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 parse_hostpci);
-use PVE::QemuServer::USB qw(parse_usb_device);
+use PVE::QemuServer::USB;
my $have_sdn;
eval {
description => "Run fstrim after moving a disk or migrating the VM.",
type => 'boolean',
optional => 1,
- default => 0
+ default => 0,
+ },
+ 'freeze-fs-on-backup' => {
+ description => "Freeze/thaw guest filesystems on backup for consistency.",
+ type => 'boolean',
+ optional => 1,
+ default => 1,
},
type => {
description => "Select the agent type",
verbose_description => <<EODESCR,
Arbitrary arguments passed to kvm, for example:
-args: -no-reboot -no-hpet
+args: -no-reboot -smbios 'type=0,vendor=FOO'
NOTE: this option is for experts only.
EODESCR
.' recommended. Use ssh keys instead. Also note that older cloud-init versions do not'
.' support hashed passwords.',
},
+ ciupgrade => {
+ optional => 1,
+ type => 'boolean',
+ description => 'cloud-init: do an automatic package upgrade after the first boot.'
+ },
cicustom => {
optional => 1,
type => 'string',
PVE::JSONSchema::register_standard_option("pve-qm-$k", $v);
}
-my $MAX_USB_DEVICES = 14;
my $MAX_NETS = 32;
my $MAX_SERIAL_PORTS = 4;
my $MAX_PARALLEL_PORTS = 3;
return $volid;
}
-my $usb_fmt = {
- host => {
- default_key => 1,
- type => 'string', format => 'pve-qm-usb-device',
- format_description => 'HOSTUSBDEVICE|spice',
- description => <<EODESCR,
-The Host USB device or port or the value 'spice'. HOSTUSBDEVICE syntax is:
-
- 'bus-port(.port)*' (decimal numbers) or
- 'vendor_id:product_id' (hexadeciaml numbers) or
- 'spice'
-
-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.
-
-The value 'spice' can be used to add a usb redirection devices for spice.
-EODESCR
- },
- usb3 => {
- optional => 1,
- type => 'boolean',
- description => "Specifies whether if given host option is a USB3 device or port."
- ." For modern guests (machine version >= 7.1 and ostype l26 and windows > 7), this flag"
- ." is irrelevant (all devices are plugged into a xhci controller).",
- default => 0,
- },
-};
-
-my $usbdesc = {
- optional => 1,
- type => 'string', format => $usb_fmt,
- description => "Configure an USB device (n is 0 to 4, for machine version >= 7.1 and ostype"
- ." l26 or windows > 7, n can be up to 14).",
-};
-PVE::JSONSchema::register_standard_option("pve-qm-usb", $usbdesc);
-
my $serialdesc = {
optional => 1,
type => 'string',
$confdesc->{$key} = $PVE::QemuServer::Drive::drivedesc_hash->{$key};
}
-for (my $i = 0; $i < $MAX_USB_DEVICES; $i++) {
- $confdesc->{"usb$i"} = $usbdesc;
+for (my $i = 0; $i < $PVE::QemuServer::USB::MAX_USB_DEVICES; $i++) {
+ $confdesc->{"usb$i"} = $PVE::QemuServer::USB::usbdesc;
}
my $boot_fmt = {
return $initiator;
}
+my sub storage_allows_io_uring_default {
+ my ($scfg, $cache_direct) = @_;
+
+ # io_uring with cache mode writeback or writethrough on krbd will hang...
+ return if $scfg && $scfg->{type} eq 'rbd' && $scfg->{krbd} && !$cache_direct;
+
+ # io_uring with cache mode writeback or writethrough on LVM will hang, without cache only
+ # sometimes, just plain disable...
+ return if $scfg && $scfg->{type} eq 'lvm';
+
+ # io_uring causes problems when used with CIFS since kernel 5.15
+ # Some discussion: https://www.spinics.net/lists/linux-cifs/msg26734.html
+ return if $scfg && $scfg->{type} eq 'cifs';
+
+ return 1;
+}
+
+my sub drive_uses_cache_direct {
+ my ($drive, $scfg) = @_;
+
+ my $cache_direct = 0;
+
+ if (my $cache = $drive->{cache}) {
+ $cache_direct = $cache =~ /^(?:off|none|directsync)$/;
+ } elsif (!drive_is_cdrom($drive) && !($scfg && $scfg->{type} eq 'btrfs' && !$scfg->{nocow})) {
+ $cache_direct = 1;
+ }
+
+ return $cache_direct;
+}
+
sub print_drive_commandline_full {
my ($storecfg, $vmid, $drive, $pbs_name, $io_uring) = @_;
$opts .= ",format=$format";
}
- my $cache_direct = 0;
-
- if (my $cache = $drive->{cache}) {
- $cache_direct = $cache =~ /^(?:off|none|directsync)$/;
- } elsif (!drive_is_cdrom($drive) && !($scfg && $scfg->{type} eq 'btrfs' && !$scfg->{nocow})) {
- $opts .= ",cache=none";
- $cache_direct = 1;
- }
-
- # io_uring with cache mode writeback or writethrough on krbd will hang...
- my $rbd_no_io_uring = $scfg && $scfg->{type} eq 'rbd' && $scfg->{krbd} && !$cache_direct;
+ my $cache_direct = drive_uses_cache_direct($drive, $scfg);
- # io_uring with cache mode writeback or writethrough on LVM will hang, without cache only
- # sometimes, just plain disable...
- my $lvm_no_io_uring = $scfg && $scfg->{type} eq 'lvm';
-
- # io_uring causes problems when used with CIFS since kernel 5.15
- # Some discussion: https://www.spinics.net/lists/linux-cifs/msg26734.html
- my $cifs_no_io_uring = $scfg && $scfg->{type} eq 'cifs';
+ $opts .= ",cache=none" if !$drive->{cache} && $cache_direct;
if (!$drive->{aio}) {
- if ($io_uring && !$rbd_no_io_uring && !$lvm_no_io_uring && !$cifs_no_io_uring) {
+ if ($io_uring && storage_allows_io_uring_default($scfg, $cache_direct)) {
# io_uring supports all cache modes
$opts .= ",aio=io_uring";
} else {
return;
}
-PVE::JSONSchema::register_format('pve-qm-usb-device', \&verify_usb_device);
-sub verify_usb_device {
- my ($value, $noerr) = @_;
-
- return $value if parse_usb_device($value);
-
- return if $noerr;
-
- die "unable to parse usb device\n";
-}
-
# add JSON properties for create and set function
sub json_config_properties {
my ($prop, $with_disk_alloc) = @_;
optional => 1,
},
qmpstatus => {
- description => "QEMU QMP agent status.",
+ description => "VM run state from the 'query-status' QMP monitor command.",
type => 'string',
optional => 1,
},
});
}
+ # Used to distinguish different invocations in the log.
+ my $log_prefix = "[id=" . int(time()) . "] ";
+
my $emulator_cmd = [
"swtpm",
"socket",
"file=$paths->{pid}",
"--terminate", # terminate on QEMU disconnect
"--daemon",
+ "--log",
+ "file=/run/qemu-server/$vmid-swtpm.log,level=1,prefix=$log_prefix",
];
push @$emulator_cmd, "--tpm2" if $tpm->{version} eq 'v2.0';
run_command($emulator_cmd, outfunc => sub { print $1; });
or die "no OVMF images known for architecture '$arch'\n";
my $type = 'default';
- if (defined($efidisk->{efitype}) && $efidisk->{efitype} eq '4m') {
+ if ($arch ne "aarch64" && defined($efidisk->{efitype}) && $efidisk->{efitype} eq '4m') {
$type = $smm ? "4m" : "4m-no-smm";
$type .= '-ms' if $efidisk->{'pre-enrolled-keys'};
}
# Since commit 277d33454f77ec1d1e0bc04e37621e4dd2424b67 in pve-qemu, smm is not off by default
# anymore. But smm=off seems to be required when using SeaBIOS and serial display.
my sub should_disable_smm {
- my ($conf, $vga) = @_;
+ my ($conf, $vga, $machine) = @_;
+
+ return if $machine =~ m/^virt/; # there is no smm flag that could be disabled
return (!defined($conf->{bios}) || $conf->{bios} eq 'seabios') &&
$vga->{type} && $vga->{type} =~ m/^(serial\d+|none)$/;
# add usb controllers
my @usbcontrollers = PVE::QemuServer::USB::get_usb_controllers(
- $conf, $bridges, $arch, $machine_type, $usbdesc->{format}, $MAX_USB_DEVICES, $machine_version);
+ $conf, $bridges, $arch, $machine_type, $machine_version);
push @$devices, @usbcontrollers if @usbcontrollers;
my $vga = parse_vga($conf->{vga});
$usb_dev_features->{spice_usb3} = 1 if min_version($machine_version, 4, 0);
my @usbdevices = PVE::QemuServer::USB::get_usb_devices(
- $conf, $usbdesc->{format}, $MAX_USB_DEVICES, $usb_dev_features, $bootorder, $machine_version);
+ $conf, $usb_dev_features, $bootorder, $machine_version);
push @$devices, @usbdevices if @usbdevices;
# serial devices
}
} else {
die "no such serial device\n" if ! -c $path;
- push @$devices, '-chardev', "tty,id=serial$i,path=$path";
+ push @$devices, '-chardev', "serial,id=serial$i,path=$path";
push @$devices, '-device', "isa-serial,chardev=serial$i";
}
}
for (my $i = 0; $i < $MAX_PARALLEL_PORTS; $i++) {
if (my $path = $conf->{"parallel$i"}) {
die "no such parallel device\n" if ! -c $path;
- my $devtype = $path =~ m!^/dev/usb/lp! ? 'tty' : 'parport';
+ my $devtype = $path =~ m!^/dev/usb/lp! ? 'serial' : 'parallel';
push @$devices, '-chardev', "$devtype,id=parallel$i,path=$path";
push @$devices, '-device', "isa-parallel,chardev=parallel$i";
}
push @$cmd, '-boot', "menu=on,strict=on,reboot-timeout=1000,splash=/usr/share/qemu-server/bootsplash.jpg";
- push @$cmd, '-no-acpi' if defined($conf->{acpi}) && $conf->{acpi} == 0;
+ push $machineFlags->@*, 'acpi=off' if defined($conf->{acpi}) && $conf->{acpi} == 0;
push @$cmd, '-no-reboot' if defined($conf->{reboot}) && $conf->{reboot} == 0;
if ($winversion >= 6) {
push @$globalFlags, 'kvm-pit.lost_tick_policy=discard';
- push @$cmd, '-no-hpet';
+ push @$machineFlags, 'hpet=off';
}
push @$rtcFlags, 'driftfix=slew' if $tdf;
push @$cmd, get_cpu_options($conf, $arch, $kvm, $kvm_off, $machine_version, $winversion, $gpu_passthrough);
}
- PVE::QemuServer::Memory::config($conf, $vmid, $sockets, $cores, $defaults, $hotplug_features, $cmd);
+ PVE::QemuServer::Memory::config(
+ $conf, $vmid, $sockets, $cores, $defaults, $hotplug_features->{memory}, $cmd);
push @$cmd, '-S' if $conf->{freeze};
push @$machineFlags, 'accel=tcg';
}
- push @$machineFlags, 'smm=off' if should_disable_smm($conf, $vga);
+ push @$machineFlags, 'smm=off' if should_disable_smm($conf, $vga, $machine_type);
my $machine_type_min = $machine_type;
if ($add_pve_version) {
qemu_deviceadd($vmid, PVE::QemuServer::USB::print_qemu_xhci_controller($pciaddr));
}
- # print_usbdevice_full expects the parsed device
- my $d = parse_usb_device($device->{host});
- $d->{usb3} = $device->{usb3};
-
# add the new one
- vm_deviceplug($storecfg, $conf, $vmid, $deviceid, $d, $arch, $machine_type);
+ vm_deviceplug($storecfg, $conf, $vmid, $deviceid, $device, $arch, $machine_type);
}
sub qemu_cpu_hotplug {
my $running = check_running($vmid);
- $size = 0 if !PVE::Storage::volume_resize($storecfg, $volid, $size, $running);
+ PVE::Storage::volume_resize($storecfg, $volid, $size, $running);
return if !$running;
}
my $fast_plug_option = {
+ 'description' => 1,
+ 'hookscript' => 1,
'lock' => 1,
+ 'migrate_downtime' => 1,
+ 'migrate_speed' => 1,
'name' => 1,
'onboot' => 1,
+ 'protection' => 1,
'shares' => 1,
'startup' => 1,
- 'description' => 1,
- 'protection' => 1,
- 'vmstatestorage' => 1,
- 'hookscript' => 1,
'tags' => 1,
+ 'vmstatestorage' => 1,
};
for my $opt (keys %$confdesc_cloudinit) {
} elsif ($opt =~ m/^usb(\d+)$/) {
my $index = $1;
die "skip\n" if !$usb_hotplug;
- my $d = eval { parse_property_string($usbdesc->{format}, $value) };
+ my $d = eval { parse_property_string('pve-qm-usb', $value) };
my $id = $opt;
- if ($d->{host} eq 'spice') {
+ if ($d->{host} =~ m/^spice$/i) {
$id = "usbredirdev$index";
}
qemu_usb_hotplug($storecfg, $conf, $vmid, $id, $d, $arch, $machine_type);
# unplug xhci controller if no usb device is left
if ($usb_hotplug) {
my $has_usb = 0;
- for (my $i = 0; $i < $MAX_USB_DEVICES; $i++) {
+ for (my $i = 0; $i < $PVE::QemuServer::USB::MAX_USB_DEVICES; $i++) {
next if !defined($conf->{"usb$i"});
$has_usb = 1;
last;
if (safe_string_ne($oldnet->{model}, $newnet->{model}) ||
safe_string_ne($oldnet->{macaddr}, $newnet->{macaddr}) ||
safe_num_ne($oldnet->{queues}, $newnet->{queues}) ||
+ safe_num_ne($oldnet->{mtu}, $newnet->{mtu}) ||
!($newnet->{bridge} && $oldnet->{bridge})) { # bridge/nat mode change
# for non online change, we try to hot-unplug
for my $dev ($d->{pciid}->@*) {
my $info = PVE::QemuServer::PCI::prepare_pci_device($vmid, $dev->{id}, $id, $d->{mdev});
- # nvidia grid needs the uuid of the mdev as qemu parameter
+ # nvidia grid needs the qemu parameter '-uuid' set
+ # use smbios uuid or mdev uuid as fallback for that
if ($d->{mdev} && !defined($uuid) && $info->{vendor} eq '10de') {
- $uuid = PVE::QemuServer::PCI::generate_mdev_uuid($vmid, $id);
+ if (defined($conf->{smbios1})) {
+ my $smbios_conf = parse_smbios1($conf->{smbios1});
+ $uuid = $smbios_conf->{uuid} if defined($smbios_conf->{uuid});
+ }
+ $uuid = PVE::QemuServer::PCI::generate_mdev_uuid($vmid, $id) if !defined($uuid);
}
}
}
if ($conf->{hugepages}) {
my $code = sub {
- my $hugepages_topology = PVE::QemuServer::Memory::hugepages_topology($conf);
+ my $hotplug_features =
+ parse_hotplug_features(defined($conf->{hotplug}) ? $conf->{hotplug} : '1');
+ my $hugepages_topology =
+ PVE::QemuServer::Memory::hugepages_topology($conf, $hotplug_features->{memory});
+
my $hugepages_host_topology = PVE::QemuServer::Memory::hugepages_host_topology();
PVE::QemuServer::Memory::hugepages_mount();
}
if (!defined($conf->{balloon}) || $conf->{balloon}) {
- mon_cmd(
- $vmid,
- 'qom-set',
- path => "machine/peripheral/balloon0",
- property => "guest-stats-polling-interval",
- value => 2
- );
+ eval {
+ mon_cmd(
+ $vmid,
+ 'qom-set',
+ path => "machine/peripheral/balloon0",
+ property => "guest-stats-polling-interval",
+ value => 2
+ );
+ };
+ log_warn("could not set polling interval for ballooning - $@") if $@;
}
if ($resume) {
# NOTE: avoid PVE::SysFSTools::pci_cleanup_mdev_device as it requires PCI ID and we
# don't want to break ABI just for this two liner
my $dev_sysfs_dir = "/sys/bus/mdev/devices/$uuid";
+
+ # some nvidia vgpu driver versions want to clean the mdevs up themselves, and error
+ # out when we do it first. so wait for 10 seconds and then try it
+ my $pciid = $d->{pciid}->[0]->{id};
+ my $info = PVE::SysFSTools::pci_device_info("$pciid");
+ if ($info->{vendor} eq '10de') {
+ sleep 10;
+ }
+
PVE::SysFSTools::file_write("$dev_sysfs_dir/remove", "1") if -e $dev_sysfs_dir;
}
}
});
}
+sub check_bridge_access {
+ my ($rpcenv, $authuser, $conf) = @_;
+
+ return 1 if $authuser eq 'root@pam';
+
+ for my $opt (sort keys $conf->%*) {
+ next if $opt !~ m/^net\d+$/;
+ my $net = parse_net($conf->{$opt});
+ my ($bridge, $tag, $trunks) = $net->@{'bridge', 'tag', 'trunks'};
+ PVE::GuestHelpers::check_vnet_access($rpcenv, $authuser, $bridge, $tag, $trunks);
+ }
+ return 1;
+};
+
# vzdump restore implementaion
sub tar_archive_read_firstfile {
}
my $new_conf = $restore_merge_config->($conffile, $new_conf_raw, $options->{override_conf});
+ check_bridge_access($rpcenv, $user, $new_conf);
PVE::QemuConfig->write_config($vmid, $new_conf);
eval { rescan($vmid, 1); };
}
my $new_conf = $restore_merge_config->($conffile, $new_conf_raw, $opts->{override_conf});
+ check_bridge_access($rpcenv, $user, $new_conf);
PVE::QemuConfig->write_config($vmid, $new_conf);
eval { rescan($vmid, 1); };
}
sub qemu_img_convert {
- my ($src_volid, $dst_volid, $size, $snapname, $is_zero_initialized) = @_;
+ my ($src_volid, $dst_volid, $size, $snapname, $is_zero_initialized, $bwlimit) = @_;
my $storecfg = PVE::Storage::config();
my ($src_storeid, $src_volname) = PVE::Storage::parse_volume_id($src_volid, 1);
if $snapname && $src_format && $src_format eq "qcow2";
push @$cmd, '-t', 'none' if $dst_scfg->{type} eq 'zfspool';
push @$cmd, '-T', $cachemode if defined($cachemode);
+ push @$cmd, '-r', "${bwlimit}K" if defined($bwlimit);
if ($src_is_iscsi) {
push @$cmd, '--image-opts';
}
}
+# Check for bug #4525: drive-mirror will open the target drive with the same aio setting as the
+# source, but some storages have problems with io_uring, sometimes even leading to crashes.
+my sub clone_disk_check_io_uring {
+ my ($src_drive, $storecfg, $src_storeid, $dst_storeid, $use_drive_mirror) = @_;
+
+ return if !$use_drive_mirror;
+
+ # Don't complain when not changing storage.
+ # Assume if it works for the source, it'll work for the target too.
+ return if $src_storeid eq $dst_storeid;
+
+ my $src_scfg = PVE::Storage::storage_config($storecfg, $src_storeid);
+ my $dst_scfg = PVE::Storage::storage_config($storecfg, $dst_storeid);
+
+ my $cache_direct = drive_uses_cache_direct($src_drive);
+
+ my $src_uses_io_uring;
+ if ($src_drive->{aio}) {
+ $src_uses_io_uring = $src_drive->{aio} eq 'io_uring';
+ } else {
+ $src_uses_io_uring = storage_allows_io_uring_default($src_scfg, $cache_direct);
+ }
+
+ die "target storage is known to cause issues with aio=io_uring (used by current drive)\n"
+ if $src_uses_io_uring && !storage_allows_io_uring_default($dst_scfg, $cache_direct);
+}
+
sub clone_disk {
my ($storecfg, $source, $dest, $full, $newvollist, $jobs, $completion, $qga, $bwlimit) = @_;
$newvolid = PVE::Storage::vdisk_clone($storecfg, $drive->{file}, $newvmid, $snapname);
push @$newvollist, $newvolid;
} else {
-
- my ($storeid, $volname) = PVE::Storage::parse_volume_id($drive->{file});
- $storeid = $storage if $storage;
+ my ($src_storeid, $volname) = PVE::Storage::parse_volume_id($drive->{file});
+ my $storeid = $storage || $src_storeid;
my $dst_format = resolve_dst_disk_format($storecfg, $storeid, $volname, $format);
$dst_format = 'raw';
$size = PVE::QemuServer::Drive::TPMSTATE_DISK_SIZE;
} else {
- ($size) = PVE::Storage::volume_size_info($storecfg, $drive->{file}, 10);
+ clone_disk_check_io_uring($drive, $storecfg, $src_storeid, $storeid, $use_drive_mirror);
+
+ $size = PVE::Storage::volume_size_info($storecfg, $drive->{file}, 10);
}
$newvolid = PVE::Storage::vdisk_alloc(
$storecfg, $storeid, $newvmid, $dst_format, $name, ($size/1024)
qemu_drive_mirror($vmid, $src_drivename, $newvolid, $newvmid, $sparseinit, $jobs,
$completion, $qga, $bwlimit);
} else {
- # TODO: handle bwlimits
if ($dst_drivename eq 'efidisk0') {
# the relevant data on the efidisk may be smaller than the source
# e.g. on RBD/ZFS, so we use dd to copy only the amount
push $cmd->@*, "bs=$bs", "osize=$size", "if=$src_path", "of=$dst_path";
run_command($cmd);
} else {
- qemu_img_convert($drive->{file}, $newvolid, $size, $snapname, $sparseinit);
+ qemu_img_convert($drive->{file}, $newvolid, $size, $snapname, $sparseinit, $bwlimit);
}
}
}
no_data_clone:
- my ($size) = eval { PVE::Storage::volume_size_info($storecfg, $newvolid, 10) };
+ my $size = eval { PVE::Storage::volume_size_info($storecfg, $newvolid, 10) };
my $disk = dclone($drive);
delete $disk->{format};
PVE::Storage::activate_volumes($storecfg, [$volid]);
qemu_img_convert($ovmf_vars, $volid, $vars_size_b, undef, 0);
- my ($size) = PVE::Storage::volume_size_info($storecfg, $volid, 3);
+ my $size = PVE::Storage::volume_size_info($storecfg, $volid, 3);
return ($volid, $size/1024);
}
}
my $bridge = $net->{bridge};
+ if (!$bridge) {
+ log_warn("Interface '$iface' not attached to any bridge.");
+ next;
+ }
if ($have_sdn) {
PVE::Network::SDN::Zones::add_bridge_fdb($iface, $mac, $bridge, $net->{firewall});
} elsif (-d "/sys/class/net/$bridge/bridge") { # avoid fdb management with OVS for now