Sometimes, we want to know how many tcp sockets are in use
different _net_ namespaces. It's a key resource metric. With
this patch, we can get it via /proc/net/sockstat.
The 'alloc' show in /proc/net/sockstat is the total tcp
sockets in the kernel. This patch moves it to namespace,
via adding a counter because the previous counter is used
in proto(e.g tcp, udp and sctp) memory management.
Signed-off-by: Tonghao Zhang <xiangxia.m....@gmail.com>
---
include/net/netns/ipv4.h | 3 +++
include/net/tcp.h | 2 ++
net/ipv4/proc.c | 2 +-
net/ipv4/tcp.c | 2 ++
net/ipv4/tcp_ipv4.c | 34 ++++++++++++++++++++++++++++++++++
net/ipv4/tcp_minisocks.c | 3 +++
6 files changed, 45 insertions(+), 1 deletion(-)
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 44668c2..85e91bd 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -68,6 +68,9 @@ struct netns_ipv4 {
struct inet_peer_base *peers;
struct sock * __percpu *tcp_sk;
+#ifdef CONFIG_PROC_FS
+ int __percpu *tcp_sock_allocated;
+#endif
struct netns_frags frags;
#ifdef CONFIG_NETFILTER
struct xt_table *iptable_filter;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 093e967..4b24b6e 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1780,6 +1780,8 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb,
int tcp_gro_complete(struct sk_buff *skb);
void __tcp_v4_send_check(struct sk_buff *skb, __be32 saddr, __be32 daddr);
+void tcp_sock_allocated_add(struct net *net, int val);
+int tcp_sock_allocated_get(struct net *net);
static inline u32 tcp_notsent_lowat(const struct tcp_sock *tp)
{
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index dc5edc8..8a147f7 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -58,7 +58,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
int orphans, sockets;
orphans = percpu_counter_sum_positive(&tcp_orphan_count);
- sockets = proto_sockets_allocated_sum_positive(&tcp_prot);
+ sockets = tcp_sock_allocated_get(net);
socket_seq_show(seq);
seq_printf(seq, "TCP: inuse %d orphan %d tw %d alloc %d mem %ld\n",
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 874c931..77fe4a5 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -453,6 +453,8 @@ void tcp_init_sock(struct sock *sk)
sk->sk_rcvbuf = sock_net(sk)->ipv4.sysctl_tcp_rmem[1];
sk_sockets_allocated_inc(sk);
+ if (likely(sk->sk_net_refcnt))
+ tcp_sock_allocated_add(sock_net(sk), 1);
}
EXPORT_SYMBOL(tcp_init_sock);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 95738aa..a7bd0c4 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1928,6 +1928,8 @@ void tcp_v4_destroy_sock(struct sock *sk)
tcp_saved_syn_free(tp);
sk_sockets_allocated_dec(sk);
+ if (likely(sk->sk_net_refcnt))
+ tcp_sock_allocated_add(sock_net(sk), -1);
}
EXPORT_SYMBOL(tcp_v4_destroy_sock);
@@ -2446,6 +2448,28 @@ struct proto tcp_prot = {
};
EXPORT_SYMBOL(tcp_prot);
+void tcp_sock_allocated_add(struct net *net, int val)
+{
+#ifdef CONFIG_PROC_FS
+ this_cpu_add(*net->ipv4.tcp_sock_allocated, val);
+#endif
+}
+EXPORT_SYMBOL(tcp_sock_allocated_add);
+
+int tcp_sock_allocated_get(struct net *net)
+{
+#ifdef CONFIG_PROC_FS
+ int cpu, res = 0;
+
+ for_each_possible_cpu(cpu)
+ res += *per_cpu_ptr(net->ipv4.tcp_sock_allocated, cpu);
+ return res;
+#else
+ return 0;
+#endif
+}
+EXPORT_SYMBOL(tcp_sock_allocated_get);
+
static void __net_exit tcp_sk_exit(struct net *net)
{
int cpu;
@@ -2455,6 +2479,10 @@ static void __net_exit tcp_sk_exit(struct net *net)
for_each_possible_cpu(cpu)
inet_ctl_sock_destroy(*per_cpu_ptr(net->ipv4.tcp_sk, cpu));
free_percpu(net->ipv4.tcp_sk);
+
+#ifdef CONFIG_PROC_FS
+ free_percpu(net->ipv4.tcp_sock_allocated);
+#endif
}
static int __net_init tcp_sk_init(struct net *net)
@@ -2465,6 +2493,12 @@ static int __net_init tcp_sk_init(struct net *net)
if (!net->ipv4.tcp_sk)
return -ENOMEM;
+#ifdef CONFIG_PROC_FS
+ net->ipv4.tcp_sock_allocated = alloc_percpu(int);
+ if (!net->ipv4.tcp_sock_allocated)
+ goto fail;
+#endif
+
for_each_possible_cpu(cpu) {
struct sock *sk;
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index a8384b0c..573fe43 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -559,6 +559,9 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
newtp->rack.reo_wnd_persist = 0;
newtp->rack.dsack_seen = 0;
+ if (likely(newsk->sk_net_refcnt))
+ tcp_sock_allocated_add(sock_net(newsk), 1);
+
__TCP_INC_STATS(sock_net(sk), TCP_MIB_PASSIVEOPENS);
}
return newsk;
--
1.8.3.1
