From owner-freebsd-arch@FreeBSD.ORG Wed Jul 26 18:27:03 2006 Return-Path: X-Original-To: freebsd-arch@freebsd.org Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E975E16A4E5 for ; Wed, 26 Jul 2006 18:27:03 +0000 (UTC) (envelope-from asmrookie@gmail.com) Received: from wx-out-0102.google.com (wx-out-0102.google.com [66.249.82.204]) by mx1.FreeBSD.org (Postfix) with ESMTP id D3D0043D55 for ; Wed, 26 Jul 2006 18:27:02 +0000 (GMT) (envelope-from asmrookie@gmail.com) Received: by wx-out-0102.google.com with SMTP id s18so1142348wxc for ; Wed, 26 Jul 2006 11:27:01 -0700 (PDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references:x-google-sender-auth; b=H1+zjsNuBWJvyuLliL1I+6ClPbNgCwlhptBibOB6d+5mELKeIZ094fsBM+bKmOQTtGGDfMqNg/1tpjSiVMDQIEBLS0cYHixB8qDXof5RkQVj8Rr9EN6sL0WcDjB30abQbVHvVuyZmPI3yYyxzUNH2qwnsrSYATtxbPd2zOw8ue0= Received: by 10.70.115.17 with SMTP id n17mr6990443wxc; Wed, 26 Jul 2006 11:27:01 -0700 (PDT) Received: by 10.70.11.18 with HTTP; Wed, 26 Jul 2006 11:27:01 -0700 (PDT) Message-ID: <3bbf2fe10607261127p3f01a6c3w80027754f7d4e594@mail.gmail.com> Date: Wed, 26 Jul 2006 20:27:01 +0200 From: "Attilio Rao" Sender: asmrookie@gmail.com To: "John Baldwin" In-Reply-To: <3bbf2fe10607251004wf94e238xb5ea7a31c973817f@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <3bbf2fe10607250813w8ff9e34pc505bf290e71758@mail.gmail.com> <3bbf2fe10607250814m1a476f09p2d962dedc0c99be1@mail.gmail.com> <200607251232.51230.jhb@freebsd.org> <3bbf2fe10607251004wf94e238xb5ea7a31c973817f@mail.gmail.com> X-Google-Sender-Auth: d3e34ec819ecb3ac Cc: freebsd-arch@freebsd.org Subject: Re: [PATCH] Mantaining turnstile aligned to 128 bytes in i386 CPUs X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Jul 2006 18:27:04 -0000 2006/7/25, Attilio Rao : > 2006/7/25, John Baldwin : > > On Tuesday 25 July 2006 11:14, Attilio Rao wrote: > > > 2006/7/25, Attilio Rao : > > > > Hi, > > > > Intel documentation points out that having a 128-bytes aligned > > > > syncronizing primitive (which fits in a cache line) will minimize the > > > > traffic for cache bus, so this patch implements an alignment for i386 > > > > on turnstiles. > > > > > > > > Any comments, feedbacks? > > > > > > Oh, sorry, I've unforgotten the diff. > > > > > > Attilio > > > > I think a better approach would be to stick turnstiles (and sleepqueues) in a > > UMA zone and specify cache-size alignment to the zone. However, turnstiles > > aren't really sychronization primitives in that you don't spin on a variable > > inside the structure, and I think it's the spinning and avoiding bouncing > > cache lines around that Intel's documentation is really about. In that case, > > the things you want aligned are things like mutexes, rwlocks, etc. > > Well, I think that this is referred in particular to the latter issue > you mentioned. > Spinning is not really concerned to cache bus issues (more, in > particular, datapath latency). > With this point of view, turnstiles (as sleepqueues) are passed around > CPUs more than a mutex/rwlock (or a cv), I guess, so I was thinking > that it's better optimizing turnstile than the real syncronizing > primitive itself. This is a patch which let turnstiles/sleepqueues using an UMA zone. I've tried in my 6.1R branch and it works quite fine, so this HEAD version might be alright (I've not tried yet, so please test): http://users.gufi.org/~rookie/works/patches/uma_sync.diff It, obviously, set default alignment for i386 at 128 bytes. Any comments, feedbacks, ideas, are welcome. Attilio PS: I know that I could simplify *_alloc(), *_free() routines implementing init/fini but it is simpler and more optimized having things like so. -- Peace can only be achieved by understanding - A. Einstein