From 1f2556916d974cfb62b6af51660186b5f58bd869 Mon Sep 17 00:00:00 2001 From: Priyaranjan Jha Date: Fri, 3 Nov 2017 16:38:48 -0700 Subject: tcp: higher throughput under reordering with adaptive RACK reordering wnd Currently TCP RACK loss detection does not work well if packets are being reordered beyond its static reordering window (min_rtt/4).Under such reordering it may falsely trigger loss recoveries and reduce TCP throughput significantly. This patch improves that by increasing and reducing the reordering window based on DSACK, which is now supported in major TCP implementations. It makes RACK's reo_wnd adaptive based on DSACK and no. of recoveries. - If DSACK is received, increment reo_wnd by min_rtt/4 (upper bounded by srtt), since there is possibility that spurious retransmission was due to reordering delay longer than reo_wnd. - Persist the current reo_wnd value for TCP_RACK_RECOVERY_THRESH (16) no. of successful recoveries (accounts for full DSACK-based loss recovery undo). After that, reset it to default (min_rtt/4). - At max, reo_wnd is incremented only once per rtt. So that the new DSACK on which we are reacting, is due to the spurious retx (approx) after the reo_wnd has been updated last time. - reo_wnd is tracked in terms of steps (of min_rtt/4), rather than absolute value to account for change in rtt. In our internal testing, we observed significant increase in throughput, in scenarios where reordering exceeds min_rtt/4 (previous static value). Signed-off-by: Priyaranjan Jha Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- include/linux/tcp.h | 9 +++++++-- include/net/tcp.h | 2 ++ 2 files changed, 9 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 8c431385b272..22f40c96a15b 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -210,8 +210,13 @@ struct tcp_sock { u64 mstamp; /* (Re)sent time of the skb */ u32 rtt_us; /* Associated RTT */ u32 end_seq; /* Ending TCP sequence of the skb */ - u8 advanced; /* mstamp advanced since last lost marking */ - u8 reord; /* reordering detected */ + u32 last_delivered; /* tp->delivered at last reo_wnd adj */ + u8 reo_wnd_steps; /* Allowed reordering window */ +#define TCP_RACK_RECOVERY_THRESH 16 + u8 reo_wnd_persist:5, /* No. of recovery since last adj */ + dsack_seen:1, /* Whether DSACK seen after last adj */ + advanced:1, /* mstamp advanced since last lost marking */ + reord:1; /* reordering detected */ } rack; u16 advmss; /* Advertised MSS */ u32 chrono_start; /* Start time in jiffies of a TCP chrono */ diff --git a/include/net/tcp.h b/include/net/tcp.h index c2bf2a822b10..babfd4da1515 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -246,6 +246,7 @@ extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ +#define TCP_RACK_STATIC_REO_WND 0x2 /* Use static RACK reo wnd */ extern atomic_long_t tcp_memory_allocated; extern struct percpu_counter tcp_sockets_allocated; @@ -1901,6 +1902,7 @@ extern void tcp_rack_mark_lost(struct sock *sk); extern void tcp_rack_advance(struct tcp_sock *tp, u8 sacked, u32 end_seq, u64 xmit_time); extern void tcp_rack_reo_timeout(struct sock *sk); +extern void tcp_rack_update_reo_wnd(struct sock *sk, struct rate_sample *rs); /* At how many usecs into the future should the RTO fire? */ static inline s64 tcp_rto_delta_us(const struct sock *sk) -- cgit v1.2.3