Skip to contentSkip to conversation

IPv6 implementation in kernel 2.4.4 oopses

DISCUSSION David Gordon (LMC) 3 participants 5 messages May 23, 2001 – May 24, 2001

lore
Share & tools
lore
#0 lore raw

Hi,

Can I be cc'ed at [email protected] (as per a normal reply) ? Thank you.

Included at the end is the ksymoops output after the crash (one of them anyhow :-)

My setup involves 2 PII installed with Linux RedHat 7.0 and 2.4.4 kernel with IPv6 enabled:
CONFIG_EXPERIMENTAL=y
CONFIG_IPV6=y
CONFIG_IPV6_EUI64=y
CONFIG_IPV6_NO_PB=y

Also, the following application web servers were installed (enabled for IPv6) :
Jigsaw 2.2.0
Tomcat 3.2.1
Apache 1.3.19

To enable IPv6 with the Java based web servers, I used jipsy 0.2.1. Lastly, netkit 0.5.1 was installed containing in particular ping6 utility.

To reproduce the crash, I ping6'ed one machine from the other in command-line mode, no servers running. That's it. Originally, the first crash occurred in a X server environment while querying the web servers in IPv6.

Thank you,

David Gordon
Ericsson Research

Here's the output:
**************

ksymoops 2.3.4 on i686 2.4.4. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.4.4/ (default)
-m /usr/src/linux/System.map (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.

No modules in ksyms, skipping objects
Warning (read_lsmod): no symbols in lsmod, is /proc/modules a valid lsmod file?
Warning (compare_maps): ksyms_base symbol __VERSIONED_SYMBOL(shmem_file_setup) not found in System.map. Ignoring ksyms_base entry c0237bc4
*pde = 00000000

Oops: 0000
CPU:    0
EIP:    0010:[<c0237bc4>]

Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010206
eax: c157c040 ebx: fffffffd ecx: 00000000 edx: 00000000
esi: 00000000 edi: 00000018 ebp: c15ac800 esp: c02f1e84
ds: 0018 es: 0018 ss: 0018
Process swapper (pid: 0, stackpage=c02f100)

Stack: c023f913 00000000 00000000 c157c040 cfb8ebd4 c022fc6b cfcab780 00000000
       c020ce8d cfcab780 010000e0 7d4985ce cfcab780 fffffffd ccf90d84 
00000000
       c15ac800 c023fe74 c15ac800 ccf90d20 ccf90d84 ccf90d84 00000000 
c15ac800
Call Trace: [<c023f913>] [<c022fc6b>] [<c020ce8d>] [<c023fe74>] [<c020e986>] [<c023fdc0>] [<c02088a0>]
       [<c011ac90>] [<c0208710>] [<c011afb6>] [<c0117dfc>] [<c0117cb5>] 
[<c0117b2d>] [<c0108a07>] [<c01051a0>]
       [<c0106ef8>] [<c01051a0>] [<c0100018>] [<c01051cc>] [<c0105252>] 
[<c0105000>] [<c01001cf>]
Code: 8b 11 f7 c2 e0 00 00 00 74 14 89 d0 25 e0 00 00 00 3d e0 00
Quoted 1 line
Quoted 1 line

EIP; c0237bc4 <ipv6_addr_type+4/e0> <=====

Trace; c023f913 <ndisc_send_ns+33/250>
Trace; c022fc6b <igmp_rcv+15b/170>
Trace; c020ce8d <ip_route_input+11d/190>
Trace; c023fe74 <ndisc_solicit+b4/c0>
Trace; c020e986 <ip_rcv+2f6/380>
Trace; c023fdc0 <ndisc_solicit+0/c0>
Trace; c02088a0 <neigh_timer_handler+190/1d0>
Trace; c011ac90 <update_process_times+20/a0>
Trace; c0208710 <neigh_timer_handler+0/1d0>
Trace; c011afb6 <timer_bh+256/2b0>
Trace; c0117dfc <bh_action+4c/b0>
Trace; c0117cb5 <tasklet_hi_action+55/90>
Trace; c0117b2d <do_softirq+6d/a0>
Trace; c0108a07 <do_IRQ+e7/100>
Trace; c01051a0 <default_idle+0/40>
Trace; c0106ef8 <ret_from_intr+0/20>
Trace; c01051a0 <default_idle+0/40>
Trace; c0100018 <startup_32+18/cb>
Trace; c01051cc <default_idle+2c/40>
Trace; c0105252 <cpu_idle+52/70>
Trace; c0105000 <init+0/180>
Trace; c01001cf <L6+0/2>
Code; c0237bc4 <ipv6_addr_type+4/e0>
00000000 <_EIP>:
Code; c0237bc4 <ipv6_addr_type+4/e0> <=====
0: 8b 11 mov (%ecx),%edx <=====
Code; c0237bc6 <ipv6_addr_type+6/e0>
2: f7 c2 e0 00 00 00 test $0xe0,%edx
Code; c0237bcc <ipv6_addr_type+c/e0>
8: 74 14 je 1e <_EIP+0x1e> c0237be2
<ipv6_addr_type+22/e0>
Code; c0237bce <ipv6_addr_type+e/e0>
a: 89 d0 mov %edx,%eax
Code; c0237bd0 <ipv6_addr_type+10/e0>
c: 25 e0 00 00 00 and $0xe0,%eax
Code; c0237bd5 <ipv6_addr_type+15/e0>
11: 3d e0 00 00 00 cmp $0xe0,%eax

Kernel panic: Aiee, killing interrupt handler!

3 warnings issued. Results may not be reliable.
***********************

End email message

#1 lore raw
Quoted 1 line
Quoted 1 line
Quoted 1 line

EIP; c0237bc4 <ipv6_addr_type+4/e0> <=====

Problem is already fixed in the latest pre kernels.

-Andi

#3 lore raw
In reply to David Gordon (LMC)
Quoted 3 lines · David Gordon (LMC)
Quoted 1 line
Quoted 1 line

EIP; c0237bc4 <ipv6_addr_type+4/e0> <=====

What exactly was the problem that was fixed in the latest pre kernel ?

A coding mistake was fixed.

Here is the patch if you're interested (cut'n'pasted so not applicable)

RCS file: /cvs/linux/net/ipv6/ndisc.c,v
retrieving revision 1.49
retrieving revision 1.50

ipv6/ndisc.c 2001/05/05 01:59:27 1.50 renamed from ipv6/ndisc.c 2001/05/03 07:02:47 1.49 +1 -1
diff -u -u -r1.49 -r1.50
@@ -382,7 +382,7 @@
        int send_llinfo;
 
        len = sizeof(struct icmp6hdr) + sizeof(struct in6_addr);
-       send_llinfo = dev->addr_len && ipv6_addr_type(saddr) != IPV6_ADDR_ANY;
+       send_llinfo = dev->addr_len && saddr && ipv6_addr_type(saddr) != IPV6_AD
DR_ANY;
        if (send_llinfo)
                len += NDISC_OPT_SPACE(dev->addr_len);
#4 lore raw
Quoted 3 lines · Andi Kleen

A coding mistake was fixed.

Here is the patch if you're interested (cut'n'pasted so not applicable)

That's not the final fix Andi. The final fix was to add this chunk about the send_llinfo test:

	if (saddr == NULL) {
		if (ipv6_get_lladdr(dev, &addr_buf))
			return;
		saddr = &addr_buf;
	}

And remove that chunk of code later in the function. This is what you'll find in 2.4.5-preX current.

You made the diff with a CVS tree, so I have no idea how you could have gotten the old change.

Later,
David S. Miller
[email protected]