From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-x232.google.com (mail-qk0-x232.google.com [IPv6:2607:f8b0:400d:c09::232]) (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 43E0D3B2AB for ; Sun, 30 Oct 2016 19:37:29 -0400 (EDT) Received: by mail-qk0-x232.google.com with SMTP id q130so25222328qke.1 for ; Sun, 30 Oct 2016 16:37:29 -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 :content-transfer-encoding; bh=oErIsgZXKHAUrfdm7IIJeKYcC2KuPI6HVEasJjJfWQ4=; b=fIPjUWp1t8LCm/6TqF5XD5MK8yeicVn0ofD4fc8uCKxns7+kjHmFgPS3WTQ3q95MG9 jxULE4uG9311FhGoX42Kdd/wbnnwE4AKuRWhN5SpqoqyOnpfYIJBEC9iYYWGMizo7bo+ q1X59Ls2Yp5EcLGA8sOBEDEfjNw5XM7WcTZcxxBXwExoB5GtDBfz1oxV42v6sy9joeJu YMQ4R+hv0ApN4Yj2T/cQMQTjA/joWR2Lcd6RhetuxvlEGzwnFh4qtrgZ3kij5NygvS7j U1WjrxK9wvwDwh5FoSKY6HRdjQPFBMbKwD33utYJUL1X14WfenDRSWO/ovbCZpcELJFi RR8Q== 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:content-transfer-encoding; bh=oErIsgZXKHAUrfdm7IIJeKYcC2KuPI6HVEasJjJfWQ4=; b=jB26g4ML4Tx0b6GgjVjjgLd/4Gk+RGJtgOcYUP7IOXvDbDt0gzQ6k5XR5fA4cWI82z aVxjPTySWxg6so7Y0IUwt2wRaJgYtTlgfufBMxnUIVVPo37wiJ0kj2Av6Jvcj/nIFbDG qnyrgXAMwlm83CZ5+Ra9v/SkAkT3qMI9d1X/2dPpTrZGmqBivcH532zR/kOzWn2IT4dS Bd+ZzDqvvxcERanGEP/Djf+h0d2ZPXtmShFJZHxRkKd0TxUQvYTBoMDXX2/DFpN7SOy/ Vnm5JX7Xf85/AQBnvjDXK/B0yysh4moX5Z3mvYQV6c+H/m8QqzPWtXJam2LYhayfZAmm F0rg== X-Gm-Message-State: ABUngvcexciVF2eHiykNPKeUlHmmCmBDLhjCJHiBUGUippuN1FBG6edqXWufoHNfd2/0fUQScoN7DU5Giv9gPg== X-Received: by 10.55.162.79 with SMTP id l76mr21055073qke.17.1477870648754; Sun, 30 Oct 2016 16:37:28 -0700 (PDT) MIME-Version: 1.0 Received: by 10.12.137.198 with HTTP; Sun, 30 Oct 2016 16:37:28 -0700 (PDT) In-Reply-To: References: <1477286912.4085.1.camel@sipsolutions.net> <1477379678.4390.2.camel@sipsolutions.net> <1477461362.4059.17.camel@sipsolutions.net> From: Dave Taht Date: Sun, 30 Oct 2016 16:37:28 -0700 Message-ID: To: =?UTF-8?Q?Bj=C3=B6rn_Smedman?= , make-wifi-fast@lists.bufferbloat.net Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Make-wifi-fast] Bayesian rate control 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: Sun, 30 Oct 2016 23:37:29 -0000 Dear Bjorn: I took a quick look at your patches in http://www.openias.org/bayesian-wifi-materials-and-methods sampling minstrel at such a high rate is going to hurt you, I think, on a lower end architecture, especially. We have a sampler that can run at about 20ms intervals (tops) on limited hardware, that can grab mcs rates, in flent, if that helps. To me, key to fixing rate control is first the other stuff for queue management, then doing "smart things" fast when a station pops up to grab a web page and then goes silent. Right now I've seen minstrel take a minute or more to find the right rate, an ideal one would be closer to 100ms. On Sat, Oct 29, 2016 at 1:50 PM, Bj=C3=B6rn Smedman wro= te: > Thomas, Dave, Adrian, Johannes, > > Thanks for comments and encouragement. I bought the TP-Link TL-WA901ND > access point and TP-Link TL-WDN4800 PCIe card. Had no problem getting > them talking to each other with ath9k, and the rate table contains 52 > entries, so plenty to start out with. > > I've written a follow-up post about it if anybody's interested: > http://www.openias.org/bayesian-wifi-materials-and-methods > > Cheers, > > Bj=C3=B6rn > > On Wed, Oct 26, 2016 at 7:56 AM, Johannes Berg > wrote: >> >>> The intel 7260 and later parts also allow user controllable rate >>> control and provide transmit completion feedback, but I don't know >>> whether it's enough for your needs. >> >> Perhaps. However, existing rate control is *very* tightly coupled to >> the driver, and it'd be fairly pointless to disentangle just for the >> sake of playing with a rate control algorithm. >> >> Also, the device doesn't support per-frame control nor any kind of >> sampling-with-table-fallback, only the rate table that you give to the >> device and update. >> >> Btw, mac80211_hwsim with wmediumd doing some medium simulation might >> also be something to look at for just extending to VHT. >> >> And come to think of it, there's this new driver Felix et al have been >> working on, mt7601u, which also should support proper rate control >> APIs. >> >> johannes --=20 Dave T=C3=A4ht Let's go make home routers and wifi faster! With better software! http://blog.cerowrt.org