname=tap0 \
534
               -net nic,vlan=1 -net tap,vlan=1,ifname=tap1
535
@end example
536

    
537

    
538
@item -net socket[,vlan=@var{n}][,fd=@var{h}][,listen=[@var{host}]:@var{port}][,connect=@var{host}:@var{port}]
539

    
540
Connect the VLAN @var{n} to a remote VLAN in another QEMU virtual
541
machine using a TCP socket connection. If @option{listen} is
542
specified, QEMU waits for incoming connections on @var{port}
543
(@var{host} is optional). @option{connect} is used to connect to
544
another QEMU instance using the @option{listen} option. @option{fd}=@var{h}
545
specifies an already opened TCP socket.
546

    
547
Example:
548
@example
549
# launch a first QEMU instance
550
qemu linux.img -net nic,macaddr=52:54:00:12:34:56 \
551
               -net socket,listen=:1234
552
# connect the VLAN 0 of this instance to the VLAN 0
553
# of the first instance
554
qemu linux.img -net nic,macaddr=52:54:00:12:34:57 \
555
               -net socket,connect=127.0.0.1:1234
556
@end example
557

    
558
@item -net socket[,vlan=@var{n}][,fd=@var{h}][,mcast=@var{maddr}:@var{port}]
559

    
560
Create a VLAN @var{n} shared with another QEMU virtual
561
machines using a UDP multicast socket, effectively making a bus for
562
every QEMU with same multicast address @var{maddr} and @var{port}.
563
NOTES:
564
@enumerate
565
@item
566
Several QEMU can be running on different hosts and share same bus (assuming
567
correct multicast setup for these hosts).
568
@item
569
mcast support is compatible with User Mode Linux (argument @option{eth@var{N}=mcast}), see
570
@url{http://user-mode-linux.sf.net}.
571
@item
572
Use @option{fd=h} to specify an already opened UDP multicast socket.
573
@end enumerate
574

    
575
Example:
576
@example
577
# launch one QEMU instance
578
qemu linux.img -net nic,macaddr=52:54:00:12:34:56 \
579
               -net socket,mcast=230.0.0.1:1234
580
# launch another QEMU instance on same "bus"
581
qemu linux.img -net nic,macaddr=52:54:00:12:34:57 \
582
               -net socket,mcast=230.0.0.1:1234
583
# launch yet another QEMU instance on same "bus"
584
qemu linux.img -net nic,macaddr=52:54:00:12:34:58 \
585
               -net socket,mcast=230.0.0.1:1234
586
@end example
587

    
588
Example (User Mode Linux compat.):
589
@example
590
# launch QEMU instance (note mcast address selected
591
# is UML's default)
592
qemu linux.img -net nic,macaddr=52:54:00:12:34:56 \
593
               -net socket,mcast=239.192.168.1:1102
594
# launch UML
595
/path/to/linux ubd0=/path/to/root_fs eth0=mcast
596
@end example
597

    
598
@item -net none
599
Indicate that no network devices should be configured. It is used to
600
override the default configuration (@option{-net nic -net user}) which
601
is activated if no @option{-net} options are provided.
602

    
603
@item -tftp @var{dir}
604
When using the user mode network stack, activate a built-in TFTP
605
server. The files in @var{dir} will be exposed as the root of a TFTP server.
606
The TFTP client on the guest must be configured in binary mode (use the command
607
@code{bin} of the Unix TFTP client). The host IP address on the guest is as
608
usual 10.0.2.2.
609

    
610
@item -bootp @var{file}
611
When using the user mode network stack, broadcast @var{file} as the BOOTP
612
filename.  In conjunction with @option{-tftp}, this can be used to network boot
613
a guest from a local directory.
614

    
615
Example (using pxelinux):
616
@example
617
qemu -hda linux.img -boot n -tftp /path/to/tftp/files -bootp /pxelinux.0
618
@end example
619

    
620
@item -smb @var{dir}
621
When using the user mode network stack, activate a built-in SMB
622
server so that Windows OSes can access to the host files in @file{@var{dir}}
623
transparently.
624

    
625
In the guest Windows OS, the line:
626
@example
627
10.0.2.4 smbserver
628
@end example
629
must be added in the file @file{C:\WINDOWS\LMHOSTS} (for windows 9x/Me)
630
or @file{C:\WINNT\SYSTEM32\DRIVERS\ETC\LMHOSTS} (Windows NT/2000).
631

    
632
Then @file{@var{dir}} can be accessed in @file{\\smbserver\qemu}.
633

    
634
Note that a SAMBA server must be installed on the host OS in
635
@file{/usr/sbin/smbd}. QEMU was tested successfully with smbd version
636
2.2.7a from the Red Hat 9 and version 3.0.10-1.fc3 from Fedora Core 3.
637

    
638
@item -redir [tcp|udp]:@var{host-port}:[@var{guest-host}]:@var{guest-port}
639

    
640
When using the user mode network stack, redirect incoming TCP or UDP
641
connections to the host port @var{host-port} to the guest
642
@var{guest-host} on guest port @var{guest-port}. If @var{guest-host}
643
is not specified, its value is 10.0.2.15 (default address given by the
644
built-in DHCP server).
645

    
646
For example, to redirect host X11 connection from screen 1 to guest
647
screen 0, use the following:
648

    
649
@example
650
# on the host
651
qemu -redir tcp:6001::6000 [...]
652
# this host xterm should open in the guest X11 server
653
xterm -display :1
654
@end example
655

    
656
To redirect telnet connections from host port 5555 to telnet port on
657
the guest, use the following:
658

    
659
@example
660
# on the host
661
qemu -redir tcp:5555::23 [...]
662
telnet localhost 5555
663
@end example
664

    
665
Then when you use on the host @code{telnet localhost 5555}, you
666
connect to the guest telnet server.
667

    
668
@end table
669

    
670
Linux boot specific: When using these options, you can use a given
671
Linux kernel without installing it in the disk image. It can be useful
672
for easier testing of various kernels.
673

    
674
@table @option
675

    
676
@item -kernel @var{bzImage}
677
Use @var{bzImage} as kernel image.
678

    
679
@item -append @var{cmdline}
680
Use @var{cmdline} as kernel command line
681

    
682
@item -initrd @var{file}
683
Use @var{file} as initial ram disk.
684

    
685
@end table
686

    
687
Debug/Expert options:
688
@table @option
689

    
690
@item -serial @var{dev}
691
Redirect the virtual serial port to host character device
692
@var{dev}. The default device is @code{vc} in graphical mode and
693
@code{stdio} in non graphical mode.
694

    
695
This option can be used several times to simulate up to 4 serials
696
ports.
697

    
698
Use @code{-serial none} to disable all serial ports.
699

    
700
Available character devices are:
701
@table @code
702
@item vc[:WxH]
703
Virtual console. Optionally, a width and height can be given in pixel with
704
@example
705
vc:800x600
706
@end example
707
It is also possible to specify width or height in characters:
708
@example
709
vc:80Cx24C
710
@end example
711
@item pty
712
[Linux only] Pseudo TTY (a new PTY is automatically allocated)
713
@item none
714
No device is allocated.
715
@item null
716
void device
717
@item /dev/XXX
718
[Linux only] Use host tty, e.g. @file{/dev/ttyS0}. The host serial port
719
parameters are set according to the emulated ones.
720
@item /dev/parport@var{N}
721
[Linux only, parallel port only] Use host parallel port
722
@var{N}. Currently SPP and EPP parallel port features can be used.
723
@item file:@var{filename}
724
Write output to @var{filename}. No character can be read.
725
@item stdio
726
[Unix only] standard input/output
727
@item pipe:@var{filename}
728
name pipe @var{filename}
729
@item COM@var{n}
730
[Windows only] Use host serial port @var{n}
731
@item udp:[@var{remote_host}]:@var{remote_port}[@@[@var{src_ip}]:@var{src_port}]
732
This implements UDP Net Console.
733
When @var{remote_host} or @var{src_ip} are not specified
734
they default to @code{0.0.0.0}.
735
When not using a specified @var{src_port} a random port is automatically chosen.
736

    
737
If you just want a simple readonly console you can use @code{netcat} or
738
@code{nc}, by starting qemu with: @code{-serial udp::4555} and nc as:
739
@code{nc -u -l -p 4555}. Any time qemu writes something to that port it
740
will appear in the netconsole session.
741

    
742
If you plan to send characters back via netconsole or you want to stop
743
and start qemu a lot of times, you should have qemu use the same
744
source port each time by using something like @code{-serial
745
udp::4555@@:4556} to qemu. Another approach is to use a patched
746
version of netcat which can listen to a TCP port and send and receive
747
characters via udp.  If you have a patched version of netcat which
748
activates telnet remote echo and single char transfer, then you can
749
use the following options to step up a netcat redirector to allow
750
telnet on port 5555 to access the qemu port.
751
@table @code
752
@item Qemu Options:
753
-serial udp::4555@@:4556
754
@item netcat options:
755
-u -P 4555 -L 0.0.0.0:4556 -t -p 5555 -I -T
756
@item telnet options:
757
localhost 5555
758
@end table
759

    
760

    
761
@item tcp:[@var{host}]:@var{port}[,@var{server}][,nowait][,nodelay]
762
The TCP Net Console has two modes of operation.  It can send the serial
763
I/O to a location or wait for a connection from a location.  By default
764
the TCP Net Console is sent to @var{host} at the @var{port}.  If you use
765
the @var{server} option QEMU will wait for a client socket application
766
to connect to the port before continuing, unless the @code{nowait}
767
option was specified.  The @code{nodelay} option disables the Nagle buffering
768
algorithm.  If @var{host} is omitted, 0.0.0.0 is assumed. Only
769
one TCP connection at a time is accepted. You can use @code{telnet} to
770
connect to the corresponding character device.
771
@table @code
772
@item Example to send tcp console to 192.168.0.2 port 4444
773
-serial tcp:192.168.0.2:4444
774
@item Example to listen and wait on port 4444 for connection
775
-serial tcp::4444,server
776
@item Example to not wait and listen on ip 192.168.0.100 port 4444
777
-serial tcp:192.168.0.100:4444,server,nowait
778
@end table
779

    
780
@item telnet:@var{host}:@var{port}[,server][,nowait][,nodelay]
781
The telnet protocol is used instead of raw tcp sockets.  The options
782
work the same as if you had specified @code{-serial tcp}.  The
783
difference is that the port acts like a telnet server or client using
784
telnet option negotiation.  This will also allow you to send the
785
MAGIC_SYSRQ sequence if you use a telnet that supports sending the break
786
sequence.  Typically in unix telnet you do it with Control-] and then
787
type "send break" followed by pressing the enter key.
788

    
789
@item unix:@var{path}[,server][,nowait]
790
A unix domain socket is used instead of a tcp socket.  The option works the
791
same as if you had specified @code{-serial tcp} except the unix domain socket
792
@var{path} is used for connections.
793

    
794
@item mon:@var{dev_string}
795
This is a special option to allow the monitor to be multiplexed onto
796
another serial port.  The monitor is accessed with key sequence of
797
@key{Control-a} and then pressing @key{c}. See monitor access
798
@ref{pcsys_keys} in the -nographic section for more keys.
799
@var{dev_string} should be any one of the serial devices specified
800
above.  An example to multiplex the monitor onto a telnet server
801
listening on port 4444 would be:
802
@table @code
803
@item -serial mon:telnet::4444,server,nowait
804
@end table
805

    
806
@end table
807

    
808
@item -parallel @var{dev}
809
Redirect the virtual parallel port to host device @var{dev} (same
810
devices as the serial port). On Linux hosts, @file{/dev/parportN} can
811
be used to use hardware devices connected on the corresponding host
812
parallel port.
813

    
814
This option can be used several times to simulate up to 3 parallel
815
ports.
816

    
817
Use @code{-parallel none} to disable all parallel ports.
818

    
819
@item -monitor @var{dev}
820
Redirect the monitor to host device @var{dev} (same devices as the
821
serial port).
822
The default device is @code{vc} in graphical mode and @code{stdio} in
823
non graphical mode.
824

    
825
@item -echr numeric_ascii_value
826
Change the escape character used for switching to the monitor when using
827
monitor and serial sharing.  The default is @code{0x01} when using the
828
@code{-nographic} option.  @code{0x01} is equal to pressing
829
@code{Control-a}.  You can select a different character from the ascii
830
control keys where 1 through 26 map to Control-a through Control-z.  For
831
instance you could use the either of the following to change the escape
832
character to Control-t.
833
@table @code
834
@item -echr 0x14
835
@item -echr 20
836
@end table
837

    
838
@item -s
839
Wait gdb connection to port 1234 (@pxref{gdb_usage}).
840
@item -p @var{port}
841
Change gdb connection port.  @var{port} can be either a decimal number
842
to specify a TCP port, or a host device (same devices as the serial port).
843
@item -S
844
Do not start CPU at startup (you must type 'c' in the monitor).
845
@item -d
846
Output log in /tmp/qemu.log
847
@item -hdachs @var{c},@var{h},@var{s},[,@var{t}]
848
Force hard disk 0 physical geometry (1 <= @var{c} <= 16383, 1 <=
849
@var{h} <= 16, 1 <= @var{s} <= 63) and optionally force the BIOS
850
translation mode (@var{t}=none, lba or auto). Usually QEMU can guess
851
all those parameters. This option is useful for old MS-DOS disk
852
images.
853

    
854
@item -L path
855
Set the directory for the BIOS, VGA BIOS and keymaps.
856

    
857
@item -std-vga
858
Simulate a standard VGA card with Bochs VBE extensions (default is
859
Cirrus Logic GD5446 PCI VGA). If your guest OS supports the VESA 2.0
860
VBE extensions (e.g. Windows XP) and if you want to use high
861
resolution modes (>= 1280x1024x16) then you should use this option.
862

    
863
@item -no-acpi
864
Disable ACPI (Advanced Configuration and Power Interface) support. Use
865
it if your guest OS complains about ACPI problems (PC target machine
866
only).
867

    
868
@item -no-reboot
869
Exit instead of rebooting.
870

    
871
@item -loadvm file
872
Start right away with a saved state (@code{loadvm} in monitor)
873

    
874
@item -semihosting
875
Enable semihosting syscall emulation (ARM and M68K target machines only).
876

    
877
On ARM this implements the "Angel" interface.
878
On M68K this implements the "ColdFire GDB" interface used by libgloss.
879

    
880
Note that this allows guest direct access to the host filesystem,
881
so should only be used with trusted guest OS.
882
@end table
883

    
884
@c man end
885

    
886
@node pcsys_keys
887
@section Keys
888

    
889
@c man begin OPTIONS
890

    
891
During the graphical emulation, you can use the following keys:
892
@table @key
893
@item Ctrl-Alt-f
894
Toggle full screen
895

    
896
@item Ctrl-Alt-n
897
Switch to virtual console 'n'. Standard console mappings are:
898
@table @emph
899
@item 1
900
Target system display
901
@item 2
902
Monitor
903
@item 3
904
Serial port
905
@end table
906

    
907
@item Ctrl-Alt
908
Toggle mouse and keyboard grab.
909
@end table
910

    
911
In the virtual consoles, you can use @key{Ctrl-Up}, @key{Ctrl-Down},
912
@key{Ctrl-PageUp} and @key{Ctrl-PageDown} to move in the back log.
913

    
914
During emulation, if you are using the @option{-nographic} option, use
915
@key{Ctrl-a h} to get terminal commands:
916

    
917
@table @key
918
@item Ctrl-a h
919
Print this help
920
@item Ctrl-a x
921
Exit emulator
922
@item Ctrl-a s
923
Save disk data back to file (if -snapshot)
924
@item Ctrl-a t
925
toggle console timestamps
926
@item Ctrl-a b
927
Send break (magic sysrq in Linux)
928
@item Ctrl-a c
929
Switch between console and monitor
930
@item Ctrl-a Ctrl-a
931
Send Ctrl-a
932
@end table
933
@c man end
934

    
935
@ignore
936

    
937
@c man begin SEEALSO
938
The HTML documentation of QEMU for more precise information and Linux
939
user mode emulator invocation.
940
@c man end
941

    
942
@c man begin AUTHOR
943
Fabrice Bellard
944
@c man end
945

    
946
@end ignore
947

    
948
@node pcsys_monitor
949
@section QEMU Monitor
950

    
951
The QEMU monitor is used to give complex commands to the QEMU
952
emulator. You can use it to:
953

    
954
@itemize @minus
955

    
956
@item
957
Remove or insert removable media images
958
(such as CD-ROM or floppies).
959

    
960
@item
961
Freeze/unfreeze the Virtual Machine (VM) and save or restore its state
962
from a disk file.
963

    
964
@item Inspect the VM state without an external debugger.
965

    
966
@end itemize
967

    
968
@subsection Commands
969

    
970
The following commands are available:
971

    
972
@table @option
973

    
974
@item help or ? [@var{cmd}]
975
Show the help for all commands or just for command @var{cmd}.
976

    
977
@item commit
978
Commit changes to the disk images (if -snapshot is used).
979

    
980
@item info @var{subcommand}
981
Show various information about the system state.
982

    
983
@table @option
984
@item info network
985
show the various VLANs and the associated devices
986
@item info block
987
show the block devices
988
@item info registers
989
show the cpu registers
990
@item info history
991
show the command line history
992
@item info pci
993
show emulated PCI device
994
@item info usb
995
show USB devices plugged on the virtual USB hub
996
@item info usbhost
997
show all USB host devices
998
@item info capture
999
show information about active capturing
1000
@item info snapshots
1001
show list of VM snapshots
1002
@item info mice
1003
show which guest mouse is receiving events
1004
@end table
1005

    
1006
@item q or quit
1007
Quit the emulator.
1008

    
1009
@item eject [-f] @var{device}
1010
Eject a removable medium (use -f to force it).
1011

    
1012
@item change @var{device} @var{setting}
1013

    
1014
Change the configuration of a device.
1015

    
1016
@table @option
1017
@item change @var{diskdevice} @var{filename}
1018
Change the medium for a removable disk device to point to @var{filename}. eg
1019

    
1020
@example
1021
(qemu) change cdrom /path/to/some.iso
1022
@end example
1023

    
1024
@item change vnc @var{display},@var{options}
1025
Change the configuration of the VNC server. The valid syntax for @var{display}
1026
and @var{options} are described at @ref{sec_invocation}. eg
1027

    
1028
@example
1029
(qemu) change vnc localhost:1
1030
@end example
1031

    
1032
@item change vnc password
1033

    
1034
Change the password associated with the VNC server. The monitor will prompt for
1035
the new password to be entered. VNC passwords are only significant upto 8 letters.
1036
eg.
1037

    
1038
@example
1039
(qemu) change vnc password
1040
Password: ********
1041
@end example
1042

    
1043
@end table
1044

    
1045
@item screendump @var{filename}
1046
Save screen into PPM image @var{filename}.
1047

    
1048
@item mouse_move @var{dx} @var{dy} [@var{dz}]
1049
Move the active mouse to the specified coordinates @var{dx} @var{dy}
1050
with optional scroll axis @var{dz}.
1051

    
1052
@item mouse_button @var{val}
1053
Change the active mouse button state @var{val} (1=L, 2=M, 4=R).
1054

    
1055
@item mouse_set @var{index}
1056
Set which mouse device receives events at given @var{index}, index
1057
can be obtained with
1058
@example
1059
info mice
1060
@end example
1061

    
1062
@item wavcapture @var{filename} [@var{frequency} [@var{bits} [@var{channels}]]]
1063
Capture audio into @var{filename}. Using sample rate @var{frequency}
1064
bits per sample @var{bits} and number of channels @var{channels}.
1065

    
1066
Defaults:
1067
@itemize @minus
1068
@item Sample rate = 44100 Hz - CD quality
1069
@item Bits = 16
1070
@item Number of channels = 2 - Stereo
1071
@end itemize
1072

    
1073
@item stopcapture @var{index}
1074
Stop capture with a given @var{index}, index can be obtained with
1075
@example
1076
info capture
1077
@end example
1078

    
1079
@item log @var{item1}[,...]
1080
Activate logging of the specified items to @file{/tmp/qemu.log}.
1081

    
1082
@item savevm [@var{tag}|@var{id}]
1083
Create a snapshot of the whole virtual machine. If @var{tag} is
1084
provided, it is used as human readable identifier. If there is already
1085
a snapshot with the same tag or ID, it is replaced. More info at
1086
@ref{vm_snapshots}.
1087

    
1088
@item loadvm @var{tag}|@var{id}
1089
Set the whole virtual machine to the snapshot identified by the tag
1090
@var{tag} or the unique snapshot ID @var{id}.
1091

    
1092
@item delvm @var{tag}|@var{id}
1093
Delete the snapshot identified by @var{tag} or @var{id}.
1094

    
1095
@item stop
1096
Stop emulation.
1097

    
1098
@item c or cont
1099
Resume emulation.
1100

    
1101
@item gdbserver [@var{port}]
1102
Start gdbserver session (default @var{port}=1234)
1103

    
1104
@item x/fmt @var{addr}
1105
Virtual memory dump starting at @var{addr}.
1106

    
1107
@item xp /@var{fmt} @var{addr}
1108
Physical memory dump starting at @var{addr}.
1109

    
1110
@var{fmt} is a format which tells the command how to format the
1111
data. Its syntax is: @option{/@{count@}@{format@}@{size@}}
1112

    
1113
@table @var
1114
@item count
1115
is the number of items to be dumped.
1116

    
1117
@item format
1118
can be x (hex), d (signed decimal), u (unsigned decimal), o (octal),
1119
c (char) or i (asm instruction).
1120

    
1121
@item size
1122
can be b (8 bits), h (16 bits), w (32 bits) or g (64 bits). On x86,
1123
@code{h} or @code{w} can be specified with the @code{i} format to
1124
respectively select 16 or 32 bit code instruction size.
1125

    
1126
@end table
1127

    
1128
Examples:
1129
@itemize
1130
@item
1131
Dump 10 instructions at the current instruction pointer:
1132
@example
1133
(qemu) x/10i $eip
1134
0x90107063:  ret
1135
0x90107064:  sti
1136
0x90107065:  lea    0x0(%esi,1),%esi
1137
0x90107069:  lea    0x0(%edi,1),%edi
1138
0x90107070:  ret
1139
0x90107071:  jmp    0x90107080
1140
0x90107073:  nop
1141
0x90107074:  nop
1142
0x90107075:  nop
1143
0x90107076:  nop
1144
@end example
1145

    
1146
@item
1147
Dump 80 16 bit values at the start of the video memory.
1148
@smallexample
1149
(qemu) xp/80hx 0xb8000
1150
0x000b8000: 0x0b50 0x0b6c 0x0b65 0x0b78 0x0b38 0x0b36 0x0b2f 0x0b42
1151
0x000b8010: 0x0b6f 0x0b63 0x0b68 0x0b73 0x0b20 0x0b56 0x0b47 0x0b41
1152
0x000b8020: 0x0b42 0x0b69 0x0b6f 0x0b73 0x0b20 0x0b63 0x0b75 0x0b72
1153
0x000b8030: 0x0b72 0x0b65 0x0b6e 0x0b74 0x0b2d 0x0b63 0x0b76 0x0b73
1154
0x000b8040: 0x0b20 0x0b30 0x0b35 0x0b20 0x0b4e 0x0b6f 0x0b76 0x0b20
1155
0x000b8050: 0x0b32 0x0b30 0x0b30 0x0b33 0x0720 0x0720 0x0720 0x0720
1156
0x000b8060: 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720
1157
0x000b8070: 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720
1158
0x000b8080: 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720
1159
0x000b8090: 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720
1160
@end smallexample
1161
@end itemize
1162

    
1163
@item p or print/@var{fmt} @var{expr}
1164

    
1165
Print expression value. Only the @var{format} part of @var{fmt} is
1166
used.
1167

    
1168
@item sendkey @var{keys}
1169

    
1170
Send @var{keys} to the emulator. Use @code{-} to press several keys
1171
simultaneously. Example:
1172
@example
1173
sendkey ctrl-alt-f1
1174
@end example
1175

    
1176
This command is useful to send keys that your graphical user interface
1177
intercepts at low level, such as @code{ctrl-alt-f1} in X Window.
1178

    
1179
@item system_reset
1180

    
1181
Reset the system.
1182

    
1183
@item usb_add @var{devname}
1184

    
1185
Add the USB device @var{devname}.  For details of available devices see
1186
@ref{usb_devices}
1187

    
1188
@item usb_del @var{devname}
1189

    
1190
Remove the USB device @var{devname} from the QEMU virtual USB
1191
hub. @var{devname} has the syntax @code{bus.addr}. Use the monitor
1192
command @code{info usb} to see the devices you can remove.
1193

    
1194
@end table
1195

    
1196
@subsection Integer expressions
1197

    
1198
The monitor understands integers expressions for every integer
1199
argument. You can use register names to get the value of specifics
1200
CPU registers by prefixing them with @emph{$}.
1201

    
1202
@node disk_images
1203
@section Disk Images
1204

    
1205
Since version 0.6.1, QEMU supports many disk image formats, including
1206
growable disk images (their size increase as non empty sectors are
1207
written), compressed and encrypted disk images. Version 0.8.3 added
1208
the new qcow2 disk image format which is essential to support VM
1209
snapshots.
1210

    
1211
@menu
1212
* disk_images_quickstart::    Quick start for disk image creation
1213
* disk_images_snapshot_mode:: Snapshot mode
1214
* vm_snapshots::              VM snapshots
1215
* qemu_img_invocation::       qemu-img Invocation
1216
* host_drives::               Using host drives
1217
* disk_images_fat_images::    Virtual FAT disk images
1218
@end menu
1219

    
1220
@node disk_images_quickstart
1221
@subsection Quick start for disk image creation
1222

    
1223
You can create a disk image with the command:
1224
@example
1225
qemu-img create myimage.img mysize
1226
@end example
1227
where @var{myimage.img} is the disk image filename and @var{mysize} is its
1228
size in kilobytes. You can add an @code{M} suffix to give the size in
1229
megabytes and a @code{G} suffix for gigabytes.
1230

    
1231
See @ref{qemu_img_invocation} for more information.
1232

    
1233
@node disk_images_snapshot_mode
1234
@subsection Snapshot mode
1235

    
1236
If you use the option @option{-snapshot}, all disk images are
1237
considered as read only. When sectors in written, they are written in
1238
a temporary file created in @file{/tmp}. You can however force the
1239
write back to the raw disk images by using the @code{commit} monitor
1240
command (or @key{C-a s} in the serial console).
1241

    
1242
@node vm_snapshots
1243
@subsection VM snapshots
1244

    
1245
VM snapshots are snapshots of the complete virtual machine including
1246
CPU state, RAM, device state and the content of all the writable
1247
disks. In order to use VM snapshots, you must have at least one non
1248
removable and writable block device using the @code{qcow2} disk image
1249
format. Normally this device is the first virtual hard drive.
1250

    
1251
Use the monitor command @code{savevm} to create a new VM snapshot or
1252
replace an existing one. A human readable name can be assigned to each
1253
snapshot in addition to its numerical ID.
1254

    
1255
Use @code{loadvm} to restore a VM snapshot and @code{delvm} to remove
1256
a VM snapshot. @code{info snapshots} lists the available snapshots
1257
with their associated information:
1258

    
1259
@example
1260
(qemu) info snapshots
1261
Snapshot devices: hda
1262
Snapshot list (from hda):
1263
ID        TAG                 VM SIZE                DATE       VM CLOCK
1264
1         start                   41M 2006-08-06 12:38:02   00:00:14.954
1265
2                                 40M 2006-08-06 12:43:29   00:00:18.633
1266
3         msys                    40M 2006-08-06 12:44:04   00:00:23.514
1267
@end example
1268

    
1269
A VM snapshot is made of a VM state info (its size is shown in
1270
@code{info snapshots}) and a snapshot of every writable disk image.
1271
The VM state info is stored in the first @code{qcow2} non removable
1272
and writable block device. The disk image snapshots are stored in
1273
every disk image. The size of a snapshot in a disk image is difficult
1274
to evaluate and is not shown by @code{info snapshots} because the
1275
associated disk sectors are shared among all the snapshots to save
1276
disk space (otherwise each snapshot would need a full copy of all the
1277
disk images).
1278

    
1279
When using the (unrelated) @code{-snapshot} option
1280
(@ref{disk_images_snapshot_mode}), you can always make VM snapshots,
1281
but they are deleted as soon as you exit QEMU.
1282

    
1283
VM snapshots currently have the following known limitations:
1284
@itemize
1285
@item
1286
They cannot cope with removable devices if they are removed or
1287
inserted after a snapshot is done.
1288
@item
1289
A few device drivers still have incomplete snapshot support so their
1290
state is not saved or restored properly (in particular USB).
1291
@end itemize
1292

    
1293
@node qemu_img_invocation
1294
@subsection @code{qemu-img} Invocation
1295

    
1296
@include qemu-img.texi
1297

    
1298
@node host_drives
1299
@subsection Using host drives
1300

    
1301
In addition to disk image files, QEMU can directly access host
1302
devices. We describe here the usage for QEMU version >= 0.8.3.
1303

    
1304
@subsubsection Linux
1305

    
1306
On Linux, you can directly use the host device filename instead of a
1307
disk image filename provided you have enough privileges to access
1308
it. For example, use @file{/dev/cdrom} to access to the CDROM or
1309
@file{/dev/fd0} for the floppy.
1310

    
1311
@table @code
1312
@item CD
1313
You can specify a CDROM device even if no CDROM is loaded. QEMU has
1314
specific code to detect CDROM insertion or removal. CDROM ejection by
1315
the guest OS is supported. Currently only data CDs are supported.
1316
@item Floppy
1317
You can specify a floppy device even if no floppy is loaded. Floppy
1318
removal is currently not detected accurately (if you change floppy
1319
without doing floppy access while the floppy is not loaded, the guest
1320
OS will think that the same floppy is loaded).
1321
@item Hard disks
1322
Hard disks can be used. Normally you must specify the whole disk
1323
(@file{/dev/hdb} instead of @file{/dev/hdb1}) so that the guest OS can
1324
see it as a partitioned disk. WARNING: unless you know what you do, it
1325
is better to only make READ-ONLY accesses to the hard disk otherwise
1326
you may corrupt your host data (use the @option{-snapshot} command
1327
line option or modify the device permissions accordingly).
1328
@end table
1329

    
1330
@subsubsection Windows
1331

    
1332
@table @code
1333
@item CD
1334
The preferred syntax is the drive letter (e.g. @file{d:}). The
1335
alternate syntax @file{\\.\d:} is supported. @file{/dev/cdrom} is
1336
supported as an alias to the first CDROM drive.
1337

    
1338
Currently there is no specific code to handle removable media, so it
1339
is better to use the @code{change} or @code{eject} monitor commands to
1340
change or eject media.
1341
@item Hard disks
1342
Hard disks can be used with the syntax: @file{\\.\PhysicalDrive@var{N}}
1343
where @var{N} is the drive number (0 is the first hard disk).
1344

    
1345
WARNING: unless you know what you do, it is better to only make
1346
READ-ONLY accesses to the hard disk otherwise you may corrupt your
1347
host data (use the @option{-snapshot} command line so that the
1348
modifications are written in a temporary file).
1349
@end table
1350

    
1351

    
1352
@subsubsection Mac OS X
1353

    
1354
@file{/dev/cdrom} is an alias to the first CDROM.
1355

    
1356
Currently there is no specific code to handle removable media, so it
1357
is better to use the @code{change} or @code{eject} monitor commands to
1358
change or eject media.
1359

    
1360
@node disk_images_fat_images
1361
@subsection Virtual FAT disk images
1362

    
1363
QEMU can automatically create a virtual FAT disk image from a
1364
directory tree. In order to use it, just type:
1365

    
1366
@example
1367
qemu linux.img -hdb fat:/my_directory
1368
@end example
1369

    
1370
Then you access access to all the files in the @file{/my_directory}
1371
directory without having to copy them in a disk image or to export
1372
them via SAMBA or NFS. The default access is @emph{read-only}.
1373

    
1374
Floppies can be emulated with the @code{:floppy:} option:
1375

    
1376
@example
1377
qemu linux.img -fda fat:floppy:/my_directory
1378
@end example
1379

    
1380
A read/write support is available for testing (beta stage) with the
1381
@code{:rw:} option:
1382

    
1383
@example
1384
qemu linux.img -fda fat:floppy:rw:/my_directory
1385
@end example
1386

    
1387
What you should @emph{never} do:
1388
@itemize
1389
@item use non-ASCII filenames ;
1390
@item use "-snapshot" together with ":rw:" ;
1391
@item expect it to work when loadvm'ing ;
1392
@item write to the FAT directory on the host system while accessing it with the guest system.
1393
@end itemize
1394

    
1395
@node pcsys_network
1396
@section Network emulation
1397

    
1398
QEMU can simulate several network cards (PCI or ISA cards on the PC
1399
target) and can connect them to an arbitrary number of Virtual Local
1400
Area Networks (VLANs). Host TAP devices can be connected to any QEMU
1401
VLAN. VLAN can be connected between separate instances of QEMU to
1402
simulate large networks. For simpler usage, a non privileged user mode
1403
network stack can replace the TAP device to have a basic network
1404
connection.
1405

    
1406
@subsection VLANs
1407

    
1408
QEMU simulates several VLANs. A VLAN can be symbolised as a virtual
1409
connection between several network devices. These devices can be for
1410
example QEMU virtual Ethernet cards or virtual Host ethernet devices
1411
(TAP devices).
1412

    
1413
@subsection Using TAP network interfaces
1414

    
1415
This is the standard way to connect QEMU to a real network. QEMU adds
1416
a virtual network device on your host (called @code{tapN}), and you
1417
can then configure it as if it was a real ethernet card.
1418

    
1419
@subsubsection Linux host
1420

    
1421
As an example, you can download the @file{linux-test-xxx.tar.gz}
1422
archive and copy the script @file{qemu-ifup} in @file{/etc} and
1423
configure properly @code{sudo} so that the command @code{ifconfig}
1424
contained in @file{qemu-ifup} can be executed as root. You must verify
1425
that your host kernel supports the TAP network interfaces: the
1426
device @file{/dev/net/tun} must be present.
1427

    
1428
See @ref{sec_invocation} to have examples of command lines using the
1429
TAP network interfaces.
1430

    
1431
@subsubsection Windows host
1432

    
1433
There is a virtual ethernet driver for Windows 2000/XP systems, called
1434
TAP-Win32. But it is not included in standard QEMU for Windows,
1435
so you will need to get it separately. It is part of OpenVPN package,
1436
so download OpenVPN from : @url{http://openvpn.net/}.
1437

    
1438
@subsection Using the user mode network stack
1439

    
1440
By using the option @option{-net user} (default configuration if no
1441
@option{-net} option is specified), QEMU uses a completely user mode
1442
network stack (you don't need root privilege to use the virtual
1443
network). The virtual network configuration is the following:
1444

    
1445
@example
1446

    
1447
         QEMU VLAN      <------>  Firewall/DHCP server <-----> Internet
1448
                           |          (10.0.2.2)
1449
                           |
1450
                           ---->  DNS server (10.0.2.3)
1451
                           |
1452
                           ---->  SMB server (10.0.2.4)
1453
@end example
1454

    
1455
The QEMU VM behaves as if it was behind a firewall which blocks all
1456
incoming connections. You can use a DHCP client to automatically
1457
configure the network in the QEMU VM. The DHCP server assign addresses
1458
to the hosts starting from 10.0.2.15.
1459

    
1460
In order to check that the user mode network is working, you can ping
1461
the address 10.0.2.2 and verify that you got an address in the range
1462
10.0.2.x from the QEMU virtual DHCP server.
1463

    
1464
Note that @code{ping} is not supported reliably to the internet as it
1465
would require root privileges. It means you can only ping the local
1466
router (10.0.2.2).
1467

    
1468
When using the built-in TFTP server, the router is also the TFTP
1469
server.
1470

    
1471
When using the @option{-redir} option, TCP or UDP connections can be
1472
redirected from the host to the guest. It allows for example to
1473
redirect X11, telnet or SSH connections.
1474

    
1475
@subsection Connecting VLANs between QEMU instances
1476

    
1477
Using the @option{-net socket} option, it is possible to make VLANs
1478
that span several QEMU instances. See @ref{sec_invocation} to have a
1479
basic example.
1480

    
1481
@node direct_linux_boot
1482
@section Direct Linux Boot
1483

    
1484
This section explains how to launch a Linux kernel inside QEMU without
1485
having to make a full bootable image. It is very useful for fast Linux
1486
kernel testing.
1487

    
1488
The syntax is:
1489
@example
1490
qemu -kernel arch/i386/boot/bzImage -hda root-2.4.20.img -append "root=/dev/hda"
1491
@end example
1492

    
1493
Use @option{-kernel} to provide the Linux kernel image and
1494
@option{-append} to give the kernel command line arguments. The
1495
@option{-initrd} option can be used to provide an INITRD image.
1496

    
1497
When using the direct Linux boot, a disk image for the first hard disk
1498
@file{hda} is required because its boot sector is used to launch the
1499
Linux kernel.
1500

    
1501
If you do not need graphical output, you can disable it and redirect
1502
the virtual serial port and the QEMU monitor to the console with the
1503
@option{-nographic} option. The typical command line is:
1504
@example
1505
qemu -kernel arch/i386/boot/bzImage -hda root-2.4.20.img \
1506
     -append "root=/dev/hda console=ttyS0" -nographic
1507
@end example
1508

    
1509
Use @key{Ctrl-a c} to switch between the serial console and the
1510
monitor (@pxref{pcsys_keys}).
1511

    
1512
@node pcsys_usb
1513
@section USB emulation
1514

    
1515
QEMU emulates a PCI UHCI USB controller. You can virtually plug
1516
virtual USB devices or real host USB devices (experimental, works only
1517
on Linux hosts).  Qemu will automatically create and connect virtual USB hubs
1518
as necessary to connect multiple USB devices.
1519

    
1520
@menu
1521
* usb_devices::
1522
* host_usb_devices::
1523
@end menu
1524
@node usb_devices
1525
@subsection Connecting USB devices
1526

    
1527
USB devices can be connected with the @option{-usbdevice} commandline option
1528
or the @code{usb_add} monitor command.  Available devices are:
1529

    
1530
@table @var
1531
@item @code{mouse}
1532
Virtual Mouse.  This will override the PS/2 mouse emulation when activated.
1533
@item @code{tablet}
1534
Pointer device that uses absolute coordinates (like a touchscreen).
1535
This means qemu is able to report the mouse position without having
1536
to grab the mouse.  Also overrides the PS/2 mouse emulation when activated.
1537
@item @code{disk:@var{file}}
1538
Mass storage device based on @var{file} (@pxref{disk_images})
1539
@item @code{host:@var{bus.addr}}
1540
Pass through the host device identified by @var{bus.addr}
1541
(Linux only)
1542
@item @code{host:@var{vendor_id:product_id}}
1543
Pass through the host device identified by @var{vendor_id:product_id}
1544
(Linux only)
1545
@item @code{wacom-tablet}
1546
Virtual Wacom PenPartner tablet.  This device is similar to the @code{tablet}
1547
above but it can be used with the tslib library because in addition to touch
1548
coordinates it reports touch pressure.
1549
@item @code{keyboard}
1550
Standard USB keyboard.  Will override the PS/2 keyboard (if present).
1551
@end table
1552

    
1553
@node host_usb_devices
1554
@subsection Using host USB devices on a Linux host
1555

    
1556
WARNING: this is an experimental feature. QEMU will slow down when
1557
using it. USB devices requiring real time streaming (i.e. USB Video
1558
Cameras) are not supported yet.
1559

    
1560
@enumerate
1561
@item If you use an early Linux 2.4 kernel, verify that no Linux driver
1562
is actually using the USB device. A simple way to do that is simply to
1563
disable the corresponding kernel module by renaming it from @file{mydriver.o}
1564
to @file{mydriver.o.disabled}.
1565

    
1566
@item Verify that @file{/proc/bus/usb} is working (most Linux distributions should enable it by default). You should see something like that:
1567
@example
1568
ls /proc/bus/usb
1569
001  devices  drivers
1570
@end example
1571

    
1572
@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:
1573
@example
1574
chown -R myuid /proc/bus/usb
1575
@end example
1576

    
1577
@item Launch QEMU and do in the monitor:
1578
@example
1579
info usbhost
1580
  Device 1.2, speed 480 Mb/s
1581
    Class 00: USB device 1234:5678, USB DISK
1582
@end example
1583
You should see the list of the devices you can use (Never try to use
1584
hubs, it won't work).
1585

    
1586
@item Add the device in QEMU by using:
1587
@example
1588
usb_add host:1234:5678
1589
@end example
1590

    
1591
Normally the guest OS should report that a new USB device is
1592
plugged. You can use the option @option{-usbdevice} to do the same.
1593

    
1594
@item Now you can try to use the host USB device in QEMU.
1595

    
1596
@end enumerate
1597

    
1598
When relaunching QEMU, you may have to unplug and plug again the USB
1599
device to make it work again (this is a bug).
1600

    
1601
@node vnc_security
1602
@section VNC security
1603

    
1604
The VNC server capability provides access to the graphical console
1605
of the guest VM across the network. This has a number of security
1606
considerations depending on the deployment scenarios.
1607

    
1608
@menu
1609
* vnc_sec_none::
1610
* vnc_sec_password::
1611
* vnc_sec_certificate::
1612
* vnc_sec_certificate_verify::
1613
* vnc_sec_certificate_pw::
1614
* vnc_generate_cert::
1615
@end menu
1616
@node vnc_sec_none
1617
@subsection Without passwords
1618

    
1619
The simplest VNC server setup does not include any form of authentication.
1620
For this setup it is recommended to restrict it to listen on a UNIX domain
1621
socket only. For example
1622

    
1623
@example
1624
qemu [...OPTIONS...] -vnc unix:/home/joebloggs/.qemu-myvm-vnc
1625
@end example
1626

    
1627
This ensures that only users on local box with read/write access to that
1628
path can access the VNC server. To securely access the VNC server from a
1629
remote machine, a combination of netcat+ssh can be used to provide a secure
1630
tunnel.
1631

    
1632
@node vnc_sec_password
1633
@subsection With passwords
1634

    
1635
The VNC protocol has limited support for password based authentication. Since
1636
the protocol limits passwords to 8 characters it should not be considered
1637
to provide high security. The password can be fairly easily brute-forced by
1638
a client making repeat connections. For this reason, a VNC server using password
1639
authentication should be restricted to only listen on the loopback interface
1640
or UNIX domain sockets. Password ayuthentication is requested with the @code{password}
1641
option, and then once QEMU is running the password is set with the monitor. Until
1642
the monitor is used to set the password all clients will be rejected.
1643

    
1644
@example
1645
qemu [...OPTIONS...] -vnc :1,password -monitor stdio
1646
(qemu) change vnc password
1647
Password: ********
1648
(qemu)
1649
@end example
1650

    
1651
@node vnc_sec_certificate
1652
@subsection With x509 certificates
1653

    
1654
The QEMU VNC server also implements the VeNCrypt extension allowing use of
1655
TLS for encryption of the session, and x509 certificates for authentication.
1656
The use of x509 certificates is strongly recommended, because TLS on its
1657
own is susceptible to man-in-the-middle attacks. Basic x509 certificate
1658
support provides a secure session, but no authentication. This allows any
1659
client to connect, and provides an encrypted session.
1660

    
1661
@example
1662
qemu [...OPTIONS...] -vnc :1,tls,x509=/etc/pki/qemu -monitor stdio
1663
@end example
1664

    
1665
In the above example @code{/etc/pki/qemu} should contain at least three files,
1666
@code{ca-cert.pem}, @code{server-cert.pem} and @code{server-key.pem}. Unprivileged
1667
users will want to use a private directory, for example @code{$HOME/.pki/qemu}.
1668
NB the @code{server-key.pem} file should be protected with file mode 0600 to
1669
only be readable by the user owning it.
1670

    
1671
@node vnc_sec_certificate_verify
1672
@subsection With x509 certificates and client verification
1673

    
1674
Certificates can also provide a means to authenticate the client connecting.
1675
The server will request that the client provide a certificate, which it will
1676
then validate against the CA certificate. This is a good choice if deploying
1677
in an environment with a private internal certificate authority.
1678

    
1679
@example
1680
qemu [...OPTIONS...] -vnc :1,tls,x509verify=/etc/pki/qemu -monitor stdio
1681
@end example
1682

    
1683

    
1684
@node vnc_sec_certificate_pw
1685
@subsection With x509 certificates, client verification and passwords
1686

    
1687
Finally, the previous method can be combined with VNC password authentication
1688
to provide two layers of authentication for clients.
1689

    
1690
@example
1691
qemu [...OPTIONS...] -vnc :1,password,tls,x509verify=/etc/pki/qemu -monitor stdio
1692
(qemu) change vnc password
1693
Password: ********
1694
(qemu)
1695
@end example
1696

    
1697
@node vnc_generate_cert
1698
@subsection Generating certificates for VNC
1699

    
1700
The GNU TLS packages provides a command called @code{certtool} which can
1701
be used to generate certificates and keys in PEM format. At a minimum it
1702
is neccessary to setup a certificate authority, and issue certificates to
1703
each server. If using certificates for authentication, then each client
1704
will also need to be issued a certificate. The recommendation is for the
1705
server to keep its certificates in either @code{/etc/pki/qemu} or for
1706
unprivileged users in @code{$HOME/.pki/qemu}.
1707

    
1708
@menu
1709
* vnc_generate_ca::
1710
* vnc_generate_server::
1711
* vnc_generate_client::
1712
@end menu
1713
@node vnc_generate_ca
1714
@subsubsection Setup the Certificate Authority
1715

    
1716
This step only needs to be performed once per organization / organizational
1717
unit. First the CA needs a private key. This key must be kept VERY secret
1718
and secure. If this key is compromised the entire trust chain of the certificates
1719
issued with it is lost.
1720

    
1721
@example
1722
# certtool --generate-privkey > ca-key.pem
1723
@end example
1724

    
1725
A CA needs to have a public certificate. For simplicity it can be a self-signed
1726
certificate, or one issue by a commercial certificate issuing authority. To
1727
generate a self-signed certificate requires one core piece of information, the
1728
name of the organization.
1729

    
1730
@example
1731
# cat > ca.info <<EOF
1732
cn = Name of your organization
1733
ca
1734
cert_signing_key
1735
EOF
1736
# certtool --generate-self-signed \
1737
           --load-privkey ca-key.pem
1738
           --template ca.info \
1739
           --outfile ca-cert.pem
1740
@end example
1741

    
1742
The @code{ca-cert.pem} file should be copied to all servers and clients wishing to utilize
1743
TLS support in the VNC server. The @code{ca-key.pem} must not be disclosed/copied at all.
1744

    
1745
@node vnc_generate_server
1746
@subsubsection Issuing server certificates
1747

    
1748
Each server (or host) needs to be issued with a key and certificate. When connecting
1749
the certificate is sent to the client which validates it against the CA certificate.
1750
The core piece of information for a server certificate is the hostname. This should
1751
be the fully qualified hostname that the client will connect with, since the client
1752
will typically also verify the hostname in the certificate. On the host holding the
1753
secure CA private key:
1754

    
1755
@example
1756
# cat > server.info <<EOF
1757
organization = Name  of your organization
1758
cn = server.foo.example.com
1759
tls_www_server
1760
encryption_key
1761
signing_key
1762
EOF
1763
# certtool --generate-privkey > server-key.pem
1764
# certtool --generate-certificate \
1765
           --load-ca-certificate ca-cert.pem \
1766
           --load-ca-privkey ca-key.pem \
1767
           --load-privkey server server-key.pem \
1768
           --template server.info \
1769
           --outfile server-cert.pem
1770
@end example
1771

    
1772
The @code{server-key.pem} and @code{server-cert.pem} files should now be securely copied
1773
to the server for which they were generated. The @code{server-key.pem} is security
1774
sensitive and should be kept protected with file mode 0600 to prevent disclosure.
1775

    
1776
@node vnc_generate_client
1777
@subsubsection Issuing client certificates
1778

    
1779
If the QEMU VNC server is to use the @code{x509verify} option to validate client
1780
certificates as its authentication mechanism, each client also needs to be issued
1781
a certificate. The client certificate contains enough metadata to uniquely identify
1782
the client, typically organization, state, city, building, etc. On the host holding
1783
the secure CA private key:
1784

    
1785
@example
1786
# cat > client.info <<EOF
1787
country = GB
1788
state = London
1789
locality = London
1790
organiazation = Name of your organization
1791
cn = client.foo.example.com
1792
tls_www_client
1793
encryption_key
1794
signing_key
1795
EOF
1796
# certtool --generate-privkey > client-key.pem
1797
# certtool --generate-certificate \
1798
           --load-ca-certificate ca-cert.pem \
1799
           --load-ca-privkey ca-key.pem \
1800
           --load-privkey client-key.pem \
1801
           --template client.info \
1802
           --outfile client-cert.pem
1803
@end example
1804

    
1805
The @code{client-key.pem} and @code{client-cert.pem} files should now be securely
1806
copied to the client for which they were generated.
1807

    
1808
@node gdb_usage
1809
@section GDB usage
1810

    
1811
QEMU has a primitive support to work with gdb, so that you can do
1812
'Ctrl-C' while the virtual machine is running and inspect its state.
1813

    
1814
In order to use gdb, launch qemu with the '-s' option. It will wait for a
1815
gdb connection:
1816
@example
1817
> qemu -s -kernel arch/i386/boot/bzImage -hda root-2.4.20.img \
1818
       -append "root=/dev/hda"
1819
Connected to host network interface: tun0
1820
Waiting gdb connection on port 1234
1821
@end example
1822

    
1823
Then launch gdb on the 'vmlinux' executable:
1824
@example
1825
> gdb vmlinux
1826
@end example
1827

    
1828
In gdb, connect to QEMU:
1829
@example
1830
(gdb) target remote localhost:1234
1831
@end example
1832

    
1833
Then you can use gdb normally. For example, type 'c' to launch the kernel:
1834
@example
1835
(gdb) c
1836
@end example
1837

    
1838
Here are some useful tips in order to use gdb on system code:
1839

    
1840
@enumerate
1841
@item
1842
Use @code{info reg} to display all the CPU registers.
1843
@item
1844
Use @code{x/10i $eip} to display the code at the PC position.
1845
@item
1846
Use @code{set architecture i8086} to dump 16 bit code. Then use
1847
@code{x/10i $cs*16+$eip} to dump the code at the PC position.
1848
@end enumerate
1849

    
1850
@node pcsys_os_specific
1851
@section Target OS specific information
1852

    
1853
@subsection Linux
1854

    
1855
To have access to SVGA graphic modes under X11, use the @code{vesa} or
1856
the @code{cirrus} X11 driver. For optimal performances, use 16 bit
1857
color depth in the guest and the host OS.
1858

    
1859
When using a 2.6 guest Linux kernel, you should add the option
1860
@code{clock=pit} on the kernel command line because the 2.6 Linux
1861
kernels make very strict real time clock checks by default that QEMU
1862
cannot simulate exactly.
1863

    
1864
When using a 2.6 guest Linux kernel, verify that the 4G/4G patch is
1865
not activated because QEMU is slower with this patch. The QEMU
1866
Accelerator Module is also much slower in this case. Earlier Fedora
1867
Core 3 Linux kernel (< 2.6.9-1.724_FC3) were known to incorporate this
1868
patch by default. Newer kernels don't have it.
1869

    
1870
@subsection Windows
1871

    
1872
If you have a slow host, using Windows 95 is better as it gives the
1873
best speed. Windows 2000 is also a good choice.
1874

    
1875
@subsubsection SVGA graphic modes support
1876

    
1877
QEMU emulates a Cirrus Logic GD5446 Video
1878
card. All Windows versions starting from Windows 95 should recognize
1879
and use this graphic card. For optimal performances, use 16 bit color
1880
depth in the guest and the host OS.
1881

    
1882
If you are using Windows XP as guest OS and if you want to use high
1883
resolution modes which the Cirrus Logic BIOS does not support (i.e. >=
1884
1280x1024x16), then you should use the VESA VBE virtual graphic card
1885
(option @option{-std-vga}).
1886

    
1887
@subsubsection CPU usage reduction
1888

    
1889
Windows 9x does not correctly use the CPU HLT
1890
instruction. The result is that it takes host CPU cycles even when
1891
idle. You can install the utility from
1892
@url{http://www.user.cityline.ru/~maxamn/amnhltm.zip} to solve this
1893
problem. Note that no such tool is needed for NT, 2000 or XP.
1894

    
1895
@subsubsection Windows 2000 disk full problem
1896

    
1897
Windows 2000 has a bug which gives a disk full problem during its
1898
installation. When installing it, use the @option{-win2k-hack} QEMU
1899
option to enable a specific workaround. After Windows 2000 is
1900
installed, you no longer need this option (this option slows down the
1901
IDE transfers).
1902

    
1903
@subsubsection Windows 2000 shutdown
1904

    
1905
Windows 2000 cannot automatically shutdown in QEMU although Windows 98
1906
can. It comes from the fact that Windows 2000 does not automatically
1907
use the APM driver provided by the BIOS.
1908

    
1909
In order to correct that, do the following (thanks to Struan
1910
Bartlett): go to the Control Panel => Add/Remove Hardware & Next =>
1911
Add/Troubleshoot a device => Add a new device & Next => No, select the
1912
hardware from a list & Next => NT Apm/Legacy Support & Next => Next
1913
(again) a few times. Now the driver is installed and Windows 2000 now
1914
correctly instructs QEMU to shutdown at the appropriate moment.
1915

    
1916
@subsubsection Share a directory between Unix and Windows
1917

    
1918
See @ref{sec_invocation} about the help of the option @option{-smb}.
1919

    
1920
@subsubsection Windows XP security problem
1921

    
1922
Some releases of Windows XP install correctly but give a security
1923
error when booting:
1924
@example
1925
A problem is preventing Windows from accurately checking the
1926
license for this computer. Error code: 0x800703e6.
1927
@end example
1928

    
1929
The workaround is to install a service pack for XP after a boot in safe
1930
mode. Then reboot, and the problem should go away. Since there is no
1931
network while in safe mode, its recommended to download the full
1932
installation of SP1 or SP2 and transfer that via an ISO or using the
1933
vvfat block device ("-hdb fat:directory_which_holds_the_SP").
1934

    
1935
@subsection MS-DOS and FreeDOS
1936

    
1937
@subsubsection CPU usage reduction
1938

    
1939
DOS does not correctly use the CPU HLT instruction. The result is that
1940
it takes host CPU cycles even when idle. You can install the utility
1941
from @url{http://www.vmware.com/software/dosidle210.zip} to solve this
1942
problem.
1943

    
1944
@node QEMU System emulator for non PC targets
1945
@chapter QEMU System emulator for non PC targets
1946

    
1947
QEMU is a generic emulator and it emulates many non PC
1948
machines. Most of the options are similar to the PC emulator. The
1949
differences are mentioned in the following sections.
1950

    
1951
@menu
1952
* QEMU PowerPC System emulator::
1953
* Sparc32 System emulator::
1954
* Sparc64 System emulator::
1955
* MIPS System emulator::
1956
* ARM System emulator::
1957
* ColdFire System emulator::
1958
@end menu
1959

    
1960
@node QEMU PowerPC System emulator
1961
@section QEMU PowerPC System emulator
1962

    
1963
Use the executable @file{qemu-system-ppc} to simulate a complete PREP
1964
or PowerMac PowerPC system.
1965

    
1966
QEMU emulates the following PowerMac peripherals:
1967

    
1968
@itemize @minus
1969
@item
1970
UniNorth PCI Bridge
1971
@item
1972
PCI VGA compatible card with VESA Bochs Extensions
1973
@item
1974
2 PMAC IDE interfaces with hard disk and CD-ROM support
1975
@item
1976
NE2000 PCI adapters
1977
@item
1978
Non Volatile RAM
1979
@item
1980
VIA-CUDA with ADB keyboard and mouse.
1981
@end itemize
1982

    
1983
QEMU emulates the following PREP peripherals:
1984

    
1985
@itemize @minus
1986
@item
1987
PCI Bridge
1988
@item
1989
PCI VGA compatible card with VESA Bochs Extensions
1990
@item
1991
2 IDE interfaces with hard disk and CD-ROM support
1992
@item
1993
Floppy disk
1994
@item
1995
NE2000 network adapters
1996
@item
1997
Serial port
1998
@item
1999
PREP Non Volatile RAM
2000
@item
2001
PC compatible keyboard and mouse.
2002
@end itemize
2003

    
2004
QEMU uses the Open Hack'Ware Open Firmware Compatible BIOS available at
2005
@url{http://perso.magic.fr/l_indien/OpenHackWare/index.htm}.
2006

    
2007
@c man begin OPTIONS
2008

    
2009
The following options are specific to the PowerPC emulation:
2010

    
2011
@table @option
2012

    
2013
@item -g WxH[xDEPTH]
2014

    
2015
Set the initial VGA graphic mode. The default is 800x600x15.
2016

    
2017
@end table
2018

    
2019
@c man end
2020

    
2021

    
2022
More information is available at
2023
@url{http://perso.magic.fr/l_indien/qemu-ppc/}.
2024

    
2025
@node Sparc32 System emulator
2026
@section Sparc32 System emulator
2027

    
2028
Use the executable @file{qemu-system-sparc} to simulate a SPARCstation
2029
5, SPARCstation 10, SPARCstation 20, SPARCserver 600MP (sun4m architecture),
2030
SPARCserver 1000, or SPARCcenter 2000 (sun4d architecture). The
2031
emulation is somewhat complete.  SMP up to 16 CPUs is supported, but
2032
Linux limits the number of usable CPUs to 4.
2033

    
2034
QEMU emulates the following sun4m/sun4d peripherals:
2035

    
2036
@itemize @minus
2037
@item
2038
IOMMU or IO-UNITs
2039
@item
2040
TCX Frame buffer
2041
@item
2042
Lance (Am7990) Ethernet
2043
@item
2044
Non Volatile RAM M48T08
2045
@item
2046
Slave I/O: timers, interrupt controllers, Zilog serial ports, keyboard
2047
and power/reset logic
2048
@item
2049
ESP SCSI controller with hard disk and CD-ROM support
2050
@item
2051
Floppy drive (not on SS-600MP)
2052
@item
2053
CS4231 sound device (only on SS-5, not working yet)
2054
@end itemize
2055

    
2056
The number of peripherals is fixed in the architecture.  Maximum
2057
memory size depends on the machine type, for SS-5 it is 256MB and for
2058
others 2047MB.
2059

    
2060
Since version 0.8.2, QEMU uses OpenBIOS
2061
@url{http://www.openbios.org/}. OpenBIOS is a free (GPL v2) portable
2062
firmware implementation. The goal is to implement a 100% IEEE
2063
1275-1994 (referred to as Open Firmware) compliant firmware.
2064

    
2065
A sample Linux 2.6 series kernel and ram disk image are available on
2066
the QEMU web site. Please note that currently NetBSD, OpenBSD or
2067
Solaris kernels don't work.
2068

    
2069
@c man begin OPTIONS
2070

    
2071
The following options are specific to the Sparc32 emulation:
2072

    
2073
@table @option
2074

    
2075
@item -g WxHx[xDEPTH]
2076

    
2077
Set the initial TCX graphic mode. The default is 1024x768x8, currently
2078
the only other possible mode is 1024x768x24.
2079

    
2080
@item -prom-env string
2081

    
2082
Set OpenBIOS variables in NVRAM, for example:
2083

    
2084
@example
2085
qemu-system-sparc -prom-env 'auto-boot?=false' \
2086
 -prom-env 'boot-device=sd(0,2,0):d' -prom-env 'boot-args=linux single'
2087
@end example
2088

    
2089
@item -M [SS-5|SS-10|SS-20|SS-600MP|SS-1000|SS-2000]
2090

    
2091
Set the emulated machine type. Default is SS-5.
2092

    
2093
@end table
2094

    
2095
@c man end
2096

    
2097
@node Sparc64 System emulator
2098
@section Sparc64 System emulator
2099

    
2100
Use the executable @file{qemu-system-sparc64} to simulate a Sun4u machine.
2101
The emulator is not usable for anything yet.
2102

    
2103
QEMU emulates the following sun4u peripherals:
2104

    
2105
@itemize @minus
2106
@item
2107
UltraSparc IIi APB PCI Bridge
2108
@item
2109
PCI VGA compatible card with VESA Bochs Extensions
2110
@item
2111
Non Volatile RAM M48T59
2112
@item
2113
PC-compatible serial ports
2114
@end itemize
2115

    
2116
@node MIPS System emulator
2117
@section MIPS System emulator
2118

    
2119
Four executables cover simulation of 32 and 64-bit MIPS systems in
2120
both endian options, @file{qemu-system-mips}, @file{qemu-system-mipsel}
2121
@file{qemu-system-mips64} and @file{qemu-system-mips64el}.
2122
Four different machine types are emulated:
2123

    
2124
@itemize @minus
2125
@item
2126
A generic ISA PC-like machine "mips"
2127
@item
2128
The MIPS Malta prototype board "malta"
2129
@item
2130
An ACER Pica "pica61". This machine needs the 64-bit emulator.
2131
@item
2132
MIPS emulator pseudo board "mipssim"
2133
@end itemize
2134

    
2135
The generic emulation is supported by Debian 'Etch' and is able to
2136
install Debian into a virtual disk image. The following devices are
2137
emulated:
2138

    
2139
@itemize @minus
2140
@item
2141
A range of MIPS CPUs, default is the 24Kf
2142
@item
2143
PC style serial port
2144
@item
2145
PC style IDE disk
2146
@item
2147
NE2000 network card
2148
@end itemize
2149

    
2150
The Malta emulation supports the following devices:
2151

    
2152
@itemize @minus
2153
@item
2154
Core board with MIPS 24Kf CPU and Galileo system controller
2155
@item
2156
PIIX4 PCI/USB/SMbus controller
2157
@item
2158
The Multi-I/O chip's serial device
2159
@item
2160
PCnet32 PCI network card
2161
@item
2162
Malta FPGA serial device
2163
@item
2164
Cirrus VGA graphics card
2165
@end itemize
2166

    
2167
The ACER Pica emulation supports:
2168

    
2169
@itemize @minus
2170
@item
2171
MIPS R4000 CPU
2172
@item
2173
PC-style IRQ and DMA controllers
2174
@item
2175
PC Keyboard
2176
@item
2177
IDE controller
2178
@end itemize
2179

    
2180
The mipssim pseudo board emulation provides an environment similiar
2181
to what the proprietary MIPS emulator uses for running Linux.
2182
It supports:
2183

    
2184
@itemize @minus
2185
@item
2186
A range of MIPS CPUs, default is the 24Kf
2187
@item
2188
PC style serial port
2189
@item
2190
MIPSnet network emulation
2191
@end itemize
2192

    
2193
@node ARM System emulator
2194
@section ARM System emulator
2195

    
2196
Use the executable @file{qemu-system-arm} to simulate a ARM
2197
machine. The ARM Integrator/CP board is emulated with the following
2198
devices:
2199

    
2200
@itemize @minus
2201
@item
2202
ARM926E, ARM1026E, ARM946E, ARM1136 or Cortex-A8 CPU
2203
@item
2204
Two PL011 UARTs
2205
@item
2206
SMC 91c111 Ethernet adapter
2207
@item
2208
PL110 LCD controller
2209
@item
2210
PL050 KMI with PS/2 keyboard and mouse.
2211
@item
2212
PL181 MultiMedia Card Interface with SD card.
2213
@end itemize
2214

    
2215
The ARM Versatile baseboard is emulated with the following devices:
2216

    
2217
@itemize @minus
2218
@item
2219
ARM926E, ARM1136 or Cortex-A8 CPU
2220
@item
2221
PL190 Vectored Interrupt Controller
2222
@item
2223
Four PL011 UARTs
2224
@item
2225
SMC 91c111 Ethernet adapter
2226
@item
2227
PL110 LCD controller
2228
@item
2229
PL050 KMI with PS/2 keyboard and mouse.
2230
@item
2231
PCI host bridge.  Note the emulated PCI bridge only provides access to
2232
PCI memory space.  It does not provide access to PCI IO space.
2233
This means some devices (eg. ne2k_pci NIC) are not usable, and others
2234
(eg. rtl8139 NIC) are only usable when the guest drivers use the memory
2235
mapped control registers.
2236
@item
2237
PCI OHCI USB controller.
2238
@item
2239
LSI53C895A PCI SCSI Host Bus Adapter with hard disk and CD-ROM devices.
2240
@item
2241
PL181 MultiMedia Card Interface with SD card.
2242
@end itemize
2243

    
2244
The ARM RealView Emulation baseboard is emulated with the following devices:
2245

    
2246
@itemize @minus
2247
@item
2248
ARM926E, ARM1136, ARM11MPCORE(x4) or Cortex-A8 CPU
2249
@item
2250
ARM AMBA Generic/Distributed Interrupt Controller
2251
@item
2252
Four PL011 UARTs
2253
@item
2254
SMC 91c111 Ethernet adapter
2255
@item
2256
PL110 LCD controller
2257
@item
2258
PL050 KMI with PS/2 keyboard and mouse
2259
@item
2260
PCI host bridge
2261
@item
2262
PCI OHCI USB controller
2263
@item
2264
LSI53C895A PCI SCSI Host Bus Adapter with hard disk and CD-ROM devices
2265
@item
2266
PL181 MultiMedia Card Interface with SD card.
2267
@end itemize
2268

    
2269
The XScale-based clamshell PDA models ("Spitz", "Akita", "Borzoi"
2270
and "Terrier") emulation includes the following peripherals:
2271

    
2272
@itemize @minus
2273
@item
2274
Intel PXA270 System-on-chip (ARM V5TE core)
2275
@item
2276
NAND Flash memory
2277
@item
2278
IBM/Hitachi DSCM microdrive in a PXA PCMCIA slot - not in "Akita"
2279
@item
2280
On-chip OHCI USB controller
2281
@item
2282
On-chip LCD controller
2283
@item
2284
On-chip Real Time Clock
2285
@item
2286
TI ADS7846 touchscreen controller on SSP bus
2287
@item
2288
Maxim MAX1111 analog-digital converter on I@math{^2}C bus
2289
@item
2290
GPIO-connected keyboard controller and LEDs
2291
@item
2292
Secure Digital card connected to PXA MMC/SD host
2293
@item
2294
Three on-chip UARTs
2295
@item
2296
WM8750 audio CODEC on I@math{^2}C and I@math{^2}S busses
2297
@end itemize
2298

    
2299
The Palm Tungsten|E PDA (codename "Cheetah") emulation includes the
2300
following elements:
2301

    
2302
@itemize @minus
2303
@item
2304
Texas Instruments OMAP310 System-on-chip (ARM 925T core)
2305
@item
2306
ROM and RAM memories (ROM firmware image can be loaded with -option-rom)
2307
@item
2308
On-chip LCD controller
2309
@item
2310
On-chip Real Time Clock
2311
@item
2312
TI TSC2102i touchscreen controller / analog-digital converter / Audio
2313
CODEC, connected through MicroWire and I@math{^2}S busses
2314
@item
2315
GPIO-connected matrix keypad
2316
@item
2317
Secure Digital card connected to OMAP MMC/SD host
2318
@item
2319
Three on-chip UARTs
2320
@end itemize
2321

    
2322
The Luminary Micro Stellaris LM3S811EVB emulation includes the following
2323
devices:
2324

    
2325
@itemize @minus
2326
@item
2327
Cortex-M3 CPU core.
2328
@item
2329
64k Flash and 8k SRAM.
2330
@item
2331
Timers, UARTs, ADC and I@math{^2}C interface.
2332
@item
2333
OSRAM Pictiva 96x16 OLED with SSD0303 controller on I@math{^2}C bus.
2334
@end itemize
2335

    
2336
The Luminary Micro Stellaris LM3S6965EVB emulation includes the following
2337
devices:
2338

    
2339
@itemize @minus
2340
@item
2341
Cortex-M3 CPU core.
2342
@item
2343
256k Flash and 64k SRAM.
2344
@item
2345
Timers, UARTs, ADC, I@math{^2}C and SSI interfaces.
2346
@item
2347
OSRAM Pictiva 128x64 OLED with SSD0323 controller connected via SSI.
2348
@end itemize
2349

    
2350
A Linux 2.6 test image is available on the QEMU web site. More
2351
information is available in the QEMU mailing-list archive.
2352

    
2353
@node ColdFire System emulator
2354
@section ColdFire System emulator
2355

    
2356
Use the executable @file{qemu-system-m68k} to simulate a ColdFire machine.
2357
The emulator is able to boot a uClinux kernel.
2358

    
2359
The M5208EVB emulation includes the following devices:
2360

    
2361
@itemize @minus
2362
@item
2363
MCF5208 ColdFire V2 Microprocessor (ISA A+ with EMAC).
2364
@item
2365
Three Two on-chip UARTs.
2366
@item
2367
Fast Ethernet Controller (FEC)
2368
@end itemize
2369

    
2370
The AN5206 emulation includes the following devices:
2371

    
2372
@itemize @minus
2373
@item
2374
MCF5206 ColdFire V2 Microprocessor.
2375
@item
2376
Two on-chip UARTs.
2377
@end itemize
2378

    
2379
@node QEMU User space emulator
2380
@chapter QEMU User space emulator
2381

    
2382
@menu
2383
* Supported Operating Systems ::
2384
* Linux User space emulator::
2385
* Mac OS X/Darwin User space emulator ::
2386
@end menu
2387

    
2388
@node Supported Operating Systems
2389
@section Supported Operating Systems
2390

    
2391
The following OS are supported in user space emulation:
2392

    
2393
@itemize @minus
2394
@item
2395
Linux (referred as qemu-linux-user)
2396
@item
2397
Mac OS X/Darwin (referred as qemu-darwin-user)
2398
@end itemize
2399

    
2400
@node Linux User space emulator
2401
@section Linux User space emulator
2402

    
2403
@menu
2404
* Quick Start::
2405
* Wine launch::
2406
* Command line options::
2407
* Other binaries::
2408
@end menu
2409

    
2410
@node Quick Start
2411
@subsection Quick Start
2412

    
2413
In order to launch a Linux process, QEMU needs the process executable
2414
itself and all the target (x86) dynamic libraries used by it.
2415

    
2416
@itemize
2417

    
2418
@item On x86, you can just try to launch any process by using the native
2419
libraries:
2420

    
2421
@example
2422
qemu-i386 -L / /bin/ls
2423
@end example
2424

    
2425
@code{-L /} tells that the x86 dynamic linker must be searched with a
2426
@file{/} prefix.
2427

    
2428
@item Since QEMU is also a linux process, you can launch qemu with
2429
qemu (NOTE: you can only do that if you compiled QEMU from the sources):
2430

    
2431
@example
2432
qemu-i386 -L / qemu-i386 -L / /bin/ls
2433
@end example
2434

    
2435
@item On non x86 CPUs, you need first to download at least an x86 glibc
2436
(@file{qemu-runtime-i386-XXX-.tar.gz} on the QEMU web page). Ensure that
2437
@code{LD_LIBRARY_PATH} is not set:
2438

    
2439
@example
2440
unset LD_LIBRARY_PATH
2441
@end example
2442

    
2443
Then you can launch the precompiled @file{ls} x86 executable:
2444

    
2445
@example
2446
qemu-i386 tests/i386/ls
2447
@end example
2448
You can look at @file{qemu-binfmt-conf.sh} so that
2449
QEMU is automatically launched by the Linux kernel when you try to
2450
launch x86 executables. It requires the @code{binfmt_misc} module in the
2451
Linux kernel.
2452

    
2453
@item The x86 version of QEMU is also included. You can try weird things such as:
2454
@example
2455
qemu-i386 /usr/local/qemu-i386/bin/qemu-i386 \
2456
          /usr/local/qemu-i386/bin/ls-i386
2457
@end example
2458

    
2459
@end itemize
2460

    
2461
@node Wine launch
2462
@subsection Wine launch
2463

    
2464
@itemize
2465

    
2466
@item Ensure that you have a working QEMU with the x86 glibc
2467
distribution (see previous section). In order to verify it, you must be
2468
able to do:
2469

    
2470
@example
2471
qemu-i386 /usr/local/qemu-i386/bin/ls-i386
2472
@end example
2473

    
2474
@item Download the binary x86 Wine install
2475
(@file{qemu-XXX-i386-wine.tar.gz} on the QEMU web page).
2476

    
2477
@item Configure Wine on your account. Look at the provided script
2478
@file{/usr/local/qemu-i386/@/bin/wine-conf.sh}. Your previous
2479
@code{$@{HOME@}/.wine} directory is saved to @code{$@{HOME@}/.wine.org}.
2480

    
2481
@item Then you can try the example @file{putty.exe}:
2482

    
2483
@example
2484
qemu-i386 /usr/local/qemu-i386/wine/bin/wine \
2485
          /usr/local/qemu-i386/wine/c/Program\ Files/putty.exe
2486
@end example
2487

    
2488
@end itemize
2489

    
2490
@node Command line options
2491
@subsection Command line options
2492

    
2493
@example
2494
usage: qemu-i386 [-h] [-d] [-L path] [-s size] program [arguments...]
2495
@end example
2496

    
2497
@table @option
2498
@item -h
2499
Print the help
2500
@item -L path
2501
Set the x86 elf interpreter prefix (default=/usr/local/qemu-i386)
2502
@item -s size
2503
Set the x86 stack size in bytes (default=524288)
2504
@end table
2505

    
2506
Debug options:
2507

    
2508
@table @option
2509
@item -d
2510
Activate log (logfile=/tmp/qemu.log)
2511
@item -p pagesize
2512
Act as if the host page size was 'pagesize' bytes
2513
@end table
2514

    
2515
Environment variables:
2516

    
2517
@table @env
2518
@item QEMU_STRACE
2519
Print system calls and arguments similar to the 'strace' program
2520
(NOTE: the actual 'strace' program will not work because the user
2521
space emulator hasn't implemented ptrace).  At the moment this is
2522
incomplete.  All system calls that don't have a specific argument
2523
format are printed with information for six arguments.  Many
2524
flag-style arguments don't have decoders and will show up as numbers.
2525
@end table
2526

    
2527
@node Other binaries
2528
@subsection Other binaries
2529

    
2530
@command{qemu-arm} is also capable of running ARM "Angel" semihosted ELF
2531
binaries (as implemented by the arm-elf and arm-eabi Newlib/GDB
2532
configurations), and arm-uclinux bFLT format binaries.
2533

    
2534
@command{qemu-m68k} is capable of running semihosted binaries using the BDM
2535
(m5xxx-ram-hosted.ld) or m68k-sim (sim.ld) syscall interfaces, and
2536
coldfire uClinux bFLT format binaries.
2537

    
2538
The binary format is detected automatically.
2539

    
2540
@command{qemu-sparc32plus} can execute Sparc32 and SPARC32PLUS binaries
2541
(Sparc64 CPU, 32 bit ABI).
2542

    
2543
@command{qemu-sparc64} can execute some Sparc64 (Sparc64 CPU, 64 bit ABI) and
2544
SPARC32PLUS binaries (Sparc64 CPU, 32 bit ABI).
2545

    
2546
@node Mac OS X/Darwin User space emulator
2547
@section Mac OS X/Darwin User space emulator
2548

    
2549
@menu
2550
* Mac OS X/Darwin Status::
2551
* Mac OS X/Darwin Quick Start::
2552
* Mac OS X/Darwin Command line options::
2553
@end menu
2554

    
2555
@node Mac OS X/Darwin Status
2556
@subsection Mac OS X/Darwin Status
2557

    
2558
@itemize @minus
2559
@item
2560
target x86 on x86: Most apps (Cocoa and Carbon too) works. [1]
2561
@item
2562
target PowerPC on x86: Not working as the ppc commpage can't be mapped (yet!)
2563
@item
2564
target PowerPC on PowerPC: Most apps (Cocoa and Carbon too) works. [1]
2565
@item
2566
target x86 on PowerPC: most utilities work. Cocoa and Carbon apps are not yet supported.
2567
@end itemize
2568

    
2569
[1] If you're host commpage can be executed by qemu.
2570

    
2571
@node Mac OS X/Darwin Quick Start
2572
@subsection Quick Start
2573

    
2574
In order to launch a Mac OS X/Darwin process, QEMU needs the process executable
2575
itself and all the target dynamic libraries used by it. If you don't have the FAT
2576
libraries (you're running Mac OS X/ppc) you'll need to obtain it from a Mac OS X
2577
CD or compile them by hand.
2578

    
2579
@itemize
2580

    
2581
@item On x86, you can just try to launch any process by using the native
2582
libraries:
2583

    
2584
@example
2585
qemu-i386 /bin/ls
2586
@end example
2587

    
2588
or to run the ppc version of the executable:
2589

    
2590
@example
2591
qemu-ppc /bin/ls
2592
@end example
2593

    
2594
@item On ppc, you'll have to tell qemu where your x86 libraries (and dynamic linker)
2595
are installed:
2596

    
2597
@example
2598
qemu-i386 -L /opt/x86_root/ /bin/ls
2599
@end example
2600

    
2601
@code{-L /opt/x86_root/} tells that the dynamic linker (dyld) path is in
2602
@file{/opt/x86_root/usr/bin/dyld}.
2603

    
2604
@end itemize
2605

    
2606
@node Mac OS X/Darwin Command line options
2607
@subsection Command line options
2608

    
2609
@example
2610
usage: qemu-i386 [-h] [-d] [-L path] [-s size] program [arguments...]
2611
@end example
2612

    
2613
@table @option
2614
@item -h
2615
Print the help
2616
@item -L path
2617
Set the library root path (default=/)
2618
@item -s size
2619
Set the stack size in bytes (default=524288)
2620
@end table
2621

    
2622
Debug options:
2623

    
2624
@table @option
2625
@item -d
2626
Activate log (logfile=/tmp/qemu.log)
2627
@item -p pagesize
2628
Act as if the host page size was 'pagesize' bytes
2629
@end table
2630

    
2631
@node compilation
2632
@chapter Compilation from the sources
2633

    
2634
@menu
2635
* Linux/Unix::
2636
* Windows::
2637
* Cross compilation for Windows with Linux::
2638
* Mac OS X::
2639
@end menu
2640

    
2641
@node Linux/Unix
2642
@section Linux/Unix
2643

    
2644
@subsection Compilation
2645

    
2646
First you must decompress the sources:
2647
@example
2648
cd /tmp
2649
tar zxvf qemu-x.y.z.tar.gz
2650
cd qemu-x.y.z
2651
@end example
2652

    
2653
Then you configure QEMU and build it (usually no options are needed):
2654
@example
2655
./configure
2656
make
2657
@end example
2658

    
2659
Then type as root user:
2660
@example
2661
make install
2662
@end example
2663
to install QEMU in @file{/usr/local}.
2664

    
2665
@subsection GCC version
2666

    
2667
In order to compile QEMU successfully, it is very important that you
2668
have the right tools. The most important one is gcc. On most hosts and
2669
in particular on x86 ones, @emph{gcc 4.x is not supported}. If your
2670
Linux distribution includes a gcc 4.x compiler, you can usually
2671
install an older version (it is invoked by @code{gcc32} or
2672
@code{gcc34}). The QEMU configure script automatically probes for
2673
these older versions so that usually you don't have to do anything.
2674

    
2675
@node Windows
2676
@section Windows
2677

    
2678
@itemize
2679
@item Install the current versions of MSYS and MinGW from
2680
@url{http://www.mingw.org/}. You can find detailed installation
2681
instructions in the download section and the FAQ.
2682

    
2683
@item Download
2684
the MinGW development library of SDL 1.2.x
2685
(@file{SDL-devel-1.2.x-@/mingw32.tar.gz}) from
2686
@url{http://www.libsdl.org}. Unpack it in a temporary place, and
2687
unpack the archive @file{i386-mingw32msvc.tar.gz} in the MinGW tool
2688
directory. Edit the @file{sdl-config} script so that it gives the
2689
correct SDL directory when invoked.
2690

    
2691
@item Extract the current version of QEMU.
2692

    
2693
@item Start the MSYS shell (file @file{msys.bat}).
2694

    
2695
@item Change to the QEMU directory. Launch @file{./configure} and
2696
@file{make}.  If you have problems using SDL, verify that
2697
@file{sdl-config} can be launched from the MSYS command line.
2698

    
2699
@item You can install QEMU in @file{Program Files/Qemu} by typing
2700
@file{make install}. Don't forget to copy @file{SDL.dll} in
2701
@file{Program Files/Qemu}.
2702

    
2703
@end itemize
2704

    
2705
@node Cross compilation for Windows with Linux
2706
@section Cross compilation for Windows with Linux
2707

    
2708
@itemize
2709
@item
2710
Install the MinGW cross compilation tools available at
2711
@url{http://www.mingw.org/}.
2712

    
2713
@item
2714
Install the Win32 version of SDL (@url{http://www.libsdl.org}) by
2715
unpacking @file{i386-mingw32msvc.tar.gz}. Set up the PATH environment
2716
variable so that @file{i386-mingw32msvc-sdl-config} can be launched by
2717
the QEMU configuration script.
2718

    
2719
@item
2720
Configure QEMU for Windows cross compilation:
2721
@example
2722
./configure --enable-mingw32
2723
@end example
2724
If necessary, you can change the cross-prefix according to the prefix
2725
chosen for the MinGW tools with --cross-prefix. You can also use
2726
--prefix to set the Win32 install path.
2727

    
2728
@item You can install QEMU in the installation directory by typing
2729
@file{make install}. Don't forget to copy @file{SDL.dll} in the
2730
installation directory.
2731

    
2732
@end itemize
2733

    
2734
Note: Currently, Wine does not seem able to launch
2735
QEMU for Win32.
2736

    
2737
@node Mac OS X
2738
@section Mac OS X
2739

    
2740
The Mac OS X patches are not fully merged in QEMU, so you should look
2741
at the QEMU mailing list archive to have all the necessary
2742
information.
2743

    
2744
@node Index
2745
@chapter Index
2746
@printindex cp
2747

    
2748
@bye