netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Octavian Purdila <octavian.purdila@intel.com>
To: netdev@vger.kernel.org
Cc: Octavian Purdila <octavian.purdila@intel.com>
Subject: [net-next v2 00/13] remove code duplication in tcp_v[46]_conn_request
Date: Wed, 25 Jun 2014 17:09:49 +0300	[thread overview]
Message-ID: <1403705402-14388-1-git-send-email-octavian.purdila@intel.com> (raw)

This patch series unifies the TCPv4 and TCPv6 connection request flow
in a single new function (tcp_conn_request).

The first 3 patches are small cleanups and fixes found during the code
merge process.

The next patches add new methods in tcp_request_sock_ops to abstract
the IPv4/IPv6 operations and keep the TCP connection request flow
common.

To identify potential performance issues this patch has been tested
by measuring the connection per second rate with nginx and a httperf
like client (to allow for concurrent connection requests - 256 CC were
used during testing) using the loopback interface. A dual-core i5 Ivy
Bridge processor was used and each process was bounded to a different
core to make results consistent.

Results for IPv4, unit is connections per second, higher is better, 20
measurements have been collected:

		before		after
min		27917		27962
max		28262		28366
avg		28094.1		28212.75
stdev		87.35		97.26

Results for IPv6, unit is connections per second, higher is better, 20
measurements have been collected:

		before		after
min		24813		24877
max		25029		25119
avg		24935.5		25017
stdev		64.13		62.93


Changes since v1:

 * add benchmarking datapoints

 * fix a few issues in the last patch (IPv6 related)


Octavian Purdila (13):
  tcp: cookie_v4_init_sequence: skb should be const
  tcp: tcp_v[46]_conn_request: fix snt_synack initialization
  net: remove inet6_reqsk_alloc
  tcp: add init_req method to tcp_request_sock_ops
  tcp: add init_cookie_seq method to tcp_request_sock_ops
  tcp: add route_req method to tcp_request_sock_ops
  tcp: move around a few calls in tcp_v6_conn_request
  tcp: add init_seq method to tcp_request_sock_ops
  tcp: add send_synack method to tcp_request_sock_ops
  tcp: unify tcp_v4_rtx_synack and tcp_v6_rtx_synack
  tcp: add mss_clamp to tcp_request_sock_ops
  tcp: add queue_add_hash to tcp_request_sock_ops
  tcp: add tcp_conn_request

 include/linux/ipv6.h    |  10 ---
 include/linux/tcp.h     |   3 -
 include/net/inet_sock.h |   6 +-
 include/net/tcp.h       |  54 +++++++++----
 net/dccp/ipv6.c         |   2 +-
 net/ipv4/syncookies.c   |   3 +-
 net/ipv4/tcp_input.c    | 148 ++++++++++++++++++++++++++++++++++
 net/ipv4/tcp_ipv4.c     | 188 ++++++++++---------------------------------
 net/ipv4/tcp_output.c   |  15 ++++
 net/ipv6/syncookies.c   |   2 +-
 net/ipv6/tcp_ipv6.c     | 206 +++++++++++++-----------------------------------
 11 files changed, 306 insertions(+), 331 deletions(-)

-- 
1.8.3.2

             reply	other threads:[~2014-06-25 14:10 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-06-25 14:09 Octavian Purdila [this message]
2014-06-25 14:09 ` [net-next v2 01/13] tcp: cookie_v4_init_sequence: skb should be const Octavian Purdila
2014-06-25 14:09 ` [net-next v2 02/13] tcp: tcp_v[46]_conn_request: fix snt_synack initialization Octavian Purdila
2014-06-28  2:16   ` Neal Cardwell
2014-06-28  2:24     ` Neal Cardwell
2014-06-30 11:50     ` Octavian Purdila
2014-06-30 19:11       ` Neal Cardwell
2014-06-30 21:00         ` Octavian Purdila
2014-07-01  2:24           ` Neal Cardwell
2014-06-25 14:09 ` [net-next v2 03/13] net: remove inet6_reqsk_alloc Octavian Purdila
2014-06-25 17:59   ` Cong Wang
2014-06-25 18:04     ` Octavian Purdila
2014-06-25 14:09 ` [net-next v2 04/13] tcp: add init_req method to tcp_request_sock_ops Octavian Purdila
2014-06-25 14:09 ` [net-next v2 05/13] tcp: add init_cookie_seq " Octavian Purdila
2014-06-25 14:09 ` [net-next v2 06/13] tcp: add route_req " Octavian Purdila
2014-06-25 14:09 ` [net-next v2 07/13] tcp: move around a few calls in tcp_v6_conn_request Octavian Purdila
2014-06-25 14:29   ` Paul Moore
2014-06-25 14:09 ` [net-next v2 08/13] tcp: add init_seq method to tcp_request_sock_ops Octavian Purdila
2014-06-25 14:09 ` [net-next v2 09/13] tcp: add send_synack " Octavian Purdila
2014-06-25 14:09 ` [net-next v2 10/13] tcp: unify tcp_v4_rtx_synack and tcp_v6_rtx_synack Octavian Purdila
2014-06-25 14:10 ` [net-next v2 11/13] tcp: add mss_clamp to tcp_request_sock_ops Octavian Purdila
2014-06-25 14:10 ` [net-next v2 12/13] tcp: add queue_add_hash " Octavian Purdila
2014-06-25 19:05   ` Cong Wang
2014-06-25 14:10 ` [net-next v2 13/13] tcp: add tcp_conn_request Octavian Purdila
2014-06-27 22:54 ` [net-next v2 00/13] remove code duplication in tcp_v[46]_conn_request 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=1403705402-14388-1-git-send-email-octavian.purdila@intel.com \
    --to=octavian.purdila@intel.com \
    --cc=netdev@vger.kernel.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).