From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ot1-x32e.google.com (mail-ot1-x32e.google.com [IPv6:2607:f8b0:4864:20::32e]) (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 BD3183B29D for ; Tue, 22 Jun 2021 14:15:13 -0400 (EDT) Received: by mail-ot1-x32e.google.com with SMTP id t4-20020a9d66c40000b029045e885b18deso1946001otm.6 for ; Tue, 22 Jun 2021 11:15:13 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to; bh=dK6CLv4fk5m4wn0TUgGl3T+Ddg6il9bq+G5FI1QGpWE=; b=g2v836DCW3SDPofDTkheyThUd3M9sS59Lc10uUmYj7UHBxHkkgd11QrrKmFzUhIwOk DxCgsSofO8hpU015/shk+mx4QpMe9aWCYXnm6NEObnkDQJMaEDcE92mX57/MBQtjuDoD xKR0534DEbJdw7B2uCqT0j5JtssTdvwElJ498C4ee9cvEpwJlB11l2NIXtt9BbcFVtY8 CGazDJkpw/G60PxOVm5iFAGkLeYUBk7jGA/xZdoanAc+/EPDuox4pAQzygkP6e1MVGzO cDHdVLVKorptzeZc1NmEM0aepLsl2q+ua6DydIl5tCBdvt4ltESYcR7XvCdiCUDrEf0j fQPQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to; bh=dK6CLv4fk5m4wn0TUgGl3T+Ddg6il9bq+G5FI1QGpWE=; b=Lucm3lOAt1/fBHp3oICAFlKUs/AnqCJq9nxXQFOps1ReNGr2ZX1AySXohSGIX3Go1p iv31yhR/vbOARhEdcJD75H0CVPSMJpwf5JmiJfPqtx3aZjs0NbERzaSXoKC/lprHQKkT ETLtZtOwmnEwJtP0l5dszOwfdSpJ1Cxf8JlaygZJ2RwOjci5lbh2Rz+It9FqFV+bvmHv KoEEgoyK7I/orc/+1rYFYpmx8pWFYOrB06BrAjfomb+63LpSYBaZs4z4gbEcfbUSbY7h 2YQ7oY/z3BWffYxyH7ufJQrGJlfWgVSBr+oUuExzPY7tX9f0JBs7gwVMCx9NlggsmqNX B+vA== X-Gm-Message-State: AOAM5324dhS2121L2DFm7WGxx5ydzT/P02/for90nF40/t4RV4d5+pcU PNZeUrI4lkuTrbCD+BhAul8pxI24xP3wpHRmmZvfmKaT X-Google-Smtp-Source: ABdhPJzpZ8q9hwvwa+r5WJCwT5e4CKatyuy0A6CKeWkElVQKXzph4nU5UqTMXeU1N17ANcLgaCSLiyVyxkVQM3mYehM= X-Received: by 2002:a9d:4689:: with SMTP id z9mr4295000ote.129.1624385712866; Tue, 22 Jun 2021 11:15:12 -0700 (PDT) MIME-Version: 1.0 References: In-Reply-To: From: Jon Pike Date: Tue, 22 Jun 2021 11:15:01 -0700 Message-ID: To: make-wifi-fast@lists.bufferbloat.net Content-Type: multipart/alternative; boundary="00000000000064c7ff05c55ec5cb" Subject: Re: [Make-wifi-fast] openwifi presentation 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: Tue, 22 Jun 2021 18:15:13 -0000 --00000000000064c7ff05c55ec5cb Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Interesting stuff... One thing that occurred to me, that they didnt mention... doing an intermediate step, i.e. make a stripped down FPGA dev board, vs the Xylinx or whoever's standard dev board. Even with a radio chipset and parts also on it, could be done much cheaper, and still be a dev platform. I also understand that if you could raise the $$$$, and push an ASIC out, theres a much greater chance of large scale acceptance. But, it is a lot of $$$$, and for a single version. Putting out a dedicated FPGA dev board for $100-200 would let a lot of people play with it, students and SW developers. On Tue, Jun 22, 2021, 9:00 AM wrote: > Send Make-wifi-fast mailing list submissions to > make-wifi-fast@lists.bufferbloat.net > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.bufferbloat.net/listinfo/make-wifi-fast > or, via email, send a message with subject or body 'help' to > make-wifi-fast-request@lists.bufferbloat.net > > You can reach the person managing the list at > make-wifi-fast-owner@lists.bufferbloat.net > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Make-wifi-fast digest..." > Today's Topics: > > 1. the myth of spectrum scarcity (Dave Taht) > 2. openwifi presentation (Dave Taht) > 3. Re: [PATCH mac80211-next v8] mac80211: Switch to a virtual > time-based airtime scheduler (Toke H=C3=B8iland-J=C3=B8rgensen) > > > > ---------- Forwarded message ---------- > From: Dave Taht > To: Make-Wifi-fast > Cc: > Bcc: > Date: Mon, 21 Jun 2021 09:55:24 -0700 > Subject: [Make-wifi-fast] the myth of spectrum scarcity > this was really good: > > https://ecfsapi.fcc.gov/file/7020396128.pdf > > > -- > Latest Podcast: > https://www.linkedin.com/feed/update/urn:li:activity:6791014284936785920/ > > Dave T=C3=A4ht CTO, TekLibre, LLC > > > > > ---------- Forwarded message ---------- > From: Dave Taht > To: Make-Wifi-fast > Cc: > Bcc: > Date: Mon, 21 Jun 2021 10:07:23 -0700 > Subject: [Make-wifi-fast] openwifi presentation > > https://media.libreplanet.org/u/libreplanet/m/openwifi-project-the-dawn-o= f-the-free-libre-wifi-chip/ > > -- > Latest Podcast: > https://www.linkedin.com/feed/update/urn:li:activity:6791014284936785920/ > > Dave T=C3=A4ht CTO, TekLibre, LLC > > > > > ---------- Forwarded message ---------- > From: "Toke H=C3=B8iland-J=C3=B8rgensen" > To: Felix Fietkau > Cc: make-wifi-fast@lists.bufferbloat.net, Rajkumar Manoharan < > rmanohar@codeaurora.org>, Kan Yan , Yibo Zhao < > yiboz@codeaurora.org>, linux-wireless@vger.kernel.org > Bcc: > Date: Tue, 22 Jun 2021 16:25:03 +0200 > Subject: Re: [Make-wifi-fast] [PATCH mac80211-next v8] mac80211: Switch t= o > a virtual time-based airtime scheduler > Toke H=C3=B8iland-J=C3=B8rgensen writes: > > > This switches the airtime scheduler in mac80211 to use a virtual > time-based > > scheduler instead of the round-robin scheduler used before. This has a > > couple of advantages: > > > > - No need to sync up the round-robin scheduler in firmware/hardware wit= h > > the round-robin airtime scheduler. > > > > - If several stations are eligible for transmission we can schedule bot= h > of > > them; no need to hard-block the scheduling rotation until the head of > the > > queue has used up its quantum. > > > > - The check of whether a station is eligible for transmission becomes > > simpler (in ieee80211_txq_may_transmit()). > > > > The drawback is that scheduling becomes slightly more expensive, as we > need > > to maintain an rbtree of TXQs sorted by virtual time. This means that > > ieee80211_register_airtime() becomes O(logN) in the number of currently > > scheduled TXQs because it can change the order of the scheduled station= s. > > We mitigate this overhead by only resorting when a station changes > position > > in the tree, and hopefully N rarely grows too big (it's only TXQs > currently > > backlogged, not all associated stations), so it shouldn't be too big of > an > > issue. > > > > To prevent divisions in the fast path, we maintain both station sums an= d > > pre-computed reciprocals of the sums. This turns the fast-path operatio= n > > into a multiplication, with divisions only happening as the number of > > active stations change (to re-compute the current sum of all active > station > > weights). To prevent this re-computation of the reciprocal from happeni= ng > > too frequently, we use a time-based notion of station activity, instead > of > > updating the weight every time a station gets scheduled or de-scheduled= . > As > > queues can oscillate between empty and occupied quite frequently, this > can > > significantly cut down on the number of re-computations. It also has th= e > > added benefit of making the station airtime calculation independent on > > whether the queue happened to have drained at the time an airtime value > was > > accounted. > > > > Co-developed-by: Yibo Zhao > > Signed-off-by: Yibo Zhao > > Signed-off-by: Toke H=C3=B8iland-J=C3=B8rgensen > > Hey Felix > > Had a chance to look at this yet? :) > > -Toke > > > _______________________________________________ > Make-wifi-fast mailing list > Make-wifi-fast@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/make-wifi-fast --00000000000064c7ff05c55ec5cb Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Interesting stuff...=C2=A0 =C2=A0One thing that occurred = to me,=C2=A0 that they didnt mention...=C2=A0 doing an intermediate step,= =C2=A0 i.e.=C2=A0 make a stripped down FPGA dev board,=C2=A0 vs the Xylinx = or whoever's standard dev board.=C2=A0 Even with a radio chipset and pa= rts also on it, could be done much cheaper, and still be a dev platform.
I also understand that if you co= uld raise=C2=A0the $$$$, and push=C2=A0an ASIC out,=C2=A0 theres a much greater chance of large scale ac= ceptance.=C2=A0 But, it is a lot of $$$$, and for a single version.

Putting out a dedicated FPGA de= v board for $100-200 would let a lot of people play with it,=C2=A0 students= and SW developers.

On Tue, Jun 22, 2021, 9:00 AM <make-wifi-fast-request@lis= ts.bufferbloat.net> wrote:
S= end Make-wifi-fast mailing list submissions to
=C2=A0 =C2=A0 =C2=A0 =C2=A0 make-wifi-fast@lists.bufferbl= oat.net

To subscribe or unsubscribe via the World Wide Web, visit
=C2=A0 =C2=A0 =C2=A0 =C2=A0 https://= lists.bufferbloat.net/listinfo/make-wifi-fast
or, via email, send a message with subject or body 'help' to
=C2=A0 =C2=A0 =C2=A0 =C2=A0 make-wifi-fast-reques= t@lists.bufferbloat.net

You can reach the person managing the list at
=C2=A0 =C2=A0 =C2=A0 =C2=A0 make-wifi-fast-owner@li= sts.bufferbloat.net

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Make-wifi-fast digest..."
Today's Topics:

=C2=A0 =C2=A01. the myth of spectrum scarcity (Dave Taht)
=C2=A0 =C2=A02. openwifi presentation (Dave Taht)
=C2=A0 =C2=A03. Re: [PATCH mac80211-next v8] mac80211: Switch to a virtual<= br> =C2=A0 =C2=A0 =C2=A0 time-based airtime scheduler (Toke H=C3=B8iland-J=C3= =B8rgensen)



---------- Forwarded message ----------
From:=C2=A0Dave Taht= <dave.taht@gmail.com>
To:=C2=A0Make-Wifi-fast <make-wifi-fast@lists.bufferbloat.net>
Cc:=C2=A0
Bcc:= =C2=A0
Date:=C2=A0Mon, 21 Jun 2021 09:55:24 -0700
Subject:=C2=A0[Make= -wifi-fast] the myth of spectrum scarcity
this was really good:

https://ecfsapi.fcc.gov/file/7020396128.pdf


--
Latest Podcast:
https://www.link= edin.com/feed/update/urn:li:activity:6791014284936785920/

Dave T=C3=A4ht CTO, TekLibre, LLC




---------- Forwarded message ----------
From:=C2=A0Dave Taht= <dave.taht@gmail.com>
To:=C2=A0Make-Wifi-fast <make-wifi-fast@lists.bufferbloat.net>
Cc:=C2=A0
Bcc:= =C2=A0
Date:=C2=A0Mon, 21 Jun 2021 10:07:23 -0700
Subject:=C2=A0[Make= -wifi-fast] openwifi presentation
https://media.libreplanet.o= rg/u/libreplanet/m/openwifi-project-the-dawn-of-the-free-libre-wifi-chip/

--
Latest Podcast:
https://www.link= edin.com/feed/update/urn:li:activity:6791014284936785920/

Dave T=C3=A4ht CTO, TekLibre, LLC




---------- Forwarded message ----------
From:=C2=A0"Tok= e H=C3=B8iland-J=C3=B8rgensen" <toke@redhat.com>
To:=C2=A0Fe= lix Fietkau <nbd@nbd.name>
Cc:=C2=A0make-wifi-fast= @lists.bufferbloat.net, Rajkumar Manoharan <rmanohar@codeaurora= .org>, Kan Yan <kyan@google.com>, Yibo Zhao <yiboz@codea= urora.org>, linux-wireless@vger.kernel.org
Bcc:= =C2=A0
Date:=C2=A0Tue, 22 Jun 2021 16:25:03 +0200
Subject:=C2=A0Re: [= Make-wifi-fast] [PATCH mac80211-next v8] mac80211: Switch to a virtual time= -based airtime scheduler
Toke H=C3=B8iland-J=C3=B8rgensen <toke@redhat.c= om> writes:

> This switches the airtime scheduler in mac80211 to use a virtual time-= based
> scheduler instead of the round-robin scheduler used before. This has a=
> couple of advantages:
>
> - No need to sync up the round-robin scheduler in firmware/hardware wi= th
>=C2=A0 =C2=A0the round-robin airtime scheduler.
>
> - If several stations are eligible for transmission we can schedule bo= th of
>=C2=A0 =C2=A0them; no need to hard-block the scheduling rotation until = the head of the
>=C2=A0 =C2=A0queue has used up its quantum.
>
> - The check of whether a station is eligible for transmission becomes<= br> >=C2=A0 =C2=A0simpler (in ieee80211_txq_may_transmit()).
>
> The drawback is that scheduling becomes slightly more expensive, as we= need
> to maintain an rbtree of TXQs sorted by virtual time. This means that<= br> > ieee80211_register_airtime() becomes O(logN) in the number of currentl= y
> scheduled TXQs because it can change the order of the scheduled statio= ns.
> We mitigate this overhead by only resorting when a station changes pos= ition
> in the tree, and hopefully N rarely grows too big (it's only TXQs = currently
> backlogged, not all associated stations), so it shouldn't be too b= ig of an
> issue.
>
> To prevent divisions in the fast path, we maintain both station sums a= nd
> pre-computed reciprocals of the sums. This turns the fast-path operati= on
> into a multiplication, with divisions only happening as the number of<= br> > active stations change (to re-compute the current sum of all active st= ation
> weights). To prevent this re-computation of the reciprocal from happen= ing
> too frequently, we use a time-based notion of station activity, instea= d of
> updating the weight every time a station gets scheduled or de-schedule= d. As
> queues can oscillate between empty and occupied quite frequently, this= can
> significantly cut down on the number of re-computations. It also has t= he
> added benefit of making the station airtime calculation independent on=
> whether the queue happened to have drained at the time an airtime valu= e was
> accounted.
>
> Co-developed-by: Yibo Zhao <yiboz@codeaurora.org>
> Signed-off-by: Yibo Zhao <yiboz@codeaurora.org>
> Signed-off-by: Toke H=C3=B8iland-J=C3=B8rgensen <toke@redhat.com&g= t;

Hey Felix

Had a chance to look at this yet? :)

-Toke


_______________________________________________
Make-wifi-fast mailing list
Make-wifi-fast@lists.bufferbloat.net
https://lists.bufferbloat.net/listin= fo/make-wifi-fast
--00000000000064c7ff05c55ec5cb--