Re: [Cfrg] Benchmarks: 384 vs 389 vs Goldilocks vs ... on Haswell

Tony Arcieri <bascule@gmail.com> Sat, 03 January 2015 23:33 UTC

Return-Path: <bascule@gmail.com>
X-Original-To: cfrg@ietfa.amsl.com
Delivered-To: cfrg@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 76A731A19FE for <cfrg@ietfa.amsl.com>; Sat, 3 Jan 2015 15:33:35 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.999
X-Spam-Level:
X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id k_dlnET5oG1c for <cfrg@ietfa.amsl.com>; Sat, 3 Jan 2015 15:33:33 -0800 (PST)
Received: from mail-oi0-x22e.google.com (mail-oi0-x22e.google.com [IPv6:2607:f8b0:4003:c06::22e]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D2D261A19F8 for <cfrg@irtf.org>; Sat, 3 Jan 2015 15:33:32 -0800 (PST)
Received: by mail-oi0-f46.google.com with SMTP id a3so13210088oib.5 for <cfrg@irtf.org>; Sat, 03 Jan 2015 15:33:32 -0800 (PST)
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:content-type; bh=lyf1FSZwJr/VBgTPLM9AbMCmpuPOFM5gJJmEin6SzsY=; b=0XQYnB08Sn7yrtGDHztTEh3zEgFOikAJMK2lgaAsGBXOwO/u11IFzifIB4AMBOVxxr baNe4yzIgudH4cqEhMA0LnoatEjlANf/rEJT6BkwhW79faXz1YgBOUYYpUqne/UD2K17 deEBevddqNnhoUXmtD7ieFSaZkNU1vrYluJI9ZOw2E7xoozNs6FG4p5l+l8NHXqkdPzm uwzT7KM4qm/xGNFxQwnrRvye3I2l9Ze+oZnxGJyVi1QzyPNOwSUMN/5IgjgQ+1iCGHbP 46rcp3bKBSuo+py93zalPWJDlesd33ONInnA3YQYdpV5fwqGi/do/Svp+RgzRltyIKEG BXQA==
X-Received: by 10.60.133.4 with SMTP id oy4mr48990813oeb.28.1420328012049; Sat, 03 Jan 2015 15:33:32 -0800 (PST)
MIME-Version: 1.0
Received: by 10.60.227.225 with HTTP; Sat, 3 Jan 2015 15:33:11 -0800 (PST)
In-Reply-To: <CAFewVt53366Axg-WZVGtVsTvipeHrCjtNBGcdmgR58CZwz47rg@mail.gmail.com>
References: <54A1E049.9000404@shiftleft.org> <D0CA0568.3B27A%kenny.paterson@rhul.ac.uk> <CAFewVt53366Axg-WZVGtVsTvipeHrCjtNBGcdmgR58CZwz47rg@mail.gmail.com>
From: Tony Arcieri <bascule@gmail.com>
Date: Sat, 3 Jan 2015 15:33:11 -0800
Message-ID: <CAHOTMVKS-yUcAxyhC3heeEbVfUEaN0OGoTSBfkYKSGRYW-Gz6A@mail.gmail.com>
To: Brian Smith <brian@briansmith.org>
Content-Type: multipart/alternative; boundary=047d7b471fe29f74f8050bc7e212
Archived-At: http://mailarchive.ietf.org/arch/msg/cfrg/kv93DlHuOYwDYL8sx8jMFKocuaY
Cc: "cfrg@irtf.org" <cfrg@irtf.org>
Subject: Re: [Cfrg] Benchmarks: 384 vs 389 vs Goldilocks vs ... on Haswell
X-BeenThere: cfrg@irtf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: Crypto Forum Research Group <cfrg.irtf.org>
List-Unsubscribe: <http://www.irtf.org/mailman/options/cfrg>, <mailto:cfrg-request@irtf.org?subject=unsubscribe>
List-Archive: <http://www.irtf.org/mail-archive/web/cfrg/>
List-Post: <mailto:cfrg@irtf.org>
List-Help: <mailto:cfrg-request@irtf.org?subject=help>
List-Subscribe: <http://www.irtf.org/mailman/listinfo/cfrg>, <mailto:cfrg-request@irtf.org?subject=subscribe>
X-List-Received-Date: Sat, 03 Jan 2015 23:33:35 -0000

On Sat, Jan 3, 2015 at 11:54 AM, Brian Smith <brian@briansmith.org>; wrote:

> FWIW, I don't think it's fair to assume that P384-mers 8-9%
> performance disadvantage can be optimized away, especially when the
> thing that's 9% faster is also slightly stronger


If future optimizations are a concern, given the P389 implementation is
completely new, I think it should probably be given the benefit of the
doubt as to how much it can be optimized. Especially given remarks like
this:

On Mon, Dec 29, 2014 at 3:14 PM, Mike Hamburg <mike@shiftleft.org>; wrote:

> The P389 code is probably less tuned than its competitors. The MS code is
> written in pure asm, but it doesn't take advantage of BMI2.  The Goldilocks
> and P389 code are C with an asm multiply/accumulate intrinsic, but I spent
> more time tuning the Goldilocks code.


-- 
Tony Arcieri