tcp: fix stretch ACK bugs in Scalable (5415e3c3) · Commits · 戴 / test

Change Scalable to properly handle stretch ACKs in additive increase mode by passing in the count of ACKed packets to tcp_cong_avoid_ai(). In addition, because we are now precisely accounting for stretch ACKs, including delayed ACKs, we can now change TCP_SCALABLE_AI_CNT to 100. Signed-off-by:

Pengcheng Yang <yangpc@wangsu.com> Acked-by:

Neal Cardwell <ncardwell@google.com> Signed-off-by:

David S. Miller <davem@davemloft.net>

net/ipv4/tcp_scalable.c

+9 −8

Original line number	Diff line number	Diff line
		@@ -10,10 +10,9 @@
		#include <net/tcp.h>

		/* These factors derived from the recommended values in the aer:
		* .01 and and 7/8. We use 50 instead of 100 to account for
		* delayed ack.
		* .01 and and 7/8.
		*/
		#define TCP_SCALABLE_AI_CNT 50U
		#define TCP_SCALABLE_AI_CNT 100U
		#define TCP_SCALABLE_MD_SCALE 3

		static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 acked)
		@@ -23,11 +22,13 @@ static void tcp_scalable_cong_avoid(struct sock *sk, u32 ack, u32 acked)
		if (!tcp_is_cwnd_limited(sk))
		return;

		if (tcp_in_slow_start(tp))
		tcp_slow_start(tp, acked);
		else
		if (tcp_in_slow_start(tp)) {
		acked = tcp_slow_start(tp, acked);
		if (!acked)
		return;
		}
		tcp_cong_avoid_ai(tp, min(tp->snd_cwnd, TCP_SCALABLE_AI_CNT),
		1);
		acked);
		}

		static u32 tcp_scalable_ssthresh(struct sock *sk)

Admin message