Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751863AbWCRAj7 (ORCPT ); Fri, 17 Mar 2006 19:39:59 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751882AbWCRAj7 (ORCPT ); Fri, 17 Mar 2006 19:39:59 -0500 Received: from e33.co.us.ibm.com ([32.97.110.151]:15779 "EHLO e33.co.us.ibm.com") by vger.kernel.org with ESMTP id S1751863AbWCRAj6 (ORCPT ); Fri, 17 Mar 2006 19:39:58 -0500 Date: Fri, 17 Mar 2006 17:39:56 -0700 From: john stultz To: lkml Cc: john stultz , george@wildturkeyranch.net, Steven Rostedt , Thomas Gleixner , Ulrich Windl , Roman Zippel , Ingo Molnar Message-Id: <20060318003955.32251.80532.sendpatchset@cog.beaverton.ibm.com> Subject: [PATCHSET 0/10] Time: Generic Timekeeping (v.C0) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2853 Lines: 68 All, So after after the last round of feedback on this series, I've gone back to the drawing board and have tried to further cut the patch set down to the bare minimum. I've realized that much of the trouble getting this into mainline have come from my falling into the well known "here's a new subsystem that duplicates the old one, but is better and cleaner and fixes bugs" trap. Thus now I'm trying the time-tested "fix what is there with small patches" method (which I would have used from the start, but I didn't think I could make such drastic changes without breaking everyone first). The majority of this new incremental design should be credited to Roman Zippel, but it is my implementation, so he gets the credit and I get the blame. :) This is the rough first cut, so it will probably need a few extra passes to be cleaned up and ready for mainline. And while there should not be any behaviour changes, since this method also affects all arches from the start, extra testing will be needed to make sure it has been done correctly. Also more work is needed before this can be re-integrated w/ the -RT tree, as timekeeping is still done from interrupt context. Summary: This patchset provides a generic timekeeping interfaces that are independent of the timer interrupt. This allows for robust and correct behavior in cases of late or lost ticks, avoids interpolation errors, reduces duplication in arch specific code, and allows or assists future changes such as high-res timers, dynamic ticks, or realtime preemption. Additionally, it provides finer nanosecond resolution values to the clock_gettime functions. Changes since the B20 release: o Threw out old NTP changes o Threw out all code cleanups o New minimal core implementation for all arches o Added sysfs clocksource fix o Added slow TSC detection On my TODO list: o Get x86-64 and powerpc patches back up and running o Try to restore cleanups via small patches The patchset applies against the current 2.6.16-rc6-git. The complete patchset can be found here: http://sr71.net/~jstultz/tod/ I'd like to thank the following people who have contributed ideas, criticism, testing and code that has helped shape this work: George Anzinger, Nish Aravamudan, Max Asbock, Serge Belyshev, Dominik Brodowski, Adrian Bunk, Thomas Gleixner, Darren Hart, Christoph Lameter, Matt Mackal, Keith Mannthey, Ingo Molnar, Andrew Morton, Paul Munt, Martin Schwidefsky, Frank Sorenson, Ulrich Windl, Jonathan Woithe, Darrick Wong, Roman Zippel and any others whom I've accidentally left off this list. thanks -john - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/