DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Thomas Monjalon <thomas@monjalon.net>
Cc: John McNamara <john.mcnamara@intel.com>,
	Marko Kovacevic <marko.kovacevic@intel.com>,
	dev@dpdk.org, stable@dpdk.org,
	Andrew Rybchenko <arybchenko@solarflare.com>,
	Adrien Mazarguil <adrien.mazarguil@6wind.com>,
	Ajit Khaparde <ajit.khaparde@broadcom.com>,
	Jerin Jacob <jerinj@marvell.com>
Subject: Re: [dpdk-dev] [PATCH 1/2] doc: add PMD filtering features back
Date: Fri, 10 Apr 2020 09:53:10 +0100	[thread overview]
Message-ID: <faf0f7b2-566d-677b-d53f-270042fcc0b3@intel.com> (raw)
In-Reply-To: <2221169.ijzf7UYKas@xps>

On 10/25/2019 2:23 PM, Thomas Monjalon wrote:
> 25/10/2019 14:51, Ferruh Yigit:
>> Also filter definitions should be clarified more in the features
>> documentation so that all PMDs can easily figure out to announce or
>> not these filtering features, also users can understand better what to
>> expect from this feature.
> 
> OK let's try to dig in some descriptions.
> 
> [...]
>> +N-tuple filter
>> +--------------
>> +
>> +Supports filtering on N-tuple values.
> 
> What is n-tuple? 3-tuple is enough?

I don't know why these are added historically, and myself not aware of the all
filtering to classify them properly.

Just saying "Flow API" is supported is not really says anything,
either we can try to group the filtering features, as done here and advertise
these filtering features, or create a rte flow based pattern/action
documentation which I believe will be too big and detailed.

So I think it is good to have some grouping for the filtering features, but if
current grouping is not good, what it should be like, and how we can define them?

I will send a v2 based on other comments, but it won't be addressing the groping
questions.

> 
>> +Tunnel filter
>> +-------------
>> +
>> +Supports tunnel filtering.
> 
> There are so many kinds of tunnels and filtering.
> What is minimum to declare such feature?
> Isn't it more relevant to list the supported tunnels?
> 
>> +Flexible filter
>> +---------------
>> +
>> +Supports a flexible (non-tuple or Ethertype) filter.
> 
> This is meaningless and should be dropped for sure.
> 
>> +* **[implements] eth_dev_ops**: ``filter_ctrl:RTE_ETH_FILTER_FLEXIBLE``.
>> +* **[related]    API**: ``rte_eth_dev_filter_ctrl()``, ``rte_eth_dev_filter_supported()``.
> 
> You are referring to the deprecated API here and in other places.

+1 to remove the deprecated APIs, I will convert them to rte flow.

> 
>> +Hash filter
>> +-----------
>> +
>> +Supports Hash filtering.
> 
> Which hash?
> 
>> +Flow director
>> +-------------
>> +
>> +Supports Flow Director style filtering to queues.
> 
> I think it is an Intel wording.
> Could you describe it with networking words?
> 
> 
> 


  reply	other threads:[~2020-04-10  8:53 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-25 12:51 Ferruh Yigit
2019-10-25 12:51 ` [dpdk-dev] [PATCH 2/2] doc: remove flow API from the feature list Ferruh Yigit
2019-10-25 13:26   ` Thomas Monjalon
2019-10-25 13:39     ` Jerin Jacob
2020-04-10  8:44       ` Ferruh Yigit
2020-04-10  8:56         ` Thomas Monjalon
2020-04-10  9:04           ` Jerin Jacob
2020-04-10  9:20             ` Thomas Monjalon
2020-04-10  9:27               ` Ferruh Yigit
2019-10-25 13:23 ` [dpdk-dev] [PATCH 1/2] doc: add PMD filtering features back Thomas Monjalon
2020-04-10  8:53   ` Ferruh Yigit [this message]
2020-04-10  9:29 ` [dpdk-dev] [PATCH v2] doc: remove flow API from the feature list Ferruh Yigit
2020-04-10  9:29 ` [dpdk-dev] [PATCH v2 1/2] doc: add PMD filtering features back Ferruh Yigit
2020-04-10  9:29   ` [dpdk-dev] [PATCH v2 2/2] doc: remove flow API from the feature list Ferruh Yigit
2020-04-10 12:10     ` Jerin Jacob
2020-04-10 12:34       ` Thomas Monjalon
2020-04-10 12:38         ` Jerin Jacob
2020-04-10 12:22   ` [dpdk-dev] [PATCH v2 1/2] doc: add PMD filtering features back Jerin Jacob

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=faf0f7b2-566d-677b-d53f-270042fcc0b3@intel.com \
    --to=ferruh.yigit@intel.com \
    --cc=adrien.mazarguil@6wind.com \
    --cc=ajit.khaparde@broadcom.com \
    --cc=arybchenko@solarflare.com \
    --cc=dev@dpdk.org \
    --cc=jerinj@marvell.com \
    --cc=john.mcnamara@intel.com \
    --cc=marko.kovacevic@intel.com \
    --cc=stable@dpdk.org \
    --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).