netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Florian Westphal <fw@strlen.de>
To: <netdev@vger.kernel.org>
Subject: [RFC ipsec-next] flow cache removal
Date: Wed, 28 Jun 2017 15:26:43 +0200	[thread overview]
Message-ID: <20170628132652.1275-1-fw@strlen.de> (raw)

Here is an updated version of the flow cache removal
set.

Changes since last iteration:
 - rebase
 - split removal into multiple gradual chunks to ease review
 - add a small pcpu xdst cache to reduce alloc/free overhead
   when subsequent packet can re-use previous xdst

I did some sanity testing and ran a few netperf tests.
The most severe hit is with pure UDP_RR workload.
TCP_STREAM is ok-ish, UDP_STREAM is marginally faster with
the simpler pcpu cache (we only instantiate one xfrm_dst and then
reuse it).

We can discuss fine print and possible further work (avoid this_cpu_xchg,
xfrm_genid removal, etc) at NFWS if needed.

Thanks,
Florian

 Documentation/networking/ip-sysctl.txt |    6 
 /include/net/flow.h                     |   34 --
 /include/net/netns/xfrm.h               |   11 
 /include/net/xfrm.h                     |    9 
 /net/core/Makefile                      |    1 
 /net/ipv4/ip_vti.c                      |   31 -
 /net/ipv4/xfrm4_policy.c                |   11 
 /net/ipv6/ip6_vti.c                     |   31 -
 /net/ipv6/xfrm6_policy.c                |   11 
 /net/key/af_key.c                       |    4 
 /net/xfrm/xfrm_device.c                 |    3 
 /net/xfrm/xfrm_policy.c                 |  334 ++++----------------
 /net/xfrm/xfrm_user.c                   |    2 
 /security/selinux/include/xfrm.h        |    4 
 include/net/flowcache.h                  |   25 -
 net/core/flow.c                          |  516 -------------------------------
 16 files changed, 82 insertions(+), 951 deletions(-)

             reply	other threads:[~2017-06-28 13:26 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-28 13:26 Florian Westphal [this message]
2017-06-28 13:26 ` [RFC net-next 1/9] vti: revert flush x-netns xfrm cache when vti interface is removed Florian Westphal
2017-06-28 13:26 ` [RFC net-next 2/9] net: xfrm: revert to lower xfrm dst gc limit Florian Westphal
2017-06-28 13:26 ` [RFC net-next 3/9] xfrm_policy: bypass flow_cache_lookup Florian Westphal
2017-06-28 13:26 ` [RFC net-next 4/9] xfrm_policy: remove always true/false branches Florian Westphal
2017-06-28 13:26 ` [RFC net-next 5/9] xfrm_policy: kill flow to policy dir conversion Florian Westphal
2017-06-28 13:26 ` [RFC net-next 6/9] xfrm_policy: remove xfrm_policy_lookup Florian Westphal
2017-06-28 13:26 ` [RFC net-next 7/9] xfrm_policy: make xfrm_bundle_lookup return xfrm dst object Florian Westphal
2017-06-28 13:26 ` [RFC net-next 8/9] xfrm: remove flow cache Florian Westphal
2017-06-28 13:26 ` [RFC net-next 9/9] xfrm: add a small xdst pcpu cache Florian Westphal
2017-06-29 13:06   ` Ilan Tayari
2017-06-29 13:17     ` Florian Westphal
2017-07-05  9:01     ` Ilan Tayari

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=20170628132652.1275-1-fw@strlen.de \
    --to=fw@strlen.de \
    --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).