Cake - FQ_codel the next generation
 help / color / mirror / Atom feed
From: Sebastian Gottschall <s.gottschall@newmedia-net.de>
To: "Toke Høiland-Jørgensen" <toke@redhat.com>,
	"Sebastian Moeller" <moeller0@gmx.de>,
	cake@lists.bufferbloat.net, "Dave Taht" <dave.taht@gmail.com>,
	"Jonathan Morton" <chromatix99@gmail.com>
Subject: Re: [Cake] pie in dd-wrt
Date: Thu, 22 Aug 2019 11:17:36 +0200	[thread overview]
Message-ID: <0cd1c3ea-6ef5-3293-0b65-58cb1e4871eb@newmedia-net.de> (raw)
In-Reply-To: <87wof6p595.fsf@toke.dk>


Am 21.08.2019 um 23:39 schrieb Toke Høiland-Jørgensen:
> Sebastian Gottschall <s.gottschall@newmedia-net.de> writes:
>
>> Am 21.08.2019 um 18:21 schrieb Sebastian Moeller:
>>> On August 21, 2019 6:12:03 PM GMT+02:00, Sebastian Gottschall <s.gottschall@newmedia-net.de> wrote:
>>>> thats rather old. i rewrote all the qos code in the last 4 or 5 days.
>>>> so
>>>> things might be changed. next phase is bringing all the link level
>>>> detail configuration stuff into the gui which will be done
>>>> tomorrow or at least still within this week.
>>>> i also added now cake to some smaller low budged routers with limited
>>>> resources, so see how it runs. i had bad experiences with fq_codel in
>>>> the past due some high memory usage.
>>>> especially since its hard coded somewhat into the wireless ath9k
>>>> driver.
>>>> so i already modded it for more efficient handling. 4 mb max per queue
>>>> is simply too much for  a 32 mb ram based router.
>>> This is why I'm sqm we reduced the queued packet maximum m to around 1000, and also why cake has an explicit memlimit keyword.
>> yeah but does this help if fq_codel is hardcoded into mac80211? fq_codel
>> has a memlimit  keyword too btw. its fixed to 4MB. i reduced it to 256kb
>> on low memory architectures. no other way to get around OOM problems
>> mac80211  does always make use of fq_codel. it has a own build in
>> implementation
> The mac80211 implementation also has a memlimit parameter. You can set
> it through debugfs - 2MB would be:
>
> echo 2097152 > /sys/kernel/debug/ieee80211/phy0/aqm
>
> or through iw:
>
> iw phy phy0 set txq memory_limit 2097152
>
> The nl80211 attribute is called NL80211_ATTR_TXQ_MEMORY_LIMIT.
as i said i already modified mac80211 in a way that it sets sane memory 
limits depending on the architecture. devices with just 32 mb ram run 
only stable with 256kb memory limit.
so i configured different defaults. but the point is still that for a 
standard user (lets say in openwrt) the current implementation is not 
good. somewhere in the openwrt community i was reading that such devices 
should not be used anymore for openwrt due the memory limitations. but 
thats no solution for me if it was working before the introduction of 
fq_codel in mac80211
>
> -Toke
>

  reply	other threads:[~2019-08-22  9:18 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-18 16:33 [Cake] cake " Dave Taht
2019-08-20 12:08 ` Sebastian Gottschall
2019-08-20 16:24   ` Dave Taht
2019-08-20 16:47     ` Sebastian Gottschall
2019-08-20 16:58       ` Toke Høiland-Jørgensen
2019-08-20 17:29         ` Sebastian Gottschall
2019-08-20 17:47           ` Toke Høiland-Jørgensen
2019-08-20 18:10             ` Sebastian Gottschall
2019-08-20 18:31               ` Toke Høiland-Jørgensen
2019-08-20 18:39                 ` Sebastian Gottschall
2019-08-20 18:49                   ` Toke Høiland-Jørgensen
2019-08-21  7:25                     ` Sebastian Gottschall
2019-08-20 19:07                   ` Jonathan Morton
2019-08-20 23:43                     ` Dave Taht
2019-08-21 10:21                       ` Toke Høiland-Jørgensen
2019-08-21 11:03                         ` Sebastian Moeller
2019-08-21 13:10                         ` Dave Taht
2019-08-21 14:52                         ` Jonathan Morton
2019-08-21 15:42                           ` [Cake] pie " Dave Taht
2019-08-21 16:12                             ` Sebastian Gottschall
2019-08-21 16:21                               ` Sebastian Moeller
2019-08-21 16:28                                 ` Sebastian Gottschall
2019-08-21 16:50                                   ` Dave Taht
2019-08-21 21:40                                     ` Toke Høiland-Jørgensen
2019-08-21 21:53                                       ` Dave Taht
2019-08-22  9:18                                         ` Sebastian Gottschall
2019-08-22 13:15                                           ` [Cake] Wifi Memory limits in small platforms Dave Taht
2019-08-22 14:59                                             ` Dave Taht
2019-08-22 15:48                                             ` Sebastian Gottschall
2019-08-22 17:03                                               ` Dave Taht
2019-08-22 17:37                                                 ` Sebastian Gottschall
2019-08-22 18:23                                                   ` Toke Høiland-Jørgensen
2019-08-22 18:56                                                     ` Dave Taht
2019-08-22 19:37                                                       ` [Cake] [Battlemesh] " Toke Høiland-Jørgensen
2019-08-22 20:10                                                         ` Sebastian Moeller
2019-08-22 20:30                                                       ` [Cake] " Sebastian Gottschall
2019-08-22 23:39                                                         ` Dave Taht
2019-08-23  6:25                                                           ` Sebastian Gottschall
2019-08-23  6:48                                                           ` Sebastian Moeller
2019-08-22 20:32                                                       ` [Cake] fq_codel_fast crash/lockup Sebastian Gottschall
2019-08-21 21:39                                   ` [Cake] pie in dd-wrt Toke Høiland-Jørgensen
2019-08-22  9:17                                     ` Sebastian Gottschall [this message]
2019-08-22 10:12                                       ` Toke Høiland-Jørgensen
2019-08-21  7:30                     ` [Cake] cake " Sebastian Gottschall
2019-08-20 18:05       ` Sebastian Gottschall
2019-08-20 23:43         ` Dave Taht
2019-08-20 23:34       ` Dave Taht
2019-08-21  7:44         ` Sebastian Gottschall
2019-08-21  7:44       ` Sebastian Moeller
2019-08-21  7:50         ` Sebastian Gottschall
2019-08-21  7:56           ` Sebastian Moeller
2019-08-21  9:04             ` Sebastian Gottschall
2019-08-21  9:17               ` Sebastian Moeller
2019-08-21 13:20           ` Dave Taht
2019-08-21 16:06             ` Sebastian Gottschall
2019-08-20 18:18 ` Sebastian Gottschall
2019-08-20 23:50   ` Dave Taht
2019-08-21  7:47     ` Sebastian Gottschall

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=0cd1c3ea-6ef5-3293-0b65-58cb1e4871eb@newmedia-net.de \
    --to=s.gottschall@newmedia-net.de \
    --cc=cake@lists.bufferbloat.net \
    --cc=chromatix99@gmail.com \
    --cc=dave.taht@gmail.com \
    --cc=moeller0@gmx.de \
    --cc=toke@redhat.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