diff options
author | Damian Lukowski <damian@tvk.rwth-aachen.de> | 2009-12-07 01:06:15 -0500 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2009-12-08 23:56:11 -0500 |
commit | 07f29bc5bbae4e53e982ab956fed7207990a7786 (patch) | |
tree | b44506c37ce4886220439b38ef647c3cd7fbc6ce /include/net | |
parent | cb19054697e92a793f336380fd72c588521178ff (diff) |
tcp: Stalling connections: Fix timeout calculation routine
This patch fixes a problem in the TCP connection timeout calculation.
Currently, timeout decisions are made on the basis of the current
tcp_time_stamp and retrans_stamp, which is usually set at the first
retransmission.
However, if the retransmission fails in tcp_retransmit_skb(),
retrans_stamp is not updated and remains zero. This leads to wrong
decisions in retransmits_timed_out() if tcp_time_stamp is larger than
the specified timeout, which is very likely.
In this case, the TCP connection dies after the first attempted
(and unsuccessful) retransmission.
With this patch, tcp_skb_cb->when is used instead, when retrans_stamp
is not available.
This bug has been introduced together with retransmits_timed_out() in
2.6.32, as the number of retransmissions has been used for timeout
decisions before. The corresponding commit was
6fa12c85031485dff38ce550c24f10da23b0adaa (Revert Backoff [v3]:
Calculate TCP's connection close threshold as a time value.).
Thanks to Ilpo Järvinen for code suggestions and Frederic Leroy for
testing.
Reported-by: Frederic Leroy <fredo@starox.org>
Signed-off-by: Damian Lukowski <damian@tvk.rwth-aachen.de>
Acked-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'include/net')
-rw-r--r-- | include/net/tcp.h | 10 |
1 files changed, 8 insertions, 2 deletions
diff --git a/include/net/tcp.h b/include/net/tcp.h index e2d2ca2509be..e54bd85d9d40 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h | |||
@@ -1265,14 +1265,20 @@ static inline struct sk_buff *tcp_write_queue_prev(struct sock *sk, struct sk_bu | |||
1265 | * TCP connection after "boundary" unsucessful, exponentially backed-off | 1265 | * TCP connection after "boundary" unsucessful, exponentially backed-off |
1266 | * retransmissions with an initial RTO of TCP_RTO_MIN. | 1266 | * retransmissions with an initial RTO of TCP_RTO_MIN. |
1267 | */ | 1267 | */ |
1268 | static inline bool retransmits_timed_out(const struct sock *sk, | 1268 | static inline bool retransmits_timed_out(struct sock *sk, |
1269 | unsigned int boundary) | 1269 | unsigned int boundary) |
1270 | { | 1270 | { |
1271 | unsigned int timeout, linear_backoff_thresh; | 1271 | unsigned int timeout, linear_backoff_thresh; |
1272 | unsigned int start_ts; | ||
1272 | 1273 | ||
1273 | if (!inet_csk(sk)->icsk_retransmits) | 1274 | if (!inet_csk(sk)->icsk_retransmits) |
1274 | return false; | 1275 | return false; |
1275 | 1276 | ||
1277 | if (unlikely(!tcp_sk(sk)->retrans_stamp)) | ||
1278 | start_ts = TCP_SKB_CB(tcp_write_queue_head(sk))->when; | ||
1279 | else | ||
1280 | start_ts = tcp_sk(sk)->retrans_stamp; | ||
1281 | |||
1276 | linear_backoff_thresh = ilog2(TCP_RTO_MAX/TCP_RTO_MIN); | 1282 | linear_backoff_thresh = ilog2(TCP_RTO_MAX/TCP_RTO_MIN); |
1277 | 1283 | ||
1278 | if (boundary <= linear_backoff_thresh) | 1284 | if (boundary <= linear_backoff_thresh) |
@@ -1281,7 +1287,7 @@ static inline bool retransmits_timed_out(const struct sock *sk, | |||
1281 | timeout = ((2 << linear_backoff_thresh) - 1) * TCP_RTO_MIN + | 1287 | timeout = ((2 << linear_backoff_thresh) - 1) * TCP_RTO_MIN + |
1282 | (boundary - linear_backoff_thresh) * TCP_RTO_MAX; | 1288 | (boundary - linear_backoff_thresh) * TCP_RTO_MAX; |
1283 | 1289 | ||
1284 | return (tcp_time_stamp - tcp_sk(sk)->retrans_stamp) >= timeout; | 1290 | return (tcp_time_stamp - start_ts) >= timeout; |
1285 | } | 1291 | } |
1286 | 1292 | ||
1287 | static inline struct sk_buff *tcp_send_head(struct sock *sk) | 1293 | static inline struct sk_buff *tcp_send_head(struct sock *sk) |