[RFC] tcp: race in receive part

Hi,

in RHEL4 we can see a race in the tcp layer. We were not able to reproduce 
this on the upstream kernel, but since the issue occurs very rarelly
(once per 8 days), we just might not be lucky.

I'm affraid this might be a long email, I'll try to structure it nicely.. :)



RACE DESCRIPTION
================

There's a nice pdf describing the issue (and sollution using locks) on
https://bugzilla.redhat.com/attachment.cgi?id=345014


The race fires, when following code paths meet, and the tp->rcv_nxt and
__add_wait_queue updates stay in CPU caches.

CPU1                         CPU2


sys_select                   receive packet
  ...                        ...
  __add_wait_queue           update tp->rcv_nxt
  ...                        ...
  tp->rcv_nxt check          sock_def_readable
  ...                        {
  schedule                      ...
                                if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
                                        wake_up_interruptible(sk->sk_sleep)
                                ...
                             }

If there were no cache the code would work ok, since the wait_queue and
rcv_nxt are opposit to each other.

Meaning that once tp->rcv_nxt is updated by CPU2, the CPU1 either already
passed the tp->rcv_nxt check and sleeps, or will get the new value for
tp->rcv_nxt and will return with new data mask.  
In both cases the process (CPU1) is being added to the wait queue, so the
waitqueue_active (CPU2) call cannot miss and will wake up CPU1.

The bad case is when the __add_wait_queue changes done by CPU1 stay in its
cache , and so does the tp->rcv_nxt update on CPU2 side.  The CPU1 will then
endup calling schedule and sleep forever if there are no more data on the
socket.

Adding smp_mb() calls before sock_def_readable call and after __add_wait_queue
should prevent the above bad scenario.

The upstream patch is attached. It seems to prevent the issue.



CPU BUGS
========

The customer has been able to reproduce this problem only on one CPU model:
Xeon E5345*2. They didn't reproduce on XEON MV, for example.

That CPU model happens to have 2 possible issues, that might cause the issue:
(see errata http://www.intel.com/Assets/PDF/specupdate/315338.pdf)

AJ39 and AJ18. The first one can be workarounded by BIOS upgrade,
the other one has following notes:

      Software should ensure at least one of the following is true when
      modifying shared data by multiple agents:
             • The shared data is aligned
             • Proper semaphores or barriers are used in order to
                prevent concurrent data accesses.



RFC
===

I'm aware that not having this issue reproduced on upstream lowers the odds
having this checked in. However AFAICS the issue is present. I'd appreciate
any comment/ideas.


thanks,
jirka


Signed-off-by: Jiri Olsa <redacted>

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 17b89c5..f5d9dbf 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c

@@ -340,6 +340,11 @@ unsigned int tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
 	struct tcp_sock *tp = tcp_sk(sk);
 
 	poll_wait(file, sk->sk_sleep, wait);
+
+	/* Get in sync with tcp_data_queue, tcp_urg
+	   and tcp_rcv_established function. */
+	smp_mb();
+
 	if (sk->sk_state == TCP_LISTEN)
 		return inet_csk_listen_poll(sk);
 

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 2bdb0da..0606e5e 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c

@@ -4362,8 +4362,11 @@ queue_and_out:
 
 		if (eaten > 0)
 			__kfree_skb(skb);
-		else if (!sock_flag(sk, SOCK_DEAD))
+		else if (!sock_flag(sk, SOCK_DEAD)) {
+			/* Get in sync with tcp_poll function. */
+			smp_mb();
 			sk->sk_data_ready(sk, 0);
+		}
 		return;
 	}
 

@@ -4967,8 +4970,11 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, struct tcphdr *th)
 			if (skb_copy_bits(skb, ptr, &tmp, 1))
 				BUG();
 			tp->urg_data = TCP_URG_VALID | tmp;
-			if (!sock_flag(sk, SOCK_DEAD))
+			if (!sock_flag(sk, SOCK_DEAD)) {
+				/* Get in sync with tcp_poll function. */
+				smp_mb();
 				sk->sk_data_ready(sk, 0);
+			}
 		}
 	}
 }

@@ -5317,8 +5323,11 @@ no_ack:
 #endif
 			if (eaten)
 				__kfree_skb(skb);
-			else
+			else {
+				/* Get in sync with tcp_poll function. */
+				smp_mb();
 				sk->sk_data_ready(sk, 0);
+			}
 			return 0;
 		}
 	}