From owner-freebsd-net@FreeBSD.ORG Tue Dec 31 17:53:26 2013 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id D6449292 for ; Tue, 31 Dec 2013 17:53:26 +0000 (UTC) Received: from mail-qc0-x230.google.com (mail-qc0-x230.google.com [IPv6:2607:f8b0:400d:c01::230]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by mx1.freebsd.org (Postfix) with ESMTPS id 95B9213AB for ; Tue, 31 Dec 2013 17:53:26 +0000 (UTC) Received: by mail-qc0-f176.google.com with SMTP id i8so12123119qcq.7 for ; Tue, 31 Dec 2013 09:53:25 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:date:message-id:subject :from:to:content-type; bh=EW5T+EWu3en0CMwcHaCMHlRZvsGLM70RvbOxHDvvsYE=; b=yGKKbjVdE9GWqrpvmyi6opESJWIMJg8kSpQn8cQD44pJ4jO/jMf/C/Vg5lGp5N4cFk ImixALfEDUwUdGVpM3gbnTo20G2ELgiwQ2mT3y+EalW+yiYms3YcUJv8DlfiRzQi6NuQ cmJct7TtaX+BiCNpgjlriicRpArj3Cz/vPWqNWoDTV5VadO0sedSJpsSV6hKPbNhWQ+I t8n6qb9udSAb2tiR8hr966kuXLApxVHya3mtDJ7nFiE3p1276sMJTiwGepgaiobDtGcA Q2056jISXUFcyBwgD98F05fStZv+WjCV10WaN048DOmtIcMiaucYbJd4IiHHXRNW5mEf nfNQ== MIME-Version: 1.0 X-Received: by 10.224.14.132 with SMTP id g4mr14868094qaa.26.1388512405703; Tue, 31 Dec 2013 09:53:25 -0800 (PST) Sender: adrian.chadd@gmail.com Received: by 10.224.52.8 with HTTP; Tue, 31 Dec 2013 09:53:25 -0800 (PST) In-Reply-To: References: Date: Tue, 31 Dec 2013 09:53:25 -0800 X-Google-Sender-Auth: Tn-ta3GzaBj91UsxaO7yFoKtUFc Message-ID: Subject: Re: ipv6 lock contention with parallel socket io From: Adrian Chadd To: FreeBSD Net Content-Type: text/plain; charset=ISO-8859-1 X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 31 Dec 2013 17:53:26 -0000 On 30 December 2013 23:35, Adrian Chadd wrote: > Hi, > > I've noticed a bunch of lock contention occurs when doing highly > parallel (> 4096 sockets) TCP IPv6 traffic. > > The contention is here: > [snip] > > .. it's the IF_ADATA lock surrounding the lla_lookup() calls. > > Now: > > * is there any reason this isn't an rmlock? > * the instance early on in nd6_output_lle() isn't taking the read > lock, it's taking the full lock. Is there any reason for this? > > I don't have much experience or time to spend on optimising the ipv6 > code to scale better but this seems like one of those things that'll > bite us in the ass as the amount of ipv6 deployed increases. > > Does anyone have any ideas/suggestions on how we could improve things? This improves things quite a bit - from 1.9gbyte/sec @ 100% cpu usage to 2.7gbyte/sec @ 85% CPU usage. It's not perfect - the lock contention is still there - but it's much less of an issue now. Are there any issues with it? Index: sys/netinet6/nd6.c =================================================================== --- sys/netinet6/nd6.c (revision 259475) +++ sys/netinet6/nd6.c (working copy) @@ -1891,9 +1891,9 @@ flags = ((m != NULL) || (lle != NULL)) ? LLE_EXCLUSIVE : 0; if (ln == NULL) { retry: - IF_AFDATA_LOCK(ifp); + IF_AFDATA_RLOCK(ifp); ln = lla_lookup(LLTABLE6(ifp), flags, (struct sockaddr *)dst); - IF_AFDATA_UNLOCK(ifp); + IF_AFDATA_RUNLOCK(ifp); if ((ln == NULL) && nd6_is_addr_neighbor(dst, ifp)) { /* * Since nd6_is_addr_neighbor() internally calls nd6_lookup(), Thanks, -a