Cake - FQ_codel the next generation
 help / color / mirror / Atom feed
From: Sebastian Moeller <moeller0@gmx.de>
To: "Dave Täht" <dave.taht@gmail.com>
Cc: Jonathan Morton <chromatix99@gmail.com>,
	Cake List <cake@lists.bufferbloat.net>
Subject: Re: [Cake] set filtered acks to bulk mode
Date: Wed, 6 Dec 2017 09:10:00 +0100	[thread overview]
Message-ID: <7740436B-FB24-49D2-999C-DAFC0BA6890C@gmx.de> (raw)
In-Reply-To: <CAA93jw45bzSb8EPj6Xvo=LB5abJ-f_4ove8Uhb1tq3N0m-oLmA@mail.gmail.com>


> On Dec 6, 2017, at 08:05, Dave Taht <dave.taht@gmail.com> wrote:
> 
> On Tue, Dec 5, 2017 at 10:08 PM, Jonathan Morton <chromatix99@gmail.com> wrote:
>> This is all sorts of wrong.
> 
> Going 1 out of 2 on untested patches for the day is not too horrible.
> 
> (sorry for crashing your router Ryan!!)
> 
>> First, setting the enum isn't sufficient to actually make the flow bulk.
>> You need to move it to the bulk queue-of-queues and update the stats
>> counters too.  That's probably why it's crashing for Ryan.
> 
> I had figured that would happen on the next round. :(
> 
>> 
>> Second, I fail to see why this is necessary or even desirable.  If there are
>> enough acks to activate the filter, then I assume it should be treated as
>> bulk already.  Or, if the filter itself thins out the acks enough to make it
>> sparse, then it still makes sense to deliver those acks timely
> 
> Consider the case where you are emptying the queue on a regular basis.
> 
> As this is indeed, a bulk flow at this point, there is no need to give
> it any extra boost.

	Mmmh, I guess I see Jonathan's point more clearly, if the (egress) ack traffic is sparse treat it like that if it is not then not. The fact that the non-filtered ACK queue would be treated as bulk, is I believe immaterial.
Well, unless we have data showing that your proposed mode actually works better in real life. Data, I realize, for the acquisition of which you would need the proposed change... This might be too complicated for me ;)

Best Regards
	Sebastian


> 
>> so as to
>> maximise the reduced information they carry.  Remember too that Codel
>> implements head drop precisely to minimise the feedback lag via ack.
>> 
>> - Jonathan Morton
> 
> 
> 
> -- 
> 
> Dave Täht
> CEO, TekLibre, LLC
> http://www.teklibre.com
> Tel: 1-669-226-2619
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake


      reply	other threads:[~2017-12-06  8:10 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-05 22:00 Dave Taht
2017-12-06  4:59 ` Ryan Mounce
2017-12-06  6:08 ` Jonathan Morton
2017-12-06  7:05   ` Dave Taht
2017-12-06  8:10     ` Sebastian Moeller [this message]

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

  List information: https://lists.bufferbloat.net/postorius/lists/cake.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=7740436B-FB24-49D2-999C-DAFC0BA6890C@gmx.de \
    --to=moeller0@gmx.de \
    --cc=cake@lists.bufferbloat.net \
    --cc=chromatix99@gmail.com \
    --cc=dave.taht@gmail.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