Initial v2 resources.unified systemd support

In case systemd is used as cgroups manager, and a user sets some
resources using unified resource map (as per [1]), systemd is not
aware of any parameters, so there will be a discrepancy between
the cgroupfs state and systemd unit state.

Let's try to fix that by converting known unified resources to systemd
properties.

Currently, this is only implemented for pids.max as a POC.

Some other parameters (that might or might not have systemd unit
property equivalents) are:

$ ls -l | grep w-
-rw-r--r--. 1 root root 0 Oct 10 13:57 cgroup.freeze
-rw-r--r--. 1 root root 0 Oct 10 13:57 cgroup.max.depth
-rw-r--r--. 1 root root 0 Oct 10 13:57 cgroup.max.descendants
-rw-r--r--. 1 root root 0 Oct 10 13:57 cgroup.procs
-rw-r--r--. 1 root root 0 Oct 21 09:43 cgroup.subtree_control
-rw-r--r--. 1 root root 0 Oct 10 13:57 cgroup.threads
-rw-r--r--. 1 root root 0 Oct 10 13:57 cgroup.type
-rw-r--r--. 1 root root 0 Oct 22 10:30 cpu.max
-rw-r--r--. 1 root root 0 Oct 10 13:57 cpu.pressure
-rw-r--r--. 1 root root 0 Oct 22 10:30 cpuset.cpus
-rw-r--r--. 1 root root 0 Oct 22 10:30 cpuset.cpus.partition
-rw-r--r--. 1 root root 0 Oct 22 10:30 cpuset.mems
-rw-r--r--. 1 root root 0 Oct 22 10:30 cpu.weight
-rw-r--r--. 1 root root 0 Oct 22 10:30 cpu.weight.nice
-rw-r--r--. 1 root root 0 Oct 22 10:30 hugetlb.1GB.max
-rw-r--r--. 1 root root 0 Oct 22 10:30 hugetlb.1GB.rsvd.max
-rw-r--r--. 1 root root 0 Oct 22 10:30 hugetlb.2MB.max
-rw-r--r--. 1 root root 0 Oct 22 10:30 hugetlb.2MB.rsvd.max
-rw-r--r--. 1 root root 0 Oct 22 10:30 io.bfq.weight
-rw-r--r--. 1 root root 0 Oct 22 10:30 io.latency
-rw-r--r--. 1 root root 0 Oct 22 10:30 io.max
-rw-r--r--. 1 root root 0 Oct 10 13:57 io.pressure
-rw-r--r--. 1 root root 0 Oct 22 10:30 io.weight
-rw-r--r--. 1 root root 0 Oct 10 13:57 memory.high
-rw-r--r--. 1 root root 0 Oct 10 13:57 memory.low
-rw-r--r--. 1 root root 0 Oct 10 13:57 memory.max
-rw-r--r--. 1 root root 0 Oct 10 13:57 memory.min
-rw-r--r--. 1 root root 0 Oct 10 13:57 memory.oom.group
-rw-r--r--. 1 root root 0 Oct 10 13:57 memory.pressure
-rw-r--r--. 1 root root 0 Oct 10 13:57 memory.swap.high
-rw-r--r--. 1 root root 0 Oct 10 13:57 memory.swap.max

Surely, it is a manual conversion for every such case...

[1] https://github.com/opencontainers/runtime-spec/pull/1040

Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>

libcontainer/cgroups/systemd/v2.go

@@ -3,6 +3,8 @@
 package systemd
 
 import (
+	"fmt"
+	"math"
 	"os"
 	"path/filepath"
 	"strconv"
@@ -34,6 +36,51 @@ func NewUnifiedManager(config *configs.Cgroup, path string, rootless bool) cgrou
 	}
 }
 
+// unifiedResToSystemdProps tries to convert from Cgroup.Resources.Unified
+// key/value map (where key is cgroupfs file name) to systemd unit properties.
+// This is on a best-effort basis, so the properties that are not known
+// (to this function and/or systemd) are ignored (but logged with "debug"
+// log level).
+//
+// For the list of keys, see https://www.kernel.org/doc/Documentation/cgroup-v2.txt
+//
+// For the list of systemd unit properties, see systemd.resource-control(5).
+func unifiedResToSystemdProps(res map[string]string) (props []systemdDbus.Property, _ error) {
+	for k, v := range res {
+		if strings.Contains(k, "/") {
+			return nil, fmt.Errorf("unified resource %q must be a file name (no slashes)", k)
+		}
+		sk := strings.SplitN(k, ".", 2)
+		if len(sk) != 2 {
+			return nil, fmt.Errorf("unified resource %q must be in the form CONTROLLER.PARAMETER", k)
+		}
+		// Kernel is quite forgiving to extra whitespace
+		// around the value, and so should we.
+		v = strings.TrimSpace(v)
+		switch k {
+		case "pids.max":
+			num := uint64(math.MaxUint64)
+			if v != "max" {
+				var err error
+				num, err = strconv.ParseUint(v, 10, 64)
+				if err != nil {
+					return nil, fmt.Errorf("unified resource %q value conversion error: %w", k, err)
+				}
+			}
+			props = append(props,
+				newProp("TasksAccounting", true),
+				newProp("TasksMax", num))
+
+		default:
+			// Ignore the unknown resource here -- will still be
+			// applied in Set which calls fs2.Set.
+			logrus.Debugf("don't know how to convert unified resource %q=%q to systemd unit property; skipping (will still be applied to cgroupfs)", k, v)
+		}
+	}
+
+	return props, nil
+}
+
 func genV2ResourcesProperties(c *configs.Cgroup, conn *systemdDbus.Conn) ([]systemdDbus.Property, error) {
 	var properties []systemdDbus.Property
 	r := c.Resources
@@ -82,6 +129,15 @@ func genV2ResourcesProperties(c *configs.Cgroup, conn *systemdDbus.Conn) ([]syst
 
 	// ignore r.KernelMemory
 
+	// convert Resources.Unified map to systemd properties
+	if r.Unified != nil {
+		unifiedProps, err := unifiedResToSystemdProps(r.Unified)
+		if err != nil {
+			return nil, err
+		}
+		properties = append(properties, unifiedProps...)
+	}
+
 	return properties, nil
 }
 

tests/integration/cgroups.bats

@@ -216,6 +216,8 @@ function setup() {
 	echo "$output" | grep -q '^memory.max:10485760$'
 	echo "$output" | grep -q '^pids.max:99$'
 	echo "$output" | grep -q '^cpu.max:10000 100000$'
+
+	check_systemd_value "TasksMax" 99
 }
 
 @test "runc run (cgroup v2 resources.unified override)" {
@@ -226,7 +228,8 @@ function setup() {
 			| .linux.resources.memorySwap |= {"limit": 33554432}
 			| .linux.resources.unified |= {
 				"memory.min": "131072",
-				"memory.max": "10485760"
+				"memory.max": "10485760",
+				"pids.max": "42"
 			}' "$BUSYBOX_BUNDLE"
 
 	runc run -d --console-socket "$CONSOLE_SOCKET" test_cgroups_unified
@@ -239,4 +242,9 @@ function setup() {
 	runc exec test_cgroups_unified cat /sys/fs/cgroup/memory.max
 	[ "$status" -eq 0 ]
 	[ "$output" = '10485760' ]
+
+	runc exec test_cgroups_unified cat /sys/fs/cgroup/pids.max
+	[ "$status" -eq 0 ]
+	[ "$output" = '42' ]
+	check_systemd_value "TasksMax" 42
 }