From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030184AbWGYUHW (ORCPT ); Tue, 25 Jul 2006 16:07:22 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S964852AbWGYUHV (ORCPT ); Tue, 25 Jul 2006 16:07:21 -0400 Received: from ozlabs.tip.net.au ([203.10.76.45]:55263 "EHLO ozlabs.org") by vger.kernel.org with ESMTP id S964846AbWGYUHU (ORCPT ); Tue, 25 Jul 2006 16:07:20 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <17606.31023.273943.551848@cargo.ozlabs.ibm.com> Date: Wed, 26 Jul 2006 06:03:59 +1000 From: Paul Mackerras To: "H. Peter Anvin" Cc: Neil Horman , Segher Boessenkool , linux-kernel@vger.kernel.org, a.zummo@towertech.it, jg@freedesktop.org Subject: Re: [PATCH] RTC: Add mmap method to rtc character driver In-Reply-To: <44C66C91.8090700@zytor.com> References: <20060725174100.GA4608@hmsreliant.homelinux.net> <03BCDC7F-13D9-42FC-86FC-30C76FD3B3B8@kernel.crashing.org> <20060725182833.GE4608@hmsreliant.homelinux.net> <44C66C91.8090700@zytor.com> X-Mailer: VM 7.19 under Emacs 21.4.1 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org H. Peter Anvin writes: > Not really. This introduces a potentially very difficult support > user-visible interface. Consider a tickless kernel -- you might end up > taking tick interrupts ONLY to update this page, since you don't have > any way of knowing when userspace wants to look at it. It's not that bad; if userspace is running, the cpu isn't idle, so there isn't the motivation to go tickless on that cpu. In other words, if every cpu has suspended ticks, then no cpu can be running stuff that needs to look at this page. Paul.