| 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
| ||
|
Message-ID: <alpine.LFD.2.11.1405232323460.1913@ja.home.ssi.bg>
Date: Fri, 23 May 2014 23:56:29 +0300 (EEST)
From: Julian Anastasov <ja@....bg>
To: Peter Christensen <pch@...bogen.com>
cc: Wensong Zhang <wensong@...ux-vs.org>,
Simon Horman <horms@...ge.net.au>, netdev@...r.kernel.org,
lvs-devel@...r.kernel.org
Subject: Re: [PATCH] ipvs: Fix panic due to non-linear skb
Hello,
On Fri, 23 May 2014, Peter Christensen wrote:
> ICMP responses to IPIP packets are deencapsulated using __skb_pull, but if the
> skb is non-linear, this may cause a panic. Instead, we use pskb_pull and
> store relevant ICMP info in variables in case pskb_pull decides to reallocate.
Thanks for fixing the problem. It comes from
commit f2edb9f7706dcb2c0d9a362b2ba849efe3a97f5e
("ipvs: implement passive PMTUD for IPIP packets").
You can mention it in the comments.
Can you post v2 with removing the cih->protocol
protection, I guess it is not needed and is not part of
the problem. Just keep the changes in the if (ipip) {} block.
Also, above lines can be wrapped early.
One more request below...
> kernel BUG at include/linux/skbuff.h:1430!
> invalid opcode: 0000 [#1] SMP
> Modules linked in:
> CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.10.40-director-15 #1
> Hardware name: Dell Inc. PowerEdge C5220/0NVH5D, BIOS 2.0.3 06/22/2012
> task: ffffffff816ea440 ti: ffffffff816da000 task.ti: ffffffff816da000
> RIP: 0010:[<ffffffff81498a60>] [<ffffffff81498a60>] ip_vs_in_icmp+0x610/0x620
> RSP: 0018:ffff8807ffc03af0 EFLAGS: 00010287
> RAX: 00000000000000ea RBX: ffff8807d5a3c300 RCX: 0000000000000000
> RDX: 000000005ca90000 RSI: 0000000000000000 RDI: ffff8807d9778128
> RBP: 0000000000000014 R08: 0000000000000000 R09: 0000000000000000
> R10: ffff8807d5a3c300 R11: ffff8807dbb3f180 R12: ffff8807ffc03b40
> R13: 0000000000000001 R14: 000000000000001c R15: ffff8807ffc03b38
> FS: 0000000000000000(0000) GS:ffff8807ffc00000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: ffffffffff600400 CR3: 00000000016e5000 CR4: 00000000001407f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Stack:
> ffff88070000011a ffffffff816dbfd8 ffff8807dbda2000 ffff8807dbb3f180
> 0000000000000014 00000044ffc03bd8 0100000000000030 ffffffff81740080
> ffff8807d5a3c300 00000000a0050303 00401710ea000045 33f8905b30e8067a
> Call Trace:
> <IRQ>
> [<ffffffff81498d9b>] ? ip_vs_in.constprop.28+0x2bb/0x450
> [<ffffffff814992e2>] ? ip_vs_out.constprop.29+0x272/0x4c0
> [<ffffffff814a9df0>] ? ip_rcv_finish+0x2e0/0x2e0
> [<ffffffff814a9df0>] ? ip_rcv_finish+0x2e0/0x2e0
> [<ffffffff8149255d>] ? nf_iterate+0x8d/0xc0
> [<ffffffff814a9df0>] ? ip_rcv_finish+0x2e0/0x2e0
> [<ffffffff814925fe>] ? nf_hook_slow+0x6e/0x130
> [<ffffffff814a9df0>] ? ip_rcv_finish+0x2e0/0x2e0
> [<ffffffff814aa042>] ? ip_local_deliver+0x52/0x90
> [<ffffffff8146e322>] ? __netif_receive_skb_core+0x4b2/0x630
> [<ffffffff81374e40>] ? macvlan_broadcast+0x150/0x150
> [<ffffffff814713a4>] ? netif_receive_skb+0x24/0x80
> [<ffffffff814720c8>] ? napi_gro_receive+0x98/0xd0
> [<ffffffff813b81e0>] ? igb_poll+0x6c0/0xf40
> [<ffffffff8146e322>] ? __netif_receive_skb_core+0x4b2/0x630
> [<ffffffff81056ce5>] ? hrtimer_get_next_event+0xd5/0xe0
> [<ffffffff81471671>] ? net_rx_action+0xf1/0x190
> [<ffffffff81041b3e>] ? get_next_timer_interrupt+0x1be/0x250
> [<ffffffff81068092>] ? do_timer+0x352/0x670
> [<ffffffff8103b7c6>] ? __do_softirq+0xd6/0x1a0
> [<ffffffff8107e23e>] ? handle_irq_event_percpu+0x7e/0x140
> [<ffffffff815113fc>] ? call_softirq+0x1c/0x30
> [<ffffffff81003c8d>] ? do_softirq+0x4d/0x80
> [<ffffffff8103b9ae>] ? irq_exit+0x7e/0xa0
> [<ffffffff81003afb>] ? do_IRQ+0x5b/0xd0
> [<ffffffff8150fb2a>] ? common_interrupt+0x6a/0x6a
> <EOI>
> [<ffffffff81056a38>] ? __hrtimer_start_range_ns+0x1a8/0x360
> [<ffffffff8143deb6>] ? cpuidle_enter_state+0x56/0xe0
> [<ffffffff8143deb2>] ? cpuidle_enter_state+0x52/0xe0
> [<ffffffff8143dfde>] ? cpuidle_idle_call+0x9e/0x140
> [<ffffffff8100a6f9>] ? arch_cpu_idle+0x9/0x30
> [<ffffffff810666f6>] ? cpu_startup_entry+0x86/0x160
> [<ffffffff81761cf6>] ? start_kernel+0x2d9/0x2e4
> [<ffffffff81761834>] ? repair_env_string+0x5b/0x5b
> Code: 48 ec 83 f8 59 0f 42 c8 8b 44 24 30 0f c9 29 e8 83 e8 08 89 44 24 30 e9 27 fc ff ff 83 78 14 00 0f 84 8d fb ff ff e9 10 fd ff ff <0f> 0b 66 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 48 89 f1 8b b6
> RIP [<ffffffff81498a60>] ip_vs_in_icmp+0x610/0x620
> RSP <ffff8807ffc03af0>
> ---[ end trace fb17b5b5910de306 ]---
> Kernel panic - not syncing: Fatal exception in interrupt
>
> Signed-off-by: Peter Christensen <pch@...bogen.com>
> ---
> net/netfilter/ipvs/ip_vs_core.c | 22 +++++++++++++++-------
> 1 file changed, 15 insertions(+), 7 deletions(-)
>
> diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c
> index 4f26ee4..8e20e12 100644
> --- a/net/netfilter/ipvs/ip_vs_core.c
> +++ b/net/netfilter/ipvs/ip_vs_core.c
> @@ -1300,6 +1300,7 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
> struct ip_vs_proto_data *pd;
> unsigned int offset, offset2, ihl, verdict;
> bool ipip;
> + __u8 protocol;
>
> *related = 1;
>
> @@ -1390,17 +1391,23 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
> goto out;
> }
>
> + protocol = cih->protocol;
> +
> if (ipip) {
> __be32 info = ic->un.gateway;
> + int type = ic->type;
> + int code = ic->code;
>
> /* Update the MTU */
> if (ic->type == ICMP_DEST_UNREACH &&
> ic->code == ICMP_FRAG_NEEDED) {
> struct ip_vs_dest *dest = cp->dest;
> u32 mtu = ntohs(ic->un.frag.mtu);
> + __be16 frag_off = cih->frag_off;
>
> /* Strip outer IP and ICMP, go to IPIP header */
> - __skb_pull(skb, ihl + sizeof(_icmph));
> + if (pskb_pull(skb, ihl + sizeof(_icmph)) == NULL)
> + goto ignore_ipip;
> offset2 -= ihl + sizeof(_icmph);
> skb_reset_network_header(skb);
> IP_VS_DBG(12, "ICMP for IPIP %pI4->%pI4: mtu=%u\n",
> @@ -1408,7 +1415,7 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
> ipv4_update_pmtu(skb, dev_net(skb->dev),
> mtu, 0, 0, 0, 0);
> /* Client uses PMTUD? */
> - if (!(cih->frag_off & htons(IP_DF)))
> + if (!(frag_off & htons(IP_DF)))
> goto ignore_ipip;
> /* Prefer the resulting PMTU */
> if (dest) {
> @@ -1427,12 +1434,13 @@ ip_vs_in_icmp(struct sk_buff *skb, int *related, unsigned int hooknum)
> /* Strip outer IP, ICMP and IPIP, go to IP header of
> * original request.
> */
> - __skb_pull(skb, offset2);
> + if (pskb_pull(skb, offset2) == NULL)
> + goto ignore_ipip;
> skb_reset_network_header(skb);
> IP_VS_DBG(12, "Sending ICMP for %pI4->%pI4: t=%u, c=%u, i=%u\n",
t and c are now signed: t=%d, c=%d
> &ip_hdr(skb)->saddr, &ip_hdr(skb)->daddr,
> - ic->type, ic->code, ntohl(info));
> - icmp_send(skb, ic->type, ic->code, info);
> + type, code, ntohl(info));
> + icmp_send(skb, type, code, info);
> /* ICMP can be shorter but anyways, account it */
> ip_vs_out_stats(cp, skb);
>
> @@ -1444,8 +1452,8 @@ ignore_ipip:
>
> /* do the statistics and put it back */
> ip_vs_in_stats(cp, skb);
> - if (IPPROTO_TCP == cih->protocol || IPPROTO_UDP == cih->protocol ||
> - IPPROTO_SCTP == cih->protocol)
> + if (IPPROTO_TCP == protocol || IPPROTO_UDP == protocol ||
> + IPPROTO_SCTP == protocol)
> offset += 2 * sizeof(__u16);
> verdict = ip_vs_icmp_xmit(skb, cp, pp, offset, hooknum, &ciph);
>
> --
> 1.7.10.4
Regards
--
Julian Anastasov <ja@....bg>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists