All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vladislav Yasevich <vyasevich@gmail.com>
To: netdev@vger.kernel.org
Cc: nhorman@tuxdriver.com, mst@redhat.com,
	virtualization@lists.linux-foundation.org,
	linux-sctp@vger.kernel.org
Subject: [PATCH net-next 0/5] virtio-net:  Add SCTP checksum offload support
Date: Mon,  2 Apr 2018 09:40:01 -0400	[thread overview]
Message-ID: <20180402134006.10111-1-vyasevic@redhat.com> (raw)

Now that we have SCTP offload capabilities in the kernel, we can add
them to virtio as well.  First step is SCTP checksum.

We need a new freature in virtio to negotiate this support since
SCTP is excluded with the stardard checksum and requires a little
bit extra.  This series proposes VIRTIO_NET_F_SCTP_CSUM feature bit.

As the "little bit extra",  the kernel uses a new bit in the skb
(skb->csum_not_inet) to determine whether to use standard inet checksum
or the SCTP CRC32c checksum.  This bit has to be communicated between
the host and the guest.  This bit is carried in the vnet header.

Tap and macvtap support is added through an extra feature for the
TUNSETOFFLOAD ioctl.  Additionally macvtap will no correctly
do sctp checksumming if the receive doesn't support SCTP offload.
This also turns on sctp offloading for macvlan devices.

As for the perf numbers, I am seeing about a 5% increase in vm-to-vm
and vm-to-hos throughput which is the same as manually disabling
sctp checksumming,since this is exactly what we are emulatting.
Sending outside the host,  the increase about 2.5-3%.

As for GSO, the way sctp GSO is currently implemented buys us nothing
in added support to virtio.  To add true GSO, would require a lot of
re-work inside of SCTP and would require extensions to the virtio
net header to carry extra sctp data.


Vladislav Yasevich (5):
  virtio: Add support for SCTP checksum offloading
  sctp: Handle sctp packets with CHECKSUM_PARTIAL
  sctp: Build sctp offload support into the base kernel
  tun: Add support for SCTP checksum offload
  macvlan/macvtap: Add support for SCTP checksum offload.

 drivers/net/macvlan.c           |  5 +++--
 drivers/net/tap.c               |  8 +++++---
 drivers/net/tun.c               |  5 +++++
 drivers/net/virtio_net.c        | 10 +++++++---
 include/linux/virtio_net.h      |  6 ++++++
 include/net/sctp/sctp.h         |  5 -----
 include/uapi/linux/if_tun.h     |  1 +
 include/uapi/linux/virtio_net.h |  2 ++
 net/Kconfig                     |  1 +
 net/sctp/Kconfig                |  1 -
 net/sctp/Makefile               |  3 ++-
 net/sctp/input.c                | 11 ++++++++++-
 net/sctp/offload.c              |  4 +++-
 net/sctp/protocol.c             |  3 ---
 14 files changed, 45 insertions(+), 20 deletions(-)

-- 
2.9.5

WARNING: multiple messages have this Message-ID (diff)
From: Vladislav Yasevich <vyasevich@gmail.com>
To: netdev@vger.kernel.org
Cc: nhorman@tuxdriver.com, mst@redhat.com,
	virtualization@lists.linux-foundation.org,
	linux-sctp@vger.kernel.org
Subject: [PATCH net-next 0/5] virtio-net:  Add SCTP checksum offload support
Date: Mon, 02 Apr 2018 13:40:01 +0000	[thread overview]
Message-ID: <20180402134006.10111-1-vyasevic@redhat.com> (raw)

Now that we have SCTP offload capabilities in the kernel, we can add
them to virtio as well.  First step is SCTP checksum.

We need a new freature in virtio to negotiate this support since
SCTP is excluded with the stardard checksum and requires a little
bit extra.  This series proposes VIRTIO_NET_F_SCTP_CSUM feature bit.

As the "little bit extra",  the kernel uses a new bit in the skb
(skb->csum_not_inet) to determine whether to use standard inet checksum
or the SCTP CRC32c checksum.  This bit has to be communicated between
the host and the guest.  This bit is carried in the vnet header.

Tap and macvtap support is added through an extra feature for the
TUNSETOFFLOAD ioctl.  Additionally macvtap will no correctly
do sctp checksumming if the receive doesn't support SCTP offload.
This also turns on sctp offloading for macvlan devices.

As for the perf numbers, I am seeing about a 5% increase in vm-to-vm
and vm-to-hos throughput which is the same as manually disabling
sctp checksumming,since this is exactly what we are emulatting.
Sending outside the host,  the increase about 2.5-3%.

As for GSO, the way sctp GSO is currently implemented buys us nothing
in added support to virtio.  To add true GSO, would require a lot of
re-work inside of SCTP and would require extensions to the virtio
net header to carry extra sctp data.


Vladislav Yasevich (5):
  virtio: Add support for SCTP checksum offloading
  sctp: Handle sctp packets with CHECKSUM_PARTIAL
  sctp: Build sctp offload support into the base kernel
  tun: Add support for SCTP checksum offload
  macvlan/macvtap: Add support for SCTP checksum offload.

 drivers/net/macvlan.c           |  5 +++--
 drivers/net/tap.c               |  8 +++++---
 drivers/net/tun.c               |  5 +++++
 drivers/net/virtio_net.c        | 10 +++++++---
 include/linux/virtio_net.h      |  6 ++++++
 include/net/sctp/sctp.h         |  5 -----
 include/uapi/linux/if_tun.h     |  1 +
 include/uapi/linux/virtio_net.h |  2 ++
 net/Kconfig                     |  1 +
 net/sctp/Kconfig                |  1 -
 net/sctp/Makefile               |  3 ++-
 net/sctp/input.c                | 11 ++++++++++-
 net/sctp/offload.c              |  4 +++-
 net/sctp/protocol.c             |  3 ---
 14 files changed, 45 insertions(+), 20 deletions(-)

-- 
2.9.5


             reply	other threads:[~2018-04-02 13:40 UTC|newest]

Thread overview: 74+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-04-02 13:40 Vladislav Yasevich [this message]
2018-04-02 13:40 ` [PATCH net-next 0/5] virtio-net: Add SCTP checksum offload support Vladislav Yasevich
2018-04-02 13:40 ` [PATCH net-next 1/5] virtio: Add support for SCTP checksum offloading Vladislav Yasevich
2018-04-02 13:40   ` Vladislav Yasevich
2018-04-10  3:17   ` Jason Wang
2018-04-10  3:17   ` Jason Wang
2018-04-10  3:17     ` Jason Wang
2018-04-11 22:39   ` Marcelo Ricardo Leitner
2018-04-11 22:39     ` Marcelo Ricardo Leitner
2018-04-11 22:49   ` Michael S. Tsirkin
2018-04-11 22:49     ` Michael S. Tsirkin
2018-04-16 13:45     ` Vlad Yasevich
2018-04-16 13:45       ` Vlad Yasevich
2018-04-16 15:52       ` Michael S. Tsirkin
2018-04-16 15:52         ` Michael S. Tsirkin
2018-04-16 17:09       ` Michael S. Tsirkin
2018-04-16 17:09         ` Michael S. Tsirkin
2018-04-17 19:06         ` Vlad Yasevich
2018-04-17 19:06           ` Vlad Yasevich
2018-04-17 19:06         ` Vlad Yasevich
2018-04-16 17:09       ` Michael S. Tsirkin
2018-04-16 13:45     ` Vlad Yasevich
2018-04-11 22:49   ` Michael S. Tsirkin
2018-04-16 17:07   ` Michael S. Tsirkin
2018-04-16 17:07   ` Michael S. Tsirkin
2018-04-16 17:07     ` Michael S. Tsirkin
2018-04-02 13:40 ` Vladislav Yasevich
2018-04-02 13:40 ` [PATCH net-next 2/5] sctp: Handle sctp packets with CHECKSUM_PARTIAL Vladislav Yasevich
2018-04-02 13:40 ` Vladislav Yasevich
2018-04-02 13:40   ` Vladislav Yasevich
2018-04-02 13:40 ` [PATCH net-next 3/5] sctp: Build sctp offload support into the base kernel Vladislav Yasevich
2018-04-02 13:40 ` Vladislav Yasevich
2018-04-02 13:40   ` Vladislav Yasevich
2018-04-02 13:40 ` [PATCH net-next 4/5] tun: Add support for SCTP checksum offload Vladislav Yasevich
2018-04-02 13:40 ` Vladislav Yasevich
2018-04-02 13:40   ` Vladislav Yasevich
2018-04-03  0:49   ` kbuild test robot
2018-04-03  0:49     ` kbuild test robot
2018-04-03  0:49   ` kbuild test robot
2018-04-16 17:12   ` Michael S. Tsirkin
2018-04-16 17:12     ` Michael S. Tsirkin
2018-04-16 17:12   ` Michael S. Tsirkin
2018-04-02 13:40 ` [PATCH net-next 5/5] macvlan/macvtap: " Vladislav Yasevich
2018-04-02 13:40   ` Vladislav Yasevich
2018-04-04 15:31   ` Davide Caratti
2018-04-04 15:31     ` Davide Caratti
2018-04-16 17:14   ` Michael S. Tsirkin
2018-04-16 17:14     ` Michael S. Tsirkin
2018-04-16 17:14   ` Michael S. Tsirkin
2018-04-02 13:40 ` Vladislav Yasevich
2018-04-02 14:16 ` [PATCH net-next 0/5] virtio-net: Add SCTP checksum offload support David Miller
2018-04-02 14:16 ` David Miller
2018-04-02 14:16   ` David Miller
2018-04-02 14:47 ` Marcelo Ricardo Leitner
2018-04-02 14:47   ` Marcelo Ricardo Leitner
2018-04-17 20:35   ` Vlad Yasevich
2018-04-17 20:35     ` Vlad Yasevich
2018-04-18  1:33     ` Marcelo Ricardo Leitner
2018-04-18  1:33       ` Marcelo Ricardo Leitner
2018-04-18  6:57       ` Xin Long
2018-04-18  6:57         ` Xin Long
2018-04-18  6:57       ` Xin Long
2018-04-18 14:06     ` Michael S. Tsirkin
2018-04-18 14:06       ` Michael S. Tsirkin
2018-04-20 17:22       ` Marcelo Ricardo Leitner
2018-04-20 17:22         ` Marcelo Ricardo Leitner
2018-04-20 18:32         ` Michael S. Tsirkin
2018-04-20 18:32         ` Michael S. Tsirkin
2018-04-20 18:32           ` Michael S. Tsirkin
2018-04-23 13:17         ` Vlad Yasevich
2018-04-23 13:17           ` Vlad Yasevich
2018-04-23 13:17         ` Vlad Yasevich
2018-04-18 14:06     ` Michael S. Tsirkin
2018-04-17 20:35   ` Vlad Yasevich

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=20180402134006.10111-1-vyasevic@redhat.com \
    --to=vyasevich@gmail.com \
    --cc=linux-sctp@vger.kernel.org \
    --cc=mst@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@tuxdriver.com \
    --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 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.