From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] NET : change layout of ehash table Date: Fri, 09 Feb 2007 00:40:19 -0800 (PST) Message-ID: <20070209.004019.75431263.davem@davemloft.net> References: <200702071159.34935.dada1@cosmosbay.com> <20070208.145632.74749802.davem@davemloft.net> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: dada1@cosmosbay.com, linux@horizon.com, akepner@sgi.com, netdev@vger.kernel.org To: ak@suse.de Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:50858 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1946201AbXBIIkV (ORCPT ); Fri, 9 Feb 2007 03:40:21 -0500 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org From: Andi Kleen Date: 09 Feb 2007 10:18:03 +0100 > David Miller writes: > > > > I've applied this, but I _REALLY_ don't like the new multiply > > instructions that are used now in the hash indexing paths when > > CONFIG_SMP is set. > > > > I think that's a higher cost than the memory waste. > > You're serious? multiply on a modern CPU is _much_ cheaper than a cache miss > e.g. a K8 can do a arbitary 64bit multiplication in 3-7 cycles. > Any cache miss will be in the three to four digits at least. I'm not thinking of modern CPUs, I'm think of the little guys :-)