From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751350AbWHWAcF (ORCPT ); Tue, 22 Aug 2006 20:32:05 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751330AbWHWAcF (ORCPT ); Tue, 22 Aug 2006 20:32:05 -0400 Received: from dsl027-180-168.sfo1.dsl.speakeasy.net ([216.27.180.168]:41190 "EHLO sunset.davemloft.net") by vger.kernel.org with ESMTP id S1751266AbWHWAcC (ORCPT ); Tue, 22 Aug 2006 20:32:02 -0400 Date: Tue, 22 Aug 2006 17:32:00 -0700 (PDT) Message-Id: <20060822.173200.126578369.davem@davemloft.net> To: sundell.software@gmail.com Cc: kuznet@ms2.inr.ac.ru, johnpol@2ka.mipt.ru, nmiell@comcast.net, linux-kernel@vger.kernel.org, drepper@redhat.com, akpm@osdl.org, netdev@vger.kernel.org, zach.brown@oracle.com, hch@infradead.org Subject: Re: [take12 0/3] kevent: Generic event handling mechanism. From: David Miller In-Reply-To: References: <20060822231129.GA18296@ms2.inr.ac.ru> X-Mailer: Mew version 4.2 on Emacs 21.4 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: "Jari Sundell" Date: Wed, 23 Aug 2006 02:28:32 +0200 > There are system calls that take timespec, so I assume the magic is > already available for handling the timeout argument of kevent. System calls are one thing, they can be translated for these kinds of situations. But this doesn't help, and nothing at all can be done, for datastructures exposed to userspace via mmap()'d buffers, which is what kevent will be doing. This is what Alexey is trying to explain to you.