Lets make wifi fast again!
 help / color / mirror / Atom feed
From: Dave Taht <dave.taht@gmail.com>
To: Tim Higgins <tim@timhiggins.com>
Cc: Erkki Lintunen <erkki.lintunen@iki.fi>,
	 Make-Wifi-fast <make-wifi-fast@lists.bufferbloat.net>
Subject: Re: [Make-wifi-fast] SmallNetBuilder article: Does OFDMA Really Work?
Date: Fri, 15 May 2020 09:25:31 -0700	[thread overview]
Message-ID: <CAA93jw72nBDNxs3P6QiQn-HWnRdmrd4TDwiJYA6BVbgYkuHirA@mail.gmail.com> (raw)
In-Reply-To: <29e9babb-b894-635d-4e60-332324d35565@timhiggins.com>

On Fri, May 15, 2020 at 8:34 AM Tim Higgins <tim@timhiggins.com> wrote:
>
> Hi Erkki,
>
> Thanks for your comments.
>
> A simple test is exactly what I'm after. The Part 1 article was basically a look behind the scenes to describe why I used the benchmark I did for the Part2 article. It will compare an AC router and multiple AX routers. If there's an off-the-shelf consumer router with that implements SQM/cake on Wi-Fi, I'm happy to try it. I have an evenroute IQrouter here, but I'm not sure if it implements SQM on Wi-Fi. I'll ask.

"SQM" isn't the thing. SQM is for better managing the ISP uplinks/downlinks.

it's the fq_codel and aqm scheduler for wifi. I'm under the impression
that the evenroute v2 production version now has some of the
aql/fq_codel related code backported to it. I've been off tuning that
up over here: https://forum.openwrt.org/t/aql-and-the-ath10k-is-lovely/
but that code has not made it back into mainline.

v3 has long had it, but was never optimized properly because I didn't
have one. Still I've never had a testbed as nice as yours,
it would be great, absolutely great, to get some feedback on your
methods you are applying to see how decent it is today! :)

The principal mis-feature in the current codebases for openwrt in
wifi's case is some code we left in there keeping the codel target
too high for 5ghz wifi, and far too large hw retries in the firmware
and driver for these two chipsets. However there is no express
supportfor mu-mimo in our versions as, like you, we discoverd the
concept didn't work - and couldn't ever work as designed - so we
discarded it in favor of better per station scheduling in the first
place.

https://arxiv.org/pdf/1703.00064.pdf
There were so many things left unfinished in that project. We could
have handled 802.11e much better than we do.





-- 
"For a successful technology, reality must take precedence over public
relations, for Mother Nature cannot be fooled" - Richard Feynman

dave@taht.net <Dave Täht> CTO, TekLibre, LLC Tel: 1-831-435-0729

  reply	other threads:[~2020-05-15 16:25 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-14 16:43 Tim Higgins
2020-05-14 21:38 ` Bob McMahon
2020-05-14 21:42   ` Bob McMahon
2020-05-15 15:20     ` Tim Higgins
2020-05-15 19:36       ` Bob McMahon
2020-05-15 19:50         ` Tim Higgins
2020-05-15 20:05           ` Bob McMahon
2020-05-15 20:24           ` Toke Høiland-Jørgensen
2020-05-15 20:30           ` Dave Taht
2020-05-15 21:35             ` Bob McMahon
2020-05-16  5:02               ` Dave Taht
     [not found]           ` <mailman.342.1589573120.24343.make-wifi-fast@lists.bufferbloat.net>
2020-05-15 20:38             ` Toke Høiland-Jørgensen
2020-05-15  6:47 ` Erkki Lintunen
2020-05-15 15:34   ` Tim Higgins
2020-05-15 16:25     ` Dave Taht [this message]
2020-05-15 16:41       ` Tim Higgins

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/make-wifi-fast.lists.bufferbloat.net/

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

  git send-email \
    --in-reply-to=CAA93jw72nBDNxs3P6QiQn-HWnRdmrd4TDwiJYA6BVbgYkuHirA@mail.gmail.com \
    --to=dave.taht@gmail.com \
    --cc=erkki.lintunen@iki.fi \
    --cc=make-wifi-fast@lists.bufferbloat.net \
    --cc=tim@timhiggins.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