From: "Richardson, Bruce" <bruce.richardson@intel.com>
To: John Fastabend <john.fastabend@gmail.com>,
"Zhang, Qi Z" <qi.z.zhang@intel.com>,
"Xing, Beilei" <beilei.xing@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
"Zhang, Helin" <helin.zhang@intel.com>,
"Yigit, Ferruh" <ferruh.yigit@intel.com>
Subject: Re: [dpdk-dev] [PATCH v2 0/2] AVX2 Vectorized Rx/Tx functions for i40e
Date: Wed, 10 Jan 2018 09:27:42 +0000 [thread overview]
Message-ID: <59AF69C657FD0841A61C55336867B5B0722203E8@IRSMSX103.ger.corp.intel.com> (raw)
In-Reply-To: <d87f5ec9-9d37-0606-d715-bc816eb4e29e@gmail.com>
> -----Original Message-----
> From: John Fastabend [mailto:john.fastabend@gmail.com]
> Sent: Tuesday, January 9, 2018 4:31 PM
> To: Richardson, Bruce; Zhang, Qi Z; Xing, Beilei
> Cc: dev@dpdk.org; Zhang, Helin; Yigit, Ferruh
> Subject: Re: [dpdk-dev] [PATCH v2 0/2] AVX2 Vectorized Rx/Tx functions for
> i40e
>
> On 01/09/2018 06:32 AM, Bruce Richardson wrote:
> > This patch adds an AVX2 vectorized path to the i40e driver, based on
> > the existing SSE4.2 version. Using AVX2 instructions gives better
> > performance than the SSE version, though the percentage increase
> > depends on the exact settings used. For example:
> >
>
> Hi Bruce,
>
> Just curious, can you provide some hints on percent increase in at least
> some representative cases? I'm just trying to get a sense of if this is
> %5, 10%, 20%, more... I know mileage will vary depending on system, setup,
> configuration, etc.
>
Best case conditions to test under are using testpmd as that is where any IO improvement will be most seen. As a ballpark figure though, on my system while testing testpmd with both 16B and 32B descriptors, (RX/TX ring sizes 1024/512) I saw ~15% performance increase, and sometimes quite a bit higher, e.g. when testing with 16B descriptors with larger burst sizes.
As you say system, setup, config all has an impact, so YMMV... :-).
/Bruce
next prev parent reply other threads:[~2018-01-10 9:27 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-11-23 16:53 [dpdk-dev] [PATCH " Bruce Richardson
2017-11-23 16:53 ` [dpdk-dev] [PATCH 1/2] net/i40e: add AVX2 Tx function Bruce Richardson
2017-11-29 2:13 ` Ferruh Yigit
2017-11-29 10:02 ` Bruce Richardson
2017-11-23 16:53 ` [dpdk-dev] [PATCH 2/2] net/i40e: add AVX2 Rx function Bruce Richardson
2017-11-23 16:56 ` [dpdk-dev] [PATCH 0/2] AVX2 Vectorized Rx/Tx functions for i40e Bruce Richardson
2017-11-27 10:45 ` Zhang, Qi Z
2018-01-09 14:32 ` [dpdk-dev] [PATCH v2 " Bruce Richardson
2018-01-09 14:32 ` [dpdk-dev] [PATCH v2 1/2] net/i40e: add AVX2 Tx function Bruce Richardson
2018-01-09 14:32 ` [dpdk-dev] [PATCH v2 2/2] net/i40e: add AVX2 Rx function Bruce Richardson
2018-01-09 16:30 ` [dpdk-dev] [PATCH v2 0/2] AVX2 Vectorized Rx/Tx functions for i40e John Fastabend
2018-01-10 9:27 ` Richardson, Bruce [this message]
2018-01-10 14:25 ` Vincent JARDIN
2018-01-10 14:38 ` Bruce Richardson
2018-01-10 7:11 ` Li, Xiaoyun
2018-01-10 7:14 ` Zhang, Qi Z
2018-01-10 13:49 ` Zhang, Helin
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=59AF69C657FD0841A61C55336867B5B0722203E8@IRSMSX103.ger.corp.intel.com \
--to=bruce.richardson@intel.com \
--cc=beilei.xing@intel.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=helin.zhang@intel.com \
--cc=john.fastabend@gmail.com \
--cc=qi.z.zhang@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).