mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Gregory CLEMENT <gregory.clement@free-electrons.com>
Cc: "David S. Miller" <davem@davemloft.net>,
	linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
	Thomas Petazzoni <thomas.petazzoni@free-electrons.com>,
	Jason Cooper <jason@lakedaemon.net>, Andrew Lunn <andrew@lunn.ch>,
	Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>,
	Ezequiel Garcia <ezequiel.garcia@free-electrons.com>,
	linux-arm-kernel@lists.infradead.org,
	Lior Amsalem <alior@marvell.com>,
	Nadav Haklai <nadavh@marvell.com>,
	Marcin Wojtas <mw@semihalf.com>,
	Simon Guinot <simon.guinot@sequanux.org>,
	Maxime Ripard <maxime.ripard@free-electrons.com>,
	Boris BREZILLON <boris.brezillon@free-electrons.com>,
	Russell King - ARM Linux <linux@arm.linux.org.uk>,
	Willy Tarreau <w@1wt.eu>, Arnd Bergmann <arnd@arndb.de>
Subject: Re: [PATCH net-next v2 4/4] net: mvneta: Spread out the TX queues management on all CPUs
Date: Fri, 04 Dec 2015 11:12:30 -0800	[thread overview]
Message-ID: <1449256350.25029.36.camel@edumazet-glaptop2.roam.corp.google.com> (raw)
In-Reply-To: <1449254700-32685-5-git-send-email-gregory.clement@free-electrons.com>

On Fri, 2015-12-04 at 19:45 +0100, Gregory CLEMENT wrote:
> With this patch each CPU is associated with its own set of TX queues. In
> the same time the SKB received in mvneta_tx is bound to the queue
> associated to the CPU sending the data. Thanks to this the next IRQ will
> be received on the same CPU allowing sending more data.
> 
> It will also allow to have a more predictable behavior regarding
> throughput and latency when having multiple threads sending out data on
> different CPUs.
> 
> As an example on Armada XP GP, with an iperf bound to a CPU and a ping
> bound to another CPU, without this patch the ping round trip was about
> 2.5ms (and could reach 3s!), whereas with this patch it was around
> 0.7ms (and sometime it went to 1.2ms).

This really looks like you need something smarter than pfifo_fast qdisc,
and maybe BQL (I did not check if this driver already implements this)

> 
> Suggested-by: Arnd Bergmann <arnd@arndb.de>
> Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>

...

> @@ -1824,13 +1835,16 @@ error:
>  static int mvneta_tx(struct sk_buff *skb, struct net_device *dev)
>  {
>  	struct mvneta_port *pp = netdev_priv(dev);
> -	u16 txq_id = skb_get_queue_mapping(skb);
> +	u16 txq_id = smp_processor_id() % txq_number;
>  	struct mvneta_tx_queue *txq = &pp->txqs[txq_id];
>  	struct mvneta_tx_desc *tx_desc;
>  	int len = skb->len;
>  	int frags = 0;
>  	u32 tx_cmd;
>  
> +	/* Use the tx queue bound to this CPU */
> +	skb_set_queue_mapping(skb, txq_id);
> +


We certainly do not want every driver implementing its own hacks.

We have a standard way to handle this, it is called XPS, and eventually
ndo_select_queue()

Documentation/networking/scaling.txt contains some hints.




  reply	other threads:[~2015-12-04 19:12 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-04 18:44 [PATCH net-next v2 0/4] mvneta: Introduce RSS support Gregory CLEMENT
2015-12-04 18:44 ` [PATCH net-next v2 1/4] net: mvneta: Make the default queue related for each port Gregory CLEMENT
2015-12-04 18:44 ` [PATCH net-next v2 2/4] net: mvneta: Associate RX queues with each CPU Gregory CLEMENT
2015-12-04 18:44 ` [PATCH net-next v2 3/4] net: mvneta: Add naive RSS support Gregory CLEMENT
2015-12-04 18:45 ` [PATCH net-next v2 4/4] net: mvneta: Spread out the TX queues management on all CPUs Gregory CLEMENT
2015-12-04 19:12   ` Eric Dumazet [this message]
2015-12-04 21:30     ` Arnd Bergmann
2015-12-05 19:14   ` Marcin Wojtas
2015-12-05 22:24     ` David Miller

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=1449256350.25029.36.camel@edumazet-glaptop2.roam.corp.google.com \
    --to=eric.dumazet@gmail.com \
    --cc=alior@marvell.com \
    --cc=andrew@lunn.ch \
    --cc=arnd@arndb.de \
    --cc=boris.brezillon@free-electrons.com \
    --cc=davem@davemloft.net \
    --cc=ezequiel.garcia@free-electrons.com \
    --cc=gregory.clement@free-electrons.com \
    --cc=jason@lakedaemon.net \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@arm.linux.org.uk \
    --cc=maxime.ripard@free-electrons.com \
    --cc=mw@semihalf.com \
    --cc=nadavh@marvell.com \
    --cc=netdev@vger.kernel.org \
    --cc=sebastian.hesselbarth@gmail.com \
    --cc=simon.guinot@sequanux.org \
    --cc=thomas.petazzoni@free-electrons.com \
    --cc=w@1wt.eu \
    /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

all inboxes | Powered by JetHome®