add sysctl to allow disabling unprivileged CLONE_NEWUSER

This is a short-term patch.  Unprivileged use of CLONE_NEWUSER
is certainly an intended feature of user namespaces.  However
for at least saucy we want to make sure that, if any security
issues are found, we have a fail-safe.

[bwh: Remove unneeded binary sysctl bits]
[bwh: Keep this sysctl, but change the default to enabled]
[heftig: correct commit subject to reduce confusion]
[heftig: for 6.17, move all code into kernel/fork.c]

Author: hallyn

kernel/fork.c

@@ -127,6 +127,12 @@
 
 #include 
 
+#ifdef CONFIG_USER_NS
+static int unprivileged_userns_clone = 1;
+#else
+#define unprivileged_userns_clone 1
+#endif
+
 /*
  * Minimum number of threads to boot the kernel
  */
@@ -2093,6 +2099,11 @@ __latent_entropy struct task_struct *copy_process(
 			return ERR_PTR(-EPERM);
 	}
 
+	if ((clone_flags & CLONE_NEWUSER) && !unprivileged_userns_clone) {
+		if (!capable(CAP_SYS_ADMIN))
+			return ERR_PTR(-EPERM);
+	}
+
 	/*
 	 * Force any signals received before this point to be delivered
 	 * before the fork happens.  Collect up signals sent to multiple
@@ -3163,6 +3174,10 @@ static int check_unshare_flags(unsigned long unshare_flags)
 		if (!current_is_single_threaded())
 			return -EINVAL;
 	}
+	if ((unshare_flags & CLONE_NEWUSER) && !unprivileged_userns_clone) {
+		if (!capable(CAP_SYS_ADMIN))
+			return -EPERM;
+	}
 
 	return 0;
 }
@@ -3398,6 +3413,15 @@ static const struct ctl_table fork_sysctl_table[] = {
 		.mode		= 0644,
 		.proc_handler	= sysctl_max_threads,
 	},
+#ifdef CONFIG_USER_NS
+	{
+		.procname	= "unprivileged_userns_clone",
+		.data		= &unprivileged_userns_clone,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+#endif
 };
 
 static int __init init_fork_sysctl(void)