linux-wireless.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Benjamin Beichler <Benjamin.Beichler@uni-rostock.de>
To: <julanhsu@google.com>, <johannes@sipsolutions.net>
Cc: <linux-wireless@vger.kernel.org>
Subject: Re: [PATCH] mac80211: mesh: use average bitrate for link metric calculation
Date: Tue, 22 Jan 2019 11:18:14 +0100	[thread overview]
Message-ID: <c0618d8a-17eb-4929-faea-56ce7bcca9f5@uni-rostock.de> (raw)
In-Reply-To: <20190115233156.225963-1-julanhsu@google.com>

Hi,
Am 16.01.2019 um 00:31 schrieb julanhsu@google.com:
> From: Julan Hsu <julanhsu@google.com>
>
> Use bitrate moving average to smooth out link metric and stablize path
> selection.
We had some investigations on the same field, but I didn't had time to
publish anything of that. The general idea is good and we implemented it
similar.
One Issue is the different rate of multi cast and management frames in
contrast to uni cast data frames. AFAIK, the rate for management frames,
especially the beacon frames of mesh, are always sent with a low
mandatory rates (although they are configurable). If the link is less
used, the beacon frames may slowly degrade the average.
But anyhow, this problem also exist in the current implementation. The
other problem is, as you may expected already, the actual used
information for the metrics are highly driver dependent.
So e.g. the ath9k driver had some issue in the past, were the update
metric function was not called for some runtime modes (no-SKB vs SKB
ect.). Or the update_metric function is not called for all kind of frames.

But in general I think it is a first good step for better metrics.

--
M.Sc. Benjamin Beichler

Universität Rostock, Fakultät für Informatik und Elektrotechnik
Institut für Angewandte Mikroelektronik und Datentechnik

University of Rostock, Department of CS and EE
Institute of Applied Microelectronics and CE

Richard-Wagner-Straße 31
18119 Rostock
Deutschland/Germany

phone: +49 (0) 381 498 - 7278
email: Benjamin.Beichler@uni-rostock.de
www: http://www.imd.uni-rostock.de/

  reply	other threads:[~2019-01-22 10:23 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-15 23:31 [PATCH] mac80211: mesh: use average bitrate for link metric calculation julanhsu
2019-01-22 10:18 ` Benjamin Beichler [this message]
2019-01-28 16:47 ` Benjamin Beichler

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=c0618d8a-17eb-4929-faea-56ce7bcca9f5@uni-rostock.de \
    --to=benjamin.beichler@uni-rostock.de \
    --cc=johannes@sipsolutions.net \
    --cc=julanhsu@google.com \
    --cc=linux-wireless@vger.kernel.org \
    /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).