From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933338AbXCBLJu (ORCPT ); Fri, 2 Mar 2007 06:09:50 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S933339AbXCBLJu (ORCPT ); Fri, 2 Mar 2007 06:09:50 -0500 Received: from relay.2ka.mipt.ru ([194.85.82.65]:37280 "EHLO 2ka.mipt.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1423027AbXCBLJs (ORCPT ); Fri, 2 Mar 2007 06:09:48 -0500 Date: Fri, 2 Mar 2007 14:08:21 +0300 From: Evgeniy Polyakov To: Ingo Molnar Cc: Pavel Machek , Theodore Tso , Linus Torvalds , Ulrich Drepper , linux-kernel@vger.kernel.org, Arjan van de Ven , Christoph Hellwig , Andrew Morton , Alan Cox , Zach Brown , "David S. Miller" , Suparna Bhattacharya , Davide Libenzi , Jens Axboe , Thomas Gleixner Subject: Re: [patch 00/13] Syslets, "Threadlets", generic AIO support, v3 Message-ID: <20070302110820.GA25908@2ka.mipt.ru> References: <20070227121116.GA31597@2ka.mipt.ru> <20070228161413.GA4319@ucw.cz> <20070301081808.GD7217@2ka.mipt.ru> <20070301092634.GB20171@elf.ucw.cz> <20070301094723.GJ7217@2ka.mipt.ru> <20070301101102.GE20171@elf.ucw.cz> <20070301111815.GB15709@2ka.mipt.ru> <20070302102714.GB27687@elf.ucw.cz> <20070302103713.GB28444@2ka.mipt.ru> <20070302105618.GA18377@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Disposition: inline In-Reply-To: <20070302105618.GA18377@elte.hu> User-Agent: Mutt/1.5.9i X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-3.0 (2ka.mipt.ru [0.0.0.0]); Fri, 02 Mar 2007 14:08:55 +0300 (MSK) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Mar 02, 2007 at 11:56:18AM +0100, Ingo Molnar (mingo@elte.hu) wrote: > > * Evgeniy Polyakov wrote: > > > Even if kevent has the same speed, it still allows to handle _any_ > > kind of events without any major surgery - a very tiny structure of > > lock and list head and you can process your own kernel event in > > userspace with timers, signals, io events, private userspace events > > and others without races and invention of differnet hacks for > > different types - _this_ is main point. > > did it ever occur to you to ... extend epoll? To speed it up? To add a > new wait syscall to it? Instead of introducing a whole new parallel > framework? Yes, I thought about its extension more than a year ago before started kevent, but epoll() is absolutely based on file structure and its file_operations with poll methodt, so it is quite impossible to work with sockets to implement network AIO. Eventually it had gathered a lot of other systems - do we really want to have per process signalfs, timerfs and so on - each simple structure must be bound to a file, which becomes too cost. > Ingo -- Evgeniy Polyakov