DPDK patches and discussions
 help / color / mirror / Atom feed
From: Maxime Coquelin <maxime.coquelin@redhat.com>
To: dev@dpdk.org, tiwei.bie@intel.com, jfreimann@redhat.com,
	zhihong.wang@intel.com, bruce.richardson@intel.com,
	konstantin.ananyev@intel.com
Cc: Maxime Coquelin <maxime.coquelin@redhat.com>
Subject: [dpdk-dev] [PATCH 0/5] vhost: I-cache pressure optimizations
Date: Fri, 17 May 2019 14:22:15 +0200	[thread overview]
Message-ID: <20190517122220.31283-1-maxime.coquelin@redhat.com> (raw)

Some OVS-DPDK PVP benchmarks show a performance drop
when switching from DPDK v17.11 to v18.11.

With the addition of packed ring layout support,
rte_vhost_enqueue_burst and rte_vhost_dequeue_burst
became very large, and only a part of the instructions
are executed (either packed or split ring used).

This series aims at improving the I-cache pressure,
first by un-inlining split and packed rings, but
also by moving parts considered as cold in dedicated
functions (dirty page logging, fragmented descriptors
buffer management added for CVE-2018-1059).

With the series applied, size of the enqueue and
dequeue split paths is reduced significantly:

+---------+--------------------+---------------------+
| Version | Enqueue split path |  Dequeue split path |
+---------+--------------------+---------------------+
| v19.05  | 16461B             | 25521B              |
| +series | 7286B              | 11285B              |
+---------+--------------------+---------------------+

Using perf tool to monitor iTLB-load-misses event
while doing PVP benchmark with testpmd as vswitch,
we can see the number of iTLB misses being reduced:

- v19.05:
# perf stat --repeat 10  -C 2,3  -e iTLB-load-miss -- sleep 10

 Performance counter stats for 'CPU(s) 2,3' (10 runs):

             2,438      iTLB-load-miss                                                ( +- 13.43% )

       10.00058928 +- 0.00000336 seconds time elapsed  ( +-  0.00% )

- +series:
# perf stat --repeat 10  -C 2,3  -e iTLB-load-miss -- sleep 10

 Performance counter stats for 'CPU(s) 2,3' (10 runs):

                55      iTLB-load-miss                                                ( +- 10.08% )

       10.00059466 +- 0.00000283 seconds time elapsed  ( +-  0.00% )

The series also force the inlining of some rte_memcpy
helpers, as by adding packed ring support, some of them
were not more inlined but embedded as functions in
the virtio_net object file, which was not expected.

Finally, the series simplifies the descriptors buffers
prefetching, by doing it in the recently introduced
descriptor buffer mapping function.

Maxime Coquelin (4):
  vhost: un-inline dirty pages logging functions
  vhost: do not inline packed and split functions
  vhost: do not inline unlikely fragmented buffers code
  vhost: simplify descriptor's buffer prefetching

root (1):
  eal/x86: force inlining of all memcpy and mov helpers

 .../common/include/arch/x86/rte_memcpy.h      |  18 +-
 lib/librte_vhost/vhost.c                      | 165 ++++++++++++++++++
 lib/librte_vhost/vhost.h                      | 164 ++---------------
 lib/librte_vhost/virtio_net.c                 | 142 +++++++--------
 4 files changed, 250 insertions(+), 239 deletions(-)

-- 
2.21.0


             reply	other threads:[~2019-05-17 12:22 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-17 12:22 Maxime Coquelin [this message]
2019-05-17 12:22 ` [dpdk-dev] [PATCH 1/5] vhost: un-inline dirty pages logging functions Maxime Coquelin
2019-05-17 12:22 ` [dpdk-dev] [PATCH 2/5] vhost: do not inline packed and split functions Maxime Coquelin
2019-05-17 13:00   ` David Marchand
2019-05-17 14:42     ` Maxime Coquelin
2019-05-17 12:22 ` [dpdk-dev] [PATCH 3/5] vhost: do not inline unlikely fragmented buffers code Maxime Coquelin
2019-05-17 12:57   ` Maxime Coquelin
2019-05-21 19:43   ` Mattias Rönnblom
2019-05-23 14:30     ` Maxime Coquelin
2019-05-23 15:17       ` Mattias Rönnblom
2019-05-23 17:40         ` Maxime Coquelin
2019-05-17 12:22 ` [dpdk-dev] [PATCH 4/5] vhost: simplify descriptor's buffer prefetching Maxime Coquelin
2019-05-17 12:22 ` [dpdk-dev] [PATCH 5/5] eal/x86: force inlining of all memcpy and mov helpers Maxime Coquelin
2019-05-17 13:04 ` [dpdk-dev] [PATCH 0/5] vhost: I-cache pressure optimizations David Marchand
2019-05-17 14:42   ` Maxime Coquelin

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=20190517122220.31283-1-maxime.coquelin@redhat.com \
    --to=maxime.coquelin@redhat.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=jfreimann@redhat.com \
    --cc=konstantin.ananyev@intel.com \
    --cc=tiwei.bie@intel.com \
    --cc=zhihong.wang@intel.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 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).