[Bridge] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[Bridge] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[Stephen Hemminger]

Could the problem be that the e100 can do IP receive checksumming on the board,
but the eepro driver doesn't enable it.  When the board is doing checksum
offload, then the csum field isn't set. 

Please try disabling receive checksumming on the e100 driver

	modprobe e100 XsumRX=0

If this is the problem, it exists both 2.4 and 2.6.


On Wed, 27 Aug 2003 18:24:57 +0200
Hannes Schulz <schulz@schwaar.com> wrote:

> setup & how to reproduce:
> HP Netserver 2000 with 2 onboard e100 NICs. Both are connected to the same hub.
> Add both eth0 and eth1 to the same bridge. Pull the plug on eth0. 
> Wait till the bridge is about to enter forwarding state on eth1.
> 
> I have reproduced this with earlier kernels (2.4.20+) too but they 
> were heavily patched. This is a pristine 2.4.22.
> 
> Strange but true: This works when using eepro100 insted of e100.
> 
> 
> details:
> uml:~# brctl addbr br0
> uml:~# brctl stp on
> uml:~# brctl addif br0 eth0
> uml:~# ip link set eth0 up
> uml:~# brctl addif br0 eth1
> uml:~# brctl setpathcost br0 eth1 1000
> uml:~# ip link set eth1 up
> uml:~# tail  /var/log/kern.log
> [junk deleted]
> Aug 27 17:07:44 uml kernel: device eth1 entered promiscuous mode
> Aug 27 17:08:43 uml kernel: br0: port 2(eth1) entering listening state
> Aug 27 17:08:44 uml kernel: eth0: received packet with  own address as source
> address
> Aug 27 17:08:44 uml kernel: br0: port 2(eth1) entering blocking state
> Aug 27 17:08:45 uml kernel: e100: eth1 NIC Link is Up 100 Mbps Half duplex
> 
> pull the plug on eth0
> 
> uml:~# tail -f /var/log/kern.log
> Aug 27 17:12:11 uml kernel: e100: eth0 NIC Link is Down
> Aug 27 17:12:30 uml kernel: br0: neighbour 8000.00:30:6e:12:dc:5d lost on port
> 2(eth1)
> [00:30:6e:12:dc:5d is the MAC-address of eth0]
> Aug 27 17:12:30 uml kernel: br0: port 2(eth1) entering listening state
> Aug 27 17:12:45 uml kernel: br0: port 2(eth1) entering learning state
> 
> and then:
> uml:~# ksymoops oops.txt
> ksymoops 2.4.8 on i686 2.4.22.  Options used
>       -V (default)
>       -k /proc/ksyms (default)
>       -l /proc/modules (default)
>       -o /lib/modules/2.4.22/ (default)
>       -m /boot/System.map-2.4.22 (default)
> 
> Warning: You did not tell me where to find symbol information.  I will
> assume that the log matches the kernel and modules that are running
> right now and I'll use the default options above for symbol resolution.
> If the current kernel and/or modules do not match the log, you can get
> more accurate output by telling me the kernel version and where to find
> map, modules, ksyms etc.  ksymoops -h explains the options.
> 
> kernel BUG at dev.c:991!
> invalid operand: 0000
> CPU:    0
> EIP:    0010:[<c0284b29>]    Not tainted
> Using defaults from ksymoops -t elf32-i386 -a i386
> EFLAGS: 00010202
> eax: 0000822b   ebx: ce55d73c   ecx: dfcca25c   edx: 000000a8
> esi: 00008229   edi: c49ca830   ebp: cf00e000   esp: c038fde0
> ds: 0018   es: 0018   ss: 0018
> Process swapper (pid: 0, stackpage=c038f000)
> Stack: ce55d73c 0000000e 000000a8 dfcca25c cef50400 ce55d73c 00000000 c0284f25
>         ce55d73c cfec1274 cfd46034 00003122 00003122 00000296 ce55d73c ce55d73c
>         cefe64bc cf00e000 d0a112de ce55d73c 00000000 cef42c9c 00000000 d0a1132b
> Call Trace:    [<c0284f25>] [<d0a112de>] [<d0a1132b>] [<d0a11503>]
> [<d0a113d0>]
>    [<d0a115e7>] [<d0a113d0>] [<d0a12137>] [<d0a122d9>] [<c0285511>]
> [<c02856dd>]
>    [<c0285845>] [<c0127436>] [<c010b64d>] [<c0118de8>] [<c0107170>]
> [<c0107170>]
>    [<c010e458>] [<c0107170>] [<c010719c>] [<c0107232>] [<c0105000>]
> Code: 0f 0b df 03 cd 82 2f c0 89 c8 c1 e0 10 81 e1 00 00 ff ff 01
> 
> 
> >>EIP; c0284b29 <skb_checksum_help+59/a0>   <=====
> 
> >>ebx; ce55d73c <_end+e138f5c/105dc880>
> >>edi; c49ca830 <_end+45a6050/105dc880>
> >>ebp; cf00e000 <_end+ebe9820/105dc880>
> >>esp; c038fde0 <init_task_union+1de0/2000>
> 
> Trace; c0284f25 <dev_queue_xmit+3b5/430>
> Trace; d0a112de <[bridge]__dev_queue_push_xmit+2e/60>
> Trace; d0a1132b <[bridge]__br_forward_finish+1b/60>
> Trace; d0a11503 <[bridge]br_flood+53/e0>
> Trace; d0a113d0 <[bridge]__br_forward+0/60>
> Trace; d0a115e7 <[bridge]br_flood_forward+27/30>
> Trace; d0a113d0 <[bridge]__br_forward+0/60>
> Trace; d0a12137 <[bridge]br_handle_frame_finish+107/180>
> Trace; d0a122d9 <[bridge]br_handle_frame+129/1dc>
> Trace; c0285511 <netif_receive_skb+c1/200>
> Trace; c02856dd <process_backlog+8d/130>
> Trace; c0285845 <net_rx_action+c5/180>
> Trace; c0127436 <do_softirq+d6/e0>
> Trace; c010b64d <do_IRQ+19d/1d0>
> Trace; c0118de8 <smp_apic_timer_interrupt+128/130>
> Trace; c0107170 <default_idle+0/50>
> Trace; c0107170 <default_idle+0/50>
> Trace; c010e458 <call_do_IRQ+5/d>
> Trace; c0107170 <default_idle+0/50>
> Trace; c010719c <default_idle+2c/50>
> Trace; c0107232 <cpu_idle+52/70>
> Trace; c0105000 <_stext+0/0>
> 
> Code;  c0284b29 <skb_checksum_help+59/a0>
> 00000000 <_EIP>:
> Code;  c0284b29 <skb_checksum_help+59/a0>   <=====
>     0:   0f 0b                     ud2a      <=====
> Code;  c0284b2b <skb_checksum_help+5b/a0>
>     2:   df 03                     fild   (%ebx)
> Code;  c0284b2d <skb_checksum_help+5d/a0>
>     4:   cd 82                     int    $0x82
> Code;  c0284b2f <skb_checksum_help+5f/a0>
>     6:   2f                        das
> Code;  c0284b30 <skb_checksum_help+60/a0>
>     7:   c0 89 c8 c1 e0 10 81      rorb   $0x81,0x10e0c1c8(%ecx)
> Code;  c0284b37 <skb_checksum_help+67/a0>
>     e:   e1 00                     loope  10 <_EIP+0x10>
> Code;  c0284b39 <skb_checksum_help+69/a0>
>    10:   00 ff                     add    %bh,%bh
> Code;  c0284b3b <skb_checksum_help+6b/a0>
>    12:   ff 01                     incl   (%ecx)
> 
>   <0>Kernel panic: Aiee, killing interrupt handler!
> 
> 1 warning issued.  Results may not be reliable.
> 
> 
> uml:~# lspci
> 00:00.0 Host bridge: ServerWorks CNB20LE Host Bridge (rev 06)
> 00:00.1 Host bridge: ServerWorks CNB20LE Host Bridge (rev 06)
> 00:02.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 08)
> 00:07.0 VGA compatible controller: ATI Technologies Inc Rage XL (rev 27)
> 00:08.0 Ethernet controller: Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 08)
> 00:0f.0 ISA bridge: ServerWorks OSB4 South Bridge (rev 50)
> 00:0f.1 IDE interface: ServerWorks OSB4 IDE Controller
> 00:0f.2 USB Controller: ServerWorks OSB4/CSB5 OHCI USB Controller (rev 04)
> 01:02.0 Ethernet controller: 3Com Corporation 3c905B 100BaseTX [Cyclone] (rev
> 34)
> 01:03.0 Unknown mass storage controller: American Megatrends Inc. MegaRAID 428
> Ultra RAID Controller (rev 03)
> 01:05.0 SCSI storage controller: LSI Logic / Symbios Logic 53c1010 Ultra3 SCSI
> Adapter (rev 01)
> 01:05.1 SCSI storage controller: LSI Logic / Symbios Logic 53c1010 Ultra3 SCSI
> Adapter (rev 01)
> 
> when e100 loaded it said:
> Intel(R) PRO/100 Network Driver - version 2.3.18-k1
> Copyright (c) 2003 Intel Corporation
> 
> e100: selftest OK.
> e100: eth0: Intel(R) PRO/100 Network Connection
>    Hardware receive checksums enabled
>    cpu cycle saver enabled
> 
> e100: selftest OK.
> e100: eth1: Intel(R) PRO/100 Network Connection
>    Hardware receive checksums enabled
>    cpu cycle saver enabled
> 
> 
> uml:~# lsmod
> Module                  Size  Used by    Not tainted
> tun                     4864  23  (autoclean)
> ipt_LOG                 3640   4  (autoclean)
> iptable_mangle          2168   0  (autoclean) (unused)
> iptable_filter          1740   1  (autoclean)
> ip_tables              13408   3  [ipt_LOG iptable_mangle iptable_filter]
> 3c59x                  28944   1  (autoclean)
> ^^^^^ this is eth2
> e100                   54888   1  (autoclean)
> bridge                 22576   3  (autoclean)
> usb-ohci               21512   0  (unused)
> rtc                     9256   0  (autoclean)
> 
> 
> Any thoughts ?
> 
> Hannes
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-net" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Bridge] [more info] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[Hannes Schulz]

>Could the problem be that the e100 can do IP receive checksumming on 
>the board,
>but the eepro driver doesn't enable it.  When the board is doing checksum
>offload, then the csum field isn't set.
>
>Please try disabling receive checksumming on the e100 driver
>
>	modprobe e100 XsumRX=0
>
>If this is the problem, it exists both 2.4 and 2.6.

Indeed: with  XsumRX=0,0 the BUG doesn't happen. I put some debugging 
code in dev.c:

=== CUT HERE ===
--- dev.c.orig	2003-08-28 20:00:22.000000000 +0200
+++ dev.c	2003-08-28 20:59:19.000000000 +0200
@@ -987,9 +987,29 @@
  	offset = skb->tail - skb->h.raw;
  	if (offset <= 0)
  		BUG();
-	if (skb->csum+2 > offset)
+/*	if (skb->csum+2 > offset)
  		BUG();
-
+*/
+	if (skb->csum+2 > offset) {
+		printk (KERN_EMERG "skb->csum+2=%d, offset=%d, 
skb->ip_summed=%d\n", skb->csum+2, offset, (int)(skb->ip_summed));
+		printk (KERN_EMERG 
"skb->mac.ethernet->h_dest=%0.2x:%0.2x:%0.2x:%0.2x:%0.2x:%0.2x\n",
+				(unsigned int)(skb->mac.ethernet->h_dest [0]),
+				(unsigned int)(skb->mac.ethernet->h_dest [1]),
+				(unsigned int)(skb->mac.ethernet->h_dest [2]),
+				(unsigned int)(skb->mac.ethernet->h_dest [3]),
+				(unsigned int)(skb->mac.ethernet->h_dest [4]),
+				(unsigned int)(skb->mac.ethernet->h_dest [5])
+		);
+		printk (KERN_EMERG 
"skb->mac.ethernet->h_source=%0.2x:%0.2x:%0.2x:%0.2x:%0.2x:%0.2x\n",
+				(unsigned 
int)(skb->mac.ethernet->h_source [0]),
+				(unsigned 
int)(skb->mac.ethernet->h_source [1]),
+				(unsigned 
int)(skb->mac.ethernet->h_source [2]),
+				(unsigned 
int)(skb->mac.ethernet->h_source [3]),
+				(unsigned 
int)(skb->mac.ethernet->h_source [4]),
+				(unsigned int)(skb->mac.ethernet->h_source [5])
+		);
+		BUG ();
+	}
  	*(u16*)(skb->h.raw + skb->csum) = csum_fold(csum);
  	skb->ip_summed = CHECKSUM_NONE;
  	return skb;
=== CUT HERE ===

It says (just before the BUG):
skb->csum+2=33323, offset=168, skb->ip_summed=1
skb->mac.ethernet->h_dest=ff:ff:ff:ff:ff:ff
skb->mac.ethernet->h_source=00:d0:b7:3c:78:0a

I also put a few lines in e100_main.c:

=== CUT HERE ===
--- e100_main.c.orig	2003-08-28 21:01:07.000000000 +0200
+++ e100_main.c	2003-08-28 21:07:10.000000000 +0200
@@ -2051,11 +2051,14 @@
  		if (bdp->flags & DF_CSUM_OFFLOAD) {
  			if (bdp->rev_id >= D102_REV_ID) {
  				skb->ip_summed = e100_D102_check_checksum(rfd);
+				printk (KERN_ERR "e100_D102: 
skb->csum+2=%d,offset=%d, skb->ip_summed=%d\n", skb->csum+2, 
skb->tail - skb->h.raw, (int)(skb->ip_summed));
  			} else {
  				skb->ip_summed = e100_D101M_checksum(bdp, skb);
+				printk (KERN_ERR "e100_D101M: 
skb->csum+2=%d,offset=%d, skb->ip_summed=%d\n", skb->csum+2, 
skb->tail - skb->h.raw, (int)(skb->ip_summed));
  			}
  		} else {
  			skb->ip_summed = CHECKSUM_NONE;
+			printk (KERN_ERR "e100_NOOFF: 
skb->csum+2=%d,offset=%d, skb->ip_summed=%d\n", skb->csum+2, 
skb->tail - skb->h.raw, (int)(skb->ip_summed));
  		}

  		bdp->drv_stats.net_stats.rx_bytes += skb->len;
=== CUT HERE ===

and my console was flooded with these:
e100_D101M: skb->csum+2=47564,offset=-2789414, skb->ip_summed=1
e100_D101M: skb->csum+2=38865,offset=3991018, skb->ip_summed=0
e100_D101M: skb->csum+2=33998,offset=4009612, skb->ip_summed=1
e100_D101M: skb->csum+2=11471,offset=845290, skb->ip_summed=1
e100_D101M: skb->csum+2=33323,offset=4036692, skb->ip_summed=1
                         ^^^^^
this line was printed just above the BUG. The bug itself is 
essentially the same as before; just different offsets.

I think the packet in question is a broadcast of linux-ha sent out by 
a completely unrelated machine that happens to be on the same network:

uml:/usr/src/linux/drivers/net/e100# tcpdump -i br0 -e -n -q ether 
host 00:d0:b7:3c:78:0a
tcpdump: listening on br0
22:11:40.413171 0:d0:b7:3c:78:a ff:ff:ff:ff:ff:ff 182: 
10.96.96.25.1025 > 10.96.96.255.694: udp 140
22:11:42.413154 0:d0:b7:3c:78:a ff:ff:ff:ff:ff:ff 182: 
10.96.96.25.1025 > 10.96.96.255.694: udp 140
[and so on; the machine is idle at that time of the day]


Q: the 'offset' looks wrong in my code in e100_main.c [I didn't 
further investigate this]; but the skb->csum shows strong 
coincidence. What is happening here ?


Thanks in advance

Hannes

[Bridge] Re: [more info] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[David S. Miller]

On Thu, 28 Aug 2003 22:24:52 +0200
Hannes Schulz <schulz@schwaar.com> wrote:

> >Could the problem be that the e100 can do IP receive checksumming on 
> >the board,
> >but the eepro driver doesn't enable it.  When the board is doing checksum
> >offload, then the csum field isn't set.
> >
> >Please try disabling receive checksumming on the e100 driver
> >
> >	modprobe e100 XsumRX=0
> >
> >If this is the problem, it exists both 2.4 and 2.6.
> 
> Indeed: with  XsumRX=0,0 the BUG doesn't happen. I put some debugging 
> code in dev.c:

Looks like bridging is taking an input packet, transmitting it but not
initializing skb->csum, skb->ip_summed, and other fields before doing
so.

Definitely a bridging bug.

[Bridge] Re: [more info] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[Stephen Hemminger]

On Sat, 30 Aug 2003 19:01:00 -0700
"David S. Miller" <davem@redhat.com> wrote:

> On Thu, 28 Aug 2003 22:24:52 +0200
> Hannes Schulz <schulz@schwaar.com> wrote:
> 
> > >Could the problem be that the e100 can do IP receive checksumming on 
> > >the board,
> > >but the eepro driver doesn't enable it.  When the board is doing checksum
> > >offload, then the csum field isn't set.
> > >
> > >Please try disabling receive checksumming on the e100 driver
> > >
> > >	modprobe e100 XsumRX=0
> > >
> > >If this is the problem, it exists both 2.4 and 2.6.
> > 
> > Indeed: with  XsumRX=0,0 the BUG doesn't happen. I put some debugging 
> > code in dev.c:
> 
> Looks like bridging is taking an input packet, transmitting it but not
> initializing skb->csum, skb->ip_summed, and other fields before doing
> so.
> 
> Definitely a bridging bug.

It is actually a bridging-filter bug, since the bridge layer is at level 2 it
doesn't know anything about IP.

Re: [Bridge] Re: [more info] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[David S. Miller]

On Tue, 2 Sep 2003 20:23:16 +0200
Bart De Schuymer <bdschuym@pandora.be> wrote:

> A pristine 2.4.22 doesn't have bridge-netfilter.

This is true.

> I don't see how bridge-netfilter would be involved anyway, altering
> a packet doesn't automagically trigger checksum recalculation.

You don't understand the problem.

When a packet gets received, if the hardware checksums the packet
it sets skb->ip_summed.  If you take that SKB and want to transmit
it you have to reinitialize the skb->ip_summed field or else things
explode.

Re: [Bridge] Re: [more info] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[Bart De Schuymer]

On Tuesday 02 September 2003 19:45, Stephen Hemminger wrote:
> > Looks like bridging is taking an input packet, transmitting it but not
> > initializing skb->csum, skb->ip_summed, and other fields before doing
> > so.
> >
> > Definitely a bridging bug.
>
> It is actually a bridging-filter bug, since the bridge layer is at level 2
> it doesn't know anything about IP.

A message ([Bridge] Re: [2.4.22] bad interaction between e100 and bridge: BUG 
at dev.c:991!) on the bridge list said this:

<quote>
> setup & how to reproduce:
> HP Netserver 2000 with 2 onboard e100 NICs. Both are connected to the same 
hub.
> Add both eth0 and eth1 to the same bridge. Pull the plug on eth0. 
> Wait till the bridge is about to enter forwarding state on eth1.
> 
> I have reproduced this with earlier kernels (2.4.20+) too but they 
> were heavily patched. This is a pristine 2.4.22.
> 
> Strange but true: This works when using eepro100 insted of e100.
<unquote>

A pristine 2.4.22 doesn't have bridge-netfilter. I don't see how 
bridge-netfilter would be involved anyway, altering a packet doesn't 
automagically trigger checksum recalculation.

cheers,
Bart

[Bridge] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[Hannes Schulz]

>Could the problem be that the e100 can do IP receive checksumming on 
>the board,
>but the eepro driver doesn't enable it.  When the board is doing checksum
>offload, then the csum field isn't set.
>
>Please try disabling receive checksumming on the e100 driver
>
>	modprobe e100 XsumRX=0
>
>If this is the problem, it exists both 2.4 and 2.6.

If have just booted into 2.6.0-test4. Unplugging works, but the 
console is flooded with:
last-F IN=br0 OUT=br0 PHYSIN=eth1 PHYSOUT=eth0 SRC=10.96.96.25 DST=10.96.96.255

[last-F ist the last rule in my FORWARDING iptable and does log+drop]

Obviously I must allow forwarding br0 <=> br0.

While I am writing this the counter on the FORWARDING rule -i br0 -o 
br0 -j ACCEPT has reaches 2000+ packets with eth0 unplugged and the 
kernel ist still running.


In both 2.4.22 and 2.6.0-test4 all of netfilter and advanced routing 
ist set to M. In 2.6.0-test4 this includes ebtables.


Hope that helps

Hannes

[Bridge] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[Stephen Hemminger]

On Wed, 3 Sep 2003 16:56:06 +0200
Hannes Schulz <schulz@schwaar.com> wrote:

> >Could the problem be that the e100 can do IP receive checksumming on 
> >the board,
> >but the eepro driver doesn't enable it.  When the board is doing checksum
> >offload, then the csum field isn't set.
> >
> >Please try disabling receive checksumming on the e100 driver
> >
> >	modprobe e100 XsumRX=0
> >
> >If this is the problem, it exists both 2.4 and 2.6.
> 
> If have just booted into 2.6.0-test4. Unplugging works, but the 
> console is flooded with:
> last-F IN=br0 OUT=br0 PHYSIN=eth1 PHYSOUT=eth0 SRC=10.96.96.25 DST=10.96.96.255
> 
> [last-F ist the last rule in my FORWARDING iptable and does log+drop]
> 
> Obviously I must allow forwarding br0 <=> br0.
> 
> While I am writing this the counter on the FORWARDING rule -i br0 -o 
> br0 -j ACCEPT has reaches 2000+ packets with eth0 unplugged and the 
> kernel ist still running.
> 
> 
> In both 2.4.22 and 2.6.0-test4 all of netfilter and advanced routing 
> ist set to M. In 2.6.0-test4 this includes ebtables.
> 
> 
> Hope that helps
> 
> Hannes

You enabled NETFILTER_DEBUG in your config

Re: [Bridge] Re: [2.4.22] bad interaction between e100 and bridge: BUG at dev.c:991!

[Bart De Schuymer]

On Wednesday 03 September 2003 16:56, Hannes Schulz wrote:
> If have just booted into 2.6.0-test4. Unplugging works, but the
> console is flooded with:
> last-F IN=br0 OUT=br0 PHYSIN=eth1 PHYSOUT=eth0 SRC=10.96.96.25
> DST=10.96.96.255
>
> [last-F ist the last rule in my FORWARDING iptable and does log+drop]
>
> Obviously I must allow forwarding br0 <=> br0.
>
> While I am writing this the counter on the FORWARDING rule -i br0 -o
> br0 -j ACCEPT has reaches 2000+ packets with eth0 unplugged and the
> kernel ist still running.
>
>
> In both 2.4.22 and 2.6.0-test4 all of netfilter and advanced routing
> ist set to M. In 2.6.0-test4 this includes ebtables.

Hi,
I don't quite get it, is the above text describing a problem?

On Wednesday 03 September 2003 17:38, Stephen Hemminger wrote:
>
> You enabled NETFILTER_DEBUG in your config

Enabling or disabling netfilter debugging has no effect on logged messages.
All messages on the bridge list about problems with these are from people 
using outdated 2.4 bridge-nf patches.

cheers,
Bart

[Bridge] [PATCH 2.4] bridging and hw checksum fix.

[Stephen Hemminger]

Could someone who was having problems with a board that does hardware
receive checksumming, test this please.

diff -Nru a/net/bridge/br_forward.c b/net/bridge/br_forward.c
--- a/net/bridge/br_forward.c	Fri Sep  5 11:28:33 2003
+++ b/net/bridge/br_forward.c	Fri Sep  5 11:28:33 2003
@@ -59,6 +59,7 @@
 
 	indev = skb->dev;
 	skb->dev = to->dev;
+	skb->ip_summed = CHECKSUM_NONE;
 
 	NF_HOOK(PF_BRIDGE, NF_BR_FORWARD, skb, indev, skb->dev,
 			__br_forward_finish);