From: Menglong Dong <menglong8.dong@gmail.com>
To: Eric Dumazet <edumazet@gmail.com>
Cc: David Ahern <dsahern@kernel.org>,
Steven Rostedt <rostedt@goodmis.org>,
Ingo Molnar <mingo@redhat.com>,
David Miller <davem@davemloft.net>,
Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>,
Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
Biao Jiang <benbjiang@tencent.com>,
Hao Peng <flyingpeng@tencent.com>,
Menglong Dong <imagedong@tencent.com>,
Eric Dumazet <edumazet@google.com>, Martin Lau <kafai@fb.com>,
Talal Ahmad <talalahmad@google.com>,
Kees Cook <keescook@chromium.org>,
Mengen Sun <mengensun@tencent.com>,
dongli.zhang@oracle.com, LKML <linux-kernel@vger.kernel.org>,
netdev <netdev@vger.kernel.org>
Subject: Re: [PATCH net-next 8/9] net: ipv6: add skb drop reasons to ip6_rcv_core()
Date: Thu, 14 Apr 2022 09:20:21 +0800 [thread overview]
Message-ID: <CADxym3Z+Ginefiw7ymbZFGa3QC5qaT5jM3DjedijzGwYoFqPaw@mail.gmail.com> (raw)
In-Reply-To: <62903323-d1c0-79cd-7cf8-d6c4b89ff848@gmail.com>
On Thu, Apr 14, 2022 at 4:41 AM Eric Dumazet <edumazet@gmail.com> wrote:
>
>
> On 4/13/22 01:15, menglong8.dong@gmail.com wrote:
> > From: Menglong Dong <imagedong@tencent.com>
> >
> > Replace kfree_skb() used in ip6_rcv_core() with kfree_skb_reason().
> > No new drop reasons are added.
> >
> > Seems now we use 'SKB_DROP_REASON_IP_INHDR' for too many case during
> > ipv6 header parse or check, just like what 'IPSTATS_MIB_INHDRERRORS'
> > do. Will it be too general and hard to know what happened?
> >
> > Signed-off-by: Menglong Dong <imagedong@tencent.com>
> > Reviewed-by: Jiang Biao <benbjiang@tencent.com>
> > Reviewed-by: Hao Peng <flyingpeng@tencent.com>
> > ---
> > net/ipv6/ip6_input.c | 24 ++++++++++++++++--------
> > 1 file changed, 16 insertions(+), 8 deletions(-)
> >
> > diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
> > index b4880c7c84eb..1b925ecb26e9 100644
> > --- a/net/ipv6/ip6_input.c
> > +++ b/net/ipv6/ip6_input.c
> > @@ -145,13 +145,14 @@ static void ip6_list_rcv_finish(struct net *net, struct sock *sk,
> > static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
> > struct net *net)
> > {
> > + enum skb_drop_reason reason;
> > const struct ipv6hdr *hdr;
> > u32 pkt_len;
> > struct inet6_dev *idev;
> >
> > if (skb->pkt_type == PACKET_OTHERHOST) {
> > dev_core_stats_rx_otherhost_dropped_inc(skb->dev);
> > - kfree_skb(skb);
> > + kfree_skb_reason(skb, SKB_DROP_REASON_OTHERHOST);
> > return NULL;
> > }
> >
> > @@ -161,9 +162,12 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
> >
> > __IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_IN, skb->len);
> >
> > + SKB_DR_SET(reason, NOT_SPECIFIED);
> > if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL ||
> > !idev || unlikely(idev->cnf.disable_ipv6)) {
> > __IP6_INC_STATS(net, idev, IPSTATS_MIB_INDISCARDS);
> > + if (unlikely(idev->cnf.disable_ipv6))
>
> idev can be NULL here (according to surrounding code), and we crash :/
Yeah, you are right :)
Thanks for your fix!
>
>
>
> > + SKB_DR_SET(reason, IPV6DISABLED);
> > goto drop;
> > }
> >
> > @@ -187,8 +191,10 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
> >
> > hdr = ipv6_hdr(skb);
> >
> > - if (hdr->version != 6)
> > + if (hdr->version != 6) {
> > + SKB_DR_SET(reason, UNHANDLED_PROTO);
> > goto err;
> > + }
> >
> > __IP6_ADD_STATS(net, idev,
> > IPSTATS_MIB_NOECTPKTS +
> > @@ -226,8 +232,10 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
> > if (!ipv6_addr_is_multicast(&hdr->daddr) &&
> > (skb->pkt_type == PACKET_BROADCAST ||
> > skb->pkt_type == PACKET_MULTICAST) &&
> > - idev->cnf.drop_unicast_in_l2_multicast)
> > + idev->cnf.drop_unicast_in_l2_multicast) {
> > + SKB_DR_SET(reason, UNICAST_IN_L2_MULTICAST);
> > goto err;
> > + }
> >
> > /* RFC4291 2.7
> > * Nodes must not originate a packet to a multicast address whose scope
> > @@ -256,12 +264,11 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
> > if (pkt_len + sizeof(struct ipv6hdr) > skb->len) {
> > __IP6_INC_STATS(net,
> > idev, IPSTATS_MIB_INTRUNCATEDPKTS);
> > + SKB_DR_SET(reason, PKT_TOO_SMALL);
> > goto drop;
> > }
> > - if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr))) {
> > - __IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
> > - goto drop;
> > - }
> > + if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr)))
> > + goto err;
> > hdr = ipv6_hdr(skb);
> > }
> >
> > @@ -282,9 +289,10 @@ static struct sk_buff *ip6_rcv_core(struct sk_buff *skb, struct net_device *dev,
> > return skb;
> > err:
> > __IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
> > + SKB_DR_OR(reason, IP_INHDR);
> > drop:
> > rcu_read_unlock();
> > - kfree_skb(skb);
> > + kfree_skb_reason(skb, reason);
> > return NULL;
> > }
> >
next prev parent reply other threads:[~2022-04-14 1:20 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-13 8:15 [PATCH net-next 0/9] net: ip: add skb drop reasons to ip ingress menglong8.dong
2022-04-13 8:15 ` [PATCH net-next 1/9] skb: add some helpers for skb drop reasons menglong8.dong
2022-04-13 8:15 ` [PATCH net-next 2/9] net: ipv4: add skb drop reasons to ip_error() menglong8.dong
2022-04-13 8:15 ` [PATCH net-next 3/9] net: ipv6: add skb drop reasons to ip6_pkt_drop() menglong8.dong
2022-04-13 8:15 ` [PATCH net-next 4/9] net: ip: add skb drop reasons to ip forwarding menglong8.dong
2022-04-13 8:15 ` [PATCH net-next 5/9] net: icmp: introduce function icmpv6_param_prob_reason() menglong8.dong
2022-04-13 8:15 ` [PATCH net-next 6/9] net: ipv6: remove redundant statistics in ipv6_hop_jumbo() menglong8.dong
2022-04-13 8:15 ` [PATCH net-next 7/9] net: ipv6: add skb drop reasons to TLV parse menglong8.dong
2022-04-13 8:15 ` [PATCH net-next 8/9] net: ipv6: add skb drop reasons to ip6_rcv_core() menglong8.dong
2022-04-13 20:40 ` Eric Dumazet
2022-04-14 1:20 ` Menglong Dong [this message]
2022-04-13 8:16 ` [PATCH net-next 9/9] net: ipv6: add skb drop reasons to ip6_protocol_deliver_rcu() menglong8.dong
2022-04-13 13:30 ` [PATCH net-next 0/9] net: ip: add skb drop reasons to ip ingress patchwork-bot+netdevbpf
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=CADxym3Z+Ginefiw7ymbZFGa3QC5qaT5jM3DjedijzGwYoFqPaw@mail.gmail.com \
--to=menglong8.dong@gmail.com \
--cc=benbjiang@tencent.com \
--cc=davem@davemloft.net \
--cc=dongli.zhang@oracle.com \
--cc=dsahern@kernel.org \
--cc=edumazet@gmail.com \
--cc=edumazet@google.com \
--cc=flyingpeng@tencent.com \
--cc=imagedong@tencent.com \
--cc=kafai@fb.com \
--cc=keescook@chromium.org \
--cc=kuba@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mengensun@tencent.com \
--cc=mingo@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=rostedt@goodmis.org \
--cc=talalahmad@google.com \
--cc=yoshfuji@linux-ipv6.org \
/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: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).