netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ben Hutchings <bhutchings@solarflare.com>
To: David Decotigny <david.decotigny@google.com>
Cc: <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	"David S. Miller" <davem@davemloft.net>,
	Ian Campbell <ian.campbell@citrix.com>,
	Eric Dumazet <eric.dumazet@gmail.com>,
	Jeff Kirsher <jeffrey.t.kirsher@intel.com>,
	Jiri Pirko <jpirko@redhat.com>, Joe Perches <joe@perches.com>,
	Szymon Janc <szymon@janc.net.pl>,
	Richard Jones <rick.jones2@hp.com>,
	Ayaz Abdulla <AAbdulla@nvidia.com>,
	Tom Herbert <therbert@google.com>
Subject: Re: [PATCH net-next v6 3/9] kbuild: document RPS/XPS network Kconfig options
Date: Wed, 16 Nov 2011 23:12:02 +0000	[thread overview]
Message-ID: <1321485122.2709.55.camel@bwh-desktop> (raw)
In-Reply-To: <c7e7b57596f9a06b1c40c8b6e671cc6ac69dd304.1321481064.git.david.decotigny@google.com>

On Wed, 2011-11-16 at 14:15 -0800, David Decotigny wrote:
> This adds a description of RPS/XPS options and allow them to be
> changed at make menuconfig time.

I'm not sure why you think this is necessary.

[...]
>  config RPS
> -	boolean
> +	boolean "Enable Receive Packet Steering"
>  	depends on SMP && SYSFS && USE_GENERIC_SMP_HELPERS
>  	default y
> +	help
> +	  RPS distributes the load of received packet processing
> +	  across multiple CPUs. If unsure, say Y.
>  
>  config RFS_ACCEL
> -	boolean
> +	boolean "Enable Hardware Acceleration of RFS"
>  	depends on RPS && GENERIC_HARDIRQS
>  	select CPU_RMAP
>  	default y
> +	help
> +	  This is the hardware version of RPS. On multi-queue network
> +	  devices, this configures the hardware to distribute the
> +	  received packets across multiple CPUs. If unsure, say Y.
[...]

There is some confusion/conflation between RPS and RFS in both code and
documentation.

RPS originaly referred to spreading out RX packet processing based on a
flow hash.  Most multiqueue devices do this in hardware, which is
commonly referred to as RSS.  RSS can be enabled independently of any
networking core features.

RFS refers to directing RX packet processsing of specific flows based on
where the corresponding sockets have been used.  RFS acceleration means
that the driver and hardware help with this by changing hardware queue
selection for specific flows.

The RPS Kconfig option controls both RPS and RFS, and various references
to 'RPS' in the code really cover RFS as well.

The RFS_ACCEL Kconfig option enables RFS to support acceleration, and
like most offload features it has no effect without a driver that
specifically supports that.  The option only exists to abstract the
slightly odd dependency on GENERIC_HARDIRQS, and I don't think it should
be manually controllable.

Ben.

-- 
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.

  reply	other threads:[~2011-11-16 23:12 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-11-16 22:15 [PATCH net-next v6 0/9] net-sysfs+forcedeth: stats & debug enhancements David Decotigny
2011-11-16 22:15 ` [PATCH net-next v6 1/9] forcedeth: fix stats on hardware without extended stats support David Decotigny
2011-11-16 22:15 ` [PATCH net-next v6 2/9] net-sysfs: fixed minor sparse warning David Decotigny
2011-11-16 22:15 ` [PATCH net-next v6 3/9] kbuild: document RPS/XPS network Kconfig options David Decotigny
2011-11-16 23:12   ` Ben Hutchings [this message]
2011-11-17  1:54     ` David Decotigny
2011-11-17  2:54       ` Ben Hutchings
2011-11-17  2:59         ` David Miller
2011-11-17  3:19           ` Eric Dumazet
2011-11-16 22:15 ` [PATCH net-next v6 4/9] net: new counter for tx_timeout errors in sysfs David Decotigny
2011-11-16 22:15 ` [PATCH net-next v6 5/9] forcedeth: Add messages to indicate using MSI or MSI-X David Decotigny
2011-11-16 22:15 ` [PATCH net-next v6 6/9] forcedeth: allow to silence "TX timeout" debug messages David Decotigny
2011-11-16 22:15 ` [PATCH net-next v6 7/9] forcedeth: implement ndo_get_stats64() API David Decotigny
2011-11-17  6:34   ` Eric Dumazet
2011-11-17 17:39     ` David Decotigny
2011-11-17 17:47       ` Eric Dumazet
2011-11-16 22:15 ` [PATCH net-next v6 8/9] forcedeth: account for dropped RX frames David Decotigny
2011-11-16 22:15 ` [PATCH net-next v6 9/9] forcedeth: stats updated with a deferrable timer David Decotigny
     [not found] <cover.1321496595.git.david.decotigny@google.com>
2011-11-17  2:39 ` [PATCH net-next v6 3/9] kbuild: document RPS/XPS network Kconfig options David Decotigny
2011-11-17  2:58   ` David Miller
2011-11-17  3:04     ` David Decotigny
2011-11-17  3:12       ` 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=1321485122.2709.55.camel@bwh-desktop \
    --to=bhutchings@solarflare.com \
    --cc=AAbdulla@nvidia.com \
    --cc=davem@davemloft.net \
    --cc=david.decotigny@google.com \
    --cc=eric.dumazet@gmail.com \
    --cc=ian.campbell@citrix.com \
    --cc=jeffrey.t.kirsher@intel.com \
    --cc=joe@perches.com \
    --cc=jpirko@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=rick.jones2@hp.com \
    --cc=szymon@janc.net.pl \
    --cc=therbert@google.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;
as well as URLs for NNTP newsgroup(s).