DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ori Kam <orika@mellanox.com>
To: Thomas Monjalon <thomas@monjalon.net>,
	Pavan Nikhilesh Bhagavatula <pbhagavatula@marvell.com>
Cc: Jerin Jacob Kollanukkaran <jerinj@marvell.com>,
	"xiang.w.wang@intel.com" <xiang.w.wang@intel.com>,
	"dev@dpdk.org" <dev@dpdk.org>,
	Shahaf Shuler <shahafs@mellanox.com>,
	"hemant.agrawal@nxp.com" <hemant.agrawal@nxp.com>,
	Opher Reviv <opher@mellanox.com>,
	Alex Rosenbaum <alexr@mellanox.com>,
	Dovrat Zifroni <dovrat@marvell.com>,
	Prasun Kapoor <pkapoor@marvell.com>,
	"nipun.gupta@nxp.com" <nipun.gupta@nxp.com>,
	"bruce.richardson@intel.com" <bruce.richardson@intel.com>,
	"yang.a.hong@intel.com" <yang.a.hong@intel.com>,
	"harry.chang@intel.com" <harry.chang@intel.com>,
	"gu.jian1@zte.com.cn" <gu.jian1@zte.com.cn>,
	"shanjiangh@chinatelecom.cn" <shanjiangh@chinatelecom.cn>,
	"zhangy.yun@chinatelecom.cn" <zhangy.yun@chinatelecom.cn>,
	"lixingfu@huachentel.com" <lixingfu@huachentel.com>,
	"wushuai@inspur.com" <wushuai@inspur.com>,
	"yuyingxia@yxlink.com" <yuyingxia@yxlink.com>,
	"fanchenggang@sunyainfo.com" <fanchenggang@sunyainfo.com>,
	"davidfgao@tencent.com" <davidfgao@tencent.com>,
	"liuzhong1@chinaunicom.cn" <liuzhong1@chinaunicom.cn>,
	"zhaoyong11@huawei.com" <zhaoyong11@huawei.com>,
	"oc@yunify.com" <oc@yunify.com>,
	"jim@netgate.com" <jim@netgate.com>,
	"hongjun.ni@intel.com" <hongjun.ni@intel.com>,
	"j.bromhead@titan-ic.com" <j.bromhead@titan-ic.com>,
	"deri@ntop.org" <deri@ntop.org>,
	"fc@napatech.com" <fc@napatech.com>,
	"arthur.su@lionic.com" <arthur.su@lionic.com>,
	"david.marchand@redhat.com" <david.marchand@redhat.com>
Subject: Re: [dpdk-dev] [EXT] [PATCH v1 4/4] regexdev: implement regex rte	level functions
Date: Mon, 6 Apr 2020 18:53:55 +0000
Message-ID: <AM6PR05MB5176D6A6CD3A71D690EA003DDBC20@AM6PR05MB5176.eurprd05.prod.outlook.com> (raw)
In-Reply-To: <9615243.0lO9i7VF8o@xps>



> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Thomas Monjalon
> Sent: Monday, April 6, 2020 5:00 PM
> To: Pavan Nikhilesh Bhagavatula <pbhagavatula@marvell.com>; Ori Kam
> <orika@mellanox.com>
> Cc: Jerin Jacob Kollanukkaran <jerinj@marvell.com>; xiang.w.wang@intel.com;
> dev@dpdk.org; Shahaf Shuler <shahafs@mellanox.com>;
> hemant.agrawal@nxp.com; Opher Reviv <opher@mellanox.com>; Alex
> Rosenbaum <alexr@mellanox.com>; Dovrat Zifroni <dovrat@marvell.com>;
> Prasun Kapoor <pkapoor@marvell.com>; nipun.gupta@nxp.com;
> bruce.richardson@intel.com; yang.a.hong@intel.com; harry.chang@intel.com;
> gu.jian1@zte.com.cn; shanjiangh@chinatelecom.cn;
> zhangy.yun@chinatelecom.cn; lixingfu@huachentel.com; wushuai@inspur.com;
> yuyingxia@yxlink.com; fanchenggang@sunyainfo.com;
> davidfgao@tencent.com; liuzhong1@chinaunicom.cn;
> zhaoyong11@huawei.com; oc@yunify.com; jim@netgate.com;
> hongjun.ni@intel.com; j.bromhead@titan-ic.com; deri@ntop.org;
> fc@napatech.com; arthur.su@lionic.com; david.marchand@redhat.com
> Subject: Re: [dpdk-dev] [EXT] [PATCH v1 4/4] regexdev: implement regex rte
> level functions
> 
> 06/04/2020 15:50, Pavan Nikhilesh Bhagavatula:
> > From: Thomas Monjalon <thomas@monjalon.net>
> > >> >06/04/2020 14:33, Pavan Nikhilesh Bhagavatula:
> > >> >> >> From: Pavan Nikhilesh Bhagavatula
> > >> >> >> > >+uint16_t
> > >> >> >> > >+rte_regexdev_enqueue_burst(uint8_t dev_id, uint16_t
> > >> >qp_id,
> > >> >> >> > >+			   struct rte_regex_ops **ops, uint16_t
> > >> >nb_ops)
> > >> >> >> > >+{
> > >> >> >> > >+	return regex_devices[dev_id]-
> > >> >> >> > >>enqueue(regex_devices[dev_id], qp_id,
> > >> >> >> > >+					      ops, nb_ops);
> > >> >> >> > >+}
> > >> >> >> >
> > >> >> >> > Move these functions to .h in-lining them.
> > >> >> >> > Also, please add debug checks @see
> > >> >> >rte_eth_rx_burst/rte_eth_tx_burst.
> > >> >> >>
> > >> >> >> O.K will update.
> > >> >> >
> > >> >> >In general, inlining is a pain for ABI compatibility.
> > >> >> >Please inline only if the gain is very significant.
> > >> >> >
> > >> >>
> > >> >> The performance gain mostly comes from hoisting
> > >> >`regex_devices[dev_id]` load above the
> > >> >> poll loop.
> > >> >> Since, the performance measurement application is still in pipeline
> > >> >and regexdev would be
> > >> >> experimental for next couple of releases I suggest inlining it now
> > >and
> > >> >worrying about ABI when
> > >> >> experimental tag needs to be removed.
> > >> >
> > >> >No, we must worry about ABI from the beginning.
> > >>
> > >> I though performance was the primary objective :-).
> > >
> > >It is.
> > >
> > >> >> We can follow the same path as done by ethdev
> > >>
> >[https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefe
> nse.proofpoint.com%2Fv2%2Furl%3Fu%3Dhttps-
> &amp;data=02%7C01%7Corika%40mellanox.com%7Cc3e07910f52846f64b5008
> d7da32d82e%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C0%7C6372177
> 84252928490&amp;sdata=8Pxq8ciUVW7bMiB1XmQBNm%2Fsmy8m1Wa7yKXK
> JRL4d%2B4%3D&amp;reserved=0
> >
> >3A__https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fww
> w.mail-
> %2F&amp;data=02%7C01%7Corika%40mellanox.com%7Cc3e07910f52846f64b
> 5008d7da32d82e%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C0%7C637
> 217784252928490&amp;sdata=sTDsN85xShTU7WOLQQ2iF8eh%2FyoVHYwDq%
> 2FTdgdHdlbM%3D&amp;reserved=0
> > >> >2Darchive.com_dev-
> > >>
> > >>40dpdk.org_msg142392.html&d=DwICAg&c=nKjWec2b6R0mOyPaz7xt
> > >f
> > >> >Q&r=E3SgYMjtKCMVsB-fmvgGV3o-
> > >> >g_fjLhk5Pupi9ijohpc&m=7Gqb6WKmZV5uY3xa7FRVrRVDz8Usrsd-
> > >>
> > >>rDjIKr6CUQQ&s=sQo2Kx9fzTNXwiQ2Fzki3s5GSuiiAEzz2VtN68_KKXo&e
> > >=
> > >> >]
> > >> >
> > >> >ethdev is not an argument.
> > >>
> > >> What about ring?
> [https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Furldefens
> e.proofpoint.com%2Fv2%2Furl%3Fu%3Dhttp-
> &amp;data=02%7C01%7Corika%40mellanox.com%7Cc3e07910f52846f64b5008
> d7da32d82e%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C0%7C6372177
> 84252928490&amp;sdata=t2qUIj51WPzqQhJngheYh1s7DWzOxd%2FcAyAQK39C
> HgQ%3D&amp;reserved=0
> > >3A__mails.dpdk.org_archives_dev_2020-
> > >2DApril_161506.html&d=DwICAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=E3S
> > >gYMjtKCMVsB-fmvgGV3o-
> > >g_fjLhk5Pupi9ijohpc&m=u9gnM_YrOJDusN4yR8YUcUuUkri4tOjnHrQ0A
> > >Qd5zTw&s=uv6AQA-
> > >Zu7o6ugyhpGHLxFOk4SfEdkHfFGDmhzANRME&e= ]
> > >>
> > >> Why do we need to prove the same performance advantage using in-
> > >lining for datapath
> > >> critical functions again and again?
> > >
> > >Because every libraries have not the same usage and load.
> > >We should compare how much cycle is saved with inline vs
> > >how much cycles is, "in average", a regex burst?
> > >
> > >If you tell me regex processing is fast, OK, let's inline.
> > >
> >
> > Regex processing speed would still be dependent on underlying device
> capabilities.
> >
> > All we are trying to do is reduce the enqueue/dequeue completion time
> which would
> > bring down the overall latency.
> 
> Take your regex HW and do a simple regex processing burst.
> How many cycles it takes to complete?
> How many cycles you lose if not inline?
> If the ratio is lower than 1/200, I think inline is not a must.
> 
> Ori, please consider the same measure for your HW.
> 
I think that the default for data path should be inline.
In this specific case maybe the benefit is small.
I guess that by definition the RegEx is slower than the net device.
It is much harder to check the performance of RegEx device due to the 
fact that it depends on the data itself and on the rules.

I don't have HW to check the numbers right now. Even if I have them
this is for Mellanox implementation other PMD may have different values.
I suggest that we change to inline, and after getting the HW we will run some tests
and based on them we can decide if the API should be changed.
Please note that the API change will be before the first integration of this patch
so it is not real API change.

 


  reply	other threads:[~2020-04-06 18:54 UTC|newest]

Thread overview: 96+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-29  6:47 [dpdk-dev] [PATCH v1 0/4] add RegEx class Ori Kam
2020-03-29  6:47 ` [dpdk-dev] [PATCH v1 1/4] regexdev: introduce regexdev subsystem Ori Kam
2020-04-04 15:04   ` [dpdk-dev] [EXT] " Pavan Nikhilesh Bhagavatula
2020-04-05 15:03     ` Ori Kam
2020-03-29  6:47 ` [dpdk-dev] [PATCH v1 2/4] regexdev: add regex core h file Ori Kam
     [not found]   ` <DM5PR18MB221411AAEAFFA9D9F373292BC6C20@DM5PR18MB2214.namprd18.prod.outlook.com>
2020-04-07  8:53     ` Guy Kaneti
2020-04-07 16:16       ` Ori Kam
2020-04-07 16:27         ` Jerin Jacob
2020-04-08  7:37           ` Ori Kam
2020-04-08  7:48             ` Jerin Jacob
2020-04-08  8:31               ` Ori Kam
2020-04-08  8:38                 ` Jerin Jacob
2020-04-08  9:51                   ` Ori Kam
2020-03-29  6:47 ` [dpdk-dev] [PATCH v1 3/4] regexdev: add regexdev core functions Ori Kam
2020-04-04 15:01   ` [dpdk-dev] [EXT] " Pavan Nikhilesh Bhagavatula
2020-04-05 15:05     ` Ori Kam
2020-04-05 17:10       ` Pavan Nikhilesh Bhagavatula
2020-04-05 20:02         ` Ori Kam
2020-04-06 12:48           ` Pavan Nikhilesh Bhagavatula
2020-04-06 13:29             ` Thomas Monjalon
2020-04-06 13:38               ` Jerin Jacob
2020-04-06 19:11                 ` Ori Kam
2020-04-07  5:49                   ` Jerin Jacob
2020-04-07  6:46                     ` Ori Kam
2020-04-07  7:22                       ` Jerin Jacob
2020-04-07 12:27                     ` Thomas Monjalon
2020-04-07 12:54                       ` Jerin Jacob
2020-04-07 14:21                   ` Guy Kaneti
2020-04-07 16:28                     ` Ori Kam
2020-04-07 16:37                       ` Guy Kaneti
2020-04-08  6:52                         ` Ori Kam
2020-04-08  8:39               ` Ori Kam
2020-04-19 10:38                 ` Guy Kaneti
2020-04-22 21:36                   ` Ori Kam
2020-04-08  9:41             ` Ori Kam
2020-03-29  6:47 ` [dpdk-dev] [PATCH v1 4/4] regexdev: implement regex rte level functions Ori Kam
2020-04-04 14:27   ` [dpdk-dev] [EXT] " Pavan Nikhilesh Bhagavatula
2020-04-05 15:04     ` Ori Kam
2020-04-05 16:48       ` Pavan Nikhilesh Bhagavatula
2020-04-05 19:46         ` Ori Kam
2020-04-06 11:16       ` Thomas Monjalon
2020-04-06 12:33         ` Pavan Nikhilesh Bhagavatula
2020-04-06 13:14           ` Thomas Monjalon
2020-04-06 13:20             ` Jerin Jacob
2020-04-06 13:22             ` Pavan Nikhilesh Bhagavatula
2020-04-06 13:36               ` Thomas Monjalon
2020-04-06 13:50                 ` Pavan Nikhilesh Bhagavatula
2020-04-06 14:00                   ` Thomas Monjalon
2020-04-06 18:53                     ` Ori Kam [this message]
2020-04-04 13:06 ` [dpdk-dev] [EXT] [PATCH v1 0/4] add RegEx class Pavan Nikhilesh Bhagavatula
2020-04-05 15:03   ` Ori Kam
2020-04-17 12:43 ` [dpdk-dev] [PATCH v2 " Ori Kam
2020-04-17 12:43   ` [dpdk-dev] [PATCH v2 1/4] regexdev: introduce regexdev subsystem Ori Kam
2020-04-17 12:43   ` [dpdk-dev] [PATCH v2 2/4] regexdev: add regex core h file Ori Kam
2020-04-20 10:48     ` Guy Kaneti
2020-04-20 15:49       ` Ori Kam
2020-04-17 12:43   ` [dpdk-dev] [PATCH v2 3/4] regexdev: add regexdev core functions Ori Kam
2020-04-17 12:43   ` [dpdk-dev] [PATCH v2 4/4] regexdev: implement regex rte level functions Ori Kam
2020-04-21 11:12     ` Guy Kaneti
2020-04-21 11:20       ` Ori Kam
2020-04-21 11:36     ` Guy Kaneti
2020-04-22 20:33       ` Ori Kam
2020-05-07  9:45 ` [dpdk-dev] [PATCH v3 0/4] add RegEx class Ori Kam
2020-05-07  9:45   ` [dpdk-dev] [PATCH v3 1/4] regexdev: introduce regexdev subsystem Ori Kam
2020-06-21 11:18     ` Ori Kam
2020-06-30 19:57       ` Ori Kam
2020-05-07  9:45   ` [dpdk-dev] [PATCH v3 2/4] regexdev: add regex core h file Ori Kam
2020-06-03  6:47     ` [dpdk-dev] [EXT] " Guy Kaneti
2020-05-07  9:45   ` [dpdk-dev] [PATCH v3 3/4] regexdev: add regexdev core functions Ori Kam
2020-06-03  6:57     ` [dpdk-dev] [EXT] " Guy Kaneti
2020-05-07  9:45   ` [dpdk-dev] [PATCH v3 4/4] regexdev: implement regex rte level functions Ori Kam
2020-06-03  6:59     ` [dpdk-dev] [EXT] " Guy Kaneti
2020-06-28 13:45     ` Guy Kaneti
2020-06-28 14:10       ` Ori Kam
2020-05-24 20:24   ` [dpdk-dev] [PATCH v3 0/4] add RegEx class Ori Kam
2020-07-02  7:45 ` [dpdk-dev] [PATCH v4 " Ori Kam
2020-07-02  7:46   ` [dpdk-dev] [PATCH v4 1/4] regexdev: introduce regexdev subsystem Ori Kam
2020-07-05 21:18     ` Thomas Monjalon
2020-07-06  7:02       ` Ori Kam
2020-07-02  7:46   ` [dpdk-dev] [PATCH v4 2/4] regexdev: add regex core h file Ori Kam
2020-07-02  7:46   ` [dpdk-dev] [PATCH v4 3/4] regexdev: add regexdev core functions Ori Kam
2020-07-05 21:08     ` Thomas Monjalon
2020-07-06  6:07       ` Ori Kam
2020-07-06  7:03         ` Thomas Monjalon
2020-07-06  8:00           ` Bruce Richardson
2020-07-02  7:46   ` [dpdk-dev] [PATCH v4 4/4] regexdev: implement regex rte level functions Ori Kam
2020-07-05 21:21   ` [dpdk-dev] [PATCH v4 0/4] add RegEx class Thomas Monjalon
2020-07-06  7:03     ` Ori Kam
2020-07-06 17:36 ` [dpdk-dev] [PATCH v5 " Ori Kam
2020-07-06 17:36   ` [dpdk-dev] [PATCH v5 1/4] regexdev: introduce regexdev subsystem Ori Kam
2020-07-06 19:38     ` Thomas Monjalon
2020-09-11 12:46     ` David Marchand
2020-07-06 17:36   ` [dpdk-dev] [PATCH v5 2/4] regexdev: add regex core h file Ori Kam
2020-07-06 17:36   ` [dpdk-dev] [PATCH v5 3/4] regexdev: add regexdev core functions Ori Kam
2020-07-06 17:36   ` [dpdk-dev] [PATCH v5 4/4] regexdev: implement regex rte level functions Ori Kam
2020-07-06 22:30   ` [dpdk-dev] [PATCH v5 0/4] add RegEx class Thomas Monjalon

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=AM6PR05MB5176D6A6CD3A71D690EA003DDBC20@AM6PR05MB5176.eurprd05.prod.outlook.com \
    --to=orika@mellanox.com \
    --cc=alexr@mellanox.com \
    --cc=arthur.su@lionic.com \
    --cc=bruce.richardson@intel.com \
    --cc=david.marchand@redhat.com \
    --cc=davidfgao@tencent.com \
    --cc=deri@ntop.org \
    --cc=dev@dpdk.org \
    --cc=dovrat@marvell.com \
    --cc=fanchenggang@sunyainfo.com \
    --cc=fc@napatech.com \
    --cc=gu.jian1@zte.com.cn \
    --cc=harry.chang@intel.com \
    --cc=hemant.agrawal@nxp.com \
    --cc=hongjun.ni@intel.com \
    --cc=j.bromhead@titan-ic.com \
    --cc=jerinj@marvell.com \
    --cc=jim@netgate.com \
    --cc=liuzhong1@chinaunicom.cn \
    --cc=lixingfu@huachentel.com \
    --cc=nipun.gupta@nxp.com \
    --cc=oc@yunify.com \
    --cc=opher@mellanox.com \
    --cc=pbhagavatula@marvell.com \
    --cc=pkapoor@marvell.com \
    --cc=shahafs@mellanox.com \
    --cc=shanjiangh@chinatelecom.cn \
    --cc=thomas@monjalon.net \
    --cc=wushuai@inspur.com \
    --cc=xiang.w.wang@intel.com \
    --cc=yang.a.hong@intel.com \
    --cc=yuyingxia@yxlink.com \
    --cc=zhangy.yun@chinatelecom.cn \
    --cc=zhaoyong11@huawei.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

DPDK patches and discussions

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ https://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git