What follows is a record of recently removed, formerly deprecated features that serves as a record for users who have encountered trouble after a recent upgrade.
The geometry defined by -hdachs c,h,s,t
should now be specified via
-device ide-hd,drive=dr,cyls=c,heads=h,secs=s,bios-chs-trans=t
(together with -drive if=none,id=dr,...
).
This option has been replaced by -net user,guestfwd=...
.
-net dump[,vlan=n][,file=filename][,len=maxlen]
has been replaced by
-object filter-dump,id=id,netdev=dev[,file=filename][,maxlen=maxlen]
.
Note that the new syntax works with netdev IDs instead of the old "vlan" hubs.
This was just a dummy option that has been ignored, since the in-kernel PIT
cannot be disabled separately from the irqchip anymore. A similar effect
(which also disables the KVM IOAPIC) can be obtained with
-M kernel_irqchip=split
.
There is no replacement, the -tdf
option has just been ignored since the
behaviour that could be changed by this option in qemu-kvm is now the default
when using the KVM PIT. It still can be requested explicitly using
-global kvm-pit.lost_tick_policy=delay
.
The drive geometry should now be specified via
-device ...,drive=dr,cyls=c,heads=h,secs=s
(together with
-drive if=none,id=dr,...
).
Use -device ...,drive=dr,serial=r,bios-chs-trans=t,addr=a
instead
(together with -drive if=none,id=dr,...
).
The term "vlan" was very confusing for most users in this context (it's about
specifying a hub ID, not about IEEE 802.1Q or something similar), so this
has been removed. To connect one NIC frontend with a network backend, either
use -nic ...
(e.g. for on-board NICs) or use -netdev ...,id=n
together
with -device ...,netdev=n
(for full control over pluggable NICs). To
connect multiple NICs or network backends via a hub device (which is what
vlan did), use -nic hubport,hubid=x,...
or
-netdev hubport,id=n,hubid=x,...
(with -device ...,netdev=n
) instead.
Use -machine kernel_irqchip=off
instead.
Use -global kvm-pit.lost_tick_policy=discard
instead.
The -balloon virtio
option has been replaced by -device virtio-balloon
.
The -balloon none
option was a no-op and has no replacement.
The -bootp /some/file
argument is replaced by either
-netdev user,id=x,bootp=/some/file
(for pluggable NICs, accompanied with
-device ...,netdev=x
), or -nic user,bootp=/some/file
(for on-board NICs).
The new syntax allows different settings to be provided per NIC.
The -redir [tcp|udp]:hostport:[guestaddr]:guestport
option is replaced
by either -netdev
user,id=x,hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport
(for pluggable NICs, accompanied with -device ...,netdev=x
) or by the option
-nic user,hostfwd=[tcp|udp]:[hostaddr]:hostport-[guestaddr]:guestport
(for on-board NICs). The new syntax allows different settings to be provided
per NIC.
The -smb /some/dir
argument is replaced by either
-netdev user,id=x,smb=/some/dir
(for pluggable NICs, accompanied with
-device ...,netdev=x
), or -nic user,smb=/some/dir
(for on-board NICs).
The new syntax allows different settings to be provided per NIC.
The -tftp /some/dir
argument is replaced by either
-netdev user,id=x,tftp=/some/dir
(for pluggable NICs, accompanied with
-device ...,netdev=x
), or -nic user,tftp=/some/dir
(for embedded NICs).
The new syntax allows different settings to be provided per NIC.
Replaced by -rtc base=localtime
.
Use -no-user-config
instead.
Use -rtc driftfix=slew
instead.
Replaced by -rtc base=date
.
The "tls-creds" option should be used instead to point to a "tls-creds-x509" object created using "-object".
If guest RAM allocation from file pointed by mem-path
failed,
QEMU was falling back to allocating from RAM, which might have resulted
in unpredictable behavior since the backing file specified by the user
as ignored. Currently, users are responsible for making sure the backing storage
specified with -mem-path
can actually provide the guest RAM configured with
-m
and QEMU fails to start up if RAM allocation is unsuccessful.
The name
parameter of the -net
option was a synonym
for the id
parameter, which should now be used instead.
The parameter mem
of -numa node
was used to assign a part of guest RAM
to a NUMA node. But when using it, it's impossible to manage a specified RAM
chunk on the host side (like bind it to a host node, setting bind policy, ...),
so the guest ends up with the fake NUMA configuration with suboptiomal
performance.
However since 2014 there is an alternative way to assign RAM to a NUMA node
using parameter memdev
, which does the same as mem
and adds
means to actually manage node RAM on the host side. Use parameter memdev
with memory-backend-ram backend as replacement for parameter mem
to achieve the same fake NUMA effect or a properly configured
memory-backend-file backend to actually benefit from NUMA configuration.
New machine versions (since 5.1) will not accept the option but it will still
work with old machine types. User can check the QAPI schema to see if the legacy
option is supported by looking at MachineInfo::numa-mem-supported property.
Splitting RAM by default between NUMA nodes had the same issues as mem
parameter with the difference that the role of the user plays QEMU using
implicit generic or board specific splitting rule.
Use memdev
with memory-backend-ram backend or mem
(if
it's supported by used machine type) to define mapping explicitly instead.
Users of existing VMs, wishing to preserve the same RAM distribution, should
configure it explicitly using -numa node,memdev
options. Current RAM
distribution can be retrieved using HMP command info numa
and if separate
memory devices (pc|nv-dimm) are present use info memory-device
and subtract
device memory from output of info numa
.
CPU topology properties should describe whole machine topology including possible CPUs.
However, historically it was possible to start QEMU with an incorrect topology where n <= sockets * cores * threads < maxcpus, which could lead to an incorrect topology enumeration by the guest. Support for invalid topologies is removed, the user must ensure topologies described with -smp include all possible cpus, i.e. sockets * cores * threads = maxcpus.
The enforce-config-section
property was replaced by the
-global migration.send-configuration={on|off}
option.
The -no-kvm
argument was a synonym for setting -machine accel=tcg
.
The -realtime mlock=on|off
argument has been replaced by the
-overcommit mem-lock=on|off
argument.
Use -display sdl,show-cursor=on
, -display gtk,show-cursor=on
or -display default,show-cursor=on
instead.
QEMU 5.0 introduced an alternative syntax to specify the size of the translation
block cache, -accel tcg,tb-size=
.
This option lacked the possibility to specify an audio backend device.
Use -device usb-audio
now instead (and specify a corresponding USB
host controller or -usb
if necessary).
The acl
option to the -vnc
argument has been replaced
by the tls-authz
and sasl-authz
options.
The pretty=on|off
switch has no effect for HMP monitors and
its use is rejected.
The 'file' driver for drives is no longer appropriate for character or host devices and will only accept regular files (S_IFREG). The correct driver for these file types is 'host_cdrom' or 'host_device' as appropriate.
Use -device floppy,...
instead. When configuring onboard floppy
controllers
-global isa-fdc.driveA=... -global sysbus-fdc.driveA=... -global SUNW,fdtwo.drive=...
become
-device floppy,unit=0,drive=...
and
-global isa-fdc.driveB=... -global sysbus-fdc.driveB=...
become
-device floppy,unit=1,drive=...
When plugging in a floppy controller
-device isa-fdc,...,driveA=...
becomes
-device isa-fdc,... -device floppy,unit=0,drive=...
and
-device isa-fdc,...,driveB=...
becomes
-device isa-fdc,... -device floppy,unit=1,drive=...
Drives with interface types other than if=none
are for onboard
devices. Drives the board doesn't pick up can no longer be used with
-device. Use if=none
instead.
This option was undocumented and not used in the field.
Use -device usb-ccid
instead.
QEMU 5.1 changes the default behaviour from -bios none
to -bios default
for the RISC-V virt
machine and sifive_u
machine.
The -no-quit
was a synonym for -display ...,window-close=off
which
should be used instead.
This option restricted usage of certain cryptographic algorithms when the host is operating in FIPS mode.
If FIPS compliance is required, QEMU should be built with the libgcrypt
or gnutls
library enabled as a cryptography provider.
Neither the nettle
library, or the built-in cryptography provider are
supported on FIPS enabled hosts.
The -writeconfig
option was not able to serialize the entire contents
of the QEMU command line. It is thus considered a failed experiment
and removed without a replacement.
The loaded=on
option in the command line or QMP object-add
either had
no effect (if loaded
was the last option) or caused options to be
effectively ignored as if they were not given. The property is therefore
useless and should simply be removed.
The opened=on
option in the command line or QMP object-add
either had
no effect (if opened
was the last option) or caused errors. The property
is therefore useless and should simply be removed.
Use -display sdl,window-close=...
instead (i.e. with a minus instead of
an underscore between "window" and "close").
Use -display sdl,grab-mod=lshift-lctrl-lalt
instead.
Use -display sdl,grab-mod=rctrl
instead.
Use -display sdl
instead.
Use -display curses
instead.
Sound card devices should be created using -device
or -audio
.
The exception is pcspk
which can be activated using -machine
pcspk-audiodev=<name>
.
Use -device
instead.
Input parameters that take a size value should only use a size suffix (such as 'k' or 'M') when the base is written in decimal, and not when the value is hexadecimal. That is, '0x20M' should be written either as '32M' or as '0x2000000'.
tty
and parport
used to be aliases for serial
and parallel
respectively. The actual backend names should be used instead.
Use -drive if=pflash
to configure the OTP device of the sifive_u
RISC-V machine instead.
This option was insecure because the SPICE password remained visible in
the process listing. This was replaced by the new password-secret
option which lets the password be securely provided on the command
line using a secret
object instance.
The -audiodev
and -audio
command line options are now the only
way to specify audio backend settings.
If no audiodev property is specified, previous versions would use the
first -audiodev
command line option as a fallback. Starting with
version 8.2, audio backends created with -audiodev
will only be
used by clients (sound cards, machines with embedded sound hardware, VNC)
that refer to it in an audiodev=
property.
In order to configure a default audio backend, use the -audio
command line option without specifying a model
; while previous
versions of QEMU required a model, starting with version 8.2
QEMU does not require a model and will not create any sound card
in this case.
Note that the default audio backend must be configured on the command
line if the -nodefaults
options is used.
The "autoload" parameter has been ignored since 2.12.0. All bitmaps are automatically loaded from qcow2 images.
Use device_add
for hotplugging vCPUs instead of cpu-add
. See
documentation of query-hotpluggable-cpus
for additional details.
Use blockdev-change-medium
or change-vnc-password
or
display-update
instead.
The query-events
command has been superseded by the more powerful
and accurate query-qmp-schema
command.
Use migrate_set_parameter
and info migrate_parameters
instead.
Use migrate_set_parameter
instead.
The query-cpus
command is replaced by the query-cpus-fast
command.
The arch
output member of the query-cpus-fast
command is
replaced by the target
output member.
Character devices creating sockets in client mode should not specify the 'wait' field, which is only applicable to sockets in server mode
Removed with no replacement.
Removed with no replacement.
The status
field of the BlockDirtyInfo
structure, returned by
these commands is removed. Two new boolean fields, recording
and
busy
effectively replace it.
The dirty-bitmaps
field of the BlockInfo
structure, returned by
the query-block command is itself now removed. The dirty-bitmaps
field of the BlockDeviceInfo
struct should be used instead, which is the
type of the inserted
field in query-block replies, as well as the
type of array items in query-named-block-nodes.
Specify the properties for the object as top-level arguments instead.
Member section-size
in the return value of query-sgx
was superseded by sections
.
Member section-size
in the return value of query-sgx-capabilities
was superseded by sections
.
Replaced by device_add
and device_del
(use device_add help
for a
list of available devices).
Replaced by netdev_add
and netdev_del
.
The [hub_id name]
parameter tuple of the 'hostfwd_add' and
'hostfwd_remove' HMP commands has been replaced by netdev_id
.
Use device_add
for hotplugging vCPUs instead of cpu-add
. See
documentation of query-hotpluggable-cpus
for additional details.
No replacement. The change vnc password
and change DEVICE MEDIUM
commands are not affected.
The acl_show
, acl_reset
, acl_policy
, acl_add
, and
acl_remove
commands were removed with no replacement. Authorization
for VNC should be performed using the pluggable QAuthZ objects.
Use migrate-set-parameters
and info migrate-parameters
instead.
Use migrate-set-parameters
instead.
This command didn't produce any output already. Removed with no replacement.
The RISC-V ISA privilege specification version 1.09.1 has been removed. QEMU supports both the newer version 1.10.0 and the ratified version 1.11.0, these should be used instead of the 1.09.1 version.
The Linux kernel has dropped support for allowing 32-bit Arm systems to host KVM guests as of the 5.7 kernel, and was thus removed from QEMU as well. Running 32-bit guests on a 64-bit Arm host remains supported.
The RISC-V cpus with the ISA version in the CPU name have been removed. The
four CPUs are: rv32gcsu-v1.9.1
, rv32gcsu-v1.10.0
, rv64gcsu-v1.9.1
and
rv64gcsu-v1.10.0
. Instead the version can be specified via the CPU priv_spec
option when using the rv32
or rv64
CPUs.
The RISC-V no MMU cpus have been removed. The two CPUs: rv32imacu-nommu
and
rv64imacu-nommu
can no longer be used. Instead the MMU status can be specified
via the CPU mmu
option when using the rv32
or rv64
CPUs.
The max-cpu-compat
property of the pseries
machine type should be used
instead.
Nobody was using this CPU emulation in QEMU, and there were no test images available to make sure that the code is still working, so it has been removed without replacement.
The only public user of this architecture was the milkymist project, which has been dead for years; there was never an upstream Linux port. Removed without replacement.
Support for this CPU was removed from the upstream Linux kernel, and there is no available upstream toolchain to build binaries for it. Removed without replacement.
There isn't ever Icelake Client CPU, it is some wrong and imaginary one.
Use Icelake-Server
instead.
-M kernel-irqchip=off
cannot be used on KVM if the CPU model includes
a local APIC. The split
setting is supported, as is using -M
kernel-irqchip=off
when the CPU does not have a local APIC.
The HAXM project has been retired (see https://github.com/intel/haxm#status). Use "whpx" (on Windows) or "hvf" (on macOS) instead.
The MIPS "Trap-and-Emulate" KVM host and guest support was removed from Linux in 2021, and is not supported anymore by QEMU either.
Use the s390-ccw-virtio
machine instead.
Use the none
machine with the loader
device instead.
The EP108 was an early access development board that is no longer used.
Use the xlnx-zcu102
machine instead.
The version specific Spike machines have been removed in favour of the
generic spike
machine. If you need to specify an older version of the RISC-V
spec you can use the -cpu rv64gcsu,priv_spec=v1.10.0
command line argument.
This machine type was very old and unmaintained. Users should use the malta
machine type instead.
This machine has been renamed fuloong2e
.
These machine types were very old and likely could not be used for live migration from old QEMU versions anymore. Use a newer machine type instead.
The Raspberry Pi machines come in various models (A, A+, B, B+). To be able
to distinguish which model QEMU is implementing, the raspi2
and raspi3
machines have been renamed raspi2b
and raspi3b
.
This machine was removed because it was unused. Alternative AST2500 based
OpenPOWER machines are witherspoon-bmc
and romulus-bmc
.
This machine was removed because it was partially emulated and 405
machines are very similar. Use the ref405ep
machine instead.
The tilegx
guest CPU support has been removed without replacement. It was
only implemented in linux-user mode, but support for this CPU was removed from
the upstream Linux kernel in 2018, and it has also been dropped from glibc, so
there is no new Linux development taking place with this architecture. For
running the old binaries, you can use older versions of QEMU.
The ppc64abi32
architecture has a number of issues which regularly
tripped up the CI testing and was suspected to be quite broken. For that
reason the maintainers strongly suspected no one actually used it.
The ability to add new TCG trace points had bit rotted and as the feature can be replicated with TCG plugins it was removed. If any user is currently using this feature and needs help with converting to using TCG plugins they should contact the qemu-devel mailing list.
The spapr-pci-vfio-host-bridge
device type has been replaced by the
spapr-pci-host-bridge
device type.
Replaced by either the ivshmem-plain
or ivshmem-doorbell
.
The 'ide-drive' device has been removed. Users should use 'ide-hd' or 'ide-cd' as appropriate to get an IDE hard disk or CD-ROM as needed.
The 'scsi-disk' device has been removed. Users should use 'scsi-hd' or 'scsi-cd' as appropriate to get a SCSI hard disk or CD-ROM as needed.
The sga
device loaded an option ROM for x86 targets which enabled
SeaBIOS to send messages to the serial console. SeaBIOS 1.11.0 onwards
contains native support for this feature and thus use of the option
ROM approach was obsolete. The native SeaBIOS support can be activated
by using -machine graphics=off
.
The qemu-nbd --partition $digit
code (also spelled -P
)
could only handle MBR partitions, and never correctly handled logical
partitions beyond partition 5. Exporting a partition can still be
done by utilizing the --image-opts
option with a raw blockdev
using the offset
and size
parameters layered on top of
any other existing blockdev. For example, if partition 1 is 100MiB
long starting at 1MiB, the old command:
qemu-nbd -t -P 1 -f qcow2 file.qcow2
can be rewritten as:
qemu-nbd -t --image-opts driver=raw,offset=1M,size=100M,file.driver=qcow2,file.file.driver=file,file.file.filename=file.qcow2
All options specified in -o
are image creation options, so
they are now rejected when used with -n
to skip image creation.
When creating an image with a backing file that could not be opened,
qemu-img create
used to issue a warning about the failure but
proceed with the image creation if an explicit size was provided.
However, as the -u
option exists for this purpose, it is safer to
enforce that any failure to open the backing image (including if the
backing file is missing or an incorrect format was specified) is an
error when -u
is not used.
The use of qemu-img amend
to modify the name or format of a qcow2
backing image was never fully documented or tested, and interferes
with other amend operations that need access to the original backing
image (such as deciding whether a v3 zero cluster may be left
unallocated when converting to a v2 image). Any changes to the
backing chain should be performed with qemu-img rebase -u
either
before or after the remaining changes being performed by amend, as
appropriate.
The use of qemu-img create
, qemu-img rebase
, or qemu-img
convert
to create or modify an image that depends on a backing file
now requires that an explicit backing format be provided. This is
for safety: if QEMU probes a different format than what you thought,
the data presented to the guest will be corrupt; similarly, presenting
a raw image to a guest allows a potential security exploit if a future
probe sees a non-raw image based on guest writes.
To avoid creating unsafe backing chains, you must pass -o
backing_fmt=
(or the shorthand -F
during create) to specify the
intended backing format. You may use qemu-img rebase -u
to
retroactively add a backing format to an existing image. However, be
aware that there are already potential security risks to blindly using
qemu-img info
to probe the format of an untrusted backing image,
when deciding what format to add into an existing image.
The VXHS code did not compile since v2.12.0. It was removed in 5.1.
The corresponding upstream server project is no longer maintained. Users are recommended to switch to an alternative distributed block device driver such as RBD.
There is a newer Rust implementation of virtiofsd
at
https://gitlab.com/virtio-fs/virtiofsd
; this has been
stable for some time and is now widely used.
The command line and feature set is very close to the removed
C implementation.