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 <[email protected]> Fixes: 95bd09e ("tcp: TSO packets automatic sizing") Signed-off-by: David S. Miller <[email protected]>
chewitt · Jan 29, 2015 · 811230c · 811230c
1 parent 150ae0e
commit 811230c
Showing 1 changed file with 1 addition and 0 deletions.
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
@@ -1517,6 +1517,7 @@ static DEFINE_PER_CPU(struct inet_sock, unicast_sock) = {
 		.sk_wmem_alloc	= ATOMIC_INIT(1),
 		.sk_allocation	= GFP_ATOMIC,
 		.sk_flags	= (1UL << SOCK_USE_WRITE_QUEUE),
+		.sk_pacing_rate = ~0U,
 	},
 	.pmtudisc	= IP_PMTUDISC_WANT,
 	.uc_ttl		= -1,