public inbox for netdev@vger.kernel.org
 help / color / mirror / Atom feed
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Zoltan Kiss <zoltan.kiss@citrix.com>
Cc: Wei Liu <wei.liu2@citrix.com>,
	Jonathan Davies <Jonathan.Davies@eu.citrix.com>,
	Ian Campbell <ian.campbell@citrix.com>,
	netdev@vger.kernel.org, xen-devel@lists.xenproject.org
Subject: Re: TSQ accounting skb->truesize degrades throughput for large packets
Date: Fri, 06 Sep 2013 09:56:44 -0700	[thread overview]
Message-ID: <1378486604.31445.34.camel@edumazet-glaptop> (raw)
In-Reply-To: <522A049A.7000105@citrix.com>

On Fri, 2013-09-06 at 17:36 +0100, Zoltan Kiss wrote:
> On 06/09/13 13:57, Eric Dumazet wrote:
> > Well, I have no problem to get line rate on 20Gb with a single flow, so
> > other drivers have no problem.
> I've made some tests on bare metal:
> Dell PE R815, Intel 82599EB 10Gb, 3.11-rc4 32 bit kernel with 3.17.3 
> ixgbe (TSO, GSO on), iperf 2.0.5
> Transmitting packets toward the remote end (so running iperf -c on this 
> host) can make 8.3 Gbps with the default 128k tcp_limit_output_bytes. 
> When I increased this to 131.506 (128k + 434 bytes) suddenly it jumped 
> to 9.4 Gbps. Iperf CPU usage also jumped a few percent from ~36 to ~40% 
> (softint percentage in top also increased from ~3 to ~5%)

Typical tradeoff between latency and throughput

If you favor throughput, then you can increase tcp_limit_output_bytes

The default is quite reasonable IMHO.

> So I guess it would be good to revisit the default value of this 
> setting. What hw you used Eric for your 20Gb results?

Mellanox CX-3

Make sure your NIC doesn't hold TX packets in TX ring too long before
signaling an interrupt for TX completion.

For example I had to patch mellanox :

commit ecfd2ce1a9d5e6376ff5c00b366345160abdbbb7
Author: Eric Dumazet <edumazet@google.com>
Date:   Mon Nov 5 16:20:42 2012 +0000

    mlx4: change TX coalescing defaults
    
    mlx4 currently uses a too high tx coalescing setting, deferring
    TX completion interrupts by up to 128 us.
    
    With the recent skb_orphan() removal in commit 8112ec3b872,
    performance of a single TCP flow is capped to ~4 Gbps, unless
    we increase tcp_limit_output_bytes.
    
    I suggest using 16 us instead of 128 us, allowing a finer control.
    
    Performance of a single TCP flow is restored to previous levels,
    while keeping TCP small queues fully enabled with default sysctl.
    
    This patch is also a BQL prereq.
    
    Reported-by: Vimalkumar <j.vimal@gmail.com>
    Signed-off-by: Eric Dumazet <edumazet@google.com>
    Cc: Yevgeny Petrilin <yevgenyp@mellanox.com>
    Cc: Or Gerlitz <ogerlitz@mellanox.com>
    Acked-by: Amir Vadai <amirv@mellanox.com>
    Signed-off-by: David S. Miller <davem@davemloft.net>

  reply	other threads:[~2013-09-06 16:56 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-06 10:16 TSQ accounting skb->truesize degrades throughput for large packets Wei Liu
2013-09-06 12:57 ` Eric Dumazet
2013-09-06 13:12   ` Wei Liu
2013-09-06 16:36   ` Zoltan Kiss
2013-09-06 16:56     ` Eric Dumazet [this message]
2013-09-09  9:27       ` Jason Wang
2013-09-09 13:47         ` Eric Dumazet
2013-09-10  7:45           ` Jason Wang
2013-09-10 12:35             ` Eric Dumazet
2013-09-06 17:00     ` Eric Dumazet
2013-09-07 17:21       ` Eric Dumazet
2013-09-09 21:41         ` Zoltan Kiss
2013-09-09 21:56           ` Eric Dumazet
     [not found]             ` <loom.20130921T045654-573@post.gmane.org>
     [not found]               ` <20130921150327.GA9078@zion.uk.xensource.com>
2013-09-22  2:36                 ` [Xen-devel] " Cong Wang
2013-09-22 14:58                   ` Eric Dumazet
2013-09-27 10:28                     ` [PATCH] tcp: TSQ can use a dynamic limit Eric Dumazet
2013-09-27 15:08                       ` Neal Cardwell
2013-09-29 15:41                       ` Cong Wang
2013-10-01  3:52                       ` David Miller
2013-09-09  5:28       ` TSQ accounting skb->truesize degrades throughput for large packets Cong Wang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1378486604.31445.34.camel@edumazet-glaptop \
    --to=eric.dumazet@gmail.com \
    --cc=Jonathan.Davies@eu.citrix.com \
    --cc=ian.campbell@citrix.com \
    --cc=netdev@vger.kernel.org \
    --cc=wei.liu2@citrix.com \
    --cc=xen-devel@lists.xenproject.org \
    --cc=zoltan.kiss@citrix.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox