All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pavel Machek <pavel@ucw.cz>
To: Niklas Cassel <niklas.cassel@axis.com>
Cc: Joao Pinto <Joao.Pinto@synopsys.com>,
	Florian Fainelli <f.fainelli@gmail.com>,
	Andy Shevchenko <andy.shevchenko@gmail.com>,
	David Miller <davem@davemloft.net>,
	Giuseppe CAVALLARO <peppe.cavallaro@st.com>,
	larper@axis.com, rabinv@axis.com, netdev <netdev@vger.kernel.org>,
	CARLOS.PALMINHA@synopsys.com, Jie.Deng1@synopsys.com,
	Stephen Warren <swarren@nvidia.com>
Subject: Re: Re: Synopsys Ethernet QoS
Date: Wed, 14 Dec 2016 12:54:24 +0100	[thread overview]
Message-ID: <20161214115424.GA27011@amd> (raw)
In-Reply-To: <1d445ec1-deb8-6e36-39c4-6813c446095f@axis.com>

[-- Attachment #1: Type: text/plain, Size: 1446 bytes --]

Hi!

> There are some performance problems with the stmmac driver though:
> 
> When running iperf3 with 3 streams:
> iperf3 -c 192.168.0.90 -P 3 -t 30
> iperf3 -c 192.168.0.90 -P 3 -t 30 -R
> 
> I get really bad fairness between the streams.
> 
> This appears to be an issue with how TX IRQ coalescing is implemented in stmmac.
> Disabling TX IRQ coalescing in the stmmac driver makes the problem go away.

Yes, I'm hitting the same problem

https://lkml.org/lkml/2016/12/11/90

> Also netperf TCP_RR and UDP_RR gives really bad results compared to the
> dwceqos driver (without IRQ coalescing).
> 2000 transactions/sec vs 9000 transactions/sec.
> Turning TX IRQ coalescing off and RX interrupt watchdog off in stmmac
> gives the same performance. I guess it's a trade off, low CPU usage
> vs low latency, so I don't know how important TCP_RR/UDP_RR really is.

Well... 75% performance hit is a bug, plain and simple, not CPU tradeoff.

> The best thing would be to get a good working TX IRQ coalesce
> implementation with HR timers in stmmac.

Actually it seems that using HR timers is not the only problem, AFAICT
the logic is wrong way around. (But yes, it needs to be HR timer, too,
and probably packet count needs to be reduced, too.)

Best regards,

									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

  parent reply	other threads:[~2016-12-14 11:54 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-09 11:29 Synopsys Ethernet QoS Joao Pinto
2016-12-09 15:33 ` David Miller
2016-12-09 15:36   ` Joao Pinto
2016-12-09 15:41     ` David Miller
2016-12-09 15:54       ` Joao Pinto
2016-12-09 22:25       ` Andy Shevchenko
2016-12-09 22:52         ` Florian Fainelli
2016-12-10  0:16           ` Andy Shevchenko
2016-12-10  1:44             ` Florian Fainelli
2016-12-12 10:19               ` Joao Pinto
2016-12-12 14:11                 ` Giuseppe CAVALLARO
2016-12-12 16:25                 ` Niklas Cassel
2016-12-12 16:46                   ` Stephen Warren
2016-12-13  7:22                   ` Giuseppe CAVALLARO
2016-12-13  9:38                     ` Niklas Cassel
2016-12-13  9:51                       ` Giuseppe CAVALLARO
2016-12-14 12:57                       ` Pavel Machek
2016-12-14 13:14                         ` Joao Pinto
2016-12-14 19:01                           ` stmmac: lockups (was Re: Synopsys Ethernet QoS) Pavel Machek
2016-12-15 11:09                           ` Synopsys Ethernet QoS Pavel Machek
2016-12-17 17:38                           ` Pavel Machek
2016-12-19 10:55                             ` Joao Pinto
2016-12-19 11:01                               ` Joao Pinto
2016-12-15 12:05                         ` Niklas Cassel
2016-12-19 17:58                         ` Niklas Cassel
2016-12-13 11:49                   ` Joao Pinto
2016-12-13 12:31                     ` Niklas Cassel
2016-12-13 12:50                       ` Lars Persson
2016-12-13 12:56                         ` Joao Pinto
2016-12-19 16:05                           ` Niklas Cassel
2016-12-14 11:54                   ` Pavel Machek [this message]
2016-12-10  2:13             ` Jie Deng

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=20161214115424.GA27011@amd \
    --to=pavel@ucw.cz \
    --cc=CARLOS.PALMINHA@synopsys.com \
    --cc=Jie.Deng1@synopsys.com \
    --cc=Joao.Pinto@synopsys.com \
    --cc=andy.shevchenko@gmail.com \
    --cc=davem@davemloft.net \
    --cc=f.fainelli@gmail.com \
    --cc=larper@axis.com \
    --cc=netdev@vger.kernel.org \
    --cc=niklas.cassel@axis.com \
    --cc=peppe.cavallaro@st.com \
    --cc=rabinv@axis.com \
    --cc=swarren@nvidia.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.