[pve-devel] [PATCH v4 container 1/1] Add device passthrough
Filip Schauer
f.schauer at proxmox.com
Mon Nov 13 11:30:37 CET 2023
Add a dev[n] argument to the container config to pass devices through to
a container. A device can be passed by its path. Additionally the access
mode, uid and gid can be specified through their respective properties.
Signed-off-by: Filip Schauer <f.schauer at proxmox.com>
---
src/PVE/LXC.pm | 57 +++++++++++++++++++++++-
src/PVE/LXC/Config.pm | 93 +++++++++++++++++++++++++++++++++++++++
src/PVE/LXC/Tools.pm | 23 +++++++---
src/lxc-pve-autodev-hook | 18 +++++++-
src/lxc-pve-prestart-hook | 60 ++++++++++++++++++++++++-
5 files changed, 242 insertions(+), 9 deletions(-)
diff --git a/src/PVE/LXC.pm b/src/PVE/LXC.pm
index 7ec816b..242b2a0 100644
--- a/src/PVE/LXC.pm
+++ b/src/PVE/LXC.pm
@@ -5,7 +5,7 @@ use warnings;
use Cwd qw();
use Errno qw(ELOOP ENOTDIR EROFS ECONNREFUSED EEXIST);
-use Fcntl qw(O_RDONLY O_WRONLY O_NOFOLLOW O_DIRECTORY);
+use Fcntl qw(O_RDONLY O_WRONLY O_NOFOLLOW O_DIRECTORY :mode);
use File::Path;
use File::Spec;
use IO::Poll qw(POLLIN POLLHUP);
@@ -639,6 +639,31 @@ sub update_lxc_config {
$raw .= "lxc.mount.auto = sys:mixed\n";
}
+ PVE::LXC::Config->foreach_passthrough_device($conf, sub {
+ my ($key, $device) = @_;
+
+ die "Path is not defined for passthrough device $key"
+ unless (defined($device->{path}));
+
+ my $absolute_path = $device->{path};
+
+ die "Device $absolute_path does not exist\n"
+ unless (-e $absolute_path);
+
+ die "$absolute_path is not a device\n"
+ unless (-b $absolute_path || -c $absolute_path);
+
+ my ($mode, $rdev) = (stat($absolute_path))[2, 6];
+
+ die "Could not get mode or device ID of $absolute_path\n"
+ if (!defined($mode) || !defined($rdev));
+
+ my $major = PVE::Tools::dev_t_major($rdev);
+ my $minor = PVE::Tools::dev_t_minor($rdev);
+ my $device_type_char = S_ISBLK($mode) ? 'b' : 'c';
+ $raw .= "lxc.cgroup2.devices.allow = $device_type_char $major:$minor rw\n";
+ });
+
# WARNING: DO NOT REMOVE this without making sure that loop device nodes
# cannot be exposed to the container with r/w access (cgroup perms).
# When this is enabled mounts will still remain in the monitor's namespace
@@ -1318,6 +1343,8 @@ sub check_ct_modify_config_perm {
$rpcenv->check_vm_perm($authuser, $vmid, $pool, ['VM.Config.Network']);
check_bridge_access($rpcenv, $authuser, $oldconf->{$opt}) if $oldconf->{$opt};
check_bridge_access($rpcenv, $authuser, $newconf->{$opt}) if $newconf->{$opt};
+ } elsif ($opt =~ m/^dev\d+$/) {
+ raise_perm_exc("configuring device passthrough is only allowed for root\@pam");
} elsif ($opt eq 'nameserver' || $opt eq 'searchdomain' || $opt eq 'hostname') {
$rpcenv->check_vm_perm($authuser, $vmid, $pool, ['VM.Config.Network']);
} elsif ($opt eq 'features') {
@@ -2367,6 +2394,34 @@ sub validate_id_maps {
}
}
+sub map_ct_id_to_host {
+ my ($id, $id_map, $id_type) = @_;
+
+ foreach my $mapping (@$id_map) {
+ my ($type, $ct, $host, $length) = @$mapping;
+
+ next if ($type ne $id_type);
+
+ if ($id >= int($ct) && $id < ($ct + $length)) {
+ return $host - $ct + $id;
+ }
+ }
+
+ return $id;
+}
+
+sub map_ct_uid_to_host {
+ my ($uid, $id_map) = @_;
+
+ return map_ct_id_to_host($uid, $id_map, 'u');
+}
+
+sub map_ct_gid_to_host {
+ my ($gid, $id_map) = @_;
+
+ return map_ct_id_to_host($gid, $id_map, 'g');
+}
+
sub userns_command {
my ($id_map) = @_;
if (@$id_map) {
diff --git a/src/PVE/LXC/Config.pm b/src/PVE/LXC/Config.pm
index 56e1f10..9f325f2 100644
--- a/src/PVE/LXC/Config.pm
+++ b/src/PVE/LXC/Config.pm
@@ -29,6 +29,7 @@ mkdir $lockdir;
mkdir "/etc/pve/nodes/$nodename/lxc";
my $MAX_MOUNT_POINTS = 256;
my $MAX_UNUSED_DISKS = $MAX_MOUNT_POINTS;
+my $MAX_DEVICES = 256;
# BEGIN implemented abstract methods from PVE::AbstractConfig
@@ -908,6 +909,71 @@ for (my $i = 0; $i < $MAX_UNUSED_DISKS; $i++) {
}
}
+PVE::JSONSchema::register_format('pve-lxc-dev-string', \&verify_lxc_dev_string);
+sub verify_lxc_dev_string {
+ my ($dev, $noerr) = @_;
+
+ if (
+ $dev =~ m@/\.\.?/@ ||
+ $dev =~ m@/\.\.?$@ ||
+ $dev !~ m!^/dev/!
+ ) {
+ return undef if $noerr;
+ die "$dev is not a valid device path\n";
+ }
+
+ return $dev;
+}
+
+PVE::JSONSchema::register_format('file-access-mode-string', \&verify_file_access_mode);
+sub verify_file_access_mode {
+ my ($mode, $noerr) = @_;
+
+ if ($mode !~ /^[0-7]*$/) {
+ return undef if $noerr;
+ die "$mode is not a valid file access mode\n";
+ }
+
+ return $mode;
+}
+
+my $dev_desc = {
+ path => {
+ optional => 1,
+ type => 'string',
+ default_key => 1,
+ format => 'pve-lxc-dev-string',
+ format_description => 'Path',
+ description => 'Device to pass through to the container',
+ verbose_description => 'Path to the device to pass through to the container',
+ },
+ mode => {
+ optional => 1,
+ type => 'integer',
+ format => 'file-access-mode-string',
+ format_description => 'Octal access mode',
+ description => 'Access mode to be set on the device node',
+ },
+ uid => {
+ optional => 1,
+ type => 'integer',
+ description => 'User ID to be assigned to the device node',
+ },
+ gid => {
+ optional => 1,
+ type => 'integer',
+ description => 'Group ID to be assigned to the device node',
+ },
+};
+
+for (my $i = 0; $i < $MAX_DEVICES; $i++) {
+ $confdesc->{"dev$i"} = {
+ optional => 1,
+ type => 'string', format => $dev_desc,
+ description => "Device to pass through to the container",
+ }
+}
+
sub parse_pct_config {
my ($filename, $raw, $strict) = @_;
@@ -1255,6 +1321,21 @@ sub parse_volume {
return;
}
+sub parse_device {
+ my ($class, $device_string, $noerr) = @_;
+
+ my $res;
+ eval { $res = PVE::JSONSchema::parse_property_string($dev_desc, $device_string) };
+ if ($@) {
+ return undef if $noerr;
+ die $@;
+ }
+
+ die "Path has to be defined" unless (defined($res->{path}));
+
+ return $res;
+}
+
sub print_volume {
my ($class, $key, $volume) = @_;
@@ -1762,4 +1843,16 @@ sub get_derived_property {
}
}
+sub foreach_passthrough_device {
+ my ($class, $conf, $func, @param) = @_;
+
+ foreach my $key (keys %$conf) {
+ next if $key !~ m/^dev(\d+)$/;
+
+ my $device = $class->parse_device($conf->{$key});
+
+ $func->($key, $device, @param);
+ }
+}
+
1;
diff --git a/src/PVE/LXC/Tools.pm b/src/PVE/LXC/Tools.pm
index 62cdbc1..c9fb834 100644
--- a/src/PVE/LXC/Tools.pm
+++ b/src/PVE/LXC/Tools.pm
@@ -81,10 +81,9 @@ sub lxc_hook($$&) {
$code->($ct_name, $common_vars, $namespaces, $args);
}
-sub for_current_devices($&) {
- my ($vmid, $code) = @_;
+sub for_devices {
+ my ($devlist_file, $vmid, $code) = @_;
- my $devlist_file = "/var/lib/lxc/$vmid/devices";
my $fd;
if (! open $fd, '<', $devlist_file) {
@@ -93,8 +92,8 @@ sub for_current_devices($&) {
}
while (defined(my $line = <$fd>)) {
- if ($line !~ m@^(b):(\d+):(\d+):/dev/(\S+)\s*$@) {
- warn "invalid .pve-devices entry: $line\n";
+ if ($line !~ m@^(b|c):(\d+):(\d+):/dev/(\S+)\s*$@) {
+ warn "invalid $devlist_file entry: $line\n";
next;
}
@@ -117,6 +116,20 @@ sub for_current_devices($&) {
close $fd;
}
+sub for_current_devices($&) {
+ my ($vmid, $code) = @_;
+
+ my $devlist_file = "/var/lib/lxc/$vmid/devices";
+ for_devices($devlist_file, $vmid, $code);
+}
+
+sub for_passthrough_devices($&) {
+ my ($vmid, $code) = @_;
+
+ my $passthrough_devlist_file = "/var/lib/lxc/$vmid/passthrough_devices";
+ for_devices($passthrough_devlist_file, $vmid, $code);
+}
+
sub cgroup_do_write($$) {
my ($path, $value) = @_;
my $fd;
diff --git a/src/lxc-pve-autodev-hook b/src/lxc-pve-autodev-hook
index 3c45949..105465f 100755
--- a/src/lxc-pve-autodev-hook
+++ b/src/lxc-pve-autodev-hook
@@ -3,17 +3,31 @@
use strict;
use warnings;
-use File::Path;
+use Fcntl qw(S_IFREG);
use File::Basename;
+use File::Path;
use PVE::LXC::Tools;
-use PVE::Tools;
+use PVE::Tools qw(MS_BIND);
PVE::LXC::Tools::lxc_hook('autodev', 'lxc', sub {
my ($vmid, $vars, undef, undef) = @_;
my $root = $vars->{ROOTFS_MOUNT};
+ PVE::LXC::Tools::for_passthrough_devices($vmid, sub {
+ my ($type, $major, $minor, $dev) = @_;
+
+ my $rel_devpath = "/dev/$dev";
+ my $rel_dir = dirname($rel_devpath);
+ File::Path::mkpath("$root/$rel_dir");
+ PVE::Tools::mknod("$root/dev/$dev", S_IFREG, 0)
+ or die("Could not mknod $root/dev/$dev");
+
+ PVE::Tools::mount("/var/lib/lxc/$vmid/passthrough/dev/$dev", "$root/dev/$dev", 0, MS_BIND, 0)
+ or die("Bind mount of device $dev into container failed\n");
+ });
+
PVE::LXC::Tools::for_current_devices($vmid, sub {
my ($type, $major, $minor, $dev) = @_;
diff --git a/src/lxc-pve-prestart-hook b/src/lxc-pve-prestart-hook
index 936d0bf..d8b0106 100755
--- a/src/lxc-pve-prestart-hook
+++ b/src/lxc-pve-prestart-hook
@@ -6,6 +6,7 @@ use strict;
use warnings;
use Fcntl qw(O_DIRECTORY :mode);
+use File::Basename;
use File::Path;
use POSIX;
@@ -62,7 +63,7 @@ PVE::LXC::Tools::lxc_hook('pre-start', 'lxc', sub {
unlink $devlist_file;
my $devices = [];
- my (undef, $rootuid, $rootgid) = PVE::LXC::parse_id_maps($conf);
+ my ($id_map, $rootuid, $rootgid) = PVE::LXC::parse_id_maps($conf);
# Unmount first when the user mounted the container with "pct mount".
eval {
@@ -118,6 +119,51 @@ PVE::LXC::Tools::lxc_hook('pre-start', 'lxc', sub {
PVE::LXC::Config->foreach_volume($conf, $setup_mountpoint);
+ # Device passthrough
+ my $passthrough_devlist_file = "/var/lib/lxc/$vmid/passthrough_devices";
+ unlink $passthrough_devlist_file;
+ my $passthrough_devices = [];
+
+ my $passthrough_dir = "/var/lib/lxc/$vmid/passthrough";
+ File::Path::make_path($passthrough_dir);
+ PVE::Tools::mount("none", $passthrough_dir, "tmpfs", 0, "size=8k")
+ or die ("Could not mount tmpfs for device passthrough at $passthrough_dir");
+
+ my $setup_passthrough_device = sub {
+ my ($key, $device) = @_;
+
+ my $absolute_path = $device->{path};
+ my ($mode, $rdev) = (stat($absolute_path))[2, 6];
+
+ die "Could not get mode or device ID of $absolute_path\n"
+ if (!defined($mode) || !defined($rdev));
+
+ my $passthrough_device_path = $passthrough_dir . $absolute_path;
+ File::Path::make_path(dirname($passthrough_device_path));
+ PVE::Tools::mknod($passthrough_device_path, $mode, $rdev)
+ or die("failed to mknod $passthrough_device_path\n");
+
+ # Use chmod because umask could mess with the access mode on mknod
+ my $passthrough_mode = 0660;
+ $passthrough_mode = oct($device->{mode}) if (defined($device->{mode}));
+ chmod $passthrough_mode, $passthrough_device_path
+ or die "failed to chmod $passthrough_mode $passthrough_device_path: $!\n";
+
+ # Set uid and gid of the device node
+ my $uid = 0;
+ my $gid = 0;
+ $uid = $device->{uid} if (defined($device->{uid}));
+ $gid = $device->{gid} if (defined($device->{gid}));
+ $uid = PVE::LXC::map_ct_uid_to_host($uid, $id_map);
+ $gid = PVE::LXC::map_ct_gid_to_host($gid, $id_map);
+ chown $uid, $gid, $passthrough_device_path
+ or die("failed to chown $uid:$gid $passthrough_device_path: $!\n");
+
+ push @$passthrough_devices, [$absolute_path, $mode, $rdev];
+ };
+
+ PVE::LXC::Config->foreach_passthrough_device($conf, $setup_passthrough_device);
+
my $lxc_setup = PVE::LXC::Setup->new($conf, $rootdir);
$lxc_setup->pre_start_hook();
@@ -140,6 +186,18 @@ PVE::LXC::Tools::lxc_hook('pre-start', 'lxc', sub {
}
PVE::Tools::file_set_contents($devlist_file, $devlist);
}
+
+ if (@$passthrough_devices) {
+ my $devlist = '';
+ foreach my $dev (@$passthrough_devices) {
+ my ($path, $mode, $rdev) = @$dev;
+ my $major = PVE::Tools::dev_t_major($rdev);
+ my $minor = PVE::Tools::dev_t_minor($rdev);
+ my $device_type_char = S_ISBLK($mode) ? 'b' : 'c';
+ $devlist .= "$device_type_char:$major:$minor:$path\n";
+ }
+ PVE::Tools::file_set_contents($passthrough_devlist_file, $devlist);
+ }
});
# Leftover cgroups prevent lxc from starting without any useful information
--
2.39.2
More information about the pve-devel
mailing list