From: Jason Wang <jasowang@redhat.com>
To: krkumar2@in.ibm.com, kvm@vger.kernel.org, mst@redhat.com,
netdev@vger.kernel.org, rusty@rustcorp.com.au,
virtualization@lists.linux-foundation.org,
levinsasha928@gmail.com, bhutchings@solarflare.com
Subject: [net-next RFC PATCH 0/5] Series short description
Date: Mon, 05 Dec 2011 16:58:37 +0800 [thread overview]
Message-ID: <20111205085603.6116.65101.stgit@dhcp-8-146.nay.redhat.com> (raw)
multiple queue virtio-net: flow steering through host/guest cooperation
Hello all:
This is a rough series adds the guest/host cooperation of flow
steering support based on Krish Kumar's multiple queue virtio-net
driver patch 3/3 (http://lwn.net/Articles/467283/).
This idea is simple, the backend pass the rxhash to the guest and
guest would tell the backend the hash to queue mapping when necessary
then backend can choose the queue based on the hash value of the
packet. The table is just a page shared bettwen userspace and the
backend.
Patch 1 enable the ability to pass the rxhash through vnet_hdr to
guest.
Patch 2,3 implement a very simple flow director for tap and
mavtap. tap part is based on the multiqueue tap patches posted by me
(http://lwn.net/Articles/459270/).
Patch 4 implement a method for virtio device to find the irq of a
specific virtqueue, in order to do device specific interrupt
optimization
Patch 5 is the part of the guest driver that using accelerate rfs to
program the flow director and with some optimizations on irq affinity
and tx queue selection.
This is just a prototype that demonstrates the idea, there are still
things need to be discussed:
- An alternative idea instead of shared page is ctrl vq, the reason
that a shared table is preferable is the delay of ctrl vq itself.
- Optimization on irq affinity and tx queue selection
Comments are welcomed, thanks!
---
Jason Wang (5):
virtio_net: passing rxhash through vnet_hdr
tuntap: simple flow director support
macvtap: flow director support
virtio: introduce a method to get the irq of a specific virtqueue
virtio-net: flow director support
drivers/lguest/lguest_device.c | 8 ++
drivers/net/macvlan.c | 4 +
drivers/net/macvtap.c | 42 ++++++++-
drivers/net/tun.c | 105 ++++++++++++++++------
drivers/net/virtio_net.c | 189 +++++++++++++++++++++++++++++++++++++++-
drivers/s390/kvm/kvm_virtio.c | 6 +
drivers/vhost/net.c | 10 +-
drivers/vhost/vhost.h | 5 +
drivers/virtio/virtio_mmio.c | 8 ++
drivers/virtio/virtio_pci.c | 12 +++
include/linux/if_macvlan.h | 1
include/linux/if_tun.h | 11 ++
include/linux/virtio_config.h | 4 +
include/linux/virtio_net.h | 16 +++
14 files changed, 377 insertions(+), 44 deletions(-)
--
Signature
next reply other threads:[~2011-12-05 8:58 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-12-05 8:58 Jason Wang [this message]
2011-12-05 8:58 ` [net-next RFC PATCH 1/5] virtio_net: passing rxhash through vnet_hdr Jason Wang
2011-12-05 8:58 ` [net-next RFC PATCH 2/5] tuntap: simple flow director support Jason Wang
2011-12-05 10:38 ` Stefan Hajnoczi
2011-12-05 20:09 ` Ben Hutchings
2011-12-06 7:21 ` Jason Wang
2011-12-06 17:31 ` Ben Hutchings
2011-12-05 8:59 ` [net-next RFC PATCH 3/5] macvtap: " Jason Wang
2011-12-05 20:11 ` Ben Hutchings
2011-12-05 8:59 ` [net-next RFC PATCH 4/5] virtio: introduce a method to get the irq of a specific virtqueue Jason Wang
2011-12-05 8:59 ` [net-next RFC PATCH 5/5] virtio-net: flow director support Jason Wang
2011-12-05 10:55 ` Stefan Hajnoczi
2011-12-06 6:33 ` Jason Wang
2011-12-06 9:18 ` Stefan Hajnoczi
2011-12-06 10:21 ` Jason Wang
2011-12-06 13:15 ` Stefan Hajnoczi
2011-12-06 15:42 ` Sridhar Samudrala
2011-12-06 16:14 ` Michael S. Tsirkin
2011-12-06 23:10 ` Sridhar Samudrala
2011-12-07 11:05 ` Jason Wang
2011-12-07 11:02 ` Jason Wang
2011-12-09 2:00 ` Sridhar Samudrala
2011-12-07 3:03 ` Jason Wang
2011-12-07 9:08 ` Stefan Hajnoczi
2011-12-07 12:10 ` Jason Wang
2011-12-07 15:04 ` Stefan Hajnoczi
2011-12-05 20:42 ` Ben Hutchings
2011-12-06 7:25 ` Jason Wang
2011-12-06 17:36 ` Ben Hutchings
2011-12-07 7:30 ` [net-next RFC PATCH 0/5] Series short description Rusty Russell
2011-12-07 11:31 ` Jason Wang
2011-12-07 17:02 ` Ben Hutchings
2011-12-08 10:06 ` Jason Wang
2011-12-09 5:31 ` Rusty Russell
2011-12-15 1:36 ` Ben Hutchings
2011-12-15 23:12 ` Rusty Russell
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=20111205085603.6116.65101.stgit@dhcp-8-146.nay.redhat.com \
--to=jasowang@redhat.com \
--cc=bhutchings@solarflare.com \
--cc=krkumar2@in.ibm.com \
--cc=kvm@vger.kernel.org \
--cc=levinsasha928@gmail.com \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=rusty@rustcorp.com.au \
--cc=virtualization@lists.linux-foundation.org \
/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).