netdev - Re: [PATCH iproute2-next] Add tc-BPF example for a TCP pure ack recognizer

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

Message-ID: <96fb372d-4fc8-77db-e971-6b9540d8ad26@gmail.com>
Date:   Tue, 1 May 2018 22:12:17 -0600
From:   David Ahern <dsahern@...il.com>
To:     Dave Taht <dave.taht@...il.com>, netdev@...r.kernel.org
Subject: Re: [PATCH iproute2-next] Add tc-BPF example for a TCP pure ack
 recognizer

On 5/1/18 12:32 PM, Dave Taht wrote:
> ack_recognize can shift pure tcp acks into another flowid.
> ---
>  examples/bpf/ack_recognize.c | 98 ++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 98 insertions(+)
>  create mode 100644 examples/bpf/ack_recognize.c
> 
> diff --git a/examples/bpf/ack_recognize.c b/examples/bpf/ack_recognize.c
> new file mode 100644
> index 0000000..5da620c
> --- /dev/null
> +++ b/examples/bpf/ack_recognize.c
> @@ -0,0 +1,98 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * Copyright 2017 Google Inc.

2017?? it's 2018. Did you write this last year and just now posting?

> + *
> + * This program is free software; you can redistribute it and/or
> + * modify it under the terms of the GNU General Public License
> + * as published by the Free Software Foundation; either version 2
> + * of the License, or (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; if not, write to the Free Software
> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
> + * 02110-1301, USA.
> + */
> +
> +/*
> + * Author: dave.taht@...il.com (Dave Taht)
> + *
> + * ack_recognizer: An eBPF program that correctly recognizes modern TCP ACKs,
> + * with tcp option fields like SACK and timestamps, and no additional data.
> + *
> + * ack_match call: Recognize "pure acks" with no data payload
> + *
> + */
> +
> +#include "bpf_api.h"
> +#include "linux/if_ether.h"
> +#include "linux/ip.h"
> +#include "linux/in.h"
> +#include "linux/ipv6.h"
> +#include "linux/tcp.h"
> +
> +/*
> + * A pure ack contains the ip header, the tcp header + options, flags with the
> + * ack field set, and no additional payload. That last bit is what every prior
> + * ack filter gets wrong, they typically assume an obsolete 64 bytes, and don't
> + * calculate the options (like sack or timestamps) to subtract from the payload.
> + */
> +
> +__section_cls_entry
> +int ack_match(struct __sk_buff *skb)
> +{
> +	void *data = (void *)(long)skb->data;
> +	void *data_end = (void *)(long)skb->data_end;
> +	struct ethhdr *eth = data;
> +	struct iphdr *iph = data + sizeof(*eth);
> +	struct tcphdr *tcp;
> +
> +	if (data + sizeof(*eth) + sizeof(*iph) + sizeof(*tcp) > data_end)
> +		return 0;
> +
> +	if (eth->h_proto == htons(ETH_P_IP) &&
> +	    iph->version == 4) {

Why not make the version == 4 under the proto check?
	if (eth->h_proto == htons(ETH_P_IP) {
		if (iph->version != 4)
			return 0;
if the eth proto is ETH_P_IP, and version != 4 something is really wrong

> +		if(iph->protocol == IPPROTO_TCP &&
> +		   iph->ihl == 5 &&
> +		   data + sizeof(*eth) + 20 + sizeof(*tcp) <= data_end) {
> +			tcp = data + sizeof(*eth) + 20;
> +			if (tcp->ack &&
> +			    htons(iph->tot_len) == 20 + tcp->doff*4)
> +				return -1;

And then here why the magic '20' and limits on ihl = 5? both are trivial
to accommodate programmatically.

> +		}
> +	} else if (eth->h_proto == htons(ETH_P_IPV6) &&
> +		   iph->version == 6) {
> +		struct ipv6hdr *iph6 = (struct ipv6hdr *) iph;
> +		if(iph6->nexthdr == IPPROTO_TCP &&
> +		   data + sizeof(*eth) + 40 + sizeof(*tcp) <= data_end ) {
> +			tcp = data + sizeof(*eth) + 40;
> +			if (tcp->ack &&
> +			    tcp->doff*4 == htons(iph6->payload_len))
> +				return -1;

ditto here.

> +		}
> +	}
> +
> +	return 0;
> +}
> +
> +/* Example: Move acks into a priority queue:
> +
> +IFACE=eth0
> +tc qdisc del dev $IFACE root 2> /dev/null
> +tc qdisc add dev $IFACE root handle 1: prio bands 3 \
> +	priomap 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
> +tc qdisc add dev $IFACE parent 1:1 handle 10:1 sfq headdrop # acks only
> +tc qdisc add dev $IFACE parent 1:2 handle 20:1 fq_codel # all other traffic
> +tc qdisc add dev $IFACE parent 1:3 handle 30:1 fq_codel # unused
> +tc filter add dev $IFACE parent 1: prio 1 bpf \
> +	object-file ack_recognize.o flowid 1:1
> +
> +Please note that a strict priority queue is not a good idea (drr would be
> +better), nor is doing any level of prioritization on acks at all....
> +*/
> +
> +BPF_LICENSE("GPL");
>