DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Hunt, David" <david.hunt@intel.com>
To: Phil Yang <phil.yang@arm.com>, dev@dpdk.org, thomas@monjalon.net
Cc: reshma.pattan@intel.com, gavin.hu@arm.com,
	honnappa.nagarahalli@arm.com, nd@arm.com
Subject: Re: [dpdk-dev] [PATCH v4 2/3] test/distributor: replace sync builtins with atomic builtins
Date: Wed, 10 Apr 2019 15:05:48 +0100	[thread overview]
Message-ID: <85718290-9c5a-01f5-21ab-0f1000e4dfde@intel.com> (raw)
In-Reply-To: <1554692551-28275-3-git-send-email-phil.yang@arm.com>

Hi Phil,

On 8/4/2019 4:02 AM, Phil Yang wrote:
> '__sync' built-in functions are deprecated, should use the '__atomic'
> built-in instead. the sync built-in functions are full barriers, while
> atomic built-in functions offer less restrictive one-way barriers,
> which help performance.
>
> Here is the example test result on TX2:
> sudo ./arm64-armv8a-linuxapp-gcc/app/test -l 112-139 \
> -n 4 --socket-mem=1024,1024 -- -i
> RTE>>distributor_perf_autotest
>
> *** distributor_perf_autotest without this patch ***
> ==== Cache line switch test ===
> Time for 33554432 iterations = 1519202730 ticks
> Ticks per iteration = 45
>
> *** distributor_perf_autotest with this patch ***
> ==== Cache line switch test ===
> Time for 33554432 iterations = 1251715496 ticks
> Ticks per iteration = 37
>
> Less ticks needed for the cache line switch test. It got 17% of
> performance improvement.


I'm seeing about an 8% performance degradation on my platform for the 
cache line switch test with the patch, however the single mode and burst 
mode tests area showing no difference, which are the more important 
tests. What kind of differences are you seeing in the single/burst mode 
tests?

Rgds,
Dave.


---snip---

  parent reply	other threads:[~2019-04-10 14:05 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-03  9:49 [dpdk-dev] [PATCH] packet_ordering: " Phil Yang
2019-03-28 18:42 ` Thomas Monjalon
2019-03-28 18:42   ` Thomas Monjalon
2019-03-29  1:34   ` Phil Yang (Arm Technology China)
2019-03-29  1:34     ` Phil Yang (Arm Technology China)
2019-03-29 10:56 ` [dpdk-dev] [PATCH v2 0/3] example and test cases optimizations Phil Yang
2019-03-29 10:56   ` Phil Yang
2019-03-29 10:56 ` [dpdk-dev] [PATCH v2 1/3] packet_ordering: add statistics for each worker thread Phil Yang
2019-03-29 10:56   ` Phil Yang
2019-03-29 16:39   ` Pattan, Reshma
2019-03-29 16:39     ` Pattan, Reshma
2019-03-30 16:55     ` Phil Yang (Arm Technology China)
2019-03-30 16:55       ` Phil Yang (Arm Technology China)
2019-04-01 12:58       ` Pattan, Reshma
2019-04-01 12:58         ` Pattan, Reshma
2019-04-02  3:33         ` Phil Yang (Arm Technology China)
2019-04-02  3:33           ` Phil Yang (Arm Technology China)
2019-03-29 10:56 ` [dpdk-dev] [PATCH v2 2/3] test/distributor: replace sync builtins with atomic builtins Phil Yang
2019-03-29 10:56   ` Phil Yang
2019-04-01 16:24   ` Honnappa Nagarahalli
2019-04-01 16:24     ` Honnappa Nagarahalli
2019-04-02  3:43     ` Phil Yang (Arm Technology China)
2019-04-02  3:43       ` Phil Yang (Arm Technology China)
2019-03-29 10:56 ` [dpdk-dev] [PATCH v2 3/3] test/ring_perf: " Phil Yang
2019-03-29 10:56   ` Phil Yang
2019-04-01 16:24   ` Honnappa Nagarahalli
2019-04-01 16:24     ` Honnappa Nagarahalli
2019-04-03  6:59 ` [dpdk-dev] [PATCH v3 0/3] example and test cases optimizations Phil Yang
2019-04-03  6:59   ` Phil Yang
2019-04-03  6:59   ` [dpdk-dev] [PATCH v3 1/3] packet_ordering: add statistics for each worker thread Phil Yang
2019-04-03  6:59     ` Phil Yang
2019-04-04 23:24     ` Thomas Monjalon
2019-04-04 23:24       ` Thomas Monjalon
2019-04-08  4:04       ` Phil Yang (Arm Technology China)
2019-04-08  4:04         ` Phil Yang (Arm Technology China)
2019-04-03  6:59   ` [dpdk-dev] [PATCH v3 2/3] test/distributor: replace sync builtins with atomic builtins Phil Yang
2019-04-03  6:59     ` Phil Yang
2019-04-04 15:30     ` Honnappa Nagarahalli
2019-04-04 15:30       ` Honnappa Nagarahalli
2019-04-03  6:59   ` [dpdk-dev] [PATCH v3 3/3] test/ring_perf: " Phil Yang
2019-04-03  6:59     ` Phil Yang
2019-04-08  3:02 ` [dpdk-dev] [PATCH v4 0/3] example and test cases optimizations Phil Yang
2019-04-08  3:02   ` Phil Yang
2019-07-04 20:15   ` Thomas Monjalon
2019-07-05  3:19     ` Phil Yang (Arm Technology China)
2019-07-08 14:38   ` Thomas Monjalon
2019-04-08  3:02 ` [dpdk-dev] [PATCH v4 1/3] packet_ordering: add statistics for each worker thread Phil Yang
2019-04-08  3:02   ` Phil Yang
2019-04-08  3:02 ` [dpdk-dev] [PATCH v4 2/3] test/distributor: replace sync builtins with atomic builtins Phil Yang
2019-04-08  3:02   ` Phil Yang
2019-04-10 14:05   ` Hunt, David [this message]
2019-04-10 14:05     ` Hunt, David
2019-04-11 11:31     ` Phil Yang (Arm Technology China)
2019-04-11 11:31       ` Phil Yang (Arm Technology China)
2019-04-08  3:02 ` [dpdk-dev] [PATCH v4 3/3] test/ring_perf: " Phil Yang
2019-04-08  3:02   ` Phil Yang

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=85718290-9c5a-01f5-21ab-0f1000e4dfde@intel.com \
    --to=david.hunt@intel.com \
    --cc=dev@dpdk.org \
    --cc=gavin.hu@arm.com \
    --cc=honnappa.nagarahalli@arm.com \
    --cc=nd@arm.com \
    --cc=phil.yang@arm.com \
    --cc=reshma.pattan@intel.com \
    --cc=thomas@monjalon.net \
    /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).