Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1946096AbWKJJql (ORCPT ); Fri, 10 Nov 2006 04:46:41 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1946221AbWKJJql (ORCPT ); Fri, 10 Nov 2006 04:46:41 -0500 Received: from pfx2.jmh.fr ([194.153.89.55]:22920 "EHLO pfx2.jmh.fr") by vger.kernel.org with ESMTP id S1946096AbWKJJqk (ORCPT ); Fri, 10 Nov 2006 04:46:40 -0500 From: Eric Dumazet To: Andi Kleen Subject: Re: [PATCH] x86_64: Make the NUMA hash function nodemap allocation dynamic and remove NODEMAPSIZE Date: Fri, 10 Nov 2006 10:46:49 +0100 User-Agent: KMail/1.9.5 Cc: Amul Shah , LKML References: <1163029076.3553.36.camel@ustr-linux-shaha1.unisys.com> <200611100748.30889.ak@suse.de> <200611101043.14749.dada1@cosmosbay.com> In-Reply-To: <200611101043.14749.dada1@cosmosbay.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-15" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200611101046.49806.dada1@cosmosbay.com> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 816 Lines: 32 On Friday 10 November 2006 10:43, Eric Dumazet wrote: > > Therefore I suggest to use a structure like that : > > struct memnode { > int shift; > unsigned int mapsize; /* no need to use 8 bytes here */ > u8 *map; > u8 embedded_map[64-8]; /* total size = 64 bytes */ > } ____cacheline_aligned; > Arg... [64 - 16] sorry > and make memnode.map point to memnode.embedded_map if mapsize <= 56 ? mapsize <= 48 > > This way, most AMD64 dual/quad processors wont waste a full PAGE to store > few bytes in it, and should use only one cache line. > > Thank you > > Eric > - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/