From: "Lu, Wenzhuo" <wenzhuo.lu@intel.com>
To: "Richardson, Bruce" <bruce.richardson@intel.com>,
"Ananyev, Konstantin" <konstantin.ananyev@intel.com>,
"Gonzalez Monroy, Sergio" <sergio.gonzalez.monroy@intel.com>
Cc: "Li, Xiaoyun" <xiaoyun.li@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>,
"Wang, Zhihong" <zhihong.wang@intel.com>,
"Zhang, Qi Z" <qi.z.zhang@intel.com>
Subject: Re: [dpdk-dev] [PATCH 1/3] eal/x86: run-time dispatch over memcpy
Date: Thu, 31 Aug 2017 01:21:15 +0000 [thread overview]
Message-ID: <6A0DE07E22DDAD4C9103DF62FEBC09093B6A2D0E@shsmsx102.ccr.corp.intel.com> (raw)
In-Reply-To: <20170830175140.GA11692@bricha3-MOBL3.ger.corp.intel.com>
Hi Bruce,
> -----Original Message-----
> From: Richardson, Bruce
> Sent: Thursday, August 31, 2017 1:52 AM
> To: Ananyev, Konstantin <konstantin.ananyev@intel.com>
> Cc: Li, Xiaoyun <xiaoyun.li@intel.com>; dev@dpdk.org; Lu, Wenzhuo
> <wenzhuo.lu@intel.com>; Wang, Zhihong <zhihong.wang@intel.com>; Zhang,
> Qi Z <qi.z.zhang@intel.com>
> Subject: Re: [dpdk-dev] [PATCH 1/3] eal/x86: run-time dispatch over memcpy
>
> On Wed, Aug 30, 2017 at 03:56:35PM +0100, Ananyev, Konstantin wrote:
> >
> >
> > > -----Original Message-----
> > > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Xiaoyun Li
> > > Sent: Friday, August 25, 2017 3:06 AM
> > > To: Richardson, Bruce <bruce.richardson@intel.com>
> > > Cc: dev@dpdk.org; Lu, Wenzhuo <wenzhuo.lu@intel.com>; Wang,
> Zhihong
> > > <zhihong.wang@intel.com>; Zhang, Qi Z <qi.z.zhang@intel.com>; Li,
> > > Xiaoyun <xiaoyun.li@intel.com>
> > > Subject: [dpdk-dev] [PATCH 1/3] eal/x86: run-time dispatch over
> > > memcpy
> > >
> > > This patch dynamically selects functions of memcpy at run-time based
> > > on CPU flags that current machine supports. This patch uses function
> > > pointers which are bind to the relative functions at constrctor time.
> > > To make AVX512 instructions pass compilation, enable the switch in
> > > makefile.
> >
> > It seems quite an overhead to add extra function call for each 16B
> movement...
> > Wouldn't it be better to have one func_ptr per implementation, i.e:
> > rte_memcpy_sse(), rte_memcpy_avx2(), rte_memcpy_avx512(), etc.?
> > Konstantin
> >
> +1 to this.
>
> Also, how big of a benefit is there for this implementation over standard libc
> memcpy (in a reasonably bleeding edge distro like e.g.
> Fedora 26)?
This patch is not an optimization. It only to make the code easier to use. So, the benefit is just the same as before.
I'm also curious about the benefit. Suppose it's better than standard libc. If not, maybe we should just use standard libc and this patch is not valuable.
+ Sergio, the maintainer of this module for more suggestion. Thanks.
>
> /Bruce
next prev parent reply other threads:[~2017-08-31 1:21 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-25 2:06 [dpdk-dev] [PATCH 0/3] dynamic linking support Xiaoyun Li
2017-08-25 2:06 ` [dpdk-dev] [PATCH 1/3] eal/x86: run-time dispatch over memcpy Xiaoyun Li
2017-08-30 14:56 ` Ananyev, Konstantin
2017-08-30 17:51 ` Bruce Richardson
2017-08-31 1:21 ` Lu, Wenzhuo [this message]
2017-08-30 18:00 ` Stephen Hemminger
2017-08-31 1:23 ` Lu, Wenzhuo
2017-08-31 5:05 ` Stephen Hemminger
2017-08-31 5:24 ` Li, Xiaoyun
2017-08-25 2:06 ` [dpdk-dev] [PATCH 2/3] app/test: run-time dispatch over memcpy perf test Xiaoyun Li
2017-08-25 2:06 ` [dpdk-dev] [PATCH 3/3] efd: run-time dispatch over x86 EFD functions Xiaoyun Li
2017-09-01 8:56 ` [dpdk-dev] [PATCH v2 0/3] dynamic linking support Xiaoyun Li
2017-09-01 8:57 ` [dpdk-dev] [PATCH v2 1/3] eal/x86: run-time dispatch over memcpy Xiaoyun Li
2017-09-01 9:16 ` Ananyev, Konstantin
2017-09-01 9:28 ` Li, Xiaoyun
2017-09-01 10:38 ` Ananyev, Konstantin
2017-09-04 1:41 ` Li, Xiaoyun
[not found] ` <B9E724F4CB7543449049E7AE7669D82F44216E@SHSMSX101.ccr.corp.intel.com>
[not found] ` <B9E724F4CB7543449049E7AE7669D82F442FE6@SHSMSX101.ccr.corp.intel.com>
2017-09-12 2:27 ` Li, Xiaoyun
2017-09-20 6:57 ` Li, Xiaoyun
2017-09-01 15:34 ` Stephen Hemminger
2017-09-01 8:57 ` [dpdk-dev] [PATCH v2 2/3] app/test: run-time dispatch over memcpy perf test Xiaoyun Li
2017-09-01 8:57 ` [dpdk-dev] [PATCH v2 3/3] efd: run-time dispatch over x86 EFD functions Xiaoyun Li
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=6A0DE07E22DDAD4C9103DF62FEBC09093B6A2D0E@shsmsx102.ccr.corp.intel.com \
--to=wenzhuo.lu@intel.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=konstantin.ananyev@intel.com \
--cc=qi.z.zhang@intel.com \
--cc=sergio.gonzalez.monroy@intel.com \
--cc=xiaoyun.li@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).