[NET]: Size listen hash tables using backlog hint

We currently allocate a fixed size (TCP_SYNQ_HSIZE=512) slots hash table for
each LISTEN socket, regardless of various parameters (listen backlog for
example)

On x86_64, this means order-1 allocations (might fail), even for 'small'
sockets, expecting few connections. On the contrary, a huge server wanting a
backlog of 50000 is slowed down a bit because of this fixed limit.

This patch makes the sizing of listen hash table a dynamic parameter,
depending of :
- net.core.somaxconn tunable (default is 128)
- net.ipv4.tcp_max_syn_backlog tunable (default : 256, 1024 or 128)
- backlog value given by user application  (2nd parameter of listen())

For large allocations (bigger than PAGE_SIZE), we use vmalloc() instead of
kmalloc().

We still limit memory allocation with the two existing tunables (somaxconn &
tcp_max_syn_backlog). So for standard setups, this patch actually reduce RAM
usage.

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 5 insertions, 5 deletions

diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index edcf0932ac6d..4a81d54a7569 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -204,7 +204,7 @@ int inet_listen(struct socket *sock, int backlog)
          * we can only allow the backlog to be adjusted.
          */
         if (old_state != TCP_LISTEN) {
-                err = inet_csk_listen_start(sk, TCP_SYNQ_HSIZE);
+                err = inet_csk_listen_start(sk, backlog);
                 if (err)
                         goto out;
         }
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 96bbe2a0aa1b..9d68837888d3 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -343,7 +343,7 @@ struct dst_entry* inet_csk_route_req(struct sock *sk,
 EXPORT_SYMBOL_GPL(inet_csk_route_req);
 
 static inline u32 inet_synq_hash(const __be32 raddr, const __be16 rport,
-                                 const u32 rnd, const u16 synq_hsize)
+                                 const u32 rnd, const u32 synq_hsize)
 {
         return jhash_2words((__force u32)raddr, (__force u32)rport, rnd) & (synq_hsize - 1);
 }
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 22ef8bd26620..5fbf96552cac 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -715,7 +715,7 @@ static struct ip_options *tcp_v4_save_options(struct sock *sk,
         return dopt;
 }
 
-struct request_sock_ops tcp_request_sock_ops = {
+struct request_sock_ops tcp_request_sock_ops __read_mostly = {
         .family         =       PF_INET,
         .obj_size       =       sizeof(struct tcp_request_sock),
         .rtx_syn_ack    =       tcp_v4_send_synack,
@@ -1385,7 +1385,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur)
         if (st->state == TCP_SEQ_STATE_OPENREQ) {
                 struct request_sock *req = cur;
 
-                icsk = inet_csk(st->syn_wait_sk);
+                icsk = inet_csk(st->syn_wait_sk);
                 req = req->dl_next;
                 while (1) {
                         while (req) {
@@ -1395,7 +1395,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur)
                                 }
                                 req = req->dl_next;
                         }
-                        if (++st->sbucket >= TCP_SYNQ_HSIZE)
+                        if (++st->sbucket >= icsk->icsk_accept_queue.listen_opt->nr_table_entries)
                                 break;
 get_req:
                         req = icsk->icsk_accept_queue.listen_opt->syn_table[st->sbucket];