From mboxrd@z Thu Jan 1 00:00:00 1970 From: Santosh Shilimkar Subject: Re: [RFC PATCH] ARM: smp: Fix the CPU hotplug race with scheduler. Date: Mon, 20 Jun 2011 15:49:57 +0530 Message-ID: <4DFF1ECD.8080305@ti.com> References: <1308561839-18407-1-git-send-email-santosh.shilimkar@ti.com> <20110620095053.GA2082@n2100.arm.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20110620095053.GA2082@n2100.arm.linux.org.uk> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linux-arm-kernel-bounces@lists.infradead.org Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=m.gmane.org@lists.infradead.org To: Russell King - ARM Linux Cc: Peter Zijlstra , Thomas Gleixner , linux-omap@vger.kernel.org, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org List-Id: linux-omap@vger.kernel.org On 6/20/2011 3:20 PM, Russell King - ARM Linux wrote: > On Mon, Jun 20, 2011 at 02:53:59PM +0530, Santosh Shilimkar wrote: >> The current ARM CPU hotplug code suffers from couple of race conditions >> in CPU online path with scheduler. >> The ARM CPU hotplug code doesn't wait for hot-plugged CPU to be marked >> active as part of cpu_notify() by the CPU which brought it up before >> enabling interrupts. > > Hmm, why not just move the set_cpu_online() call before notify_cpu_starting() > and add the wait after the set_cpu_online() ? I think that's what patch is doing. Do you mean, calling hotplug notifier chain immediately after CPU marked as online. Something like below. set_cpu_online(cpu, true); notify_cpu_starting(cpu); while (!cpumask_test_cpu(smp_processor_id(), cpu_active_mask)) cpu_relax(); /* * Enable local interrupts. */ local_irq_enable(); local_fiq_enable(); Regards Santosh From mboxrd@z Thu Jan 1 00:00:00 1970 From: santosh.shilimkar@ti.com (Santosh Shilimkar) Date: Mon, 20 Jun 2011 15:49:57 +0530 Subject: [RFC PATCH] ARM: smp: Fix the CPU hotplug race with scheduler. In-Reply-To: <20110620095053.GA2082@n2100.arm.linux.org.uk> References: <1308561839-18407-1-git-send-email-santosh.shilimkar@ti.com> <20110620095053.GA2082@n2100.arm.linux.org.uk> Message-ID: <4DFF1ECD.8080305@ti.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 6/20/2011 3:20 PM, Russell King - ARM Linux wrote: > On Mon, Jun 20, 2011 at 02:53:59PM +0530, Santosh Shilimkar wrote: >> The current ARM CPU hotplug code suffers from couple of race conditions >> in CPU online path with scheduler. >> The ARM CPU hotplug code doesn't wait for hot-plugged CPU to be marked >> active as part of cpu_notify() by the CPU which brought it up before >> enabling interrupts. > > Hmm, why not just move the set_cpu_online() call before notify_cpu_starting() > and add the wait after the set_cpu_online() ? I think that's what patch is doing. Do you mean, calling hotplug notifier chain immediately after CPU marked as online. Something like below. set_cpu_online(cpu, true); notify_cpu_starting(cpu); while (!cpumask_test_cpu(smp_processor_id(), cpu_active_mask)) cpu_relax(); /* * Enable local interrupts. */ local_irq_enable(); local_fiq_enable(); Regards Santosh From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754062Ab1FTKUH (ORCPT ); Mon, 20 Jun 2011 06:20:07 -0400 Received: from na3sys009aog117.obsmtp.com ([74.125.149.242]:35639 "EHLO na3sys009aog117.obsmtp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753885Ab1FTKUE (ORCPT ); Mon, 20 Jun 2011 06:20:04 -0400 Message-ID: <4DFF1ECD.8080305@ti.com> Date: Mon, 20 Jun 2011 15:49:57 +0530 From: Santosh Shilimkar User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.15) Gecko/20110303 Thunderbird/3.1.9 MIME-Version: 1.0 To: Russell King - ARM Linux CC: linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, Thomas Gleixner , Peter Zijlstra Subject: Re: [RFC PATCH] ARM: smp: Fix the CPU hotplug race with scheduler. References: <1308561839-18407-1-git-send-email-santosh.shilimkar@ti.com> <20110620095053.GA2082@n2100.arm.linux.org.uk> In-Reply-To: <20110620095053.GA2082@n2100.arm.linux.org.uk> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 6/20/2011 3:20 PM, Russell King - ARM Linux wrote: > On Mon, Jun 20, 2011 at 02:53:59PM +0530, Santosh Shilimkar wrote: >> The current ARM CPU hotplug code suffers from couple of race conditions >> in CPU online path with scheduler. >> The ARM CPU hotplug code doesn't wait for hot-plugged CPU to be marked >> active as part of cpu_notify() by the CPU which brought it up before >> enabling interrupts. > > Hmm, why not just move the set_cpu_online() call before notify_cpu_starting() > and add the wait after the set_cpu_online() ? I think that's what patch is doing. Do you mean, calling hotplug notifier chain immediately after CPU marked as online. Something like below. set_cpu_online(cpu, true); notify_cpu_starting(cpu); while (!cpumask_test_cpu(smp_processor_id(), cpu_active_mask)) cpu_relax(); /* * Enable local interrupts. */ local_irq_enable(); local_fiq_enable(); Regards Santosh