ipv6: BUG: unable to handle kernel paging request at 0000000101bca2be

Hi:

Since late Jun, early Jul, both using stable 3.4 and now 3.5 I am getting the following crash in the IPv6 code.

<1>[116145.466708] BUG: unable to handle kernel paging request at
0000000101bca2be
<1>[116145.467797] IP: [<ffffffff8145800b>] dst_release+0x1b/0x80
<4>[116145.468645] PGD 0
<4>[116145.468964] Oops: 0002 [#1] SMP
<4>[116145.469479] CPU 6
<4>[116145.469783] Modules linked in: bluetooth act_police cls_basic

cls_flow cls_fw cls_u32 sch_tbf sch_prio sch_htb sch_hfsc sch_ingress sch_sfq bridge stp xt_statistic xt_CT xt_realm iptable_raw xt_LOG xt_connlimit xt_addrtype xt_comment xt_recent ipt_ULOG ipt_REJECT ipt_REDIRECT ipt_NETMAP ipt_MASQUERADE ipt_ECN ipt_CLUSTERIP ipt_ah nf_nat_tftp nf_nat_snmp_basic nf_conntrack_snmp nf_nat_sip ip6_queue nf_nat_pptp nf_nat_proto_gre xt_set ip_set nf_nat_irc nf_nat_h323 nf_nat_ftp nf_nat_amanda nf_conntrack_tftp nf_conntrack_sane nf_conntrack_sip nf_conntrack_proto_udplite nf_conntrack_proto_sctp ts_kmp nf_conntrack_pptp nf_conntrack_proto_gre nf_conntrack_amanda nf_conntrack_netlink xt_time nf_conntrack_netbios_ns nf_conntrack_broadcast xt_TCPMSS nf_conntrack_irc nf_conntrack_h323 xt_sctp xt_policy ip6t_REJECT nf_conntrack_ipv6 ip6table_raw ip6table_mangle nf_conntrack_ftp xt_TPROXY nf_tproxy_core nf_defrag_ipv6 xt_tcpmss xt_pkttype xt_physdev xt_owner xt_NFQUEUE xt_NFLOG nfnetlink_log xt_multiport xt_mark xt_mac xt_limit xt_length xt_iprange xt_helper xt_hashlimit xt_DSCP xt_dscp xt_dccp xt_conntrack xt_connmark xt_CLASSIFY xt_AUDIT xt_tcpudp xt_state ip6table_filter iptable_nat nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 ip6_tables nf_conntrack iptable_mangle nfnetlink iptable_filter ip_tables x_tables cachefiles fscache af_packet edd eeepc_wmi asus_wmi sparse_keymap rfkill pci_hotplug acpi_cpufreq mperf coretemp wmi iTCO_wdt crc32c_intel e1000e mei(C) joydev pcspkr ghash_clmulni_intel iTCO_vendor_support i2c_i801 xhci_hcd aesni_intel cryptd aes_x86_64 microcode autofs4 usbhid i915 drm_kms_helper drm ehci_hcd i2c_algo_bit usbcore usb_common video button scsi_dh_hp_sw scsi_dh_alua scsi_dh_emc scsi_dh_rdac scsi_dh fan processor thermal thermal_sys megaraid_sas

<4>[116145.495352]
<4>[116145.495579] Pid: 0, comm: swapper/6 Tainted: G         C
3.4.6-1-default #1 System manufacturer System Product Name/P8B WS
<4>[116145.497313] RIP: 0010:[<ffffffff8145800b>]  [<ffffffff8145800b>]
dst_release+0x1b/0x80
<4>[116145.498527] RSP: 0018:ffff88041f383ce0  EFLAGS: 00010202
<4>[116145.499330] RAX: ffff8803c199e568 RBX: 0000000101bca23e RCX:
0000000000000000
<4>[116145.500407] RDX: 0000000101ba584e RSI: ffff8803c199e568 RDI:
0000000101bca23e
<4>[116145.501484] RBP: 00000000ffffffff R08: 0000000000000025 R09:
ffff88041f383cb0
<4>[116145.502560] R10: ffffffffa03f82e0 R11: ffff8803bf369886 R12:
ffff8803bf369886
<4>[116145.503636] R13: 0000000000000000 R14: 0000000000000500 R15:
ffff8803ff0de0c0
<4>[116145.504714] FS:  0000000000000000(0000) GS:ffff88041f380000(0000)
knlGS:0000000000000000
<4>[116145.505935] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
<4>[116145.506802] CR2: 0000000101bca2be CR3: 0000000001a0b000 CR4:
00000000000407e0
<4>[116145.507878] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
<4>[116145.508955] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
<4>[116145.510033] Process swapper/6 (pid: 0, threadinfo
ffff88040a482000, task ffff88040a4800c0)
<4>[116145.511278] Stack:
<4>[116145.511582]  0000000002000000 00000000000249f0 ffff8803bf369896
ffffffff814f088a
<4>[116145.512754]  ffff8804095fc000 ffff8803bf369896 ffff8803bf369886
ffffffff81d5fb80
<4>[116145.513926]  0000000000000500 ffff8804095fc000 000000000000003a
ffffffff814f0dca
<4>[116145.515100] Call Trace:
<4>[116145.515482]  [<ffffffff814f088a>] rt6_do_pmtu_disc+0x27a/0x330
<4>[116145.516369]  [<ffffffff814f0dca>] rt6_pmtu_discovery+0x3a/0x70
<4>[116145.517252]  [<ffffffff81500423>] icmpv6_rcv+0x3c3/0x4a0
<4>[116145.563503]  [<ffffffff814e44e7>] ip6_input_finish+0x157/0x380
<4>[116145.609526]  [<ffffffff81450e93>] __netif_receive_skb+0x493/0x510
<4>[116145.655203]  [<ffffffff81451901>] process_backlog+0xa1/0x170
<4>[116145.701481]  [<ffffffff81451761>] net_rx_action+0x121/0x220
<4>[116145.748076]  [<ffffffff810460ad>] __do_softirq+0x9d/0x1f0
<4>[116145.774285]  [<ffffffff8155be0c>] call_softirq+0x1c/0x30
<4>[116145.800009]  [<ffffffff81004195>] do_softirq+0x65/0xa0
<4>[116145.845329]  [<ffffffff8104647e>] irq_exit+0x8e/0xb0
<4>[116145.890313]  [<ffffffff8155b8da>]
call_function_single_interrupt+0x6a/0x70
<4>[116145.935988]  [<ffffffff8131d4a9>] intel_idle+0xe9/0x160
<4>[116145.981862]  [<ffffffff81418c2e>] cpuidle_idle_call+0x9e/0x280
<4>[116146.026965]  [<ffffffff8100b53f>] cpu_idle+0x7f/0xd0
<4>[116146.069060] Code: ff 86 80 00 00 00 48 89 77 58 c3 0f 1f 44 00 00
48 83 ec 18 48 85 ff 48 89 5c 24 08 48 89 6c 24 10 48 89 fb 74 16 bd ff
ff ff ff <f0> 0f c1 af 80 00 00 00 83 ed 01 78 41 85 ed 74 14 48 8b 5c 24
<1>[116146.129910] RIP  [<ffffffff8145800b>] dst_release+0x1b/0x80
<4>[116146.172426]  RSP <ffff88041f383ce0>
<4>[116146.213632] CR2: 0000000101bca2be

It apparently appeared in either 3.4.3 or 3.4.4 and persist in 3.5..

Will be cool if someone can take a look...

This looks like an error coming from commit
1716a96101c49186bb0b8491922fd3e69030235f
(ipv6: fix problem with expired dst cache)

dst_release() is called with a bogus dst pointer : 0000000101bca23e

rt6_clean_expires() seems to be called where it shouldnt.

Could you try reverting it ?

Or try following patch instead :

index 0ae759a..a85a865 100644
@@ -123,13 +123,18 @@ static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst)
 	return ((struct rt6_info *)dst)->rt6i_idev;
 }
 
+static inline void __rt6_clean_expires(struct rt6_info *rt)
+{
+	rt->rt6i_flags &= ~RTF_EXPIRES;
+	rt->dst.from = NULL;
+}
+
 static inline void rt6_clean_expires(struct rt6_info *rt)
 {
 	if (!(rt->rt6i_flags & RTF_EXPIRES) && rt->dst.from)
 		dst_release(rt->dst.from);
 
-	rt->rt6i_flags &= ~RTF_EXPIRES;
-	rt->dst.from = NULL;
+	__rt6_clean_expires(rt);
 }
 
 static inline void rt6_set_expires(struct rt6_info *rt, unsigned long expires)

index becb048..965f315 100644
@@ -969,7 +969,7 @@ struct dst_entry *ip6_blackhole_route(struct net *net, struct dst_entry *dst_ori
 
 		rt->rt6i_gateway = ort->rt6i_gateway;
 		rt->rt6i_flags = ort->rt6i_flags;
-		rt6_clean_expires(rt);
+		__rt6_clean_expires(rt);
 		rt->rt6i_metric = 0;
 
 		memcpy(&rt->rt6i_dst, &ort->rt6i_dst, sizeof(struct rt6key));

@@ -1305,7 +1305,7 @@ int ip6_route_add(struct fib6_config *cfg)
 		rt6_set_expires(rt, jiffies +
 				clock_t_to_jiffies(cfg->fc_expires));
 	else
-		rt6_clean_expires(rt);
+		__rt6_clean_expires(rt);
 
 	if (cfg->fc_protocol == RTPROT_UNSPEC)
 		cfg->fc_protocol = RTPROT_BOOT;

@@ -1837,7 +1837,7 @@ static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
 		    (RTF_DEFAULT | RTF_ADDRCONF))
 			rt6_set_from(rt, ort);
 		else
-			rt6_clean_expires(rt);
+			__rt6_clean_expires(rt);
 		rt->rt6i_metric = 0;
 
 #ifdef CONFIG_IPV6_SUBTREES