From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-it0-x230.google.com (mail-it0-x230.google.com [IPv6:2607:f8b0:4001:c0b::230]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 3D97A3B25E; Wed, 10 Aug 2016 09:11:36 -0400 (EDT) Received: by mail-it0-x230.google.com with SMTP id x130so43762662ite.1; Wed, 10 Aug 2016 06:11:36 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=bmX+ZKXSp1CezD5+7tkhodbln8iujGQdfhg3XdNnTCw=; b=iHbRiJVCB1kNHCzk/dP+BQngSk9Ux4JBFg6bHP6yssUFOsf7zAK3bKPYFG5+Bz7fGF hTQL0rJAnn6ascSCFPoi8Kyjt4YFjfVjmXHtQcJGv4kK6cjUFxQjyedLX21S2ERkgqJV zI+Ew9d6+HYNjUPl4EvPHnyBhfCOkOSH91uiXlGi6cQSHIPsl8NqVgF1tkeGlVhwOMmK ze4AHiOkHvrdCk23vQMHbCckESJTFTzVfb0jy3uUYTtSfYpVYjn+KxYpq3MTzVmw+Ytw 3mOEt5mi4P5XMPI5IvnzLc63Sqj6VB/bg5UUqzIVQ44LaxBS/N0/teDgYOkOw1omKb8Z eQTg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=bmX+ZKXSp1CezD5+7tkhodbln8iujGQdfhg3XdNnTCw=; b=IFdEGj/jJar5DK86OM5N6tYAzRXrM4eDrmUcc+zRH7yWTTLG8UMYLGEBV6ZMuGOJJ+ cxSlRqHAKbnZ5Cl2wp0Jyq7xwsPsREBlPj17b+mn3jCAajMzz2nOGu8Sw+DDfQUiV7o/ CIljwc0gmMp2Aor8W4XyngooUA1y0P1zkEbGU1Udq8t+M8Z5l769q8y//qjhOAQo/26B w946h/Xb3SWtmNc09TAsmvvKLagTRW5N7a3rGrFjFHq1ZVJ3hXM+T5o7cGc7WC5TLmSW vxQldS3/WvvznD9cy43A8JhZ5k9F8L6K7N0YM585/TaFT+JM+ZqPjuxTKrVN86P5JtXB 6NCA== X-Gm-Message-State: AEkoousjEKhmZfVwm7zgcRd7VpFOJYlfCeJoyjF71xSd/NcGB98Ng6qVZnFppnujaMgkME8R6e96cvBNRga1yA== X-Received: by 10.36.20.75 with SMTP id 72mr3051258itg.83.1470834695682; Wed, 10 Aug 2016 06:11:35 -0700 (PDT) MIME-Version: 1.0 Received: by 10.36.13.200 with HTTP; Wed, 10 Aug 2016 06:11:35 -0700 (PDT) Received: by 10.36.13.200 with HTTP; Wed, 10 Aug 2016 06:11:35 -0700 (PDT) In-Reply-To: <76eb6a6a-2f39-3b36-d4c1-04198083c0f6@gmail.com> References: <76eb6a6a-2f39-3b36-d4c1-04198083c0f6@gmail.com> From: Dave Taht Date: Wed, 10 Aug 2016 15:11:35 +0200 Message-ID: To: Noah Causin Cc: Felix Fietkau , Simon Wunderlich , "cerowrt-devel@lists.bufferbloat.net" , make-wifi-fast@lists.bufferbloat.net Content-Type: multipart/alternative; boundary=001a114469428f4e220539b76329 Subject: Re: [Make-wifi-fast] wifi airtime fairness patches could use eyeballs and testing X-BeenThere: make-wifi-fast@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 10 Aug 2016 13:11:36 -0000 --001a114469428f4e220539b76329 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On Aug 10, 2016 3:06 PM, "Noah Causin" wrote: > > Hi, > > I am aware that a performance regression with using fq in mac80211 with multiple tcp streams has been reported, and this patch in lede disables it. > > https://github.com/lede-project/source/commit/4952469ff9278288d766b28247a17= 694b1c4faaa > > Has that been resolved? I don't know! Please nuke it in your build. Some evidence pointed at an iperf threading bug. > > Also, would it work if I removed the 220-fq_disable_hack.patch from mac80211 in lede and applied those patches for airtime fairness, Yes. Gofer it! > Thank you, > > Noah Causin > > On 8/10/2016 7:28 AM, Dave Taht wrote: >> >> Simon's team successfully[1] tested the attached patches on top of felix's lede staging tree on monday against 30 stations. The initial results were lovely, at the rates he tested at. If anyone out there is daring enough to try building these for the wndr3800, wndr3700v2 and nanostation, (or anything else with ath9k in it!) it would be very good to put these through as many other scenarios as possible, notably adhoc and wds need to get looked at - but go forth! blow things up anyway you can! measure! Flent has a rtt_fair_var test if you can get multiple stations going... Virtual beer to you, if you put your build up somewhere. The staging tree: https://git.lede-project.org/?p=3Dlede/nbd/staging.git;a=3Dsu= mmary ---------- Forwarded message ---------- From: Toke H=C3=B8iland-J=C3=B8rgen= sen < toke@toke.dk> Date: Mon, Aug 8, 2016 at 12:49 PM Subject: Re: u alive today? patch To: Dave Taht Cc: Simon Wunderlich < sw@simonwunderlich.de> Dave Taht writes: >>> >>> I am hanging with simon today, have lab setup, was hoping to do some patching... >> >> These are for LEDE; just dump them in package/kernel/mac80211/patches -Toke >> >> >> >> _______________________________________________ Make-wifi-fast mailing list Make-wifi-fast@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/make-wifi-fast > > --001a114469428f4e220539b76329 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable

On Aug 10, 2016 3:06 PM, "Noah Causin" <n0manletter@gmail.com> wrote:
>
> Hi,
>
> I am aware that a performance regression with using fq in mac80211 wit= h multiple tcp streams has been reported, and this patch in lede disables i= t.
>
> https://github.com/lede-project/source/commit/= 4952469ff9278288d766b28247a17694b1c4faaa
>
> Has that been resolved?

I don't know! Please nuke it in your build.

Some evidence pointed at an iperf threading bug.
>
> Also, would it work if I removed the 220-fq_disable_hack.patch from ma= c80211 in lede and applied those patches for airtime fairness,

Yes.

Gofer it!

> Thank you,
>
> Noah Causin
>
> On 8/10/2016 7:28 AM, Dave Taht wrote:
>>
>> Simon's team successfully[1] tested the attached patches on to= p of felix's lede staging tree on monday against 30 stations. The initi= al results were lovely, at the rates he tested at. If anyone out there is d= aring enough to try building these for the wndr3800, wndr3700v2 and nanosta= tion, (or anything else with ath9k in it!) it would be very good to put the= se through as many other scenarios as possible, notably adhoc and wds need = to get looked at - but go forth! blow things up anyway you can! measure! Fl= ent has a rtt_fair_var test if you can get multiple stations going... Virtu= al beer to you, if you put your build up somewhere. The staging tree: h= ttps://git.lede-project.org/?p=3Dlede/nbd/staging.git;a=3Dsummary -----= ----- Forwarded message ---------- From: Toke H=C3=B8iland-J=C3=B8rgensen &= lt;toke@toke.dk> Date: Mon, Aug 8, 2= 016 at 12:49 PM Subject: Re: u alive today? patch To: Dave Taht <dave.taht@gmail.com> Cc: Simon Wunde= rlich <sw@simonwunderlich.de> Dave Taht <dave.taht@gmail.= com> writes:
>>>
>>> I am hanging with simon today, have lab setup, was hoping to d= o some patching...
>>
>>=C2=A0 These are for LEDE; just dump them in package/kernel/mac8021= 1/patches -Toke
>>
>>
>>
>> _______________________________________________ Make-wifi-fast mai= ling list Make-wifi= -fast@lists.bufferbloat.net https://lists.bufferbloat.net/listinfo/make-wifi-fas= t
>
>

--001a114469428f4e220539b76329--