From mboxrd@z Thu Jan 1 00:00:00 1970 From: Willem de Bruijn Subject: [PATCH net-next v2 5/6] packet: rollover huge flows before small flows Date: Sat, 9 May 2015 17:25:00 -0400 Message-ID: <1431206701-5019-6-git-send-email-willemb@google.com> References: <1431206701-5019-1-git-send-email-willemb@google.com> Cc: davem@davemloft.net, eric.dumazet@gmail.com, david.laight@aculab.com, Willem de Bruijn To: netdev@vger.kernel.org Return-path: Received: from mail-yh0-f50.google.com ([209.85.213.50]:33824 "EHLO mail-yh0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752873AbbEIVZK (ORCPT ); Sat, 9 May 2015 17:25:10 -0400 Received: by yhw29 with SMTP id 29so27650114yhw.1 for ; Sat, 09 May 2015 14:25:10 -0700 (PDT) In-Reply-To: <1431206701-5019-1-git-send-email-willemb@google.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Willem de Bruijn Migrate flows from a socket to another socket in the fanout group not only when the socket is full. Start migrating huge flows early, to divert possible 4-tuple attacks without affecting normal traffic. Introduce fanout_flow_is_huge(). This detects huge flows, which are defined as taking up more than half the load. It does so cheaply, by storing the rxhashes of the N most recent packets. If over half of these are the same rxhash as the current packet, then drop it. This only protects against 4-tuple attacks. N is chosen to fit all data in a single cache line. Tested: Ran bench_rollover for 10 sec with 1.5 Mpps of single flow input. lpbb5:/export/hda3/willemb# ./bench_rollover -l 1000 -r -s cpu rx rx.k drop.k rollover r.huge r.failed 0 14 14 0 0 0 0 1 20 20 0 0 0 0 2 16 16 0 0 0 0 3 6168824 6168824 0 4867721 4867721 0 4 4867741 4867741 0 0 0 0 5 12 12 0 0 0 0 6 15 15 0 0 0 0 7 17 17 0 0 0 0 Signed-off-by: Willem de Bruijn --- net/packet/af_packet.c | 25 ++++++++++++++++++++++--- net/packet/internal.h | 2 ++ 2 files changed, 24 insertions(+), 3 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index e6b252a..0755203 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1341,6 +1341,20 @@ static int fanout_rr_next(struct packet_fanout *f, unsigned int num) return x; } +static bool fanout_flow_is_huge(struct packet_sock *po, struct sk_buff *skb) +{ + u32 rxhash; + int i, count = 0; + + rxhash = skb_get_hash(skb); + for (i = 0; i < ROLLOVER_HLEN; i++) + if (po->rollover->history[i] == rxhash) + count++; + + po->rollover->history[prandom_u32() % ROLLOVER_HLEN] = rxhash; + return count > (ROLLOVER_HLEN >> 1); +} + static unsigned int fanout_demux_hash(struct packet_fanout *f, struct sk_buff *skb, unsigned int num) @@ -1381,11 +1395,16 @@ static unsigned int fanout_demux_rollover(struct packet_fanout *f, unsigned int num) { struct packet_sock *po, *po_next; - unsigned int i, j; + unsigned int i, j, room; po = pkt_sk(f->arr[idx]); - if (try_self && packet_rcv_has_room(po, skb) != ROOM_NONE) - return idx; + + if (try_self) { + room = packet_rcv_has_room(po, skb); + if (room == ROOM_NORMAL || + (room == ROOM_LOW && !fanout_flow_is_huge(po, skb))) + return idx; + } i = j = min_t(int, po->rollover->sock, num - 1); do { diff --git a/net/packet/internal.h b/net/packet/internal.h index 22d7d77..a9d30a1 100644 --- a/net/packet/internal.h +++ b/net/packet/internal.h @@ -89,6 +89,8 @@ struct packet_fanout { struct packet_rollover { int sock; +#define ROLLOVER_HLEN (L1_CACHE_BYTES / sizeof(u32)) + u32 history[ROLLOVER_HLEN] ____cacheline_aligned; } ____cacheline_aligned_in_smp; struct packet_sock { -- 2.2.0.rc0.207.ga3a616c