Re: [PATCH net-next] tcp: remove bad timeout logic in fast recovery

[Eric Dumazet <eric.dumazet@gmail.com>]

On Fri, 2013-05-17 at 16:45 -0700, Yuchung Cheng wrote:
> tcp_timeout_skb() was intended to trigger fast recovery on timeout,
> unfortunately in reality it often causes spurious retransmission
> storms during fast recovery. The particular sign is fast retransmit
> over highest sacked sequence (SND.FACK).
> 
> Currently the RTO timer re-arming (as in RFC6298) offers a nice cushion
> to avoid spurious timeout: when SND.UNA advances the sender re-arms
> RTO and extends the timeout by icsk_rto. The sender does not offset
> the time elapsed since the packet at SND.UNA was sent.
> 
> But if the next (DUP)ACK arrives later than ~RTTVAR and triggers
> tcp_fastretrans_alert(), then tcp_timeout_skb() will mark any packet
> sent before icsk_rto interval lost, including one that's above the
> highest sacked sequence. Most likely a large part of scorebard will
> be marked.
> 
> If most packets are not lost then the subsequence DUPACKs with new
> SACK blockes will cause the sender to continue retransmit packets
> beyond SND.FACK spuriously right. Even only one packet is lost the
> sender may falsely retransmit almost the entire window.
> 
> The situation becomes common in the world of bufferbloat: the RTT
> continues to grow as the queue builds up but RTTVAR remains small and
> close to the minimum 200ms. If a data packet is lost and the DUPACK
> triggered by the next data packet is slightly delayed, then a spurious
> retransmission storm forms.
> 
> As the original comment on tcp_timeout_skb() suggests: the usefulness
> of this feature is questionable. It also wastes cycles walking the
> sack scoreboard and is actually harmful because of the false recovery.
> It's time to remove this.
> 
> Signed-off-by: Yuchung Cheng <ycheng@google.com>
> ---
>  include/linux/tcp.h  |  1 -
>  include/net/tcp.h    |  1 -
>  net/ipv4/tcp_input.c | 65 +---------------------------------------------------
>  3 files changed, 1 insertion(+), 66 deletions(-)

Acked-by: Eric Dumazet <edumazet@google.com>