X-Git-Url: https://git.proxmox.com/?a=blobdiff_plain;f=PVE%2FQemuServer.pm;h=cfac03a92858236763b8f9637efd04af2f94f8b8;hb=5cfa9f5f7374c95c077e6f45b3943ca6388a7c53;hp=6461da35cb3d050d99585c7bf3089f1e709cd9e1;hpb=a2ec5a670596b191df82fbd562431222e4ec83a9;p=qemu-server.git diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm index 6461da3..cfac03a 100644 --- a/PVE/QemuServer.pm +++ b/PVE/QemuServer.pm @@ -31,7 +31,7 @@ use PVE::DataCenterConfig; use PVE::Exception qw(raise raise_param_exc); use PVE::GuestHelpers qw(safe_string_ne safe_num_ne safe_boolean_ne); use PVE::INotify; -use PVE::JSONSchema qw(get_standard_option); +use PVE::JSONSchema qw(get_standard_option parse_property_string); use PVE::ProcFSTools; use PVE::RPCEnvironment; use PVE::Storage; @@ -48,7 +48,7 @@ use PVE::QemuServer::Drive qw(is_valid_drivename drive_is_cloudinit drive_is_cdr use PVE::QemuServer::Machine; use PVE::QemuServer::Memory; use PVE::QemuServer::Monitor qw(mon_cmd); -use PVE::QemuServer::PCI qw(print_pci_addr print_pcie_addr print_pcie_root_port); +use PVE::QemuServer::PCI qw(print_pci_addr print_pcie_addr print_pcie_root_port parse_hostpci); use PVE::QemuServer::USB qw(parse_usb_device); my $have_sdn; @@ -441,6 +441,13 @@ EODESC description => "Enable/disable hugepages memory.", enum => [qw(any 2 1024)], }, + keephugepages => { + optional => 1, + type => 'boolean', + default => 0, + description => "Use together with hugepages. If enabled, hugepages will not not be deleted" + ." after VM shutdown and can be used for subsequent starts.", + }, vcpus => { optional => 1, type => 'integer', @@ -475,7 +482,8 @@ EODESC localtime => { optional => 1, type => 'boolean', - description => "Set the real time clock to local time. This is enabled by default if ostype indicates a Microsoft OS.", + description => "Set the real time clock to local time. This is enabled by default if ostype" + ." indicates a Microsoft OS.", }, freeze => { optional => 1, @@ -486,29 +494,28 @@ EODESC optional => 1, type => 'string', format => $vga_fmt, description => "Configure the VGA hardware.", - verbose_description => "Configure the VGA Hardware. If you want to use ". - "high resolution modes (>= 1280x1024x16) you may need to increase " . - "the vga memory option. Since QEMU 2.9 the default VGA display type " . - "is 'std' for all OS types besides some Windows versions (XP and " . - "older) which use 'cirrus'. The 'qxl' option enables the SPICE " . - "display server. For win* OS you can select how many independent " . - "displays you want, Linux guests can add displays them self.\n". - "You can also run without any graphic card, using a serial device as terminal.", + verbose_description => "Configure the VGA Hardware. If you want to use high resolution" + ." modes (>= 1280x1024x16) you may need to increase the vga memory option. Since QEMU" + ." 2.9 the default VGA display type is 'std' for all OS types besides some Windows" + ." versions (XP and older) which use 'cirrus'. The 'qxl' option enables the SPICE" + ." display server. For win* OS you can select how many independent displays you want," + ." Linux guests can add displays them self.\nYou can also run without any graphic card," + ." using a serial device as terminal.", }, watchdog => { optional => 1, type => 'string', format => 'pve-qm-watchdog', description => "Create a virtual hardware watchdog device.", - verbose_description => "Create a virtual hardware watchdog device. Once enabled" . - " (by a guest action), the watchdog must be periodically polled " . - "by an agent inside the guest or else the watchdog will reset " . - "the guest (or execute the respective action specified)", + verbose_description => "Create a virtual hardware watchdog device. Once enabled (by a guest" + ." action), the watchdog must be periodically polled by an agent inside the guest or" + ." else the watchdog will reset the guest (or execute the respective action specified)", }, startdate => { optional => 1, type => 'string', typetext => "(now | YYYY-MM-DD | YYYY-MM-DDTHH:MM:SS)", - description => "Set the initial date of the real time clock. Valid format for date are: 'now' or '2006-06-17T16:01:21' or '2006-06-17'.", + description => "Set the initial date of the real time clock. Valid format for date are:" + ."'now' or '2006-06-17T16:01:21' or '2006-06-17'.", pattern => '(now|\d{4}-\d{1,2}-\d{1,2}(T\d{1,2}:\d{1,2}:\d{1,2})?)', default => 'now', }, @@ -536,12 +543,11 @@ EODESCR type => 'boolean', default => 1, description => "Enable/disable the USB tablet device.", - verbose_description => "Enable/disable the USB tablet device. This device is " . - "usually needed to allow absolute mouse positioning with VNC. " . - "Else the mouse runs out of sync with normal VNC clients. " . - "If you're running lots of console-only guests on one host, " . - "you may consider disabling this to save some context switches. " . - "This is turned off by default if you use spice (-vga=qxl).", + verbose_description => "Enable/disable the USB tablet device. This device is usually needed" + ." to allow absolute mouse positioning with VNC. Else the mouse runs out of sync with" + ." normal VNC clients. If you're running lots of console-only guests on one host, you" + ." may consider disabling this to save some context switches. This is turned off by" + ." default if you use spice (`qm set --vga qxl`).", }, migrate_speed => { optional => 1, @@ -582,17 +588,20 @@ EODESCR vmstate => { optional => 1, type => 'string', format => 'pve-volume-id', - description => "Reference to a volume which stores the VM state. This is used internally for snapshots.", + description => "Reference to a volume which stores the VM state. This is used internally" + ." for snapshots.", }, vmstatestorage => get_standard_option('pve-storage-id', { description => "Default storage for VM state volumes/files.", optional => 1, }), runningmachine => get_standard_option('pve-qemu-machine', { - description => "Specifies the QEMU machine type of the running vm. This is used internally for snapshots.", + description => "Specifies the QEMU machine type of the running vm. This is used internally" + ." for snapshots.", }), runningcpu => { - description => "Specifies the QEMU '-cpu' parameter of the running vm. This is used internally for snapshots.", + description => "Specifies the QEMU '-cpu' parameter of the running vm. This is used" + ." internally for snapshots.", optional => 1, type => 'string', pattern => $PVE::QemuServer::CPUConfig::qemu_cmdline_cpu_re, @@ -614,7 +623,8 @@ EODESCR protection => { optional => 1, type => 'boolean', - description => "Sets the protection flag of the VM. This will disable the remove VM and remove disk operations.", + description => "Sets the protection flag of the VM. This will disable the remove VM and" + ." remove disk operations.", default => 0, }, bios => { @@ -628,17 +638,16 @@ EODESCR type => 'string', pattern => '(?:[a-fA-F0-9]{8}(?:-[a-fA-F0-9]{4}){3}-[a-fA-F0-9]{12}|[01])', format_description => 'UUID', - description => "Set VM Generation ID. Use '1' to autogenerate on create or update, pass '0' to disable explicitly.", - verbose_description => "The VM generation ID (vmgenid) device exposes a". - " 128-bit integer value identifier to the guest OS. This allows to". - " notify the guest operating system when the virtual machine is". - " executed with a different configuration (e.g. snapshot execution". - " or creation from a template). The guest operating system notices". - " the change, and is then able to react as appropriate by marking". - " its copies of distributed databases as dirty, re-initializing its". - " random number generator, etc.\n". - "Note that auto-creation only works when done throug API/CLI create". - " or update methods, but not when manually editing the config file.", + description => "Set VM Generation ID. Use '1' to autogenerate on create or update, pass '0'" + ." to disable explicitly.", + verbose_description => "The VM generation ID (vmgenid) device exposes a 128-bit integer" + ." value identifier to the guest OS. This allows to notify the guest operating system" + ." when the virtual machine is executed with a different configuration (e.g. snapshot" + ." execution or creation from a template). The guest operating system notices the" + ." change, and is then able to react as appropriate by marking its copies of" + ." distributed databases as dirty, re-initializing its random number generator, etc.\n" + ."Note that auto-creation only works when done through API/CLI create or update methods" + .", but not when manually editing the config file.", default => "1 (autogenerated)", optional => 1, }, @@ -651,7 +660,8 @@ EODESCR ivshmem => { type => 'string', format => $ivshmem_fmt, - description => "Inter-VM shared memory. Useful for direct communication between VMs, or to the host.", + description => "Inter-VM shared memory. Useful for direct communication between VMs, or to" + ." the host.", optional => 1, }, audio0 => { @@ -683,21 +693,24 @@ my $cicustom_fmt = { meta => { type => 'string', optional => 1, - description => 'Specify a custom file containing all meta data passed to the VM via cloud-init. This is provider specific meaning configdrive2 and nocloud differ.', + description => 'Specify a custom file containing all meta data passed to the VM via" + ." cloud-init. This is provider specific meaning configdrive2 and nocloud differ.', format => 'pve-volume-id', format_description => 'volume', }, network => { type => 'string', optional => 1, - description => 'Specify a custom file containing all network data passed to the VM via cloud-init.', + description => 'Specify a custom file containing all network data passed to the VM via' + .' cloud-init.', format => 'pve-volume-id', format_description => 'volume', }, user => { type => 'string', optional => 1, - description => 'Specify a custom file containing all user data passed to the VM via cloud-init.', + description => 'Specify a custom file containing all user data passed to the VM via' + .' cloud-init.', format => 'pve-volume-id', format_description => 'volume', }, @@ -708,34 +721,44 @@ my $confdesc_cloudinit = { citype => { optional => 1, type => 'string', - description => 'Specifies the cloud-init configuration format. The default depends on the configured operating system type (`ostype`. We use the `nocloud` format for Linux, and `configdrive2` for windows.', + description => 'Specifies the cloud-init configuration format. The default depends on the' + .' configured operating system type (`ostype`. We use the `nocloud` format for Linux,' + .' and `configdrive2` for windows.', enum => ['configdrive2', 'nocloud'], }, ciuser => { optional => 1, type => 'string', - description => "cloud-init: User name to change ssh keys and password for instead of the image's configured default user.", + description => "cloud-init: User name to change ssh keys and password for instead of the" + ." image's configured default user.", }, cipassword => { optional => 1, type => 'string', - description => 'cloud-init: Password to assign the user. Using this is generally not recommended. Use ssh keys instead. Also note that older cloud-init versions do not support hashed passwords.', + description => 'cloud-init: Password to assign the user. Using this is generally not' + .' recommended. Use ssh keys instead. Also note that older cloud-init versions do not' + .' support hashed passwords.', }, cicustom => { optional => 1, type => 'string', - description => 'cloud-init: Specify custom files to replace the automatically generated ones at start.', + description => 'cloud-init: Specify custom files to replace the automatically generated' + .' ones at start.', format => 'pve-qm-cicustom', }, searchdomain => { optional => 1, type => 'string', - description => "cloud-init: Sets DNS search domains for a container. Create will automatically use the setting from the host if neither searchdomain nor nameserver are set.", + description => "cloud-init: Sets DNS search domains for a container. Create will' + .' automatically use the setting from the host if neither searchdomain nor nameserver' + .' are set.", }, nameserver => { optional => 1, type => 'string', format => 'address-list', - description => "cloud-init: Sets DNS server IP address for a container. Create will automatically use the setting from the host if neither searchdomain nor nameserver are set.", + description => "cloud-init: Sets DNS server IP address for a container. Create will' + .' automatically use the setting from the host if neither searchdomain nor nameserver' + .' are set.", }, sshkeys => { optional => 1, @@ -768,7 +791,6 @@ while (my ($k, $v) = each %$confdesc) { my $MAX_USB_DEVICES = 5; my $MAX_NETS = 32; -my $MAX_HOSTPCI_DEVICES = 16; my $MAX_SERIAL_PORTS = 4; my $MAX_PARALLEL_PORTS = 3; my $MAX_NUMA = 8; @@ -833,11 +855,14 @@ __EOD__ my $net_fmt = { macaddr => get_standard_option('mac-addr', { - description => "MAC address. That address must be unique withing your network. This is automatically generated if not specified.", + description => "MAC address. That address must be unique withing your network. This is" + ." automatically generated if not specified.", }), model => { type => 'string', - description => "Network Card Model. The 'virtio' model provides the best performance with very low CPU overhead. If your guest does not support this driver, it is usually best to use 'e1000'.", + description => "Network Card Model. The 'virtio' model provides the best performance with" + ." very low CPU overhead. If your guest does not support this driver, it is usually" + ." best to use 'e1000'.", enum => $nic_model_list, default_key => 1, }, @@ -884,6 +909,12 @@ my $net_fmt = { description => 'Whether this interface should be disconnected (like pulling the plug).', optional => 1, }, + mtu => { + type => 'integer', + minimum => 1, maximum => 65520, + description => "Force MTU, for VirtIO only. Set to '1' to use the bridge MTU", + optional => 1, + }, }; my $netdesc = { @@ -937,10 +968,12 @@ cloud-init: Specify IP addresses and gateways for the corresponding interface. IP addresses use CIDR notation, gateways are optional but need an IP of the same type specified. -The special string 'dhcp' can be used for IP addresses to use DHCP, in which case no explicit gateway should be provided. +The special string 'dhcp' can be used for IP addresses to use DHCP, in which case no explicit +gateway should be provided. For IPv6 the special string 'auto' can be used to use stateless autoconfiguration. -If cloud-init is enabled and neither an IPv4 nor an IPv6 address is specified, it defaults to using dhcp on IPv4. +If cloud-init is enabled and neither an IPv4 nor an IPv6 address is specified, it defaults to using +dhcp on IPv4. EODESCR }; PVE::JSONSchema::register_standard_option("pve-qm-ipconfig", $netdesc); @@ -985,7 +1018,8 @@ The Host USB device or port or the value 'spice'. HOSTUSBDEVICE syntax is: You can use the 'lsusb -t' command to list existing usb devices. -NOTE: This option allows direct access to host hardware. So it is no longer possible to migrate such machines - use with special care. +NOTE: This option allows direct access to host hardware. So it is no longer possible to migrate such +machines - use with special care. The value 'spice' can be used to add a usb redirection devices for spice. EODESCR @@ -1005,76 +1039,6 @@ my $usbdesc = { }; PVE::JSONSchema::register_standard_option("pve-qm-usb", $usbdesc); -my $PCIRE = qr/([a-f0-9]{4}:)?[a-f0-9]{2}:[a-f0-9]{2}(?:\.[a-f0-9])?/; -my $hostpci_fmt = { - host => { - default_key => 1, - type => 'string', - pattern => qr/$PCIRE(;$PCIRE)*/, - format_description => 'HOSTPCIID[;HOSTPCIID2...]', - description => < { - type => 'boolean', - description => "Specify whether or not the device's ROM will be visible in the guest's memory map.", - optional => 1, - default => 1, - }, - romfile => { - type => 'string', - pattern => '[^,;]+', - format_description => 'string', - description => "Custom pci device rom filename (must be located in /usr/share/kvm/).", - optional => 1, - }, - pcie => { - type => 'boolean', - description => "Choose the PCI-express bus (needs the 'q35' machine model).", - optional => 1, - default => 0, - }, - 'x-vga' => { - type => 'boolean', - description => "Enable vfio-vga device support.", - optional => 1, - default => 0, - }, - 'mdev' => { - type => 'string', - format_description => 'string', - pattern => '[^/\.:]+', - optional => 1, - description => < 1, - type => 'string', format => 'pve-qm-hostpci', - description => "Map host PCI devices into guest.", - verbose_description => < 1, type => 'string', @@ -1085,7 +1049,8 @@ Create a serial device inside the VM (n is 0 to 3), and pass through a host serial device (i.e. /dev/ttyS0), or create a unix socket on the host side (use 'qm terminal' to open a terminal connection). -NOTE: If you pass through a host serial device, it is no longer possible to migrate such machines - use with special care. +NOTE: If you pass through a host serial device, it is no longer possible to migrate such machines - +use with special care. CAUTION: Experimental! User reported problems with this option. EODESCR @@ -1099,7 +1064,8 @@ my $paralleldesc= { verbose_description => <{"serial$i"} = $serialdesc; } -for (my $i = 0; $i < $MAX_HOSTPCI_DEVICES; $i++) { - $confdesc->{"hostpci$i"} = $hostpcidesc; +for (my $i = 0; $i < $PVE::QemuServer::PCI::MAX_HOSTPCI_DEVICES; $i++) { + $confdesc->{"hostpci$i"} = $PVE::QemuServer::PCI::hostpcidesc; } for my $key (keys %{$PVE::QemuServer::Drive::drivedesc_hash}) { @@ -1125,6 +1091,68 @@ for (my $i = 0; $i < $MAX_USB_DEVICES; $i++) { $confdesc->{"usb$i"} = $usbdesc; } +my $boot_fmt = { + legacy => { + optional => 1, + default_key => 1, + type => 'string', + description => "Boot on floppy (a), hard disk (c), CD-ROM (d), or network (n)." + . " Deprecated, use 'order=' instead.", + pattern => '[acdn]{1,4}', + format_description => "[acdn]{1,4}", + + # note: this is also the fallback if boot: is not given at all + default => 'cdn', + }, + order => { + optional => 1, + type => 'string', + format => 'pve-qm-bootdev-list', + format_description => "device[;device...]", + description => <{$dev}; + return 1; + }; + + return $dev if $check->("net"); + return $dev if $check->("usb"); + return $dev if $check->("hostpci"); + + return undef if $noerr; + die "invalid boot device '$dev'\n"; +} + +sub print_bootorder { + my ($devs) = @_; + my $data = { order => join(';', @$devs) }; + return PVE::JSONSchema::print_property_string($data, $boot_fmt); +} + my $kvm_api_version = 0; sub kvm_version { @@ -1168,6 +1196,11 @@ sub kvm_user_version { return $kvm_user_version->{$binary}; } +my sub extract_version { + my ($machine_type, $version) = @_; + $version = kvm_user_version() if !defined($version); + PVE::QemuServer::Machine::extract_version($machine_type, $version) +} sub kernel_has_vhost_net { return -c '/dev/vhost-net'; @@ -1250,7 +1283,8 @@ sub cleanup_drive_path { ($drive->{file} !~ m/^([^:]+):(.+)$/) && ($drive->{file} !~ m/^\d+$/)) { my ($vtype, $volid) = PVE::Storage::path_to_volume_id($storecfg, $drive->{file}); - raise_param_exc({ $opt => "unable to associate path '$drive->{file}' to any storage"}) if !$vtype; + raise_param_exc({ $opt => "unable to associate path '$drive->{file}' to any storage"}) + if !$vtype; $drive->{media} = 'cdrom' if !$drive->{media} && $vtype eq 'iso'; verify_media_type($opt, $vtype, $drive->{media}); $drive->{file} = $volid; @@ -1380,10 +1414,11 @@ sub print_drivedevice_full { my $device = ''; my $maxdev = 0; + my $drive_id = "$drive->{interface}$drive->{index}"; if ($drive->{interface} eq 'virtio') { - my $pciaddr = print_pci_addr("$drive->{interface}$drive->{index}", $bridges, $arch, $machine_type); - $device = "virtio-blk-pci,drive=drive-$drive->{interface}$drive->{index},id=$drive->{interface}$drive->{index}$pciaddr"; - $device .= ",iothread=iothread-$drive->{interface}$drive->{index}" if $drive->{iothread}; + my $pciaddr = print_pci_addr("$drive_id", $bridges, $arch, $machine_type); + $device = "virtio-blk-pci,drive=drive-$drive_id,id=${drive_id}${pciaddr}"; + $device .= ",iothread=iothread-$drive_id" if $drive->{iothread}; } elsif ($drive->{interface} eq 'scsi') { my ($maxdev, $controller, $controller_prefix) = scsihw_infos($conf, $drive); @@ -1407,7 +1442,7 @@ sub print_drivedevice_full { } # for compatibility only, we prefer scsi-hd (#2408, #2355, #2380) - my $version = PVE::QemuServer::Machine::extract_version($machine_type, kvm_user_version()); + my $version = extract_version($machine_type, kvm_user_version()); if ($path =~ m/^iscsi\:\/\// && !min_version($version, 4, 1)) { $devicetype = 'generic'; @@ -1415,10 +1450,12 @@ sub print_drivedevice_full { } if (!$conf->{scsihw} || ($conf->{scsihw} =~ m/^lsi/)){ - $device = "scsi-$devicetype,bus=$controller_prefix$controller.0,scsi-id=$unit,drive=drive-$drive->{interface}$drive->{index},id=$drive->{interface}$drive->{index}"; + $device = "scsi-$devicetype,bus=$controller_prefix$controller.0,scsi-id=$unit"; } else { - $device = "scsi-$devicetype,bus=$controller_prefix$controller.0,channel=0,scsi-id=0,lun=$drive->{index},drive=drive-$drive->{interface}$drive->{index},id=$drive->{interface}$drive->{index}"; + $device = "scsi-$devicetype,bus=$controller_prefix$controller.0,channel=0,scsi-id=0" + .",lun=$drive->{index}"; } + $device .= ",drive=drive-$drive_id,id=$drive_id"; if ($drive->{ssd} && ($devicetype eq 'block' || $devicetype eq 'hd')) { $device .= ",rotation_rate=1"; @@ -1437,7 +1474,7 @@ sub print_drivedevice_full { } else { $device .= ",bus=ahci$controller.$unit"; } - $device .= ",drive=drive-$drive->{interface}$drive->{index},id=$drive->{interface}$drive->{index}"; + $device .= ",drive=drive-$drive_id,id=$drive_id"; if ($devicetype eq 'hd') { if (my $model = $drive->{model}) { @@ -1587,12 +1624,29 @@ sub print_netdevice_full { my $pciaddr = print_pci_addr("$netid", $bridges, $arch, $machine_type); my $tmpstr = "$device,mac=$net->{macaddr},netdev=$netid$pciaddr,id=$netid"; if ($net->{queues} && $net->{queues} > 1 && $net->{model} eq 'virtio'){ - #Consider we have N queues, the number of vectors needed is 2*N + 2 (plus one config interrupt and control vq) + # Consider we have N queues, the number of vectors needed is 2 * N + 2, i.e., one per in + # and out of each queue plus one config interrupt and control vector queue my $vectors = $net->{queues} * 2 + 2; $tmpstr .= ",vectors=$vectors,mq=on"; } $tmpstr .= ",bootindex=$net->{bootindex}" if $net->{bootindex} ; + if (my $mtu = $net->{mtu}) { + if ($net->{model} eq 'virtio' && $net->{bridge}) { + my $bridge_mtu = PVE::Network::read_bridge_mtu($net->{bridge}); + if ($mtu == 1) { + $mtu = $bridge_mtu; + } elsif ($mtu < 576) { + die "netdev $netid: MTU '$mtu' is smaller than the IP minimum MTU '576'\n"; + } elsif ($mtu > $bridge_mtu) { + die "netdev $netid: MTU '$mtu' is bigger than the bridge MTU '$bridge_mtu'\n"; + } + $tmpstr .= ",host_mtu=$mtu"; + } else { + warn "WARN: netdev $netid: ignoring MTU '$mtu', not using VirtIO or no bridge configured.\n"; + } + } + if ($use_old_bios_files) { my $romfile; if ($device eq 'virtio-net-pci') { @@ -1639,7 +1693,8 @@ sub print_netdev_full { my $script = $hotplug ? "pve-bridge-hotplug" : "pve-bridge"; if ($net->{bridge}) { - $netdev = "type=tap,id=$netid,ifname=${ifname},script=/var/lib/qemu-server/$script,downscript=/var/lib/qemu-server/pve-bridgedown$vhostparam"; + $netdev = "type=tap,id=$netid,ifname=${ifname},script=/var/lib/qemu-server/$script" + .",downscript=/var/lib/qemu-server/pve-bridgedown$vhostparam"; } else { $netdev = "type=user,id=$netid,hostname=$vmname"; } @@ -1697,6 +1752,11 @@ sub print_vga_device { $memory = ",ram_size=67108864,vram_size=33554432"; } + my $edidoff = ""; + if ($type eq 'VGA' && windows_version($conf->{ostype})) { + $edidoff=",edid=off" if (!defined($conf->{bios}) || $conf->{bios} ne 'ovmf'); + } + my $q35 = PVE::QemuServer::Machine::machine_type_is_q35($conf); my $vgaid = "vga" . ($id // ''); my $pciaddr; @@ -1708,7 +1768,7 @@ sub print_vga_device { $pciaddr = print_pci_addr($vgaid, $bridges, $arch, $machine); } - return "$type,id=${vgaid}${memory}${max_outputs}${pciaddr}"; + return "$type,id=${vgaid}${memory}${max_outputs}${pciaddr}${edidoff}"; } sub parse_number_sets { @@ -1728,34 +1788,17 @@ sub parse_number_sets { sub parse_numa { my ($data) = @_; - my $res = PVE::JSONSchema::parse_property_string($numa_fmt, $data); + my $res = parse_property_string($numa_fmt, $data); $res->{cpus} = parse_number_sets($res->{cpus}) if defined($res->{cpus}); $res->{hostnodes} = parse_number_sets($res->{hostnodes}) if defined($res->{hostnodes}); return $res; } -sub parse_hostpci { - my ($value) = @_; - - return undef if !$value; - - my $res = PVE::JSONSchema::parse_property_string($hostpci_fmt, $value); - - my @idlist = split(/;/, $res->{host}); - delete $res->{host}; - foreach my $id (@idlist) { - my $devs = PVE::SysFSTools::lspci($id); - die "no PCI device found for '$id'\n" if !scalar(@$devs); - push @{$res->{pciid}}, @$devs; - } - return $res; -} - # netX: e1000=XX:XX:XX:XX:XX:XX,bridge=vmbr0,rate= sub parse_net { my ($data) = @_; - my $res = eval { PVE::JSONSchema::parse_property_string($net_fmt, $data) }; + my $res = eval { parse_property_string($net_fmt, $data) }; if ($@) { warn $@; return undef; @@ -1771,7 +1814,7 @@ sub parse_net { sub parse_ipconfig { my ($data) = @_; - my $res = eval { PVE::JSONSchema::parse_property_string($ipconfig_fmt, $data) }; + my $res = eval { parse_property_string($ipconfig_fmt, $data) }; if ($@) { warn $@; return undef; @@ -1908,7 +1951,7 @@ my $smbios1_fmt = { sub parse_smbios1 { my ($data) = @_; - my $res = eval { PVE::JSONSchema::parse_property_string($smbios1_fmt, $data) }; + my $res = eval { parse_property_string($smbios1_fmt, $data) }; warn $@ if $@; return $res; } @@ -1925,7 +1968,7 @@ sub parse_watchdog { return undef if !$value; - my $res = eval { PVE::JSONSchema::parse_property_string($watchdog_fmt, $value) }; + my $res = eval { parse_property_string($watchdog_fmt, $value) }; warn $@ if $@; return $res; } @@ -1935,7 +1978,7 @@ sub parse_guest_agent { return {} if !defined($value->{agent}); - my $res = eval { PVE::JSONSchema::parse_property_string($agent_fmt, $value->{agent}) }; + my $res = eval { parse_property_string($agent_fmt, $value->{agent}) }; warn $@ if $@; # if the agent is disabled ignore the other potentially set properties @@ -1947,7 +1990,7 @@ sub parse_vga { my ($value) = @_; return {} if !$value; - my $res = eval { PVE::JSONSchema::parse_property_string($vga_fmt, $value) }; + my $res = eval { parse_property_string($vga_fmt, $value) }; warn $@ if $@; return $res; } @@ -1957,7 +2000,7 @@ sub parse_rng { return undef if !$value; - my $res = eval { PVE::JSONSchema::parse_property_string($rng_fmt, $value) }; + my $res = eval { parse_property_string($rng_fmt, $value) }; warn $@ if $@; return $res; } @@ -2263,7 +2306,7 @@ sub write_vm_config { } foreach my $key (sort keys %$conf) { - next if $key eq 'digest' || $key eq 'description' || $key eq 'pending' || $key eq 'snapshots'; + next if $key =~ /^(digest|description|pending|snapshots)$/; $raw .= "$key: $conf->{$key}\n"; } return $raw; @@ -2740,7 +2783,7 @@ sub conf_has_audio { my $audio = $conf->{"audio$id"}; return undef if !defined($audio); - my $audioproperties = PVE::JSONSchema::parse_property_string($audio_fmt, $audio); + my $audioproperties = parse_property_string($audio_fmt, $audio); my $audiodriver = $audioproperties->{driver} // 'spice'; return { @@ -2994,7 +3037,7 @@ sub config_to_command { my $add_pve_version = min_version($kvmver, 4, 1); my $machine_type = get_vm_machine($conf, $forcemachine, $arch, $add_pve_version); - my $machine_version = PVE::QemuServer::Machine::extract_version($machine_type, $kvmver); + my $machine_version = extract_version($machine_type, $kvmver); $kvm //= 1 if is_native($arch); $machine_version =~ m/(\d+)\.(\d+)/; @@ -3003,11 +3046,13 @@ sub config_to_command { if ($kvmver =~ m/^\d+\.\d+\.(\d+)/ && $1 >= 90) { warn "warning: Installed QEMU version ($kvmver) is a release candidate, ignoring version checks\n"; } elsif (!min_version($kvmver, $machine_major, $machine_minor)) { - die "Installed QEMU version '$kvmver' is too old to run machine type '$machine_type', please upgrade node '$nodename'\n" + die "Installed QEMU version '$kvmver' is too old to run machine type '$machine_type'," + ." please upgrade node '$nodename'\n" } elsif (!PVE::QemuServer::Machine::can_run_pve_machine_version($machine_version, $kvmver)) { my $max_pve_version = PVE::QemuServer::Machine::get_pve_version($machine_version); - die "Installed qemu-server (max feature level for $machine_major.$machine_minor is pve$max_pve_version)" - ." is too old to run machine type '$machine_type', please upgrade node '$nodename'\n"; + die "Installed qemu-server (max feature level for $machine_major.$machine_minor is" + ." pve$max_pve_version) is too old to run machine type '$machine_type', please upgrade" + ." node '$nodename'\n"; } # if a specific +pve version is required for a feature, use $version_guard @@ -3023,9 +3068,9 @@ sub config_to_command { return 1; }; - if ($kvm) { - die "KVM virtualisation configured, but not available. Either disable in VM configuration or enable in BIOS.\n" - if !defined kvm_version(); + if ($kvm && !defined kvm_version()) { + die "KVM virtualisation configured, but not available. Either disable in VM configuration" + ." or enable in BIOS.\n"; } my $q35 = PVE::QemuServer::Machine::machine_type_is_q35($conf); @@ -3134,7 +3179,8 @@ sub config_to_command { } # add usb controllers - my @usbcontrollers = PVE::QemuServer::USB::get_usb_controllers($conf, $bridges, $arch, $machine_type, $usbdesc->{format}, $MAX_USB_DEVICES); + my @usbcontrollers = PVE::QemuServer::USB::get_usb_controllers( + $conf, $bridges, $arch, $machine_type, $usbdesc->{format}, $MAX_USB_DEVICES); push @$devices, @usbcontrollers if @usbcontrollers; my $vga = parse_vga($conf->{vga}); @@ -3167,83 +3213,16 @@ sub config_to_command { push @$devices, '-device', $kbd if defined($kbd); } - my $kvm_off = 0; - my $gpu_passthrough; - - # host pci devices - for (my $i = 0; $i < $MAX_HOSTPCI_DEVICES; $i++) { - my $id = "hostpci$i"; - my $d = parse_hostpci($conf->{$id}); - next if !$d; - - if (my $pcie = $d->{pcie}) { - die "q35 machine model is not enabled" if !$q35; - # win7 wants to have the pcie devices directly on the pcie bus - # instead of in the root port - if ($winversion == 7) { - $pciaddr = print_pcie_addr("${id}bus0"); - } else { - # add more root ports if needed, 4 are present by default - # by pve-q35 cfgs, rest added here on demand. - if ($i > 3) { - push @$devices, '-device', print_pcie_root_port($i); - } - $pciaddr = print_pcie_addr($id); - } - } else { - $pciaddr = print_pci_addr($id, $bridges, $arch, $machine_type); - } - - my $xvga = ''; - if ($d->{'x-vga'}) { - $xvga = ',x-vga=on' if !($conf->{bios} && $conf->{bios} eq 'ovmf'); - $kvm_off = 1; - $vga->{type} = 'none' if !defined($conf->{vga}); - $gpu_passthrough = 1; - } - - my $pcidevices = $d->{pciid}; - my $multifunction = 1 if @$pcidevices > 1; - - my $sysfspath; - if ($d->{mdev} && scalar(@$pcidevices) == 1) { - my $pci_id = $pcidevices->[0]->{id}; - my $uuid = PVE::SysFSTools::generate_mdev_uuid($vmid, $i); - $sysfspath = "/sys/bus/pci/devices/$pci_id/$uuid"; - } elsif ($d->{mdev}) { - warn "ignoring mediated device '$id' with multifunction device\n"; - } - - my $j=0; - foreach my $pcidevice (@$pcidevices) { - my $devicestr = "vfio-pci"; - - if ($sysfspath) { - $devicestr .= ",sysfsdev=$sysfspath"; - } else { - $devicestr .= ",host=$pcidevice->{id}"; - } - - my $mf_addr = $multifunction ? ".$j" : ''; - $devicestr .= ",id=${id}${mf_addr}${pciaddr}${mf_addr}"; - - if ($j == 0) { - $devicestr .= ',rombar=0' if defined($d->{rombar}) && !$d->{rombar}; - $devicestr .= "$xvga"; - $devicestr .= ",multifunction=on" if $multifunction; - $devicestr .= ",romfile=/usr/share/kvm/$d->{romfile}" if $d->{romfile}; - } - - push @$devices, '-device', $devicestr; - $j++; - } - } + # host pci device passthrough + my ($kvm_off, $gpu_passthrough, $legacy_igd) = PVE::QemuServer::PCI::print_hostpci_devices( + $vmid, $conf, $devices, $winversion, $q35, $bridges, $arch, $machine_type); # usb devices my $usb_dev_features = {}; $usb_dev_features->{spice_usb3} = 1 if min_version($machine_version, 4, 0); - my @usbdevices = PVE::QemuServer::USB::get_usb_devices($conf, $usbdesc->{format}, $MAX_USB_DEVICES, $usb_dev_features); + my @usbdevices = PVE::QemuServer::USB::get_usb_devices( + $conf, $usbdesc->{format}, $MAX_USB_DEVICES, $usb_dev_features); push @$devices, @usbdevices if @usbdevices; # serial devices for (my $i = 0; $i < $MAX_SERIAL_PORTS; $i++) { @@ -3328,7 +3307,8 @@ sub config_to_command { push @$cmd, '-no-reboot' if defined($conf->{reboot}) && $conf->{reboot} == 0; if ($vga->{type} && $vga->{type} !~ m/^serial\d+$/ && $vga->{type} ne 'none'){ - push @$devices, '-device', print_vga_device($conf, $vga, $arch, $machine_version, $machine_type, undef, $qxlnum, $bridges); + push @$devices, '-device', print_vga_device( + $conf, $vga, $arch, $machine_version, $machine_type, undef, $qxlnum, $bridges); my $socket = PVE::QemuServer::Helpers::vnc_socket($vmid); push @$cmd, '-vnc', "unix:$socket,password"; } else { @@ -3411,7 +3391,8 @@ sub config_to_command { if ($qxlnum > 1) { if ($winversion){ for (my $i = 1; $i < $qxlnum; $i++){ - push @$devices, '-device', print_vga_device($conf, $vga, $arch, $machine_version, $machine_type, $i, $qxlnum, $bridges); + push @$devices, '-device', print_vga_device( + $conf, $vga, $arch, $machine_version, $machine_type, $i, $qxlnum, $bridges); } } else { # assume other OS works like Linux @@ -3438,14 +3419,17 @@ sub config_to_command { my $localhost = PVE::Network::addr_to_ip($nodeaddrs[0]->{addr}); $spice_port = PVE::Tools::next_spice_port($pfamily, $localhost); - my $spice_enhancement = PVE::JSONSchema::parse_property_string($spice_enhancements_fmt, $conf->{spice_enhancements} // ''); + my $spice_enhancement_str = $conf->{spice_enhancements} // ''; + my $spice_enhancement = parse_property_string($spice_enhancements_fmt, $spice_enhancement_str); if ($spice_enhancement->{foldersharing}) { push @$devices, '-chardev', "spiceport,id=foldershare,name=org.spice-space.webdav.0"; push @$devices, '-device', "virtserialport,chardev=foldershare,name=org.spice-space.webdav.0"; } my $spice_opts = "tls-port=${spice_port},addr=$localhost,tls-ciphers=HIGH,seamless-migration=on"; - $spice_opts .= ",streaming-video=$spice_enhancement->{videostreaming}" if $spice_enhancement->{videostreaming}; + $spice_opts .= ",streaming-video=$spice_enhancement->{videostreaming}" + if $spice_enhancement->{videostreaming}; + push @$devices, '-spice', "$spice_opts"; } @@ -3524,43 +3508,50 @@ sub config_to_command { $queues = ",num_queues=$drive->{queues}"; } - push @$devices, '-device', "$scsihw_type,id=$controller_prefix$controller$pciaddr$iothread$queues" if !$scsicontroller->{$controller}; + push @$devices, '-device', "$scsihw_type,id=$controller_prefix$controller$pciaddr$iothread$queues" + if !$scsicontroller->{$controller}; $scsicontroller->{$controller}=1; } if ($drive->{interface} eq 'sata') { my $controller = int($drive->{index} / $PVE::QemuServer::Drive::MAX_SATA_DISKS); $pciaddr = print_pci_addr("ahci$controller", $bridges, $arch, $machine_type); - push @$devices, '-device', "ahci,id=ahci$controller,multifunction=on$pciaddr" if !$ahcicontroller->{$controller}; + push @$devices, '-device', "ahci,id=ahci$controller,multifunction=on$pciaddr" + if !$ahcicontroller->{$controller}; $ahcicontroller->{$controller}=1; } my $drive_cmd = print_drive_commandline_full($storecfg, $vmid, $drive); + $drive_cmd .= ',readonly' if PVE::QemuConfig->is_template($conf); + push @$devices, '-drive',$drive_cmd; - push @$devices, '-device', print_drivedevice_full($storecfg, $conf, $vmid, $drive, $bridges, $arch, $machine_type); + push @$devices, '-device', print_drivedevice_full( + $storecfg, $conf, $vmid, $drive, $bridges, $arch, $machine_type); }); for (my $i = 0; $i < $MAX_NETS; $i++) { - next if !$conf->{"net$i"}; - my $d = parse_net($conf->{"net$i"}); - next if !$d; + next if !$conf->{"net$i"}; + my $d = parse_net($conf->{"net$i"}); + next if !$d; - $use_virtio = 1 if $d->{model} eq 'virtio'; + $use_virtio = 1 if $d->{model} eq 'virtio'; - if ($bootindex_hash->{n}) { + if ($bootindex_hash->{n}) { $d->{bootindex} = $bootindex_hash->{n}; $bootindex_hash->{n} += 1; - } + } + + my $netdevfull = print_netdev_full($vmid, $conf, $arch, $d, "net$i"); + push @$devices, '-netdev', $netdevfull; - my $netdevfull = print_netdev_full($vmid, $conf, $arch, $d, "net$i"); - push @$devices, '-netdev', $netdevfull; + my $netdevicefull = print_netdevice_full( + $vmid, $conf, $d, "net$i", $bridges, $use_old_bios_files, $arch, $machine_type); - my $netdevicefull = print_netdevice_full($vmid, $conf, $d, "net$i", $bridges, $use_old_bios_files, $arch, $machine_type); - push @$devices, '-device', $netdevicefull; + push @$devices, '-device', $netdevicefull; } if ($conf->{ivshmem}) { - my $ivshmem = PVE::JSONSchema::parse_property_string($ivshmem_fmt, $conf->{ivshmem}); + my $ivshmem = parse_property_string($ivshmem_fmt, $conf->{ivshmem}); my $bus; if ($q35) { @@ -3573,7 +3564,8 @@ sub config_to_command { my $path = '/dev/shm/pve-shm-' . $ivshmem_name; push @$devices, '-device', "ivshmem-plain,memdev=ivshmem$bus,"; - push @$devices, '-object', "memory-backend-file,id=ivshmem,share=on,mem-path=$path,size=$ivshmem->{size}M"; + push @$devices, '-object', "memory-backend-file,id=ivshmem,share=on,mem-path=$path" + .",size=$ivshmem->{size}M"; } # pci.4 is nested in pci.1 @@ -3592,7 +3584,13 @@ sub config_to_command { for my $k (sort {$b cmp $a} keys %$bridges) { next if $q35 && $k < 4; # q35.cfg already includes bridges up to 3 - $pciaddr = print_pci_addr("pci.$k", undef, $arch, $machine_type); + + my $k_name = $k; + if ($k == 2 && $legacy_igd) { + $k_name = "$k-igd"; + } + $pciaddr = print_pci_addr("pci.$k_name", undef, $arch, $machine_type); + my $devstr = "pci-bridge,id=pci.$k,chassis_nr=$k$pciaddr"; if ($q35) { # add after -readconfig pve-q35.cfg @@ -3643,12 +3641,11 @@ sub check_rng_source { my $rng_current = '/sys/devices/virtual/misc/hw_random/rng_current'; if ($source eq '/dev/hwrng' && file_read_firstline($rng_current) eq 'none') { - # Needs to abort, otherwise QEMU crashes on first rng access. - # Note that rng_current cannot be changed to 'none' manually, so - # once the VM is past this point, it is no longer an issue. - die "Cannot start VM with passed-through RNG device: '/dev/hwrng'" - . " exists, but '$rng_current' is set to 'none'. Ensure that" - . " a compatible hardware-RNG is attached to the host.\n"; + # Needs to abort, otherwise QEMU crashes on first rng access. Note that rng_current cannot + # be changed to 'none' manually, so once the VM is past this point, it's no longer an issue. + die "Cannot start VM with passed-through RNG device: '/dev/hwrng' exists, but" + ." '$rng_current' is set to 'none'. Ensure that a compatible hardware-RNG is attached" + ." to the host.\n"; } } @@ -3718,7 +3715,8 @@ sub vm_deviceplug { my $devices_list = vm_devices_list($vmid); return 1 if defined($devices_list->{$deviceid}); - qemu_add_pci_bridge($storecfg, $conf, $vmid, $deviceid, $arch, $machine_type); # add PCI bridge if we need it for the device + # add PCI bridge if we need it for the device + qemu_add_pci_bridge($storecfg, $conf, $vmid, $deviceid, $arch, $machine_type); if ($deviceid eq 'tablet') { @@ -3793,7 +3791,8 @@ sub vm_deviceplug { my $use_old_bios_files = undef; ($use_old_bios_files, $machine_type) = qemu_use_old_bios_files($machine_type); - my $netdevicefull = print_netdevice_full($vmid, $conf, $device, $deviceid, undef, $use_old_bios_files, $arch, $machine_type); + my $netdevicefull = print_netdevice_full( + $vmid, $conf, $device, $deviceid, undef, $use_old_bios_files, $arch, $machine_type); qemu_deviceadd($vmid, $netdevicefull); eval { qemu_deviceaddverify($vmid, $deviceid); @@ -4065,6 +4064,14 @@ sub qemu_netdevadd { my $netdev = print_netdev_full($vmid, $conf, $arch, $device, $deviceid, 1); my %options = split(/[=,]/, $netdev); + if (defined(my $vhost = $options{vhost})) { + $options{vhost} = JSON::boolean(PVE::JSONSchema::parse_boolean($vhost)); + } + + if (defined(my $queues = $options{queues})) { + $options{queues} = $queues + 0; + } + mon_cmd($vmid, "netdev_add", %options); return 1; } @@ -4515,7 +4522,7 @@ sub vmconfig_hotplug_pending { # since we cannot reliably hot unplug usb devices # we are disabling it die "skip\n" if !$hotplug_features->{usb} || $value =~ m/spice/i; - my $d = eval { PVE::JSONSchema::parse_property_string($usbdesc->{format}, $value) }; + my $d = eval { parse_property_string($usbdesc->{format}, $value) }; die "skip\n" if !$d; qemu_usb_hotplug($storecfg, $conf, $vmid, $opt, $d, $arch, $machine_type); } elsif ($opt eq 'vcpus') { @@ -4717,102 +4724,100 @@ sub vmconfig_update_disk { my $drive = parse_drive($opt, $value); - if ($conf->{$opt}) { - - if (my $old_drive = parse_drive($opt, $conf->{$opt})) { + if ($conf->{$opt} && (my $old_drive = parse_drive($opt, $conf->{$opt}))) { + my $media = $drive->{media} || 'disk'; + my $oldmedia = $old_drive->{media} || 'disk'; + die "unable to change media type\n" if $media ne $oldmedia; - my $media = $drive->{media} || 'disk'; - my $oldmedia = $old_drive->{media} || 'disk'; - die "unable to change media type\n" if $media ne $oldmedia; + if (!drive_is_cdrom($old_drive)) { - if (!drive_is_cdrom($old_drive)) { + if ($drive->{file} ne $old_drive->{file}) { - if ($drive->{file} ne $old_drive->{file}) { + die "skip\n" if !$hotplug; - die "skip\n" if !$hotplug; - - # unplug and register as unused - vm_deviceunplug($vmid, $conf, $opt); - vmconfig_register_unused_drive($storecfg, $vmid, $conf, $old_drive) + # unplug and register as unused + vm_deviceunplug($vmid, $conf, $opt); + vmconfig_register_unused_drive($storecfg, $vmid, $conf, $old_drive) - } else { - # update existing disk - - # skip non hotpluggable value - if (safe_string_ne($drive->{discard}, $old_drive->{discard}) || - safe_string_ne($drive->{iothread}, $old_drive->{iothread}) || - safe_string_ne($drive->{queues}, $old_drive->{queues}) || - safe_string_ne($drive->{cache}, $old_drive->{cache}) || - safe_string_ne($drive->{ssd}, $old_drive->{ssd})) { - die "skip\n"; - } + } else { + # update existing disk + + # skip non hotpluggable value + if (safe_string_ne($drive->{discard}, $old_drive->{discard}) || + safe_string_ne($drive->{iothread}, $old_drive->{iothread}) || + safe_string_ne($drive->{queues}, $old_drive->{queues}) || + safe_string_ne($drive->{cache}, $old_drive->{cache}) || + safe_string_ne($drive->{ssd}, $old_drive->{ssd})) { + die "skip\n"; + } - # apply throttle - if (safe_num_ne($drive->{mbps}, $old_drive->{mbps}) || - safe_num_ne($drive->{mbps_rd}, $old_drive->{mbps_rd}) || - safe_num_ne($drive->{mbps_wr}, $old_drive->{mbps_wr}) || - safe_num_ne($drive->{iops}, $old_drive->{iops}) || - safe_num_ne($drive->{iops_rd}, $old_drive->{iops_rd}) || - safe_num_ne($drive->{iops_wr}, $old_drive->{iops_wr}) || - safe_num_ne($drive->{mbps_max}, $old_drive->{mbps_max}) || - safe_num_ne($drive->{mbps_rd_max}, $old_drive->{mbps_rd_max}) || - safe_num_ne($drive->{mbps_wr_max}, $old_drive->{mbps_wr_max}) || - safe_num_ne($drive->{iops_max}, $old_drive->{iops_max}) || - safe_num_ne($drive->{iops_rd_max}, $old_drive->{iops_rd_max}) || - safe_num_ne($drive->{iops_wr_max}, $old_drive->{iops_wr_max}) || - safe_num_ne($drive->{bps_max_length}, $old_drive->{bps_max_length}) || - safe_num_ne($drive->{bps_rd_max_length}, $old_drive->{bps_rd_max_length}) || - safe_num_ne($drive->{bps_wr_max_length}, $old_drive->{bps_wr_max_length}) || - safe_num_ne($drive->{iops_max_length}, $old_drive->{iops_max_length}) || - safe_num_ne($drive->{iops_rd_max_length}, $old_drive->{iops_rd_max_length}) || - safe_num_ne($drive->{iops_wr_max_length}, $old_drive->{iops_wr_max_length})) { - - qemu_block_set_io_throttle($vmid,"drive-$opt", - ($drive->{mbps} || 0)*1024*1024, - ($drive->{mbps_rd} || 0)*1024*1024, - ($drive->{mbps_wr} || 0)*1024*1024, - $drive->{iops} || 0, - $drive->{iops_rd} || 0, - $drive->{iops_wr} || 0, - ($drive->{mbps_max} || 0)*1024*1024, - ($drive->{mbps_rd_max} || 0)*1024*1024, - ($drive->{mbps_wr_max} || 0)*1024*1024, - $drive->{iops_max} || 0, - $drive->{iops_rd_max} || 0, - $drive->{iops_wr_max} || 0, - $drive->{bps_max_length} || 1, - $drive->{bps_rd_max_length} || 1, - $drive->{bps_wr_max_length} || 1, - $drive->{iops_max_length} || 1, - $drive->{iops_rd_max_length} || 1, - $drive->{iops_wr_max_length} || 1); + # apply throttle + if (safe_num_ne($drive->{mbps}, $old_drive->{mbps}) || + safe_num_ne($drive->{mbps_rd}, $old_drive->{mbps_rd}) || + safe_num_ne($drive->{mbps_wr}, $old_drive->{mbps_wr}) || + safe_num_ne($drive->{iops}, $old_drive->{iops}) || + safe_num_ne($drive->{iops_rd}, $old_drive->{iops_rd}) || + safe_num_ne($drive->{iops_wr}, $old_drive->{iops_wr}) || + safe_num_ne($drive->{mbps_max}, $old_drive->{mbps_max}) || + safe_num_ne($drive->{mbps_rd_max}, $old_drive->{mbps_rd_max}) || + safe_num_ne($drive->{mbps_wr_max}, $old_drive->{mbps_wr_max}) || + safe_num_ne($drive->{iops_max}, $old_drive->{iops_max}) || + safe_num_ne($drive->{iops_rd_max}, $old_drive->{iops_rd_max}) || + safe_num_ne($drive->{iops_wr_max}, $old_drive->{iops_wr_max}) || + safe_num_ne($drive->{bps_max_length}, $old_drive->{bps_max_length}) || + safe_num_ne($drive->{bps_rd_max_length}, $old_drive->{bps_rd_max_length}) || + safe_num_ne($drive->{bps_wr_max_length}, $old_drive->{bps_wr_max_length}) || + safe_num_ne($drive->{iops_max_length}, $old_drive->{iops_max_length}) || + safe_num_ne($drive->{iops_rd_max_length}, $old_drive->{iops_rd_max_length}) || + safe_num_ne($drive->{iops_wr_max_length}, $old_drive->{iops_wr_max_length})) { + + qemu_block_set_io_throttle( + $vmid,"drive-$opt", + ($drive->{mbps} || 0)*1024*1024, + ($drive->{mbps_rd} || 0)*1024*1024, + ($drive->{mbps_wr} || 0)*1024*1024, + $drive->{iops} || 0, + $drive->{iops_rd} || 0, + $drive->{iops_wr} || 0, + ($drive->{mbps_max} || 0)*1024*1024, + ($drive->{mbps_rd_max} || 0)*1024*1024, + ($drive->{mbps_wr_max} || 0)*1024*1024, + $drive->{iops_max} || 0, + $drive->{iops_rd_max} || 0, + $drive->{iops_wr_max} || 0, + $drive->{bps_max_length} || 1, + $drive->{bps_rd_max_length} || 1, + $drive->{bps_wr_max_length} || 1, + $drive->{iops_max_length} || 1, + $drive->{iops_rd_max_length} || 1, + $drive->{iops_wr_max_length} || 1, + ); - } + } - return 1; - } + return 1; + } - } else { # cdrom + } else { # cdrom - if ($drive->{file} eq 'none') { - mon_cmd($vmid, "eject", force => JSON::true, id => "$opt"); - if (drive_is_cloudinit($old_drive)) { - vmconfig_register_unused_drive($storecfg, $vmid, $conf, $old_drive); - } - } else { - my $path = get_iso_path($storecfg, $vmid, $drive->{file}); + if ($drive->{file} eq 'none') { + mon_cmd($vmid, "eject", force => JSON::true, id => "$opt"); + if (drive_is_cloudinit($old_drive)) { + vmconfig_register_unused_drive($storecfg, $vmid, $conf, $old_drive); + } + } else { + my $path = get_iso_path($storecfg, $vmid, $drive->{file}); - # force eject if locked - mon_cmd($vmid, "eject", force => JSON::true, id => "$opt"); + # force eject if locked + mon_cmd($vmid, "eject", force => JSON::true, id => "$opt"); - if ($path) { - mon_cmd($vmid, "blockdev-change-medium", - id => "$opt", filename => "$path"); - } + if ($path) { + mon_cmd($vmid, "blockdev-change-medium", + id => "$opt", filename => "$path"); } - - return 1; } + + return 1; } } @@ -4879,7 +4884,8 @@ sub vm_migrate_alloc_nbd_disks { $format = qemu_img_format($scfg, $volname); } - my $newvolid = PVE::Storage::vdisk_alloc($storecfg, $storeid, $vmid, $format, undef, ($drive->{size}/1024)); + my $size = $drive->{size} / 1024; + my $newvolid = PVE::Storage::vdisk_alloc($storecfg, $storeid, $vmid, $format, undef, $size); my $newdrive = $drive; $newdrive->{format} = $format; $newdrive->{file} = $newvolid; @@ -4900,12 +4906,15 @@ sub vm_start { return PVE::QemuConfig->lock_config($vmid, sub { my $conf = PVE::QemuConfig->load_config($vmid, $migrate_opts->{migratedfrom}); - die "you can't start a vm if it's a template\n" if PVE::QemuConfig->is_template($conf); + die "you can't start a vm if it's a template\n" + if !$params->{skiptemplate} && PVE::QemuConfig->is_template($conf); - $params->{resume} = PVE::QemuConfig->has_lock($conf, 'suspended'); + my $has_suspended_lock = PVE::QemuConfig->has_lock($conf, 'suspended'); PVE::QemuConfig->check_lock($conf) - if !($params->{skiplock} || $params->{resume}); + if !($params->{skiplock} || $has_suspended_lock); + + $params->{resume} = $has_suspended_lock || defined($conf->{vmstate}); die "VM $vmid already running\n" if check_running($vmid, undef, $migrate_opts->{migratedfrom}); @@ -4927,6 +4936,7 @@ sub vm_start { # params: # statefile => 'tcp', 'unix' for migration or path/volid for RAM state # skiplock => 0/1, skip checking for config lock +# skiptemplate => 0/1, skip checking whether VM is template # forcemachine => to force Qemu machine (rollback/migration) # forcecpu => a QEMU '-cpu' argument string to override get_cpu_options # timeout => in seconds @@ -5062,7 +5072,7 @@ sub vm_start_nolock { } # host pci devices - for (my $i = 0; $i < $MAX_HOSTPCI_DEVICES; $i++) { + for (my $i = 0; $i < $PVE::QemuServer::PCI::MAX_HOSTPCI_DEVICES; $i++) { my $d = parse_hostpci($conf->{"hostpci$i"}); next if !$d; my $pcidevices = $d->{pciid}; @@ -5143,11 +5153,13 @@ sub vm_start_nolock { eval { $run_qemu->() }; if (my $err = $@) { - PVE::QemuServer::Memory::hugepages_reset($hugepages_host_topology); + PVE::QemuServer::Memory::hugepages_reset($hugepages_host_topology) + if !$conf->{keephugepages}; die $err; } - PVE::QemuServer::Memory::hugepages_pre_deallocate($hugepages_topology); + PVE::QemuServer::Memory::hugepages_pre_deallocate($hugepages_topology) + if !$conf->{keephugepages}; }; eval { PVE::QemuServer::Memory::hugepages_update_locked($code); }; @@ -5185,7 +5197,13 @@ sub vm_start_nolock { my $pfamily = PVE::Tools::get_host_address_family($nodename); my $storage_migrate_port = PVE::Tools::next_migrate_port($pfamily); - mon_cmd($vmid, "nbd-server-start", addr => { type => 'inet', data => { host => "${localip}", port => "${storage_migrate_port}" } } ); + mon_cmd($vmid, "nbd-server-start", addr => { + type => 'inet', + data => { + host => "${localip}", + port => "${storage_migrate_port}", + }, + }); $localip = "[$localip]" if Net::IP::ip_is_ipv6($localip); $migrate_storage_uri = "nbd:${localip}:${storage_migrate_port}"; } @@ -5216,7 +5234,8 @@ sub vm_start_nolock { print "spice listens on port $spice_port\n"; $res->{spice_port} = $spice_port; if ($migrate_opts->{spice_ticket}) { - mon_cmd($vmid, "set_password", protocol => 'spice', password => $migrate_opts->{spice_ticket}); + mon_cmd($vmid, "set_password", protocol => 'spice', password => + $migrate_opts->{spice_ticket}); mon_cmd($vmid, "expire_password", protocol => 'spice', time => "+30"); } } @@ -5326,7 +5345,7 @@ sub vm_stop_cleanup { } if ($conf->{ivshmem}) { - my $ivshmem = PVE::JSONSchema::parse_property_string($ivshmem_fmt, $conf->{ivshmem}); + my $ivshmem = parse_property_string($ivshmem_fmt, $conf->{ivshmem}); # just delete it for now, VMs which have this already open do not # are affected, but new VMs will get a separated one. If this # becomes an issue we either add some sort of ref-counting or just @@ -5512,7 +5531,8 @@ sub vm_suspend { } - $vmstate = PVE::QemuConfig->__snapshot_save_vmstate($vmid, $conf, "suspend-$date", $storecfg, $statestorage, 1); + $vmstate = PVE::QemuConfig->__snapshot_save_vmstate( + $vmid, $conf, "suspend-$date", $storecfg, $statestorage, 1); $path = PVE::Storage::path($storecfg, $vmstate); PVE::QemuConfig->write_config($vmid, $conf); } else { @@ -5795,7 +5815,8 @@ my $restore_allocate_devices = sub { $name .= ".$d->{format}" if $d->{format} ne 'raw'; } - my $volid = PVE::Storage::vdisk_alloc($storecfg, $storeid, $vmid, $d->{format}, $name, $alloc_size); + my $volid = PVE::Storage::vdisk_alloc( + $storecfg, $storeid, $vmid, $d->{format}, $name, $alloc_size); print STDERR "new volume ID is '$volid'\n"; $d->{volid} = $volid; @@ -5924,7 +5945,7 @@ sub update_disk_config { my ($vmid, $conf, $volid_hash) = @_; my $changes; - my $prefix = "VM $vmid:"; + my $prefix = "VM $vmid"; # used and unused disks my $referenced = {}; @@ -5941,22 +5962,22 @@ sub update_disk_config { my $volid = $drive->{file}; return if !$volid; + my $volume = $volid_hash->{$volid}; # mark volid as "in-use" for next step $referenced->{$volid} = 1; - if ($volid_hash->{$volid} && - (my $path = $volid_hash->{$volid}->{path})) { + if ($volume && (my $path = $volume->{path})) { $referencedpath->{$path} = 1; } return if drive_is_cdrom($drive); - return if !$volid_hash->{$volid}; + return if !$volume; - my ($updated, $old_size, $new_size) = PVE::QemuServer::Drive::update_disksize($drive, $volid_hash); + my ($updated, $msg) = PVE::QemuServer::Drive::update_disksize($drive, $volume->{size}); if (defined($updated)) { $changes = 1; $conf->{$opt} = print_drive($updated); - print "$prefix size of disk '$volid' ($opt) updated from $old_size to $new_size\n"; + print "$prefix ($opt): $msg\n"; } }); @@ -6055,8 +6076,11 @@ sub restore_proxmox_backup_archive { my $datastore = $scfg->{datastore}; my $username = $scfg->{username} // 'root@pam'; my $fingerprint = $scfg->{fingerprint}; + my $keyfile = PVE::Storage::PBSPlugin::pbs_encryption_key_file_name($storecfg, $storeid); my $repo = "$username\@$server:$datastore"; + + # This is only used for `pbs-restore`! my $password = PVE::Storage::PBSPlugin::pbs_get_password($scfg, $storeid); local $ENV{PBS_PASSWORD} = $password; local $ENV{PBS_FINGERPRINT} = $fingerprint if defined($fingerprint); @@ -6118,7 +6142,9 @@ sub restore_proxmox_backup_archive { } } - my $is_qemu_server_backup = scalar(grep { $_->{filename} eq 'qemu-server.conf.blob' } @{$index->{files}}); + my $is_qemu_server_backup = scalar( + grep { $_->{filename} eq 'qemu-server.conf.blob' } @{$index->{files}} + ); if (!$is_qemu_server_backup) { die "backup does not look like a qemu-server backup (missing 'qemu-server.conf' file)\n"; } @@ -6137,7 +6163,7 @@ sub restore_proxmox_backup_archive { } my $fh = IO::File->new($cfgfn, "r") || - "unable to read qemu-server.conf - $!\n"; + die "unable to read qemu-server.conf - $!\n"; my $virtdev_hash = $parse_backup_hints->($rpcenv, $user, $storecfg, $fh, $devinfo, $options); @@ -6159,6 +6185,7 @@ sub restore_proxmox_backup_archive { my $path = PVE::Storage::path($storecfg, $volid); + # This is the ONLY user of the PBS_ env vars set on top of this function! my $pbs_restore_cmd = [ '/usr/bin/pbs-restore', '--repository', $repo, @@ -6168,6 +6195,9 @@ sub restore_proxmox_backup_archive { '--verbose', ]; + push @$pbs_restore_cmd, '--format', $d->{format} if $d->{format}; + push @$pbs_restore_cmd, '--keyfile', $keyfile if -e $keyfile; + if (PVE::Storage::volume_has_feature($storecfg, 'sparseinit', $volid)) { push @$pbs_restore_cmd, '--skip-zero'; } @@ -6294,7 +6324,7 @@ sub restore_vma_archive { # we can read the config - that is already extracted my $fh = IO::File->new($cfgfn, "r") || - "unable to read qemu-server.conf - $!\n"; + die "unable to read qemu-server.conf - $!\n"; my $fwcfgfn = "$tmpdir/qemu-server.fw"; if (-f $fwcfgfn) { @@ -6939,7 +6969,8 @@ sub clone_disk { } elsif ($drivename eq 'efidisk0') { $size = get_efivars_size($conf); } - $newvolid = PVE::Storage::vdisk_alloc($storecfg, $storeid, $newvmid, $dst_format, $name, ($size/1024)); + $size /= 1024; + $newvolid = PVE::Storage::vdisk_alloc($storecfg, $storeid, $newvmid, $dst_format, $name, $size); push @$newvollist, $newvolid; PVE::Storage::activate_volumes($storecfg, [$newvolid]); @@ -6957,7 +6988,8 @@ sub clone_disk { # that is given by the OVMF_VARS.fd my $src_path = PVE::Storage::path($storecfg, $drive->{file}); my $dst_path = PVE::Storage::path($storecfg, $newvolid); - run_command(['qemu-img', 'dd', '-n', '-O', $dst_format, "bs=1", "count=$size", "if=$src_path", "of=$dst_path"]); + run_command(['qemu-img', 'dd', '-n', '-O', $dst_format, "bs=1", "count=$size", + "if=$src_path", "of=$dst_path"]); } else { qemu_img_convert($drive->{file}, $newvolid, $size, $snapname, $sparseinit); } @@ -6969,7 +7001,8 @@ sub clone_disk { if $drive->{iothread}; } - qemu_drive_mirror($vmid, $drivename, $newvolid, $newvmid, $sparseinit, $jobs, $completion, $qga, $bwlimit); + qemu_drive_mirror($vmid, $drivename, $newvolid, $newvmid, $sparseinit, $jobs, + $completion, $qga, $bwlimit); } } @@ -7001,7 +7034,7 @@ sub qemu_use_old_bios_files { $machine_type = $1; $use_old_bios_files = 1; } else { - my $version = PVE::QemuServer::Machine::extract_version($machine_type, kvm_user_version()); + my $version = extract_version($machine_type, kvm_user_version()); # Note: kvm version < 2.4 use non-efi pxe files, and have problems when we # load new efi bios files on migration. So this hack is required to allow # live migration from qemu-2.2 to qemu-2.4, which is sometimes used when @@ -7076,7 +7109,9 @@ sub scsihw_infos { } my $controller = int($drive->{index} / $maxdev); - my $controller_prefix = ($conf->{scsihw} && $conf->{scsihw} eq 'virtio-scsi-single') ? "virtioscsi" : "scsihw"; + my $controller_prefix = ($conf->{scsihw} && $conf->{scsihw} eq 'virtio-scsi-single') + ? "virtioscsi" + : "scsihw"; return ($maxdev, $controller, $controller_prefix); } @@ -7179,6 +7214,74 @@ sub clear_reboot_request { return $res; } +sub bootorder_from_legacy { + my ($conf, $bootcfg) = @_; + + my $boot = $bootcfg->{legacy} || $boot_fmt->{legacy}->{default}; + my $bootindex_hash = {}; + my $i = 1; + foreach my $o (split(//, $boot)) { + $bootindex_hash->{$o} = $i*100; + $i++; + } + + my $bootorder = {}; + + PVE::QemuConfig->foreach_volume($conf, sub { + my ($ds, $drive) = @_; + + if (drive_is_cdrom ($drive, 1)) { + if ($bootindex_hash->{d}) { + $bootorder->{$ds} = $bootindex_hash->{d}; + $bootindex_hash->{d} += 1; + } + } elsif ($bootindex_hash->{c}) { + $bootorder->{$ds} = $bootindex_hash->{c} + if $conf->{bootdisk} && $conf->{bootdisk} eq $ds; + $bootindex_hash->{c} += 1; + } + }); + + if ($bootindex_hash->{n}) { + for (my $i = 0; $i < $MAX_NETS; $i++) { + my $netname = "net$i"; + next if !$conf->{$netname}; + $bootorder->{$netname} = $bootindex_hash->{n}; + $bootindex_hash->{n} += 1; + } + } + + return $bootorder; +} + +# Generate default device list for 'boot: order=' property. Matches legacy +# default boot order, but with explicit device names. This is important, since +# the fallback for when neither 'order' nor the old format is specified relies +# on 'bootorder_from_legacy' above, and it would be confusing if this diverges. +sub get_default_bootdevices { + my ($conf) = @_; + + my @ret = (); + + # harddisk + my $first = PVE::QemuServer::Drive::resolve_first_disk($conf, 0); + push @ret, $first if $first; + + # cdrom + $first = PVE::QemuServer::Drive::resolve_first_disk($conf, 1); + push @ret, $first if $first; + + # network + for (my $i = 0; $i < $MAX_NETS; $i++) { + my $netname = "net$i"; + next if !$conf->{$netname}; + push @ret, $netname; + last; + } + + return \@ret; +} + # bash completion helper sub complete_backup_archives {