From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753528Ab2FMLBC (ORCPT ); Wed, 13 Jun 2012 07:01:02 -0400 Received: from www.linutronix.de ([62.245.132.108]:48229 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751255Ab2FMLBA (ORCPT ); Wed, 13 Jun 2012 07:01:00 -0400 Message-Id: <20120613102823.373180763@linutronix.de> User-Agent: quilt/0.48-1 Date: Wed, 13 Jun 2012 11:00:53 -0000 From: Thomas Gleixner To: LKML Cc: Peter Zijlstra , Ingo Molnar , "Srivatsa S. Bhat" , Rusty Russell , "Paul E. McKenney" , Tejun Heo Subject: [RFC patch 0/5] Per cpu thread hotplug infrastructure X-Linutronix-Spam-Score: -1.0 X-Linutronix-Spam-Level: - X-Linutronix-Spam-Status: No , -1.0 points, 5.0 required, ALL_TRUSTED=-1,SHORTCIRCUIT=-0.0001 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The following series implements the infrastructure for parking and unparking kernel threads to avoid the full teardown and fork on cpu hotplug operations along with management infrastructure for hotplug and users. I've converted the most obvious (and simple) users ksoftirqd and watchdog along with a driver (the latter is untested due to lack of hardware). Looking at the other (ab)users of per cpu kernel threads: - drivers/* should be rather trivial to convert all users - RCU needs more thought, but I'm sure that Paul will figure it out in no time :) The main issue with RCU are the per node threads, but it should be simple to extend the infrastructure to allow registration and handling of per node threads as well. - workqueues. I did not even try to think about converting it. The hotplug related code in there causes strong feelings related to zombies and chainsaws. (See https://lwn.net/Articles/378036/) Thanks, tglx