From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [RFC] fib_trie: memory waste solutions Date: Thu, 10 Apr 2008 02:57:17 -0700 (PDT) Message-ID: <20080410.025717.04755660.davem@davemloft.net> References: <20080404090257.2ec38b0c@extreme> <47FA4FE5.5010500@cosmosbay.com> <20080407154800.6912ad1d@speedy> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: dada1@cosmosbay.com, shemminger@vyatta.com, Robert.Olsson@data.slu.se, netdev@vger.kernel.org To: stephen.hemminger@vyatta.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:53675 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1752198AbYDJJ5R (ORCPT ); Thu, 10 Apr 2008 05:57:17 -0400 In-Reply-To: <20080407154800.6912ad1d@speedy> Sender: netdev-owner@vger.kernel.org List-ID: From: Stephen Hemminger Date: Mon, 7 Apr 2008 15:48:00 -0700 > On Mon, 07 Apr 2008 18:46:29 +0200 > Eric Dumazet wrote: > > > I prefer my patch Stephen, as your version enlarges every tnode with an > > embedded "struct work_struct" which can be larger than a "struct rcu_head" > > The number of tnode's is small and the size growth of 2*(unsigned long) is not > worth worrying about. Also theoretically, my version could have multiple > work elements processed at once. I pretty much agree with Stephen and thus applied his patch to net-2.6.26. We can tweak this further if new data supports that.