From: <gregkh@linuxfoundation.org>
To: edumazet@google.com, davem@davemloft.net,
gregkh@linuxfoundation.org, ycai@google.com
Cc: <stable@vger.kernel.org>, <stable-commits@vger.kernel.org>
Subject: Patch "tcp/dccp: fix timewait races in timer handling" has been added to the 4.1-stable tree
Date: Tue, 26 Jan 2016 22:28:09 -0800 [thread overview]
Message-ID: <145387608984246@kroah.com> (raw)
This is a note to let you know that I've just added the patch titled
tcp/dccp: fix timewait races in timer handling
to the 4.1-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
tcp-dccp-fix-timewait-races-in-timer-handling.patch
and it can be found in the queue-4.1 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.
>From foo@baz Tue Jan 26 21:37:04 PST 2016
From: Eric Dumazet <edumazet@google.com>
Date: Sat, 19 Sep 2015 09:08:34 -0700
Subject: tcp/dccp: fix timewait races in timer handling
From: Eric Dumazet <edumazet@google.com>
[ Upstream commit ed2e923945892a8372ab70d2f61d364b0b6d9054 ]
When creating a timewait socket, we need to arm the timer before
allowing other cpus to find it. The signal allowing cpus to find
the socket is setting tw_refcnt to non zero value.
As we set tw_refcnt in __inet_twsk_hashdance(), we therefore need to
call inet_twsk_schedule() first.
This also means we need to remove tw_refcnt changes from
inet_twsk_schedule() and let the caller handle it.
Note that because we use mod_timer_pinned(), we have the guarantee
the timer wont expire before we set tw_refcnt as we run in BH context.
To make things more readable I introduced inet_twsk_reschedule() helper.
When rearming the timer, we can use mod_timer_pending() to make sure
we do not rearm a canceled timer.
Note: This bug can possibly trigger if packets of a flow can hit
multiple cpus. This does not normally happen, unless flow steering
is broken somehow. This explains this bug was spotted ~5 months after
its introduction.
A similar fix is needed for SYN_RECV sockets in reqsk_queue_hash_req(),
but will be provided in a separate patch for proper tracking.
Fixes: 789f558cfb36 ("tcp/dccp: get rid of central timewait timer")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Ying Cai <ycai@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
include/net/inet_timewait_sock.h | 14 +++++++++++++-
net/dccp/minisocks.c | 4 ++--
net/ipv4/inet_timewait_sock.c | 16 ++++++++++------
net/ipv4/tcp_minisocks.c | 13 ++++++-------
4 files changed, 31 insertions(+), 16 deletions(-)
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -112,7 +112,19 @@ struct inet_timewait_sock *inet_twsk_all
void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
struct inet_hashinfo *hashinfo);
-void inet_twsk_schedule(struct inet_timewait_sock *tw, const int timeo);
+void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo,
+ bool rearm);
+
+static void inline inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo)
+{
+ __inet_twsk_schedule(tw, timeo, false);
+}
+
+static void inline inet_twsk_reschedule(struct inet_timewait_sock *tw, int timeo)
+{
+ __inet_twsk_schedule(tw, timeo, true);
+}
+
void inet_twsk_deschedule(struct inet_timewait_sock *tw);
void inet_twsk_purge(struct inet_hashinfo *hashinfo,
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -48,8 +48,6 @@ void dccp_time_wait(struct sock *sk, int
tw->tw_ipv6only = sk->sk_ipv6only;
}
#endif
- /* Linkage updates. */
- __inet_twsk_hashdance(tw, sk, &dccp_hashinfo);
/* Get the TIME_WAIT timeout firing. */
if (timeo < rto)
@@ -60,6 +58,8 @@ void dccp_time_wait(struct sock *sk, int
timeo = DCCP_TIMEWAIT_LEN;
inet_twsk_schedule(tw, timeo);
+ /* Linkage updates. */
+ __inet_twsk_hashdance(tw, sk, &dccp_hashinfo);
inet_twsk_put(tw);
} else {
/* Sorry, if we're out of memory, just CLOSE this
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -153,13 +153,15 @@ void __inet_twsk_hashdance(struct inet_t
/*
* Step 2: Hash TW into tcp ehash chain.
* Notes :
- * - tw_refcnt is set to 3 because :
+ * - tw_refcnt is set to 4 because :
* - We have one reference from bhash chain.
* - We have one reference from ehash chain.
+ * - We have one reference from timer.
+ * - One reference for ourself (our caller will release it).
* We can use atomic_set() because prior spin_lock()/spin_unlock()
* committed into memory all tw fields.
*/
- atomic_set(&tw->tw_refcnt, 1 + 1 + 1);
+ atomic_set(&tw->tw_refcnt, 4);
inet_twsk_add_node_rcu(tw, &ehead->chain);
/* Step 3: Remove SK from hash chain */
@@ -243,7 +245,7 @@ void inet_twsk_deschedule(struct inet_ti
}
EXPORT_SYMBOL(inet_twsk_deschedule);
-void inet_twsk_schedule(struct inet_timewait_sock *tw, const int timeo)
+void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, bool rearm)
{
/* timeout := RTO * 3.5
*
@@ -271,12 +273,14 @@ void inet_twsk_schedule(struct inet_time
*/
tw->tw_kill = timeo <= 4*HZ;
- if (!mod_timer_pinned(&tw->tw_timer, jiffies + timeo)) {
- atomic_inc(&tw->tw_refcnt);
+ if (!rearm) {
+ BUG_ON(mod_timer_pinned(&tw->tw_timer, jiffies + timeo));
atomic_inc(&tw->tw_dr->tw_count);
+ } else {
+ mod_timer_pending(&tw->tw_timer, jiffies + timeo);
}
}
-EXPORT_SYMBOL_GPL(inet_twsk_schedule);
+EXPORT_SYMBOL_GPL(__inet_twsk_schedule);
void inet_twsk_purge(struct inet_hashinfo *hashinfo,
struct inet_timewait_death_row *twdr, int family)
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -163,9 +163,9 @@ kill_with_rst:
if (tcp_death_row.sysctl_tw_recycle &&
tcptw->tw_ts_recent_stamp &&
tcp_tw_remember_stamp(tw))
- inet_twsk_schedule(tw, tw->tw_timeout);
+ inet_twsk_reschedule(tw, tw->tw_timeout);
else
- inet_twsk_schedule(tw, TCP_TIMEWAIT_LEN);
+ inet_twsk_reschedule(tw, TCP_TIMEWAIT_LEN);
return TCP_TW_ACK;
}
@@ -203,7 +203,7 @@ kill:
return TCP_TW_SUCCESS;
}
}
- inet_twsk_schedule(tw, TCP_TIMEWAIT_LEN);
+ inet_twsk_reschedule(tw, TCP_TIMEWAIT_LEN);
if (tmp_opt.saw_tstamp) {
tcptw->tw_ts_recent = tmp_opt.rcv_tsval;
@@ -253,7 +253,7 @@ kill:
* Do not reschedule in the last case.
*/
if (paws_reject || th->ack)
- inet_twsk_schedule(tw, TCP_TIMEWAIT_LEN);
+ inet_twsk_reschedule(tw, TCP_TIMEWAIT_LEN);
return tcp_timewait_check_oow_rate_limit(
tw, skb, LINUX_MIB_TCPACKSKIPPEDTIMEWAIT);
@@ -324,9 +324,6 @@ void tcp_time_wait(struct sock *sk, int
} while (0);
#endif
- /* Linkage updates. */
- __inet_twsk_hashdance(tw, sk, &tcp_hashinfo);
-
/* Get the TIME_WAIT timeout firing. */
if (timeo < rto)
timeo = rto;
@@ -340,6 +337,8 @@ void tcp_time_wait(struct sock *sk, int
}
inet_twsk_schedule(tw, timeo);
+ /* Linkage updates. */
+ __inet_twsk_hashdance(tw, sk, &tcp_hashinfo);
inet_twsk_put(tw);
} else {
/* Sorry, if we're out of memory, just CLOSE this
Patches currently in stable-queue which might be from edumazet@google.com are
queue-4.1/net-sched-fix-missing-free-per-cpu-on-qstats.patch
queue-4.1/ipv6-tcp-add-rcu-locking-in-tcp_v6_send_synack.patch
queue-4.1/ipv6-update-skb-csum-when-ce-mark-is-propagated.patch
queue-4.1/tcp-dccp-fix-old-style-declarations.patch
queue-4.1/net-possible-use-after-free-in-dst_release.patch
queue-4.1/phonet-properly-unshare-skbs-in-phonet_rcv.patch
queue-4.1/tcp-dccp-fix-timewait-races-in-timer-handling.patch
queue-4.1/tcp_yeah-don-t-set-ssthresh-below-2.patch
reply other threads:[~2016-01-27 6:31 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=145387608984246@kroah.com \
--to=gregkh@linuxfoundation.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=stable-commits@vger.kernel.org \
--cc=stable@vger.kernel.org \
--cc=ycai@google.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: 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).