From: wenxu <wenxu@ucloud.cn> To: pshelar@nicira.com, davem@davemloft.net, jesse@kernel.org Cc: netdev@vger.kernel.org, dev@openvswitch.org, linux-kernel@vger.kernel.org, wenxu@ucloud.cn Subject: [PATCH] [stable 4.1.y PACTH] openvswitch: fix crash cause by non-nvgre packet Date: Tue, 22 Dec 2015 17:15:59 +0800 [thread overview] Message-ID: <1450775759-11059-1-git-send-email-wenxu@ucloud.cn> (raw) kernel BUG at include/linux/skbuff.h:1219! invalid opcode: 0000 [#1] SMP RIP: 0010:[<ffffffffa01dc89d>] ovs_flow_extract+0x8ed/0xa40 [openvswitch] Call Trace: <IRQ> ovs_dp_process_received_packet+0x44/0x80 [openvswitch] ovs_vport_receive+0x2e/0x30 [openvswitch] gre_rcv+0xac/0xd0 [openvswitch] gre_cisco_rcv+0x1c2/0x310 [openvswitch] gre_rcv+0x59/0x80 [openvswitch] ovs_flow_extract call __skb_pull to lead BUG_ON(skb->len < skb->data_len) if the gre header protocol is not TEB and most part of the packet is in the nolinear-spatial. 1. gre_rcv: pskb_may_pull(skb, 12) pull the 12 bytes to linear-spatial(skb->data). The gre header is 8 bytes only with key. 2. gre_cisco_rcv-->parse_gre_header-->iptunnel_pull_header { if (inner_proto == htons(ETH_P_TEB)) { struct ethhdr *eh; if (unlikely(!pskb_may_pull(skb, ETH_HLEN))) return -ENOMEM; ...... } } The wrong inner_proto leads no pull the Mac header to linear-spatial 3. finally It made a crash in ovs_flow_extract->__skb_pull Signed-off-by: wenxu <wenxu@ucloud.cn> --- net/openvswitch/vport-gre.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/openvswitch/vport-gre.c b/net/openvswitch/vport-gre.c index f17ac96..4a993b5 100644 --- a/net/openvswitch/vport-gre.c +++ b/net/openvswitch/vport-gre.c @@ -107,6 +107,9 @@ static int gre_rcv(struct sk_buff *skb, if (unlikely(!vport)) return PACKET_REJECT; + if (unlikely(tpi->proto != htons(ETH_P_TEB))) + return PACKET_REJECT; + key = key_to_tunnel_id(tpi->key, tpi->seq); ovs_flow_tun_info_init(&tun_info, ip_hdr(skb), 0, 0, key, filter_tnl_flags(tpi->flags), NULL, 0); -- 1.9.1
WARNING: multiple messages have this Message-ID (diff)
From: wenxu <wenxu-t3UIVg6hrfr/PtFMR13I2A@public.gmane.org> To: pshelar-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org, davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org, jesse-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org Cc: dev-yBygre7rU0TnMu66kgdUjQ@public.gmane.org, netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, wenxu-t3UIVg6hrfr/PtFMR13I2A@public.gmane.org Subject: [PATCH] [stable 4.1.y PACTH] openvswitch: fix crash cause by non-nvgre packet Date: Tue, 22 Dec 2015 17:15:59 +0800 [thread overview] Message-ID: <1450775759-11059-1-git-send-email-wenxu@ucloud.cn> (raw) kernel BUG at include/linux/skbuff.h:1219! invalid opcode: 0000 [#1] SMP RIP: 0010:[<ffffffffa01dc89d>] ovs_flow_extract+0x8ed/0xa40 [openvswitch] Call Trace: <IRQ> ovs_dp_process_received_packet+0x44/0x80 [openvswitch] ovs_vport_receive+0x2e/0x30 [openvswitch] gre_rcv+0xac/0xd0 [openvswitch] gre_cisco_rcv+0x1c2/0x310 [openvswitch] gre_rcv+0x59/0x80 [openvswitch] ovs_flow_extract call __skb_pull to lead BUG_ON(skb->len < skb->data_len) if the gre header protocol is not TEB and most part of the packet is in the nolinear-spatial. 1. gre_rcv: pskb_may_pull(skb, 12) pull the 12 bytes to linear-spatial(skb->data). The gre header is 8 bytes only with key. 2. gre_cisco_rcv-->parse_gre_header-->iptunnel_pull_header { if (inner_proto == htons(ETH_P_TEB)) { struct ethhdr *eh; if (unlikely(!pskb_may_pull(skb, ETH_HLEN))) return -ENOMEM; ...... } } The wrong inner_proto leads no pull the Mac header to linear-spatial 3. finally It made a crash in ovs_flow_extract->__skb_pull Signed-off-by: wenxu <wenxu@ucloud.cn> --- net/openvswitch/vport-gre.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/openvswitch/vport-gre.c b/net/openvswitch/vport-gre.c index f17ac96..4a993b5 100644 --- a/net/openvswitch/vport-gre.c +++ b/net/openvswitch/vport-gre.c @@ -107,6 +107,9 @@ static int gre_rcv(struct sk_buff *skb, if (unlikely(!vport)) return PACKET_REJECT; + if (unlikely(tpi->proto != htons(ETH_P_TEB))) + return PACKET_REJECT; + key = key_to_tunnel_id(tpi->key, tpi->seq); ovs_flow_tun_info_init(&tun_info, ip_hdr(skb), 0, 0, key, filter_tnl_flags(tpi->flags), NULL, 0); -- 1.9.1 _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
next reply other threads:[~2015-12-22 9:19 UTC|newest] Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top 2015-12-22 9:15 wenxu [this message] 2015-12-22 9:15 ` [PATCH] [stable 4.1.y PACTH] openvswitch: fix crash cause by non-nvgre packet wenxu
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=1450775759-11059-1-git-send-email-wenxu@ucloud.cn \ --to=wenxu@ucloud.cn \ --cc=davem@davemloft.net \ --cc=dev@openvswitch.org \ --cc=jesse@kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=netdev@vger.kernel.org \ --cc=pshelar@nicira.com \ /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: linkBe 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.