1 \input texinfo @c -*- texinfo -*-
4 @settitle QEMU CPU Emulator User Documentation
7 @center @titlefont{QEMU CPU Emulator User Documentation}
16 QEMU is a FAST! processor emulator using dynamic translation to
17 achieve good emulation speed.
19 QEMU has two operating modes:
24 Full system emulation. In this mode, QEMU emulates a full system (for
25 example a PC), including a processor and various peripherials. It can
26 be used to launch different Operating Systems without rebooting the
27 PC or to debug system code.
30 User mode emulation (Linux host only). In this mode, QEMU can launch
31 Linux processes compiled for one CPU on another CPU. It can be used to
32 launch the Wine Windows API emulator (@url{http://www.winehq.org}) or
33 to ease cross-compilation and cross-debugging.
37 As QEMU requires no host kernel driver to run, it is very safe and
40 For system emulation, the following hardware targets are supported:
42 @item PC (x86 processor)
43 @item PREP (PowerPC processor)
46 For user emulation, x86, PowerPC, ARM, and SPARC CPUs are supported.
52 If you want to compile QEMU, please read the @file{README} which gives
53 the related information. Otherwise just download the binary
54 distribution (@file{qemu-XXX-i386.tar.gz}) and untar it as root in
60 tar zxvf /tmp/qemu-XXX-i386.tar.gz
66 @item Install the current versions of MSYS and MinGW from
67 @url{http://www.mingw.org/}. You can find detailed installation
68 instructions in the download section and the FAQ.
71 the MinGW development library of SDL 1.2.x
72 (@file{SDL-devel-1.2.x-mingw32.tar.gz}) from
73 @url{http://www.libsdl.org}. Unpack it in a temporary place, and
74 unpack the archive @file{i386-mingw32msvc.tar.gz} in the MinGW tool
75 directory. Edit the @file{sdl-config} script so that it gives the
76 correct SDL directory when invoked.
78 @item Extract the current version of QEMU.
80 @item Start the MSYS shell (file @file{msys.bat}).
82 @item Change to the QEMU directory. Launch @file{./configure} and
83 @file{make}. If you have problems using SDL, verify that
84 @file{sdl-config} can be launched from the MSYS command line.
86 @item You can install QEMU in @file{Program Files/Qemu} by typing
87 @file{make install}. Don't forget to copy @file{SDL.dll} in
88 @file{Program Files/Qemu}.
92 @section Cross compilation for Windows with Linux
96 Install the MinGW cross compilation tools available at
97 @url{http://www.mingw.org/}.
100 Install the Win32 version of SDL (@url{http://www.libsdl.org}) by
101 unpacking @file{i386-mingw32msvc.tar.gz}. Set up the PATH environment
102 variable so that @file{i386-mingw32msvc-sdl-config} can be launched by
103 the QEMU configuration script.
106 Configure QEMU for Windows cross compilation:
108 ./configure --enable-mingw32
110 If necessary, you can change the cross-prefix according to the prefix
111 choosen for the MinGW tools with --cross-prefix. You can also use
112 --prefix to set the Win32 install path.
114 @item You can install QEMU in the installation directory by typing
115 @file{make install}. Don't forget to copy @file{SDL.dll} in the
116 installation directory.
120 Note: Currently, Wine does not seem able to launch
125 Mac OS X is currently not supported.
127 @chapter QEMU PC System emulator invocation
129 @section Introduction
131 @c man begin DESCRIPTION
133 The QEMU System emulator simulates a complete PC.
135 In order to meet specific user needs, two versions of QEMU are
141 @code{qemu-fast} uses the host Memory Management Unit (MMU) to simulate
142 the x86 MMU. It is @emph{fast} but has limitations because the whole 4 GB
143 address space cannot be used and some memory mapped peripherials
144 cannot be emulated accurately yet. Therefore, a specific Linux kernel
145 must be used (@xref{linux_compile}).
148 @code{qemu} uses a software MMU. It is about @emph{two times
149 slower} but gives a more accurate emulation.
153 QEMU emulates the following PC peripherials:
157 VGA (hardware level, including all non standard modes)
159 PS/2 mouse and keyboard
161 2 IDE interfaces with hard disk and CD-ROM support
165 up to 6 NE2000 network adapters
176 Download and uncompress the linux image (@file{linux.img}) and type:
182 Linux should boot and give you a prompt.
187 @c man begin SYNOPSIS
188 usage: qemu [options] [disk_image]
193 @var{disk_image} is a raw hard disk image for IDE hard disk 0.
199 Use @var{file} as floppy disk 0/1 image (@xref{disk_images}).
205 Use @var{file} as hard disk 0, 1, 2 or 3 image (@xref{disk_images}).
208 Use @var{file} as CD-ROM image (you cannot use @option{-hdc} and and
209 @option{-cdrom} at the same time).
212 Boot on floppy (a), hard disk (c) or CD-ROM (d). Hard disk boot is
216 Write to temporary files instead of disk image files. In this case,
217 the raw disk image you use is not written back. You can however force
218 the write back by pressing @key{C-a s} (@xref{disk_images}).
221 Set virtual RAM size to @var{megs} megabytes.
224 Use @var{file} as initial ram disk.
228 Normally, QEMU uses SDL to display the VGA output. With this option,
229 you can totally disable graphical output so that QEMU is a simple
230 command line application. The emulated serial port is redirected on
231 the console. Therefore, you can still use QEMU to debug a Linux kernel
232 with a serial console.
236 The SB16 emulation is disabled by default as it may give problems with
237 Windows. You can enable it manually with this option.
246 Set TUN/TAP network init script [default=/etc/qemu-ifup]. This script
247 is launched to configure the host network interface (usually tun0)
248 corresponding to the virtual NE2000 card.
252 Set the mac address of the first interface (the format is
253 aa:bb:cc:dd:ee:ff in hexa). The mac address is incremented for each
254 new network interface.
257 Assumes @var{fd} talks to a tap/tun host network interface and use
258 it. Read @url{http://bellard.org/qemu/tetrinet.html} to have an
262 (Experimental) Use the user mode network stack. This is the default if
263 no tun/tap network init script is found.
266 Use the dummy network stack: no packet will be received on the network
271 Linux boot specific. When using this options, you can use a given
272 Linux kernel without installing it in the disk image. It can be useful
273 for easier testing of various kernels.
277 @item -kernel bzImage
278 Use @var{bzImage} as kernel image.
280 @item -append cmdline
281 Use @var{cmdline} as kernel command line
284 Use @var{file} as initial ram disk.
291 Wait gdb connection to port 1234 (@xref{gdb_usage}).
293 Change gdb connection port.
295 Do not start CPU at startup (you must type 'c' in the monitor).
297 Output log in /tmp/qemu.log
300 During the graphical emulation, you can use the following keys:
303 Toggle mouse and keyboard grab.
308 During emulation, if you are using the serial console, use @key{C-a h}
309 to get terminal commands:
317 Save disk data back to file (if -snapshot)
319 Send break (magic sysrq in Linux)
321 Switch between console and monitor
330 @settitle QEMU System Emulator
333 The HTML documentation of QEMU for more precise information and Linux
334 user mode emulator invocation.
346 @section QEMU Monitor
348 The QEMU monitor is used to give complex commands to the QEMU
349 emulator. You can use it to:
354 Remove or insert removable medias images
355 (such as CD-ROM or floppies)
358 Freeze/unfreeze the Virtual Machine (VM) and save or restore its state
361 @item Inspect the VM state without an external debugger.
367 The following commands are available:
371 @item help or ? [cmd]
372 Show the help for all commands or just for command @var{cmd}.
375 Commit changes to the disk images (if -snapshot is used)
377 @item info subcommand
378 show various information about the system state
382 show the network state
384 show the block devices
386 show the cpu registers
388 show the command line history
394 @item eject [-f] device
395 Eject a removable media (use -f to force it).
397 @item change device filename
398 Change a removable media.
400 @item screendump filename
401 Save screen into PPM image @var{filename}.
403 @item log item1[,...]
404 Activate logging of the specified items to @file{/tmp/qemu.log}.
406 @item savevm filename
407 Save the whole virtual machine state to @var{filename}.
409 @item loadvm filename
410 Restore the whole virtual machine state from @var{filename}.
418 @item gdbserver [port]
419 Start gdbserver session (default port=1234)
422 Virtual memory dump starting at @var{addr}.
425 Physical memory dump starting at @var{addr}.
427 @var{fmt} is a format which tells the command how to format the
428 data. Its syntax is: @option{/@{count@}@{format@}@{size@}}
432 is the number of items to be dumped.
435 can be x (hexa), d (signed decimal), u (unsigned decimal), o (octal),
436 c (char) or i (asm instruction).
439 can be b (8 bits), h (16 bits), w (32 bits) or g (64 bits). On x86,
440 @code{h} or @code{w} can be specified with the @code{i} format to
441 respectively select 16 or 32 bit code instruction size.
448 Dump 10 instructions at the current instruction pointer:
453 0x90107065: lea 0x0(%esi,1),%esi
454 0x90107069: lea 0x0(%edi,1),%edi
456 0x90107071: jmp 0x90107080
464 Dump 80 16 bit values at the start of the video memory.
466 (qemu) xp/80hx 0xb8000
467 0x000b8000: 0x0b50 0x0b6c 0x0b65 0x0b78 0x0b38 0x0b36 0x0b2f 0x0b42
468 0x000b8010: 0x0b6f 0x0b63 0x0b68 0x0b73 0x0b20 0x0b56 0x0b47 0x0b41
469 0x000b8020: 0x0b42 0x0b69 0x0b6f 0x0b73 0x0b20 0x0b63 0x0b75 0x0b72
470 0x000b8030: 0x0b72 0x0b65 0x0b6e 0x0b74 0x0b2d 0x0b63 0x0b76 0x0b73
471 0x000b8040: 0x0b20 0x0b30 0x0b35 0x0b20 0x0b4e 0x0b6f 0x0b76 0x0b20
472 0x000b8050: 0x0b32 0x0b30 0x0b30 0x0b33 0x0720 0x0720 0x0720 0x0720
473 0x000b8060: 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720
474 0x000b8070: 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720
475 0x000b8080: 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720
476 0x000b8090: 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720 0x0720
480 @item p or print/fmt expr
482 Print expression value. Only the @var{format} part of @var{fmt} is
487 @subsection Integer expressions
489 The monitor understands integers expressions for every integer
490 argument. You can use register names to get the value of specifics
491 CPU registers by prefixing them with @emph{$}.
496 @subsection Raw disk images
498 The disk images can simply be raw images of the hard disk. You can
499 create them with the command:
501 dd of=myimage bs=1024 seek=mysize count=0
503 where @var{myimage} is the image filename and @var{mysize} is its size
506 @subsection Snapshot mode
508 If you use the option @option{-snapshot}, all disk images are
509 considered as read only. When sectors in written, they are written in
510 a temporary file created in @file{/tmp}. You can however force the
511 write back to the raw disk images by pressing @key{C-a s}.
513 NOTE: The snapshot mode only works with raw disk images.
515 @subsection Copy On Write disk images
517 QEMU also supports user mode Linux
518 (@url{http://user-mode-linux.sourceforge.net/}) Copy On Write (COW)
519 disk images. The COW disk images are much smaller than normal images
520 as they store only modified sectors. They also permit the use of the
521 same disk image template for many users.
523 To create a COW disk images, use the command:
526 qemu-mkcow -f myrawimage.bin mycowimage.cow
529 @file{myrawimage.bin} is a raw image you want to use as original disk
530 image. It will never be written to.
532 @file{mycowimage.cow} is the COW disk image which is created by
533 @code{qemu-mkcow}. You can use it directly with the @option{-hdx}
534 options. You must not modify the original raw disk image if you use
535 COW images, as COW images only store the modified sectors from the raw
536 disk image. QEMU stores the original raw disk image name and its
537 modified time in the COW disk image so that chances of mistakes are
540 If the raw disk image is not read-only, by pressing @key{C-a s} you
541 can flush the COW disk image back into the raw disk image, as in
544 COW disk images can also be created without a corresponding raw disk
545 image. It is useful to have a big initial virtual disk image without
546 using much disk space. Use:
549 qemu-mkcow mycowimage.cow 1024
552 to create a 1 gigabyte empty COW disk image.
557 COW disk images must be created on file systems supporting
558 @emph{holes} such as ext2 or ext3.
560 Since holes are used, the displayed size of the COW disk image is not
561 the real one. To know it, use the @code{ls -ls} command.
564 @section Network emulation
566 QEMU simulates up to 6 networks cards (NE2000 boards). Each card can
567 be connected to a specific host network interface.
569 @subsection Using tun/tap network interface
571 This is the standard way to emulate network. QEMU adds a virtual
572 network device on your host (called @code{tun0}), and you can then
573 configure it as if it was a real ethernet card.
575 As an example, you can download the @file{linux-test-xxx.tar.gz}
576 archive and copy the script @file{qemu-ifup} in @file{/etc} and
577 configure properly @code{sudo} so that the command @code{ifconfig}
578 contained in @file{qemu-ifup} can be executed as root. You must verify
579 that your host kernel supports the TUN/TAP network interfaces: the
580 device @file{/dev/net/tun} must be present.
582 See @ref{direct_linux_boot} to have an example of network use with a
585 @subsection Using the user mode network stack
587 This is @emph{experimental} (version 0.5.4). You must configure qemu
588 with @code{--enable-slirp}. Then by using the option
589 @option{-user-net} or if you have no tun/tap init script, QEMU uses a
590 completely user mode network stack (you don't need root priviledge to
591 use the virtual network). The virtual network configuration is the
596 QEMU Virtual Machine <------> Firewall/DHCP server <-----> Internet
597 (10.0.2.x) | (10.0.2.2)
603 The QEMU VM behaves as if it was behind a firewall which blocks all
604 incoming connections. You can use a DHCP client to automatically
605 configure the network in the QEMU VM.
607 In order to check that the user mode network is working, you can ping
608 the address 10.0.2.2 and verify that you got an address in the range
609 10.0.2.x from the QEMU virtual DHCP server.
611 @node direct_linux_boot
612 @section Direct Linux Boot
614 This section explains how to launch a Linux kernel inside QEMU without
615 having to make a full bootable image. It is very useful for fast Linux
616 kernel testing. The QEMU network configuration is also explained.
620 Download the archive @file{linux-test-xxx.tar.gz} containing a Linux
621 kernel and a disk image.
623 @item Optional: If you want network support (for example to launch X11 examples), you
624 must copy the script @file{qemu-ifup} in @file{/etc} and configure
625 properly @code{sudo} so that the command @code{ifconfig} contained in
626 @file{qemu-ifup} can be executed as root. You must verify that your host
627 kernel supports the TUN/TAP network interfaces: the device
628 @file{/dev/net/tun} must be present.
630 When network is enabled, there is a virtual network connection between
631 the host kernel and the emulated kernel. The emulated kernel is seen
632 from the host kernel at IP address 172.20.0.2 and the host kernel is
633 seen from the emulated kernel at IP address 172.20.0.1.
635 @item Launch @code{qemu.sh}. You should have the following output:
639 Connected to host network interface: tun0
640 Linux version 2.4.21 (bellard@voyager.localdomain) (gcc version 3.2.2 20030222 (Red Hat Linux 3.2.2-5)) #5 Tue Nov 11 18:18:53 CET 2003
641 BIOS-provided physical RAM map:
642 BIOS-e801: 0000000000000000 - 000000000009f000 (usable)
643 BIOS-e801: 0000000000100000 - 0000000002000000 (usable)
644 32MB LOWMEM available.
645 On node 0 totalpages: 8192
649 Kernel command line: root=/dev/hda sb=0x220,5,1,5 ide2=noprobe ide3=noprobe ide4=noprobe ide5=noprobe console=ttyS0
650 ide_setup: ide2=noprobe
651 ide_setup: ide3=noprobe
652 ide_setup: ide4=noprobe
653 ide_setup: ide5=noprobe
655 Detected 2399.621 MHz processor.
656 Console: colour EGA 80x25
657 Calibrating delay loop... 4744.80 BogoMIPS
658 Memory: 28872k/32768k available (1210k kernel code, 3508k reserved, 266k data, 64k init, 0k highmem)
659 Dentry cache hash table entries: 4096 (order: 3, 32768 bytes)
660 Inode cache hash table entries: 2048 (order: 2, 16384 bytes)
661 Mount cache hash table entries: 512 (order: 0, 4096 bytes)
662 Buffer-cache hash table entries: 1024 (order: 0, 4096 bytes)
663 Page-cache hash table entries: 8192 (order: 3, 32768 bytes)
664 CPU: Intel Pentium Pro stepping 03
665 Checking 'hlt' instruction... OK.
666 POSIX conformance testing by UNIFIX
667 Linux NET4.0 for Linux 2.4
668 Based upon Swansea University Computer Society NET3.039
669 Initializing RT netlink socket
672 Journalled Block Device driver loaded
673 Detected PS/2 Mouse Port.
674 pty: 256 Unix98 ptys configured
675 Serial driver version 5.05c (2001-07-08) with no serial options enabled
676 ttyS00 at 0x03f8 (irq = 4) is a 16450
677 ne.c:v1.10 9/23/94 Donald Becker (becker@scyld.com)
678 Last modified Nov 1, 2000 by Paul Gortmaker
679 NE*000 ethercard probe at 0x300: 52 54 00 12 34 56
680 eth0: NE2000 found at 0x300, using IRQ 9.
681 RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
682 Uniform Multi-Platform E-IDE driver Revision: 7.00beta4-2.4
683 ide: Assuming 50MHz system bus speed for PIO modes; override with idebus=xx
684 hda: QEMU HARDDISK, ATA DISK drive
685 ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
686 hda: attached ide-disk driver.
687 hda: 20480 sectors (10 MB) w/256KiB Cache, CHS=20/16/63
690 Soundblaster audio driver Copyright (C) by Hannu Savolainen 1993-1996
691 NET4: Linux TCP/IP 1.0 for NET4.0
692 IP Protocols: ICMP, UDP, TCP, IGMP
693 IP: routing cache hash table of 512 buckets, 4Kbytes
694 TCP: Hash tables configured (established 2048 bind 4096)
695 NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
696 EXT2-fs warning: mounting unchecked fs, running e2fsck is recommended
697 VFS: Mounted root (ext2 filesystem).
698 Freeing unused kernel memory: 64k freed
700 Linux version 2.4.21 (bellard@voyager.localdomain) (gcc version 3.2.2 20030222 (Red Hat Linux 3.2.2-5)) #5 Tue Nov 11 18:18:53 CET 2003
702 QEMU Linux test distribution (based on Redhat 9)
704 Type 'exit' to halt the system
710 Then you can play with the kernel inside the virtual serial console. You
711 can launch @code{ls} for example. Type @key{Ctrl-a h} to have an help
712 about the keys you can type inside the virtual serial console. In
713 particular, use @key{Ctrl-a x} to exit QEMU and use @key{Ctrl-a b} as
717 If the network is enabled, launch the script @file{/etc/linuxrc} in the
718 emulator (don't forget the leading dot):
723 Then enable X11 connections on your PC from the emulated Linux:
728 You can now launch @file{xterm} or @file{xlogo} and verify that you have
729 a real Virtual Linux system !
736 A 2.5.74 kernel is also included in the archive. Just
737 replace the bzImage in qemu.sh to try it.
740 qemu-fast creates a temporary file in @var{$QEMU_TMPDIR} (@file{/tmp} is the
741 default) containing all the simulated PC memory. If possible, try to use
742 a temporary directory using the tmpfs filesystem to avoid too many
743 unnecessary disk accesses.
746 In order to exit cleanly from qemu, you can do a @emph{shutdown} inside
747 qemu. qemu will automatically exit when the Linux shutdown is done.
750 You can boot slightly faster by disabling the probe of non present IDE
751 interfaces. To do so, add the following options on the kernel command
754 ide1=noprobe ide2=noprobe ide3=noprobe ide4=noprobe ide5=noprobe
758 The example disk image is a modified version of the one made by Kevin
759 Lawton for the plex86 Project (@url{www.plex86.org}).
764 @section Linux Kernel Compilation
766 You can use any linux kernel with QEMU. However, if you want to use
767 @code{qemu-fast} to get maximum performances, you must use a modified
768 guest kernel. If you are using a 2.6 guest kernel, you can use
769 directly the patch @file{linux-2.6-qemu-fast.patch} made by Rusty
770 Russel available in the QEMU source archive. Otherwise, you can make the
771 following changes @emph{by hand} to the Linux kernel:
775 The kernel must be mapped at 0x90000000 (the default is
776 0xc0000000). You must modify only two lines in the kernel source:
778 In @file{include/asm/page.h}, replace
780 #define __PAGE_OFFSET (0xc0000000)
784 #define __PAGE_OFFSET (0x90000000)
787 And in @file{arch/i386/vmlinux.lds}, replace
789 . = 0xc0000000 + 0x100000;
793 . = 0x90000000 + 0x100000;
797 If you want to enable SMP (Symmetric Multi-Processing) support, you
798 must make the following change in @file{include/asm/fixmap.h}. Replace
800 #define FIXADDR_TOP (0xffffX000UL)
804 #define FIXADDR_TOP (0xa7ffX000UL)
806 (X is 'e' or 'f' depending on the kernel version). Although you can
807 use an SMP kernel with QEMU, it only supports one CPU.
810 If you are not using a 2.6 kernel as host kernel but if you use a target
811 2.6 kernel, you must also ensure that the 'HZ' define is set to 100
812 (1000 is the default) as QEMU cannot currently emulate timers at
813 frequencies greater than 100 Hz on host Linux systems < 2.6. In
814 @file{include/asm/param.h}, replace:
817 # define HZ 1000 /* Internal kernel timer frequency */
821 # define HZ 100 /* Internal kernel timer frequency */
826 The file config-2.x.x gives the configuration of the example kernels.
833 As you would do to make a real kernel. Then you can use with QEMU
834 exactly the same kernel as you would boot on your PC (in
835 @file{arch/i386/boot/bzImage}).
840 QEMU has a primitive support to work with gdb, so that you can do
841 'Ctrl-C' while the virtual machine is running and inspect its state.
843 In order to use gdb, launch qemu with the '-s' option. It will wait for a
846 > qemu -s -kernel arch/i386/boot/bzImage -hda root-2.4.20.img -append "root=/dev/hda"
847 Connected to host network interface: tun0
848 Waiting gdb connection on port 1234
851 Then launch gdb on the 'vmlinux' executable:
856 In gdb, connect to QEMU:
858 (gdb) target remote localhost:1234
861 Then you can use gdb normally. For example, type 'c' to launch the kernel:
866 Here are some useful tips in order to use gdb on system code:
870 Use @code{info reg} to display all the CPU registers.
872 Use @code{x/10i $eip} to display the code at the PC position.
874 Use @code{set architecture i8086} to dump 16 bit code. Then use
875 @code{x/10i $cs*16+*eip} to dump the code at the PC position.
878 @section Target OS specific information
882 To have access to SVGA graphic modes under X11, use the @code{vesa}
883 X11 driver. For optimal performances, use the same depth as your
888 If you have a slow host, using Windows 95 is better as it gives the
889 best speed. Windows 2000 is also a good choice.
891 SVGA graphic modes support: QEMU currently supports the Bochs VESA VBE
892 extensions. It supports color depths of 8, 15, 16 and 32 bits per
893 pixel in 640x480, 800x600 and 1024x768. For optimal performances, use
894 the same depth as your native display.
898 @item Windows XP: it should be automatically detected.
900 @item Windows NT4 or 2000: use the driver
901 @url{http://www.volny.cz/xnavara/qemuvid_bin.zip} by Filip Navara.
903 @item Windows 95/98/Me: no clean solution yet (but it will change
904 soon). You can however use the shareware driver from SciTech. Here are
905 the steps recommended by Christophe Bothamy on the Bochs mailing list:
908 @item install win95 with the VGA driver.
909 @item download sdd 7 beta from @url{http://www.majorgeeks.com/download382.html}
910 @item download pmhelp.vxd from @url{http://unununium.org/viewcvs/snap/redist/release/pmhelp.vxd}
911 @item copy pmhelp.vxd to the win95 system directory
916 @chapter QEMU PREP PowerPC System emulator invocation
918 Use the executable @file{qemu-system-ppc} to simulate a complete PREP
921 QEMU emulates the following PREP peripherials:
925 2 IDE interfaces with hard disk and CD-ROM support
929 up to 6 NE2000 network adapters
933 PREP Non Volatile RAM
936 You can read the qemu PC system emulation chapter to have more
937 informations about QEMU usage.
939 More information is available at
940 @url{http://jocelyn.mayer.free.fr/qemu-ppc/}.
942 @chapter QEMU User space emulator invocation
946 In order to launch a Linux process, QEMU needs the process executable
947 itself and all the target (x86) dynamic libraries used by it.
951 @item On x86, you can just try to launch any process by using the native
955 qemu-i386 -L / /bin/ls
958 @code{-L /} tells that the x86 dynamic linker must be searched with a
961 @item Since QEMU is also a linux process, you can launch qemu with qemu (NOTE: you can only do that if you compiled QEMU from the sources):
964 qemu-i386 -L / qemu-i386 -L / /bin/ls
967 @item On non x86 CPUs, you need first to download at least an x86 glibc
968 (@file{qemu-runtime-i386-XXX-.tar.gz} on the QEMU web page). Ensure that
969 @code{LD_LIBRARY_PATH} is not set:
972 unset LD_LIBRARY_PATH
975 Then you can launch the precompiled @file{ls} x86 executable:
978 qemu-i386 tests/i386/ls
980 You can look at @file{qemu-binfmt-conf.sh} so that
981 QEMU is automatically launched by the Linux kernel when you try to
982 launch x86 executables. It requires the @code{binfmt_misc} module in the
985 @item The x86 version of QEMU is also included. You can try weird things such as:
987 qemu-i386 /usr/local/qemu-i386/bin/qemu-i386 /usr/local/qemu-i386/bin/ls-i386
996 @item Ensure that you have a working QEMU with the x86 glibc
997 distribution (see previous section). In order to verify it, you must be
1001 qemu-i386 /usr/local/qemu-i386/bin/ls-i386
1004 @item Download the binary x86 Wine install
1005 (@file{qemu-XXX-i386-wine.tar.gz} on the QEMU web page).
1007 @item Configure Wine on your account. Look at the provided script
1008 @file{/usr/local/qemu-i386/bin/wine-conf.sh}. Your previous
1009 @code{$@{HOME@}/.wine} directory is saved to @code{$@{HOME@}/.wine.org}.
1011 @item Then you can try the example @file{putty.exe}:
1014 qemu-i386 /usr/local/qemu-i386/wine/bin/wine /usr/local/qemu-i386/wine/c/Program\ Files/putty.exe
1019 @section Command line options
1022 usage: qemu-i386 [-h] [-d] [-L path] [-s size] program [arguments...]
1029 Set the x86 elf interpreter prefix (default=/usr/local/qemu-i386)
1031 Set the x86 stack size in bytes (default=524288)
1038 Activate log (logfile=/tmp/qemu.log)
1040 Act as if the host page size was 'pagesize' bytes