From: Eric Dumazet <eric.dumazet@gmail.com>
To: Julius Werner <jwerner@chromium.org>
Cc: linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
Patrick McHardy <kaber@trash.net>,
Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>,
James Morris <jmorris@namei.org>,
Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>,
"David S. Miller" <davem@davemloft.net>,
Dave Jones <davej@redhat.com>, Sameer Nanda <snanda@chromium.org>,
Mandeep Singh Baines <msb@chromium.org>
Subject: Re: [PATCH] tcp: Avoid infinite loop on recvmsg bug
Date: Wed, 07 Nov 2012 15:42:34 -0800 [thread overview]
Message-ID: <1352331754.2748.14.camel@edumazet-glaptop> (raw)
In-Reply-To: <1352331192.2748.10.camel@edumazet-glaptop>
On Wed, 2012-11-07 at 15:33 -0800, Eric Dumazet wrote:
> So you probably are fighting a bug we already fixed in upstream kernel.
>
> (commit c8628155ece363 "tcp: reduce out_of_order memory use" did not
> played well with cloned skbs.)
>
> This issue was already discussed on netdev in the past.
If you use a 3.4 kernel, you want the following patch.
(I guess you could reproduce the crash easily running a tcpdump in //)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 257b617..9f8f68c 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4496,7 +4496,9 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb)
* to avoid future tcp_collapse_ofo_queue(),
* probably the most expensive function in tcp stack.
*/
- if (skb->len <= skb_tailroom(skb1) && !tcp_hdr(skb)->fin) {
+ if (skb->len <= skb_tailroom(skb1) &&
+ !tcp_hdr(skb)->fin &&
+ !skb_cloned(skb1)) {
NET_INC_STATS_BH(sock_net(sk),
LINUX_MIB_TCPRCVCOALESCE);
BUG_ON(skb_copy_bits(skb, 0,
next prev parent reply other threads:[~2012-11-07 23:42 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-07 0:15 [PATCH] tcp: Replace infinite loop on recvmsg bug with proper crash Julius Werner
2012-11-07 1:39 ` Dave Jones
2012-11-07 1:51 ` Julius Werner
2012-11-07 15:54 ` Dave Jones
2012-11-07 16:29 ` [PATCH] tcp: Replace infinite loop on recvmsg bug with proper crashusers Eric Dumazet
2012-11-07 16:43 ` Dave Jones
2012-11-07 17:05 ` Eric Dumazet
2012-11-07 17:15 ` Dave Jones
2012-11-07 19:32 ` Julius Werner
2012-11-07 19:33 ` [PATCH] tcp: Avoid infinite loop on recvmsg bug Julius Werner
2012-11-07 19:40 ` Eric Dumazet
2012-11-07 21:14 ` Julius Werner
2012-11-07 23:33 ` Eric Dumazet
2012-11-07 23:42 ` Eric Dumazet [this message]
2012-11-08 2:25 ` Julius Werner
2012-11-09 3:29 ` Eric Dumazet
2012-12-10 19:33 ` Julius Werner
2012-12-10 20:23 ` David Miller
2012-11-07 1:51 ` [PATCH] tcp: Replace infinite loop on recvmsg bug with proper crash Eric Dumazet
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=1352331754.2748.14.camel@edumazet-glaptop \
--to=eric.dumazet@gmail.com \
--cc=davej@redhat.com \
--cc=davem@davemloft.net \
--cc=jmorris@namei.org \
--cc=jwerner@chromium.org \
--cc=kaber@trash.net \
--cc=kuznet@ms2.inr.ac.ru \
--cc=linux-kernel@vger.kernel.org \
--cc=msb@chromium.org \
--cc=netdev@vger.kernel.org \
--cc=snanda@chromium.org \
--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