[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <20141113.152202.1381528737325236329.davem@davemloft.net>
Date: Thu, 13 Nov 2014 15:22:02 -0500 (EST)
From: David Miller <davem@...emloft.net>
To: eric.dumazet@...il.com
Cc: netdev@...r.kernel.org, ncardwell@...gle.com, ycheng@...gle.com,
nanditad@...gle.com
Subject: Re: [PATCH net-next] tcp: limit GSO packets to half cwnd
From: Eric Dumazet <eric.dumazet@...il.com>
Date: Thu, 13 Nov 2014 09:45:22 -0800
> From: Eric Dumazet <edumazet@...gle.com>
>
> In DC world, GSO packets initially cooked by tcp_sendmsg() are usually
> big, as sk_pacing_rate is high.
>
> When network is congested, cwnd can be smaller than the GSO packets
> found in socket write queue. tcp_write_xmit() splits GSO packets
> using the available cwnd, and we end up sending a single GSO packet,
> consuming all available cwnd.
>
> With GRO aggregation on the receiver, we might handle a single GRO
> packet, sending back a single ACK.
>
> 1) This single ACK might be lost
> TLP or RTO are forced to attempt a retransmit.
> 2) This ACK releases a full cwnd, sender sends another big GSO packet,
> in a ping pong mode.
>
> This behavior does not fill the pipes in the best way, because of
> scheduling artifacts.
>
> Make sure we always have at least two GSO packets in flight.
>
> This allows us to safely increase GRO efficiency without risking
> spurious retransmits.
>
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
This looks fantastic, applied, thanks Eric!
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists