From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752882AbdEPSwg (ORCPT ); Tue, 16 May 2017 14:52:36 -0400 Received: from Galois.linutronix.de ([146.0.238.70]:41219 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750950AbdEPSwc (ORCPT ); Tue, 16 May 2017 14:52:32 -0400 Message-Id: <20170516184734.943149935@linutronix.de> User-Agent: quilt/0.63-1 Date: Tue, 16 May 2017 20:42:32 +0200 From: Thomas Gleixner To: LKML Cc: Peter Zijlstra , Ingo Molnar , Steven Rostedt , Mark Rutland , Greg Kroah-Hartman Subject: [patch V2 01/17] init: Pin init task to boot cpu initially References: <20170516184231.564888231@linutronix.de> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Disposition: inline; filename=init--Pin-init-task-to-boot-cpu-initially.patch Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Some of the boot code in init_kernel_freeable() which runs before SMP bringup assumes (rightfully) that it runs on the boot cpu and therefor can use smp_processor_id() in preemptible context. That works so far because the smp_processor_id() check starts to be effective after smp bringup. That's just wrong. Starting with SMP bringup and the ability to move threads around, smp_processor_id() in preemptible context is broken. Aside of that it does not make sense to allow init to run on all cpus before sched_smp_init() has been run. Pin the init to the boot cpu so the existing code can continue to use smp_processor_id() without triggering the checks when the enabling of those checks starts earlier. Signed-off-by: Thomas Gleixner --- init/main.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) --- a/init/main.c +++ b/init/main.c @@ -389,6 +389,7 @@ static __initdata DECLARE_COMPLETION(kth static noinline void __ref rest_init(void) { + struct task_struct *tsk; int pid; rcu_scheduler_starting(); @@ -397,7 +398,17 @@ static noinline void __ref rest_init(voi * the init task will end up wanting to create kthreads, which, if * we schedule it before we create kthreadd, will OOPS. */ - kernel_thread(kernel_init, NULL, CLONE_FS); + pid = kernel_thread(kernel_init, NULL, CLONE_FS); + /* + * Pin init on the boot cpu. Task migration is not properly working + * until sched_init_smp() has been run. It will set the allowed + * cpus for init to the non isolated cpus. + */ + rcu_read_lock(); + tsk = find_task_by_pid_ns(pid, &init_pid_ns); + set_cpus_allowed_ptr(tsk, cpumask_of(smp_processor_id())); + rcu_read_unlock(); + numa_default_policy(); pid = kernel_thread(kthreadd, NULL, CLONE_FS | CLONE_FILES); rcu_read_lock(); @@ -1015,10 +1026,6 @@ static noinline void __init kernel_init_ * init can allocate pages on any node */ set_mems_allowed(node_states[N_MEMORY]); - /* - * init can run on any cpu. - */ - set_cpus_allowed_ptr(current, cpu_all_mask); cad_pid = task_pid(current);