tcp/dccp: fix hashdance race for passive sessions
Multiple cpus can process duplicates of incoming ACK messages
matching a SYN_RECV request socket. This is a rare event under
normal operations, but definitely can happen.
Only one must win the race, otherwise corruption would occur.
To fix this without adding new atomic ops, we use logic in
inet_ehash_nolisten() to detect the request was present in the same
ehash bucket where we try to insert the new child.
If request socket was not found, we have to undo the child creation.
This actually removes a spin_lock()/spin_unlock() pair in
reqsk_queue_unlink() for the fast path.
Fixes: e994b2f0fb92 ("tcp: do not lock listener to process SYN packets")
Fixes: 079096f103fa ("tcp/dccp: install syn_recv requests into ehash table")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index d10aace..1994f8a 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -143,6 +143,7 @@
{
struct sock *child = NULL;
struct dccp_request_sock *dreq = dccp_rsk(req);
+ bool own_req;
/* Check for retransmitted REQUEST */
if (dccp_hdr(skb)->dccph_type == DCCP_PKT_REQUEST) {
@@ -182,14 +183,13 @@
if (dccp_parse_options(sk, dreq, skb))
goto drop;
- child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL);
- if (child == NULL)
+ child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL,
+ req, &own_req);
+ if (!child)
goto listen_overflow;
- inet_csk_reqsk_queue_drop(sk, req);
- inet_csk_reqsk_queue_add(sk, req, child);
-out:
- return child;
+ return inet_csk_complete_hashdance(sk, child, req, own_req);
+
listen_overflow:
dccp_pr_debug("listen_overflow!\n");
DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_TOO_BUSY;
@@ -198,7 +198,7 @@
req->rsk_ops->send_reset(sk, skb);
inet_csk_reqsk_queue_drop(sk, req);
- goto out;
+ return NULL;
}
EXPORT_SYMBOL_GPL(dccp_check_req);