From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [RFC 6/6] fib_trie: combine leaf and info Date: Tue, 15 Jan 2008 22:16:23 +0100 Message-ID: <478D22A7.6080506@cosmosbay.com> References: <20080112064646.282104074@linux-foundation.org> <20080112.205520.55747078.davem@davemloft.net> <4789A29C.6080000@linux-foundation.org> <20080112.214417.154179770.davem@davemloft.net> <20080114125755.6157a3bf@deepthought> <20080114164450.55f8c9b2@deepthought> <20080114164621.2bc5011f@deepthought> <20080114164727.210047f6@deepthought> <20080114185843.0981f0ff@deepthought> <20080114210716.4b09c84d@deepthought> <478C4EB7.6060807@cosmosbay.com> <478C4FBE.5040308@cosmosbay.com> <20080115081905.3b8b6f05@deepthought> <18316.58111.211387.271534@robur.slu.se> <20080115182544.98c18d08.dada1@cosmosbay.com> <20080115094753.32e35823@deepthought> <18317.5368.952926.46132@robur.slu.se> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Stephen Hemminger , David Miller , robert.olsson@its.uu.se, netdev@vger.kernel.org To: Robert Olsson Return-path: Received: from gw1.cosmosbay.com ([86.65.150.130]:44100 "EHLO gw1.cosmosbay.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750955AbYAOVQg (ORCPT ); Tue, 15 Jan 2008 16:16:36 -0500 In-Reply-To: <18317.5368.952926.46132@robur.slu.se> Sender: netdev-owner@vger.kernel.org List-ID: Robert Olsson a =E9crit : >=20 > Stephen Hemminger writes: > > This is how I did it: >=20 > Yes looks like an elegant solution. Did you even test it? > Maybe we see some effects in just dumping a full table? >=20 > Anyway lookup should be tested in some way. We can a lot > of analyzing before getting to right entry, local_table > backtracking, main lookup w. ev. backtracking etc. So > hopefully we get paid for this work. >=20 > Also it might be idea to do some analysis of the fib_aliases > list. Maybe the trick can be done again? ;) >=20 Back in 2.6.9 times, sizeof(bi_alias) was 16 bytes on i386 Nowadays, 64/128 bytes are the norm :( SLAB_HWCACHE_ALIGN is not our friend.