All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrea Rossato <mailing_list@istitutocolli.org>
To: netfilter-devel@lists.netfilter.org
Subject: Re: ECN target bug report
Date: Mon, 09 Dec 2002 13:07:16 +0100	[thread overview]
Message-ID: <3DF48774.4030603@istitutocolli.org> (raw)
In-Reply-To: <3DF1F442.806@istitutocolli.org>

[-- Attachment #1: Type: text/plain, Size: 1723 bytes --]

attached you will find 2 patches. the first one is just a hack that
provides a solution, the second is an attempt to prove the existence of
the problem.
this is far from being the solution, it's just a workaround: checksum of
the packets will be recalculated from scratch. this means that unclean
packets with ec e cwr bits set will be cleaned after passing through
this rule. that's the reason for partial checksum recalculation: if a
packet was unclean it must remain as such.

The problem, as far as I can see it, could be located in csum_partial
(arch/i386/checksum.S): i'm not a kernel hacker (i'm a lawyer, a legal
scholar actually), but i do not see any mistake in the way partial
checksum is carried out in tcp_etc_set. anyway checksum after partial or
total recalculation differ. That's a fact. Evidence of the fact  can be
gained with the second patch: in this case the kernel will log the
checksum after partial recalculation and after total recalculation (that
means that two calculations will take place). The two values differ!

now, is this the right place for such a bug submission? is this code
manteined? should i refer to the kernel mailing list?

If someone could give me directions I would very much appreciate. I
would like to see this problem solved.
Thanks for your attention.

1. check the bug:
echo 1 /proc/sys/net/ipv4/tcp_ecn
iptables -A OUTPUT -t mangle -o ppp0 -p tcp -d my.host.org --dport 80 -j
ECN --ecn-tcp-remove
iptables -A OUTPUT -o ppp0 -p tcp -d my.host.org --dport 80 -m unclean
-j DROP
packets will be dropped


2. apply one of the patches and try again:
packets will get though and the connection will be established.

andrea

please send me a cc as I'm not a member of this list.

[-- Attachment #2: ecn_csum.patch --]
[-- Type: text/plain, Size: 1225 bytes --]

--- linux-2.4.20/net/ipv4/netfilter/ipt_ECN.c.orig	2002-12-07 23:22:37.000000000 +0100
+++ linux-2.4.20/net/ipv4/netfilter/ipt_ECN.c	2002-12-09 10:13:56.000000000 +0100
@@ -11,6 +11,7 @@
 #include <linux/skbuff.h>
 #include <linux/ip.h>
 #include <net/checksum.h>
+#include <net/tcp.h>
 
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ipt_ECN.h>
@@ -62,6 +63,7 @@
 	struct tcphdr *tcph = (void *) iph + iph->ihl * 4;
 	u_int16_t *tcpflags = (u_int16_t *)tcph + 6;
 	u_int16_t diffs[2];
+	u_int32_t tcplen;
 
 	/* raw socket (tcpdump) may have clone of incoming
 	 * skb: don't disturb it --RR */
@@ -87,13 +89,14 @@
 	}
 	
 	if (diffs[0] != *tcpflags) {
-		diffs[0] = htons(diffs[0]) ^ 0xFFFF;
-		diffs[1] = htons(*tcpflags);
-		tcph->check = csum_fold(csum_partial((char *)diffs,
-		                                    sizeof(diffs),
-		                                    tcph->check^0xFFFF));
-		(*pskb)->nfcache |= NFC_ALTERED;
 
+		tcplen = (*pskb)->len - iph->ihl*4;
+		tcph->check = 0;
+		tcph->check = tcp_v4_check(tcph, tcplen, iph->saddr, iph->daddr,
+				   csum_partial((char *)tcph, tcph->doff*4,
+					   (*pskb)->csum));
+		(*pskb)->nfcache |= NFC_ALTERED;
+	
 		return 1;
 	}
 


[-- Attachment #3: ecn_csum_with_debug_info.patch --]
[-- Type: text/plain, Size: 1261 bytes --]

--- linux-2.4.20/net/ipv4/netfilter/ipt_ECN.c.orig	2002-12-09 10:44:03.000000000 +0100
+++ linux-2.4.20/net/ipv4/netfilter/ipt_ECN.c	2002-12-09 10:48:46.000000000 +0100
@@ -11,6 +11,7 @@
 #include <linux/skbuff.h>
 #include <linux/ip.h>
 #include <net/checksum.h>
+#include <net/tcp.h>
 
 #include <linux/netfilter_ipv4/ip_tables.h>
 #include <linux/netfilter_ipv4/ipt_ECN.h>
@@ -62,6 +63,7 @@
 	struct tcphdr *tcph = (void *) iph + iph->ihl * 4;
 	u_int16_t *tcpflags = (u_int16_t *)tcph + 6;
 	u_int16_t diffs[2];
+	u_int32_t tcplen;
 
 	/* raw socket (tcpdump) may have clone of incoming
 	 * skb: don't disturb it --RR */
@@ -92,6 +94,17 @@
 		tcph->check = csum_fold(csum_partial((char *)diffs,
 		                                    sizeof(diffs),
 		                                    tcph->check^0xFFFF));
+
+		printk(KERN_WARNING "ECN: checksum after partial recalculation \"%x\"\n", tcph->check );
+		
+		tcplen = (*pskb)->len - iph->ihl*4;
+		tcph->check = 0;
+		tcph->check = tcp_v4_check(tcph, tcplen, iph->saddr, iph->daddr,
+				   csum_partial((char *)tcph, tcph->doff*4,
+					   (*pskb)->csum));
+
+		printk(KERN_WARNING "ECN: checksum after total recalculation \"%x\"\n", tcph->check );
+
 		(*pskb)->nfcache |= NFC_ALTERED;
 
 		return 1;


  parent reply	other threads:[~2002-12-09 12:07 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-12-07 13:14 ECN target bug report Andrea Rossato
2002-12-09 10:13 ` Andrea Rossato
2002-12-09 12:07 ` Andrea Rossato [this message]
2002-12-09 16:23 ` Andrea Rossato
2002-12-09 16:37 ` Andrea Rossato
2002-12-09 21:19 ` Andrea Rossato
2002-12-09 23:02   ` Patrick McHardy
2002-12-10 17:27     ` Andrea Rossato
2002-12-10 18:17       ` Patrick McHardy
2002-12-10 18:30         ` Andrea Rossato
2003-01-02  9:03   ` Harald Welte

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3DF48774.4030603@istitutocolli.org \
    --to=mailing_list@istitutocolli.org \
    --cc=netfilter-devel@lists.netfilter.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.