1 = How to convert to -device & friends =
3 === Specifying Bus and Address on Bus ===
5 In qdev, each device has a parent bus. Some devices provide one or
6 more buses for children. You can specify a device's parent bus with
9 A device typically has a device address on its parent bus. For buses
10 where this address can be configured, devices provide a bus-specific
13 bus property name value format
14 PCI addr %x.%x (dev.fn, .fn optional)
18 Example: device i440FX-pcihost is on the root bus, and provides a PCI
19 bus named pci.0. To put a FOO device into its slot 4, use -device
20 FOO,bus=/i440FX-pcihost/pci.0,addr=4. The abbreviated form bus=pci.0
21 also works as long as the bus name is unique.
23 Note: the USB device address can't be controlled at this time.
27 A QEMU block device (drive) has a host and a guest part.
29 In the general case, the guest device is connected to a controller
30 device. For instance, the IDE controller provides two IDE buses, each
31 of which can have up to two ide-drive devices, and each ide-drive
32 device is a guest part, and is connected to a host part.
34 Except we sometimes lump controller, bus(es) and drive device(s) all
35 together into a single device. For instance, the ISA floppy
36 controller is connected to up to two host drives.
38 The old ways to define block devices define host and guest part
39 together. Sometimes, they can even define a controller device in
40 addition to the block device.
42 The new way keeps the parts separate: you create the host part with
43 -drive, and guest device(s) with -device.
45 The various old ways to define drives all boil down to the common form
47 -drive if=TYPE,index=IDX,bus=BUS,unit=UNIT,HOST-OPTS...
49 TYPE, BUS and UNIT identify the controller device, which of its buses
50 to use, and the drive's address on that bus. Details depend on TYPE.
51 IDX is an alternative way to specify BUS and UNIT.
53 In the new way, this becomes something like
55 -drive if=none,id=DRIVE-ID,HOST-OPTS...
56 -device DEVNAME,drive=DRIVE-ID,DEV-OPTS...
58 The -device argument differs in detail for each kind of drive:
62 -device ide-drive,drive=DRIVE-ID,bus=IDE-BUS,unit=UNIT
64 where IDE-BUS identifies an IDE bus, normally either ide.0 or ide.1,
65 and UNIT is either 0 or 1.
67 Bug: new way does not work for ide.1 unit 0 (in old terms: index=2)
68 unless you disable the default CD-ROM with -nodefaults.
72 The old way implicitly creates SCSI controllers as needed. The new
73 way makes that explicit:
75 -device lsi53c895a,id=ID
77 As for all PCI devices, you can add bus=PCI-BUS,addr=DEVFN to
78 control the PCI device address.
80 This SCSI controller a single SCSI bus, named ID.0. Put a disk on
83 -device scsi-disk,drive=DRIVE-ID,bus=ID.0,scsi-id=SCSI-ID,removable=RMB
85 The (optional) removable parameter lets you override the SCSI INQUIRY
86 removable (RMB) bit for non CD-ROM devices. It is ignored for CD-ROM devices
87 which are always removable. RMB is "on" or "off".
91 -global isa-fdc,driveA=DRIVE-ID,driveB=DRIVE-ID
93 This is -global instead of -device, because the floppy controller is
94 created automatically, and we want to configure that one, not create
95 a second one (which isn't possible anyway).
97 Omitting a drive parameter makes that drive empty.
99 Bug: driveA works only if you disable the default floppy drive with
104 -device virtio-blk-pci,drive=DRIVE-ID,class=C,vectors=V,ioeventfd=IOEVENTFD
106 This lets you control PCI device class and MSI-X vectors.
108 IOEVENTFD controls whether or not ioeventfd is used for virtqueue notify. It
109 can be set to on (default) or off.
111 As for all PCI devices, you can add bus=PCI-BUS,addr=DEVFN to
112 control the PCI device address.
114 * if=pflash, if=mtd, if=sd, if=xen are not yet available with -device
116 For USB devices, the old way is actually different:
118 -usbdevice disk:format=FMT:FILENAME
120 Provides much less control than -drive's HOST-OPTS... The new way
123 -device usb-storage,drive=DRIVE-ID,removable=RMB
125 The removable parameter gives control over the SCSI INQUIRY removable (RMB)
126 bit. USB thumbdrives usually set removable=on, while USB hard disks set
127 removable=off. See the if=scsi description above for details on the removable
128 parameter, which applies only to scsi-disk devices and not to scsi-generic.
130 === Character Devices ===
132 A QEMU character device has a host and a guest part.
134 The old ways to define character devices define host and guest part
137 The new way keeps the parts separate: you create the host part with
138 -chardev, and the guest device with -device.
140 The various old ways to define a character device are all of the
143 -FOO FOO-OPTS...,LEGACY-CHARDEV
145 where FOO-OPTS... is specific to -FOO, and the host part
146 LEGACY-CHARDEV is the same everywhere.
148 In the new way, this becomes
150 -chardev HOST-OPTS...,id=CHR-ID
151 -device DEVNAME,chardev=CHR-ID,DEV-OPTS...
153 The appropriate DEVNAME depends on the machine type. For type "pc":
155 * -serial becomes -device isa-serial,iobase=IOADDR,irq=IRQ,index=IDX
157 This lets you control I/O ports and IRQs.
159 * -parallel becomes -device isa-parallel,iobase=IOADDR,irq=IRQ,index=IDX
161 This lets you control I/O ports and IRQs.
163 * -usbdevice serial:vendorid=VID,productid=PRID becomes
164 -device usb-serial,vendorid=VID,productid=PRID
166 * -usbdevice braille doesn't support LEGACY-CHARDEV syntax. It always
167 uses "braille". With -device, this useful default is gone, so you
168 have to use something like
170 -device usb-braille,chardev=braille,vendorid=VID,productid=PRID
171 -chardev braille,id=braille
173 * -virtioconsole is still being worked on
175 LEGACY-CHARDEV translates to -chardev HOST-OPTS... as follows:
177 * null becomes -chardev null
179 * pty, msmouse, braille, stdio likewise
181 * vc:WIDTHxHEIGHT becomes -chardev vc,width=WIDTH,height=HEIGHT
183 * vc:<COLS>Cx<ROWS>C becomes -chardev vc,cols=<COLS>,rows=<ROWS>
185 * con: becomes -chardev console
187 * COM<NUM> becomes -chardev serial,path=<NUM>
189 * file:FNAME becomes -chardev file,path=FNAME
191 * pipe:FNAME becomes -chardev pipe,path=FNAME
193 * tcp:HOST:PORT,OPTS... becomes -chardev socket,host=HOST,port=PORT,OPTS...
195 * telnet:HOST:PORT,OPTS... becomes
196 -chardev socket,host=HOST,port=PORT,OPTS...,telnet=on
198 * udp:HOST:PORT@LOCALADDR:LOCALPORT becomes
199 -chardev udp,host=HOST,port=PORT,localaddr=LOCALADDR,localport=LOCALPORT
201 * unix:FNAME becomes -chardev socket,path=FNAME
203 * /dev/parportN becomes -chardev parport,file=/dev/parportN
207 * Any other /dev/FNAME becomes -chardev tty,path=/dev/FNAME
209 * mon:LEGACY-CHARDEV is special: it multiplexes the monitor onto the
210 character device defined by LEGACY-CHARDEV. -chardev provides more
211 general multiplexing instead: you can connect up to four users to a
212 single host part. You need to pass mux=on to -chardev to enable
213 switching the input focus.
215 QEMU uses LEGACY-CHARDEV syntax not just to set up guest devices, but
216 also in various other places such as -monitor or -net
217 user,guestfwd=... You can use chardev:CHR-ID in place of
218 LEGACY-CHARDEV to refer to a host part defined with -chardev.
220 === Network Devices ===
222 A QEMU network device (NIC) has a host and a guest part.
224 The old ways to define NICs define host and guest part together. It
227 -net nic,vlan=VLAN,macaddr=MACADDR,model=MODEL,name=ID,addr=STR,vectors=V
229 Except for USB it looks like this:
231 -usbdevice net:vlan=VLAN,macaddr=MACADDR,name=ID,addr=STR,vectors=V
233 The new way keeps the parts separate: you create the host part with
234 -netdev, and the guest device with -device, like this:
236 -netdev type=TYPE,id=NET-ID
237 -device DEVNAME,netdev=NET-ID,mac=MACADDR,DEV-OPTS...
239 Unlike the old way, this creates just a network device, not a VLAN.
240 If you really want a VLAN, create it the usual way, then create the
241 guest device like this:
243 -device DEVNAME,vlan=VLAN,mac=MACADDR,DEV-OPTS...
245 DEVNAME equals MODEL, except for virtio you have to name the virtio
246 device appropriate for the bus (virtio-net-pci for PCI), and for USB
247 NIC you have to use usb-net.
249 The old name=ID parameter becomes the usual id=ID with -device.
251 For PCI devices, you can add bus=PCI-BUS,addr=DEVFN to control the PCI
252 device address, as usual. The old -net nic provides parameter addr
253 for that, it is silently ignored when the NIC is not a PCI device.
255 For virtio-net-pci, you can control whether or not ioeventfd is used for
256 virtqueue notify by setting ioeventfd= to on or off (default).
258 -net nic accepts vectors=V for all models, but it's silently ignored
259 except for virtio-net-pci (model=virtio). With -device, only devices
260 that support it accept it.
262 Not all devices are available with -device at this time. All PCI
263 devices and ne2k_isa are.
265 Some PCI devices aren't available with -net nic, e.g. i82558a.
267 Bug: usb-net does not work, yet. Patch posted.
269 === Graphics Devices ===
271 Host and guest part of graphics devices have always been separate.
273 The old way to define the guest graphics device is -vga VGA.
275 The new way is -device. Map from -vga argument to -device:
278 cirrus -device cirrus-vga
279 vmware -device vmware-svga
280 xenfb not yet available with -device
282 As for all PCI devices, you can add bus=PCI-BUS,addr=DEVFN to control
283 the PCI device address.
285 -device VGA supports properties bios-offset and bios-size, but they
286 aren't used with machine type "pc".
288 Bug: -device cirrus-vga and -device vmware-svga require -nodefaults.
290 Bug: the new way requires PCI; ISA VGA is not yet available with
293 Bug: the new way doesn't work for machine type "pc", because it
294 violates obscure device initialization ordering constraints.
296 === Audio Devices ===
298 Host and guest part of audio devices have always been separate.
300 The old way to define guest audio devices is -soundhw C1,...
302 The new way is to define each guest audio device separately with
305 Map from -soundhw sound card name to -device:
308 cs4231a -device cs4231a,iobase=IOADDR,irq=IRQ,dma=DMA
309 es1370 -device ES1370
310 gus -device gus,iobase=IOADDR,irq=IRQ,dma=DMA,freq=F
311 sb16 -device sb16,iobase=IOADDR,irq=IRQ,dma=DMA,dma16=DMA16,version=V
312 adlib not yet available with -device
313 pcspk not yet available with -device
315 For PCI devices, you can add bus=PCI-BUS,addr=DEVFN to control the PCI
316 device address, as usual.
320 The old way to define a virtual USB device is -usbdevice DRIVER:OPTS...
322 The new way is -device DEVNAME,DEV-OPTS... Details depend on DRIVER:
324 * mouse -device usb-mouse
325 * tablet -device usb-tablet
326 * keyboard -device usb-kdb
327 * wacom-tablet -device usb-wacom-tablet
328 * host:... See "Host Device Assignment"
329 * disk:... See "Block Devices"
330 * serial:... See "Character Devices"
331 * braille See "Character Devices"
332 * net:... See "Network Devices"
333 * bt:... not yet available with -device
335 === Watchdog Devices ===
337 Host and guest part of watchdog devices have always been separate.
339 The old way to define a guest watchdog device is -watchdog DEVNAME.
340 The new way is -device DEVNAME. For PCI devices, you can add
341 bus=PCI-BUS,addr=DEVFN to control the PCI device address, as usual.
343 === Host Device Assignment ===
345 QEMU supports assigning host PCI devices (qemu-kvm only at this time)
346 and host USB devices.
348 The old way to assign a host PCI device is
350 -pcidevice host=ADDR,dma=none,id=ID
354 -device pci-assign,host=ADDR,iommu=IOMMU,id=ID
356 The old dma=none becomes iommu=0 with -device.
358 The old way to assign a host USB device is
360 -usbdevice host:auto:BUS.ADDR:VID:PRID
362 where any of BUS, ADDR, VID, PRID can be the wildcard *.
366 -device usb-host,hostbus=BUS,hostaddr=ADDR,vendorid=VID,productid=PRID
368 where left out or zero BUS, ADDR, VID, PRID serve as wildcard.