setns init: delay seccomp as late as possible

This mirrors the standard_init_linux.go seccomp code, which only applies
seccomp early if NoNewPrivileges is enabled. Otherwise it's done
immediately before execve to reduce the amount of syscalls necessary for
users to enable in their seccomp profiles.

Signed-off-by: Aleksa Sarai <[email protected]>

Author: cyphar

libcontainer/setns_init_linux.go

@@ -47,7 +47,10 @@ func (l *linuxSetnsInit) Init() error {
 			return err
 		}
 	}
-	if l.config.Config.Seccomp != nil {
+	// Without NoNewPrivileges seccomp is a privileged operation, so we need to
+	// do this before dropping capabilities; otherwise do it as late as possible
+	// just before execve so as few syscalls take place after it as possible.
+	if l.config.Config.Seccomp != nil && !l.config.NoNewPrivileges {
 		if err := seccomp.InitSeccomp(l.config.Config.Seccomp); err != nil {
 			return err
 		}
@@ -61,5 +64,13 @@ func (l *linuxSetnsInit) Init() error {
 	if err := label.SetProcessLabel(l.config.ProcessLabel); err != nil {
 		return err
 	}
+	// Set seccomp as close to execve as possible, so as few syscalls take
+	// place afterward (reducing the amount of syscalls that users need to
+	// enable in their seccomp profiles).
+	if l.config.Config.Seccomp != nil && l.config.NoNewPrivileges {
+		if err := seccomp.InitSeccomp(l.config.Config.Seccomp); err != nil {
+			return newSystemErrorWithCause(err, "init seccomp")
+		}
+	}
 	return system.Execv(l.config.Args[0], l.config.Args[0:], os.Environ())
 }