From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753881Ab1JPK46 (ORCPT ); Sun, 16 Oct 2011 06:56:58 -0400 Received: from mail-pz0-f42.google.com ([209.85.210.42]:64860 "EHLO mail-pz0-f42.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750750Ab1JPK44 (ORCPT ); Sun, 16 Oct 2011 06:56:56 -0400 From: Yong Zhang To: linux-kernel@vger.kernel.org Cc: linux-rt-users@vger.kernel.org, tglx@linutronix.de Subject: [RFC] [PATCH -rt 0/5] patchset focusing on cpu hotplug Date: Sun, 16 Oct 2011 18:56:42 +0800 Message-Id: <1318762607-2261-1-git-send-email-yong.zhang0@gmail.com> X-Mailer: git-send-email 1.7.1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org When tring cpu hotplug, I get some warning. And these patches are tring to fix them up. BTW, I have marked some of them as RFC: patch#3: printk: don't call printk_tick in printk_needs_cpu() I don't think we need to call printk_tick() even in mainline kernel, fix me if I'm wrong. patch#5: cpufreq: get rid of get_online_cpus()/put_online_cpus IMHO, it's about lockdep false positive, but it's a little annoying. Maybe there is other way to fix it up. And the side effect is we introduce more latency than before though it's rare path. More details in eatch patch :) Thanks, Yong --- Yong Zhang (5): hotplug: sync_unplug: No '\n' in task name hotplug: Call cpu_unplug_begin() a little early printk: don't call printk_tick in printk_needs_cpu() workqueue: hotplug fix cpufreq: get rid of get_online_cpus()/put_online_cpus drivers/cpufreq/cpufreq.c | 20 +++++++------------- kernel/cpu.c | 18 ++++++++---------- kernel/printk.c | 4 ++-- kernel/workqueue.c | 12 ++++++++---- 4 files changed, 25 insertions(+), 29 deletions(-)