From: David Laight <David.Laight@ACULAB.COM>
To: 'Christoph Lameter' <cl@linux.com>,
Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: "Jesper Dangaard Brouer" <brouer@redhat.com>,
"John Fastabend" <john.fastabend@gmail.com>,
"Mike Rapoport" <rppt@linux.vnet.ibm.com>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
linux-mm <linux-mm@kvack.org>,
"Willem de Bruijn" <willemdebruijn.kernel@gmail.com>,
"Björn Töpel" <bjorn.topel@intel.com>,
"Karlsson, Magnus" <magnus.karlsson@intel.com>,
"Alexander Duyck" <alexander.duyck@gmail.com>,
"Mel Gorman" <mgorman@techsingularity.net>,
"Tom Herbert" <tom@herbertland.com>,
"Brenden Blanco" <bblanco@plumgrid.com>,
"Tariq Toukan" <tariqt@mellanox.com>,
"Saeed Mahameed" <saeedm@mellanox.com>,
"Jesse Brandeburg" <jesse.brandeburg@intel.com>,
"Kalman Meth" <METH@il.ibm.com>,
"Vladislav Yasevich" <vyasevich@gmail.com>
Subject: RE: Designing a safe RX-zero-copy Memory Model for Networking
Date: Wed, 14 Dec 2016 17:37:37 +0000 [thread overview]
Message-ID: <063D6719AE5E284EB5DD2968C1650D6DB023FA6E@AcuExch.aculab.com> (raw)
In-Reply-To: <alpine.DEB.2.20.1612141059020.20959@east.gentwo.org>
From: Christoph Lameter
> Sent: 14 December 2016 17:00
> On Tue, 13 Dec 2016, Hannes Frederic Sowa wrote:
>
> > > Interesting. So you even imagine sockets registering memory regions
> > > with the NIC. If we had a proper NIC HW filter API across the drivers,
> > > to register the steering rule (like ibv_create_flow), this would be
> > > doable, but we don't (DPDK actually have an interesting proposal[1])
> >
> > On a side note, this is what windows does with RIO ("registered I/O").
> > Maybe you want to look at the API to get some ideas: allocating and
> > pinning down memory in user space and registering that with sockets to
> > get zero-copy IO.
>
> Yup that is also what I think. Regarding the memory registration and flow
> steering for user space RX/TX ring please look at the qpair model
> implemented by the RDMA subsystem in the kernel. The memory semantics are
> clearly established there and have been in use for more than a decade.
Isn't there a bigger problem for transmit?
If the kernel is doing ANY validation on the frames it must copy the
data to memory the application cannot modify before doing the validation.
Otherwise the application could change the data afterwards.
David
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2016-12-14 17:37 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-05 14:31 Designing a safe RX-zero-copy Memory Model for Networking Jesper Dangaard Brouer
2016-12-12 8:38 ` Mike Rapoport
2016-12-12 9:40 ` Jesper Dangaard Brouer
2016-12-12 14:14 ` Mike Rapoport
2016-12-12 14:14 ` Mike Rapoport
2016-12-12 14:49 ` John Fastabend
2016-12-12 17:13 ` Jesper Dangaard Brouer
2016-12-12 18:06 ` Christoph Lameter
2016-12-12 18:06 ` Christoph Lameter
2016-12-13 16:10 ` Jesper Dangaard Brouer
2016-12-13 16:36 ` Christoph Lameter
2016-12-13 16:36 ` Christoph Lameter
2016-12-13 17:43 ` John Fastabend
2016-12-13 17:43 ` John Fastabend
2016-12-13 19:53 ` David Miller
2016-12-13 20:08 ` John Fastabend
2016-12-14 9:39 ` Jesper Dangaard Brouer
2016-12-14 16:32 ` John Fastabend
2016-12-14 16:45 ` Alexander Duyck
2016-12-14 21:29 ` Jesper Dangaard Brouer
2016-12-14 22:45 ` Alexander Duyck
2016-12-15 8:28 ` Jesper Dangaard Brouer
2016-12-15 15:59 ` Alexander Duyck
2016-12-15 16:38 ` Christoph Lameter
2016-12-14 21:04 ` Jesper Dangaard Brouer
2016-12-13 18:39 ` Hannes Frederic Sowa
2016-12-14 17:00 ` Christoph Lameter
2016-12-14 17:00 ` Christoph Lameter
2016-12-14 17:37 ` David Laight [this message]
2016-12-14 19:43 ` Christoph Lameter
2016-12-14 19:43 ` Christoph Lameter
2016-12-14 20:37 ` Hannes Frederic Sowa
2016-12-14 20:37 ` Hannes Frederic Sowa
2016-12-14 21:22 ` Christoph Lameter
2016-12-13 9:42 ` Mike Rapoport
2016-12-12 15:10 ` Jesper Dangaard Brouer
2016-12-12 15:10 ` Jesper Dangaard Brouer
2016-12-13 8:43 ` Mike Rapoport
2016-12-13 8:43 ` Mike Rapoport
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=063D6719AE5E284EB5DD2968C1650D6DB023FA6E@AcuExch.aculab.com \
--to=david.laight@aculab.com \
--cc=METH@il.ibm.com \
--cc=alexander.duyck@gmail.com \
--cc=bblanco@plumgrid.com \
--cc=bjorn.topel@intel.com \
--cc=brouer@redhat.com \
--cc=cl@linux.com \
--cc=hannes@stressinduktion.org \
--cc=jesse.brandeburg@intel.com \
--cc=john.fastabend@gmail.com \
--cc=linux-mm@kvack.org \
--cc=magnus.karlsson@intel.com \
--cc=mgorman@techsingularity.net \
--cc=netdev@vger.kernel.org \
--cc=rppt@linux.vnet.ibm.com \
--cc=saeedm@mellanox.com \
--cc=tariqt@mellanox.com \
--cc=tom@herbertland.com \
--cc=vyasevich@gmail.com \
--cc=willemdebruijn.kernel@gmail.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.