Am 11.11.24 um 14:57 schrieb Markus Frank:
> This patch is for enabling AMD SEV (Secure Encrypted Virtualization)
> support in QEMU.
>
> VM-Config-Examples:
> amd_sev: type=std,no-debug=1,no-key-sharing=1
> amd_sev: es,no-debug=1,kernel-hashes=1
>
> kernel-hashes, reduced-phys-bits & cbitpos correspond to the variables
> with the same name in QEMU.
>
> kernel-hashes=1 adds kernel hashes to enable measured linux kernel
> launch since it is per default off for backward compatibility.
>
> reduced-phys-bios and cbitpos are system specific and are read out by
> the query-machine-capabilities c program and saved to the
> /run/qemu-server/host-hw-capabilities.json file. This file is parsed
> and than used by qemu-server to correctly start a AMD SEV VM.
>
> type=std stands for standard sev to differentiate it from sev-es (es)
> or sev-snp (snp) when support is upstream.
>
> QEMU's sev-guest policy gets calculated with the parameters no-debug
> & no-key-sharing. These parameters correspond to policy-bits 0 & 1.
> If type is 'es' than policy-bit 2 gets set to 1 to activate SEV-ES.
> Policy bit 3 (nosend) is always set to 1, because migration features
> for sev are not upstream yet and are attackable.
>
> SEV-ES is highly experimental since it could not be tested.
>
> see coherent doc patch
looks mostly OK some more minor stuff inline
>
> Signed-off-by: Markus Frank <m.fr...@proxmox.com>
> Reviewed-by: Fiona Ebner <f.eb...@proxmox.com>
> ---
> changes v12:
> * a eval for decode_json()
> * get_amd_sev_object: use the three properties as parameters instead of
> the whole config
> * removed efidisk check, as it is obsolete
> * small perl style changes
>
> PVE/QemuServer.pm | 13 +++++-
> PVE/QemuServer/CPUConfig.pm | 88 ++++++++++++++++++++++++++++++++++++-
> 2 files changed, 99 insertions(+), 2 deletions(-)
>
> diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
> index 0df3bda0..dd6ae93f 100644
> --- a/PVE/QemuServer.pm
> +++ b/PVE/QemuServer.pm
> @@ -53,7 +53,7 @@ use PVE::QemuConfig;
> use PVE::QemuServer::Helpers qw(config_aware_timeout min_version
> windows_version);
> use PVE::QemuServer::Cloudinit;
> use PVE::QemuServer::CGroup;
> -use PVE::QemuServer::CPUConfig qw(print_cpu_device get_cpu_options
> get_cpu_bitness is_native_arch);
> +use PVE::QemuServer::CPUConfig qw(print_cpu_device get_cpu_options
> get_cpu_bitness is_native_arch get_amd_sev_object);
> use PVE::QemuServer::Drive qw(is_valid_drivename drive_is_cloudinit
> drive_is_cdrom drive_is_read_only parse_drive print_drive);
> use PVE::QemuServer::Machine;
> use PVE::QemuServer::Memory qw(get_current_memory);
> @@ -358,6 +358,12 @@ my $confdesc = {
> description => "Memory properties.",
> format => $PVE::QemuServer::Memory::memory_fmt
> },
> + amd_sev => {
as of now this can be named 'amd-sev'
> + description => "Secure Encrypted Virtualization (SEV) features by AMD
> CPUs",
> + optional => 1,
> + format => 'pve-qemu-sev-fmt',
> + type => 'string',
> + },
> balloon => {
> optional => 1,
> type => 'integer',
> @@ -4159,6 +4165,11 @@ sub config_to_command {
> }
> }
>
> + if ($conf->{amd_sev}) {
> + push @$devices, '-object', get_amd_sev_object($conf->{amd_sev},
> $conf->{bios}, $conf->{efidisk0});
> + push @$machineFlags, 'confidential-guest-support=sev0';
> + }
> +
> push @$cmd, @$devices;
> push @$cmd, '-rtc', join(',', @$rtcFlags) if scalar(@$rtcFlags);
> push @$cmd, '-machine', join(',', @$machineFlags) if
> scalar(@$machineFlags);
> diff --git a/PVE/QemuServer/CPUConfig.pm b/PVE/QemuServer/CPUConfig.pm
> index 33f7524f..5110b37e 100644
> --- a/PVE/QemuServer/CPUConfig.pm
> +++ b/PVE/QemuServer/CPUConfig.pm
> @@ -3,9 +3,11 @@ package PVE::QemuServer::CPUConfig;
> use strict;
> use warnings;
>
> +use JSON;
> +
> use PVE::JSONSchema;
> use PVE::Cluster qw(cfs_register_file cfs_read_file);
> -use PVE::Tools qw(get_host_arch);
> +use PVE::Tools qw(run_command get_host_arch);
> use PVE::QemuServer::Helpers qw(min_version);
>
> use base qw(PVE::SectionConfig Exporter);
> @@ -15,6 +17,7 @@ print_cpu_device
> get_cpu_options
> get_cpu_bitness
> is_native_arch
> +get_amd_sev_object
> );
>
> # under certain race-conditions, this module might be loaded before
> pve-cluster
> @@ -225,6 +228,37 @@ my $cpu_fmt = {
> },
> };
>
> +my $sev_fmt = {
> + type => {
> + description => "Enable standard SEV with type='std' or enable"
> + ." experimental SEV-ES with the 'es' option.",
> + type => 'string',
> + default_key => 1,
> + format_description => "sev-type",
> + enum => ['std', 'es'],
> + maxLength => 3,
> + },
> + 'no-debug' => {
> + description => "Sets policy bit 0 to 1 to disallow debugging of guest",
> + type => 'boolean',
> + default => 0,
> + optional => 1,
> + },
> + 'no-key-sharing' => {
> + description => "Sets policy bit 1 to 1 to disallow key sharing with
> other guests",
> + type => 'boolean',
> + default => 0,
> + optional => 1,
> + },
> + "kernel-hashes" => {
> + description => "Add kernel hashes to guest firmware for measured linux
> kernel launch",
> + type => 'boolean',
> + default => 0,
> + optional => 1,
> + },
> +};
> +PVE::JSONSchema::register_format('pve-qemu-sev-fmt', $sev_fmt);
> +
> PVE::JSONSchema::register_format('pve-phys-bits', \&parse_phys_bits);
> sub parse_phys_bits {
> my ($str, $noerr) = @_;
> @@ -773,6 +807,58 @@ sub get_cpu_bitness {
> die "unsupported architecture '$arch'\n";
> }
>
> +sub get_hw_capabilities {
> + # Get reduced-phys-bits & cbitpos from host-hw-capabilities.json
> + my $filename = '/run/qemu-server/host-hw-capabilities.json';
hmm, rethinking this it might be good to move that to somewhere accessible for
non-root,
but can be done also in the future.
> + if (! -e $filename) {
> + run_command("/usr/libexec/qemu-server/query-machine-capabilities");
maybe once at boot through a systemd service file could be nicer than this a bit
subtle way.
> + }
> + my $json_text = PVE::Tools::file_get_contents($filename);
> + ($json_text) = $json_text =~ /(.*)/; # untaint json text
> + my $hw_capabilities;
> + eval {
> + $hw_capabilities = decode_json($json_text);
> + };
can be shortened to:
my $hw_capabilities = eval { decode_json($json_text) };
> + if (my $err = $@) {
> + die $err;
> + }
> + return $hw_capabilities;
> +}
> +
> +sub get_amd_sev_object {
> + my ($amd_sev, $bios, $efidisk0) = @_;
> +
> + my $amd_sev_conf = PVE::JSONSchema::parse_property_string($sev_fmt,
> $amd_sev);
> + my $sev_hw_caps = get_hw_capabilities()->{'amd-sev'};
> +
> + if (!$sev_hw_caps->{'sev-support'}) {
> + die "Your CPU does not support AMD SEV.\n";
> + }
> + if ($amd_sev_conf->{type} eq 'es' && !$sev_hw_caps->{'sev-support-es'}) {
> + die "Your CPU does not support AMD SEV-ES.\n";
> + }
> + if (!$bios || ($bios && $bios ne 'ovmf')) {
In general `!a || (a && b) == !a || b` so above can be shortened to
if (!$bios || $bios ne 'ovmf') {
# ...
> + die "To use AMD SEV, you need to change the BIOS to OVMF.\n";
> + }
> +
> + my $sev_mem_object = 'sev-guest,id=sev0';
> + $sev_mem_object .= ',cbitpos='.$sev_hw_caps->{cbitpos};
> + $sev_mem_object .=
> ',reduced-phys-bits='.$sev_hw_caps->{'reduced-phys-bits'};
> +
> + # guest policy bit calculation as described here:
> + #
> https://documentation.suse.com/sles/15-SP5/html/SLES-amd-sev/article-amd-sev.html#table-guestpolicy
> + my $policy = 0b0000;
> + $policy += 0b0001 if $amd_sev_conf->{'no-debug'};
> + $policy += 0b0010 if $amd_sev_conf->{'no-key-sharing'};
> + $policy += 0b0100 if $amd_sev_conf->{type} eq 'es';
> + # disable migration with bit 3 nosend to prevent amd-sev-migration-attack
> + $policy += 0b1000;
> +
> + $sev_mem_object .= ',policy='.sprintf("%#x", $policy);
> + $sev_mem_object .= ',kernel-hashes=on' if
> ($amd_sev_conf->{'kernel-hashes'});
> + return $sev_mem_object;
> +}
> +
> __PACKAGE__->register();
> __PACKAGE__->init();
>
_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
