[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <5640BEAC.10404@axis.com>
Date: Mon, 9 Nov 2015 16:41:32 +0100
From: Niklas Cassel <niklas.cassel@...s.com>
To: <edumazet@...gle.com>
CC: "netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: network stream fairness
I have a ethernet driver for a 100 Mbps NIC.
The NIC has dedicated hardware for offloading.
The driver has implemented TSO, GSO and BQL.
Since the CPU on the SoC is rather weak, I'd rather
not increase the CPU load by turning off offloading.
Since commit
605ad7f184b6 ("tcp: refine TSO autosizing")
the bandwidth is no longer fair between streams.
see output at the end of the mail, where I'm testing with 2 streams.
If I revert 605ad7f184b6 on 4.3, I get a stable 45 Mbps per stream.
I can also use vanilla 4.3 and do:
echo 3000 > /sys/class/net/eth0/queues/tx-0/byte_queue_limits/limit_max
to also get a stable 45 Mbps per stream.
My question is, am I supposed to set the BQL limit explicitly?
It is possible that I have missed something in my driver,
but my understanding is that the TCP stack sets and adjusts
the BQL limit automatically.
Perhaps the following info might help:
After running iperf3 on vanilla 4.3:
/sys/class/net/eth0/queues/tx-0/byte_queue_limits/
limit 89908
limit_max 1879048192
After running iperf3 on vanilla 4.3 + BQL explicitly set:
/sys/class/net/eth0/queues/tx-0/byte_queue_limits/
limit 3000
limit_max 3000
After running iperf3 on 4.3 + 605ad7f184b6 reverted:
/sys/class/net/eth0/queues/tx-0/byte_queue_limits/
limit 8886
limit_max 1879048192
Vanilla 4.3, no BQL limit explicitly set.
iperf3 -R -c 192.168.0.141 -t 240 -P 2
Connecting to host 192.168.0.141, port 5201
Reverse mode, remote host 192.168.0.141 is sending
[ 4] local 192.168.0.1 port 50902 connected to 192.168.0.141 port 5201
[ 6] local 192.168.0.1 port 50904 connected to 192.168.0.141 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 7.24 MBytes 60.7 Mbits/sec
[ 6] 0.00-1.00 sec 3.56 MBytes 29.9 Mbits/sec
[SUM] 0.00-1.00 sec 10.8 MBytes 90.6 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 1.00-2.00 sec 7.40 MBytes 62.1 Mbits/sec
[ 6] 1.00-2.00 sec 3.40 MBytes 28.5 Mbits/sec
[SUM] 1.00-2.00 sec 10.8 MBytes 90.6 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 2.00-3.00 sec 7.45 MBytes 62.5 Mbits/sec
[ 6] 2.00-3.00 sec 3.36 MBytes 28.2 Mbits/sec
[SUM] 2.00-3.00 sec 10.8 MBytes 90.7 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 3.00-4.00 sec 7.72 MBytes 64.8 Mbits/sec
[ 6] 3.00-4.00 sec 3.08 MBytes 25.9 Mbits/sec
[SUM] 3.00-4.00 sec 10.8 MBytes 90.7 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 4.00-5.00 sec 8.01 MBytes 67.2 Mbits/sec
[ 6] 4.00-5.00 sec 2.80 MBytes 23.5 Mbits/sec
[SUM] 4.00-5.00 sec 10.8 MBytes 90.7 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 5.00-6.00 sec 7.91 MBytes 66.3 Mbits/sec
[ 6] 5.00-6.00 sec 2.90 MBytes 24.3 Mbits/sec
[SUM] 5.00-6.00 sec 10.8 MBytes 90.7 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 6.00-7.00 sec 7.77 MBytes 65.2 Mbits/sec
[ 6] 6.00-7.00 sec 3.04 MBytes 25.5 Mbits/sec
[SUM] 6.00-7.00 sec 10.8 MBytes 90.7 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 7.00-8.00 sec 6.99 MBytes 58.6 Mbits/sec
[ 6] 7.00-8.00 sec 3.82 MBytes 32.0 Mbits/sec
[SUM] 7.00-8.00 sec 10.8 MBytes 90.7 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 8.00-9.00 sec 7.08 MBytes 59.4 Mbits/sec
[ 6] 8.00-9.00 sec 3.73 MBytes 31.3 Mbits/sec
[SUM] 8.00-9.00 sec 10.8 MBytes 90.7 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ 4] 9.00-10.00 sec 6.93 MBytes 58.1 Mbits/sec
[ 6] 9.00-10.00 sec 3.88 MBytes 32.6 Mbits/sec
[SUM] 9.00-10.00 sec 10.8 MBytes 90.7 Mbits/sec
--
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