netdev - [PATCH] fix tcp fin memory accounting

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <1426807170-19258-1-git-send-email-johunt@akamai.com>
Date:	Thu, 19 Mar 2015 19:19:30 -0400
From:	Josh Hunt <johunt@...mai.com>
To:	davem@...emloft.net, netdev@...r.kernel.org
Cc:	edumazet@...gle.com, Josh Hunt <johunt@...mai.com>
Subject: [PATCH] fix tcp fin memory accounting

tcp_send_fin() does not account for the memory it allocates properly, so
sk_forward_alloc can be negative in cases where we've sent a FIN:

ss example output (ss -amn | grep -B1 f4294):
tcp    FIN-WAIT-1 0      1            192.168.0.1:45520         192.0.2.1:8080
	skmem:(r0,rb87380,t0,tb87380,f4294966016,w1280,o0,bl0)
--
tcp    FIN-WAIT-1 0      1            192.168.0.1:8080          192.0.2.1:59710
	skmem:(r0,rb87380,t0,tb87380,f4294966016,w1280,o0,bl0)

This can be resolved by using sk_stream_alloc_skb() instead of calling alloc_skb_fclone()
directly. The same fix was made in tcp_connect() in commit 355a901e6cf1 (tcp: make connect() mem charging friendly).

With this fix applied I no longer see the negative sk_forward_alloc values (or
very large values repoted by ss) in my tests.

Signed-off-by: Josh Hunt <johunt@...mai.com>
---
 net/ipv4/tcp_output.c |    6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 5a73ad5..c2f0f60 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -2820,15 +2820,11 @@ void tcp_send_fin(struct sock *sk)
 	} else {
 		/* Socket is locked, keep trying until memory is available. */
 		for (;;) {
-			skb = alloc_skb_fclone(MAX_TCP_HEADER,
-					       sk->sk_allocation);
+			skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation);
 			if (skb)
 				break;
 			yield();
 		}
-
-		/* Reserve space for headers and prepare control bits. */
-		skb_reserve(skb, MAX_TCP_HEADER);
 		/* FIN eats a sequence byte, write_seq advanced by tcp_queue_skb(). */
 		tcp_init_nondata_skb(skb, tp->write_seq,
 				     TCPHDR_ACK | TCPHDR_FIN);
-- 
1.7.9.5

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html