[TCP]: Restrict congestion control choices.

Allow normal users to only choose among a restricted set of congestion
control choices.  The default is reno and what ever has been configured
as default. But the policy can be changed by administrator at any time.

For example, to allow any choice:
    cp /proc/sys/net/ipv4/tcp_available_congestion_control \
       /proc/sys/net/ipv4/tcp_allowed_congestion_control

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

5 files changed, 125 insertions, 0 deletions

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index db428085658..bbcc8deda17 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -351,6 +351,12 @@ tcp_frto - BOOLEAN
 	where packet loss is typically due to random radio interference
 	rather than intermediate router congestion.
 
+tcp_allowed_congestion_control - STRING
+	Show/set the congestion control choices available to non-privileged
+	processes. The list is a subset of those listed in
+	tcp_available_congestion_control.
+	Default is "reno" and the default setting (tcp_congestion_control).
+
 tcp_available_congestion_control - STRING
 	Shows the available congestion control choices that are registered.
 	More congestion control algorithms may be available as modules,
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 28a48279654..0725441621d 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -427,6 +427,7 @@ enum
 	NET_CIPSOV4_RBM_OPTFMT=120,
 	NET_CIPSOV4_RBM_STRICTVALID=121,
 	NET_TCP_AVAIL_CONG_CONTROL=122,
+	NET_TCP_ALLOWED_CONG_CONTROL=123,
 };
 
 enum {
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 6af4baf5b76..e1a5d29d0a1 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -625,6 +625,7 @@ enum tcp_ca_event {
 
 struct tcp_congestion_ops {
 	struct list_head	list;
+	int	non_restricted;
 
 	/* initialize private data (optional) */
 	void (*init)(struct sock *sk);
@@ -663,6 +664,8 @@ extern void tcp_cleanup_congestion_control(struct sock *sk);
 extern int tcp_set_default_congestion_control(const char *name);
 extern void tcp_get_default_congestion_control(char *name);
 extern void tcp_get_available_congestion_control(char *buf, size_t len);
+extern void tcp_get_allowed_congestion_control(char *buf, size_t len);
+extern int tcp_set_allowed_congestion_control(char *allowed);
 extern int tcp_set_congestion_control(struct sock *sk, const char *name);
 extern void tcp_slow_start(struct tcp_sock *tp);
 
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 2e770f45d82..dfcf47f10f8 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -146,6 +146,50 @@ static int proc_tcp_available_congestion_control(ctl_table *ctl,
 	return ret;
 }
 
+static int proc_allowed_congestion_control(ctl_table *ctl,
+					   int write, struct file * filp,
+					   void __user *buffer, size_t *lenp,
+					   loff_t *ppos)
+{
+	ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX };
+	int ret;
+
+	tbl.data = kmalloc(tbl.maxlen, GFP_USER);
+	if (!tbl.data)
+		return -ENOMEM;
+
+	tcp_get_allowed_congestion_control(tbl.data, tbl.maxlen);
+	ret = proc_dostring(&tbl, write, filp, buffer, lenp, ppos);
+	if (write && ret == 0)
+		ret = tcp_set_allowed_congestion_control(tbl.data);
+	kfree(tbl.data);
+	return ret;
+}
+
+static int strategy_allowed_congestion_control(ctl_table *table, int __user *name,
+					       int nlen, void __user *oldval,
+					       size_t __user *oldlenp,
+					       void __user *newval, size_t newlen,
+					       void **context)
+{
+	ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX };
+	int ret;
+
+	tbl.data = kmalloc(tbl.maxlen, GFP_USER);
+	if (!tbl.data)
+		return -ENOMEM;
+
+	tcp_get_available_congestion_control(tbl.data, tbl.maxlen);
+	ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen,
+			    context);
+	if (ret == 0 && newval && newlen)
+		ret = tcp_set_allowed_congestion_control(tbl.data);
+	kfree(tbl.data);
+
+	return ret;
+
+}
+
 ctl_table ipv4_table[] = {
         {
 		.ctl_name	= NET_IPV4_TCP_TIMESTAMPS,
@@ -755,6 +799,14 @@ ctl_table ipv4_table[] = {
 		.mode		= 0444,
 		.proc_handler   = &proc_tcp_available_congestion_control,
 	},
+	{
+		.ctl_name	= NET_TCP_ALLOWED_CONG_CONTROL,
+		.procname	= "tcp_allowed_congestion_control",
+		.maxlen		= TCP_CA_BUF_MAX,
+		.mode		= 0644,
+		.proc_handler   = &proc_allowed_congestion_control,
+		.strategy	= &strategy_allowed_congestion_control,
+	},
 	{ .ctl_name = 0 }
 };
 
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index d846d7b95e1..343d6197c92 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -123,6 +123,7 @@ int tcp_set_default_congestion_control(const char *name)
 #endif
 
 	if (ca) {
+		ca->non_restricted = 1;	/* default is always allowed */
 		list_move(&ca->list, &tcp_cong_list);
 		ret = 0;
 	}
@@ -168,6 +169,64 @@ void tcp_get_default_congestion_control(char *name)
 	rcu_read_unlock();
 }
 
+/* Built list of non-restricted congestion control values */
+void tcp_get_allowed_congestion_control(char *buf, size_t maxlen)
+{
+	struct tcp_congestion_ops *ca;
+	size_t offs = 0;
+
+	*buf = '\0';
+	rcu_read_lock();
+	list_for_each_entry_rcu(ca, &tcp_cong_list, list) {
+		if (!ca->non_restricted)
+			continue;
+		offs += snprintf(buf + offs, maxlen - offs,
+				 "%s%s",
+				 offs == 0 ? "" : " ", ca->name);
+
+	}
+	rcu_read_unlock();
+}
+
+/* Change list of non-restricted congestion control */
+int tcp_set_allowed_congestion_control(char *val)
+{
+	struct tcp_congestion_ops *ca;
+	char *clone, *name;
+	int ret = 0;
+
+	clone = kstrdup(val, GFP_USER);
+	if (!clone)
+		return -ENOMEM;
+
+	spin_lock(&tcp_cong_list_lock);
+	/* pass 1 check for bad entries */
+	while ((name = strsep(&clone, " ")) && *name) {
+		ca = tcp_ca_find(name);
+		if (!ca) {
+			ret = -ENOENT;
+			goto out;
+		}
+	}
+
+	/* pass 2 clear */
+	list_for_each_entry_rcu(ca, &tcp_cong_list, list)
+		ca->non_restricted = 0;
+
+	/* pass 3 mark as allowed */
+	while ((name = strsep(&val, " ")) && *name) {
+		ca = tcp_ca_find(name);
+		WARN_ON(!ca);
+		if (ca)
+			ca->non_restricted = 1;
+	}
+out:
+	spin_unlock(&tcp_cong_list_lock);
+
+	return ret;
+}
+
+
 /* Change congestion control for socket */
 int tcp_set_congestion_control(struct sock *sk, const char *name)
 {
@@ -183,6 +242,9 @@ int tcp_set_congestion_control(struct sock *sk, const char *name)
 	if (!ca)
 		err = -ENOENT;
 
+	else if (!(ca->non_restricted || capable(CAP_NET_ADMIN)))
+		err = -EPERM;
+
 	else if (!try_module_get(ca->owner))
 		err = -EBUSY;
 
@@ -284,6 +346,7 @@ EXPORT_SYMBOL_GPL(tcp_reno_min_cwnd);
 
 struct tcp_congestion_ops tcp_reno = {
 	.name		= "reno",
+	.non_restricted = 1,
 	.owner		= THIS_MODULE,
 	.ssthresh	= tcp_reno_ssthresh,
 	.cong_avoid	= tcp_reno_cong_avoid,