From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-x236.google.com (mail-wm0-x236.google.com [IPv6:2a00:1450:400c:c09::236]) (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 6982D3B29E for ; Wed, 22 Nov 2017 16:19:10 -0500 (EST) Received: by mail-wm0-x236.google.com with SMTP id b189so12846318wmd.5 for ; Wed, 22 Nov 2017 13:19:10 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:subject:from:in-reply-to:date:cc:message-id:references :to; bh=vwRCMWImB3981Id7YdHK9Ib9t6Tygd3/3jMEQV+cxYc=; b=Q16c3fkhTXwlvkZ53KwAiVPDBjEqMs0kPSf35RQ0P7fNSdclIiWio1g2HSjELxKxoD GtKIzWYnA8CitGTQWJrsPFCeZEJT9lG8XOivlE7cuNWS2Z3c4cj+PbWsgvMPh2RgKWu8 ZJZZiOB6e4+hlm0iwpAqcboaXhZXV45V2zPnwLCVyxy7ZQ5jqaSaF9WJ++6WMJj6wFtq Cp/LgRAIVR4Kt5cbbPhC09oVZYvM+z45uOXwEfpcIYf1vbMHv/PMzs/HOhlkNiwQTJ03 XjRHYmYsq5rNN0dUjHbadHc6/diAes5Uo2Np4mDRebVk9aBUbNzyC7FJGStLORbW9T63 SDcw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:subject:from:in-reply-to:date:cc :message-id:references:to; bh=vwRCMWImB3981Id7YdHK9Ib9t6Tygd3/3jMEQV+cxYc=; b=r5LbATh/Ve1xhxLPQIQqYOCx5JvBTvwdL/tR0IWQ8C3jda8ieukylWqDvd9BbpnQaK q/Ax8t31OE+j4TcxtgjT0rz+WHq2QM6ZQBkvvL8Za2bZo+kbtEfnDUH0N46Womqvqjv1 VHQ7cY5m6nHZoN1yqEBmIjKpsSGWrPGLsU9oUfRMCsFsisD/bt8L3vlH8hp+leXSR2AN EslUlwo5WvIN3roe9PpUA50AGhIvqJFkE125hFsKqjw6jideX4+V/TINWbo12ZCbUQiS m/Pu+QZPXbmnamJNU0TJBATuVPrLvppURzHldTKt9M/wuupAbwHR7hRUQmiXUhCljpeN mu7Q== X-Gm-Message-State: AJaThX4x7dm+A8YfdGeReaSJvHmEe3drE8G45S0pAgC99J5ljwix1zUl PRn1WZQINSO+n0WWOQmytfc= X-Google-Smtp-Source: AGs4zMYOzlWtx4AYGKy2djOTX49UpfFhHuEUcXspKMuNoGKAYDDcddeqB+14ilv08XmJHj0ZKG7KvA== X-Received: by 10.28.157.7 with SMTP id g7mr4713399wme.99.1511385549310; Wed, 22 Nov 2017 13:19:09 -0800 (PST) Received: from [10.72.0.130] (h-1169.lbcfree.net. [185.99.119.68]) by smtp.gmail.com with ESMTPSA id h2sm13021907wrf.47.2017.11.22.13.19.07 (version=TLS1 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Wed, 22 Nov 2017 13:19:08 -0800 (PST) Content-Type: multipart/alternative; boundary="Apple-Mail=_26597BCD-54C9-42B1-9BF8-03F126040E65" Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\)) From: Pete Heist In-Reply-To: <9550F1F5-A808-41DB-9523-D5AD565F1474@gmail.com> Date: Wed, 22 Nov 2017 22:19:07 +0100 Cc: Cake List Message-Id: <57FE2B16-D60C-4427-993F-AEE44E73021F@gmail.com> References: <107CB879-B196-40F1-A7C2-08B963B7A835@gmail.com> <9550F1F5-A808-41DB-9523-D5AD565F1474@gmail.com> To: Dave Taht X-Mailer: Apple Mail (2.3124) Subject: Re: [Cake] small cake_hash optimization? X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Cake - FQ_codel the next generation List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 22 Nov 2017 21:19:10 -0000 --Apple-Mail=_26597BCD-54C9-42B1-9BF8-03F126040E65 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=utf-8 > On Nov 22, 2017, at 7:49 PM, Pete Heist wrote: >=20 >> On Nov 22, 2017, at 7:38 PM, Dave Taht wrote: >>=20 >> It is somewhat unfair to not include the pfifo bandwidth on the test >> (a cpu cost/byte might be a better metric), also pfifo_fast has three >> tiers of classification in it. >=20 > Yeah, it=E2=80=99s probably better to not try to subtract the = pfifo_fast system time out in the way that I did. I should probably just = compare cake with and without the change, using a more accurate tool. >=20 > I don=E2=80=99t see how the change could hurt, but I also now am not = sure it helps much either. I guess it=E2=80=99s just two divs per call = to cake_hash, which is obviously going to happen more at GigE. I didn=E2=80=99t figure out =E2=80=98perf=E2=80=99 for this, but I did = instrument cake_hash in a simple way with calls to local_clock_ns using = =E2=80=98stap'. Results on stap tab: = https://docs.google.com/spreadsheets/d/1LKoq5NaswuHm9H1atXoZA1AhNDg6L4UYS3= Pn5lCsb1I/edit#gid=3D1493356365 = It=E2=80=99s a head scratcher, but I saw about a 3% mean time reduction = in cake_hash for the =E2=80=9Coptimized=E2=80=9D version when limited at = 950mbit, and a very slight slowdown when unlimited. = =E2=80=9CConfounding=E2=80=9D...(by Estee Lauder). Whether or not those results are either correct or statistically = significant, it doesn=E2=80=99t look like it=E2=80=99s worth too much = more effort, and I can leave it to you whether you want this change or = not. I don=E2=80=99t see the harm in it, and neither do I see much of a = benefit.= --Apple-Mail=_26597BCD-54C9-42B1-9BF8-03F126040E65 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=utf-8
On Nov 22, 2017, at 7:49 PM, Pete Heist <peteheist@gmail.com>= wrote:

On Nov 22, 2017, at 7:38 PM, Dave Taht <dave.taht@gmail.com>= wrote:

It is somewhat unfair to not = include the pfifo bandwidth on the test
(a cpu cost/byte = might be a better metric), also pfifo_fast has three
tiers = of classification in it.

Yeah, = it=E2=80=99s probably better to not try to subtract the pfifo_fast = system time out in the way that I did. I should probably just compare = cake with and without the change, using a more accurate tool.

I don=E2=80=99t see how the change could hurt, = but I also now am not sure it helps much either. I guess it=E2=80=99s = just two divs per call to cake_hash, which is obviously going to happen = more at GigE.

I didn=E2=80=99t figure = out =E2=80=98perf=E2=80=99 for this, but I did instrument cake_hash in a = simple way with calls to local_clock_ns using =E2=80=98stap'. Results on = stap tab:

https://docs.google.com/spreadsheets/d/1LKoq5NaswuHm9H1atXoZA1A= hNDg6L4UYS3Pn5lCsb1I/edit#gid=3D1493356365

It=E2=80=99s a head scratcher, but I = saw about a 3% mean time reduction in cake_hash for the =E2=80=9Coptimized= =E2=80=9D version when limited at 950mbit, and a very slight slowdown = when unlimited. =E2=80=9CConfounding=E2=80=9D...(by Estee = Lauder).

Whether= or not those results are either correct or statistically significant, = it doesn=E2=80=99t look like it=E2=80=99s worth too much more effort, = and I can leave it to you whether you want this change or not. I don=E2=80= =99t see the harm in it, and neither do I see much of a = benefit.
= --Apple-Mail=_26597BCD-54C9-42B1-9BF8-03F126040E65--