mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Ben Hutchings <bhutchings@solarflare.com>
To: Eliezer Tamir <eliezer.tamir@linux.intel.com>
Cc: Tom Herbert <therbert@google.com>,
	Stephen Hemminger <stephen@networkplumber.org>,
	Eliezer Tamir <eliezer.tamir@linux.jf.intel.com>,
	<linux-kernel@vger.kernel.org>, <netdev@vger.kernel.org>,
	Dave Miller <davem@davemloft.net>,
	Jesse Brandeburg <jesse.brandeburg@intel.com>,
	<e1000-devel@lists.sourceforge.net>,
	Willem de Bruijn <willemb@google.com>,
	Andi Kleen <andi@firstfloor.org>, HPA <hpa@zytor.com>,
	Eliezer Tamir <eliezer@tamir.org.il>
Subject: Re: [RFC PATCH 0/5] net: low latency Ethernet device polling
Date: Mon, 4 Mar 2013 22:34:29 +0000	[thread overview]
Message-ID: <1362436469.2956.33.camel@bwh-desktop.uk.solarflarecom.com> (raw)
In-Reply-To: <512E5BBD.2040803@linux.intel.com>

On Wed, 2013-02-27 at 21:17 +0200, Eliezer Tamir wrote:
> On 27/02/2013 20:47, Tom Herbert wrote:
> > On Wed, Feb 27, 2013 at 10:13 AM, Stephen Hemminger
> > <stephen@networkplumber.org> wrote:
> >>
> >> Have you looked at netmap? Seems like a cleaner user API for this.
> >>
> > That might be a bit orthogonal to this.  I believe the intent it to
> > allow spin polling from socket calls without API change.
> 
> Exactly.
> 
> BTW, an interesting alternative to netmap would be to map a VF into 
> userspace. (you would need to pre-allocate DMA memory regions and all of 
> that RDMA-like stuff)

Unless the hardware has separate event rings for completions, userland
is presumably going to need to be able to rewrite the DMA descriptor
rings.  At which point you have to restrict the VF's DMA with an IOMMU,
and might as well let userland allocate and assign DMA buffers
dynamically.

(In OpenOnload this is called 'scalable buffer mode' as an IOMMU allows
mapping more buffers than the on-board buffer table on the SFC9000
family.  The catch is that an IOMMU has higher latency.)

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:[~2013-03-04 22:34 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-27 17:55 Eliezer Tamir
2013-02-27 17:55 ` [RFC PATCH 1/5] net: implement support for low latency socket polling Eliezer Tamir
2013-03-03 18:35   ` Eric Dumazet
2013-03-03 19:21     ` Andi Kleen
2013-03-03 21:20       ` Eric Dumazet
2013-03-04  3:55         ` Andi Kleen
2013-03-04  8:43     ` Eliezer Tamir
2013-03-04 14:52       ` Eric Dumazet
2013-03-04 15:28         ` Eliezer Tamir
2013-03-04 16:15           ` Eric Dumazet
2013-03-04  7:23   ` Cong Wang
2013-03-05 16:43   ` Ben Hutchings
2013-03-05 17:15     ` Eliezer Tamir
2013-03-05 19:57       ` David Miller
2013-03-05 19:55     ` David Miller
2013-03-05 20:03       ` H. Peter Anvin
2013-02-27 17:56 ` [RFC PATCH 2/5] tcp: add TCP support for low latency receive poll Eliezer Tamir
2013-03-05 17:13   ` Ben Hutchings
2013-02-27 17:56 ` [RFC PATCH 3/5] ixgbe: Add support for ndo_ll_poll Eliezer Tamir
2013-02-27 18:41   ` Eric Dumazet
2013-02-27 19:20     ` Eliezer Tamir
2013-03-05 17:26   ` Ben Hutchings
2013-03-05 17:28     ` Ben Hutchings
2013-03-05 17:36       ` Eric Dumazet
2013-02-27 17:56 ` [RFC PATCH 4/5] ixgbe: add extra stats " Eliezer Tamir
2013-02-27 17:56 ` [RFC PATCH 5/5] ixgbe: kprobes latency test module Eliezer Tamir
2013-02-27 18:07 ` [RFC PATCH 0/5] net: low latency Ethernet device polling Eliezer Tamir
2013-02-27 18:13 ` Stephen Hemminger
2013-02-27 18:47   ` Tom Herbert
2013-02-27 19:17     ` Eliezer Tamir
2013-03-04 22:34       ` Ben Hutchings [this message]
2013-02-27 19:58 ` Rick Jones
2013-02-27 20:40   ` Eliezer Tamir
2013-02-27 21:42     ` Ben Greear
2013-02-28  8:38       ` Eliezer Tamir
2013-03-01 21:24 ` David Miller
2013-03-01 22:57   ` Tom Herbert
2013-03-02 17:02   ` Eliezer Tamir
2013-03-04  7:37 ` Cong Wang
2013-03-04  8:19   ` Eliezer Tamir
2013-03-04  8:46     ` Eliezer Tamir
2013-03-04 17:19 ` Stephen Hemminger

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=1362436469.2956.33.camel@bwh-desktop.uk.solarflarecom.com \
    --to=bhutchings@solarflare.com \
    --cc=andi@firstfloor.org \
    --cc=davem@davemloft.net \
    --cc=e1000-devel@lists.sourceforge.net \
    --cc=eliezer.tamir@linux.intel.com \
    --cc=eliezer.tamir@linux.jf.intel.com \
    --cc=eliezer@tamir.org.il \
    --cc=hpa@zytor.com \
    --cc=jesse.brandeburg@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=stephen@networkplumber.org \
    --cc=therbert@google.com \
    --cc=willemb@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

all inboxes | Powered by JetHome®