From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932368AbXB1TJl (ORCPT ); Wed, 28 Feb 2007 14:09:41 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932374AbXB1TJl (ORCPT ); Wed, 28 Feb 2007 14:09:41 -0500 Received: from pfx2.jmh.fr ([194.153.89.55]:45517 "EHLO pfx2.jmh.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932368AbXB1TJk (ORCPT ); Wed, 28 Feb 2007 14:09:40 -0500 From: Eric Dumazet To: Davide Libenzi Subject: Re: [patch - v3] epoll ready set loops diet ... Date: Wed, 28 Feb 2007 20:09:29 +0100 User-Agent: KMail/1.9.5 Cc: Linux Kernel Mailing List , Andrew Morton , Linus Torvalds References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200702282009.29527.dada1@cosmosbay.com> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wednesday 28 February 2007 19:37, Davide Libenzi wrote: > + list_del(&epi->rdllink); > + if (!(epi->event.events & EPOLLET) && (revents & epi->event.events)) > + list_add_tail(&epi->rdllink, &injlist); > + else { Is the ( ... & epi->event.events) really necessary ? (It seems already done) I was wrong about the size of epitem : it is now 68 bytes instead of 72. At least we now use/dirty one cache line instead of two per epitem. Do you have another brilliant idea to shrink 4 more bytes ? :) It seems to me 'nwait' is only used at init time (so that ep_ptable_queue_proc() can signal an error occured). Maybe another mechanism could let us delete nwait from epitem ? We could use a field in task_struct for example (see usage of total_link_count for example) Thank you