1
0
Files
linux/net
Eric Dumazet 811230cd85 tcp: ipv4: initialize unicast_sock sk_pacing_rate
When I added sk_pacing_rate field, I forgot to initialize its value
in the per cpu unicast_sock used in ip_send_unicast_reply()

This means that for sch_fq users, RST packets, or ACK packets sent
on behalf of TIME_WAIT sockets might be sent to slowly or even dropped
once we reach the per flow limit.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Fixes: 95bd09eb27 ("tcp: TSO packets automatic sizing")
Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-28 23:24:47 -08:00
..
2014-10-15 10:25:04 +10:30
2014-11-24 04:28:48 -05:00
2014-12-09 16:29:03 -05:00
2014-11-24 04:28:48 -05:00
2014-12-09 16:29:03 -05:00
2014-12-07 21:22:05 +01:00
2014-11-24 04:28:51 -05:00
2015-01-25 16:00:54 -08:00
2014-11-24 04:28:48 -05:00
2014-10-09 01:40:05 -04:00
2014-11-24 04:28:48 -05:00
2014-11-24 04:28:48 -05:00
2014-11-24 04:28:48 -05:00
2014-12-09 16:29:03 -05:00
2014-12-09 16:29:03 -05:00
2014-10-07 20:28:44 -04:00
2014-11-24 04:28:48 -05:00
2014-12-09 16:29:03 -05:00
2015-01-27 12:25:33 -08:00