From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751762Ab0ATMhb (ORCPT ); Wed, 20 Jan 2010 07:37:31 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751181Ab0ATMh3 (ORCPT ); Wed, 20 Jan 2010 07:37:29 -0500 Received: from mx1.redhat.com ([209.132.183.28]:65416 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750922Ab0ATMh2 (ORCPT ); Wed, 20 Jan 2010 07:37:28 -0500 From: Stanislaw Gruszka To: Thomas Gleixner , Andrew Morton Cc: Ingo Molnar , Oleg Nesterov , Peter Zijlstra , Hidetoshi Seto , Balbir Singh , linux-kernel@vger.kernel.org, Stanislaw Gruszka Subject: [PATCH 0/6 -mm] cpu-timers patches Date: Wed, 20 Jan 2010 13:34:22 +0100 Message-Id: <1263990868-8119-1-git-send-email-sgruszka@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Here are patches for cpu-timers I mainly wrote to not call thread_group_cputime() in fastpath_timer_check(), which we discuss in this thread: http://thread.gmane.org/gmane.linux.kernel/909349/focus=912327 On the way I fix some other bugs (SIGEV_NONE timer, old increase value in timer_settime()) and do other cleanups and optimizations, patch 5 however increase time when sighand->siglock is held. Compared with previous send of patches as RFC, I rebased them to -mm tree, change some changelogs and do some minor code changes. [PATCH 1/6 -mm] cpu-timers: simplify RLIMIT_CPU handling [PATCH 2/6 -mm] cpu-timers: cleanup arm_timer() [PATCH 3/6 -mm] cpu-timers: return correct previous timer reload value [PATCH 4/6 -mm] cpu-timers: change SIGEV_NONE timer implementation [PATCH 5/6 -mm] cpu-timers: assure to not iterate over all threads in fastpath_timer_check() [PATCH 6/6 -mm] cpu-timers: optimize run_posix_cpu_timers() Cheers Stanislaw