Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1348454
| From | "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [tip:smp/hotplug] rcu: Make CPU_DYING_IDLE an explicit call |
| Date | 2016-03-02 21:20 +0100 |
| Message-ID | <r8lpx-4I9-15@gated-at.bofh.it> (permalink) |
| References | <r6vCH-7sv-49@gated-at.bofh.it> <r7YCE-5hG-53@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On Tue, Mar 01, 2016 at 11:58:59AM -0800, tip-bot for Thomas Gleixner wrote:
> Commit-ID: 27d50c7eeb0f03c3d3ca72aac4d2dd487ca1f3f0
> Gitweb: http://git.kernel.org/tip/27d50c7eeb0f03c3d3ca72aac4d2dd487ca1f3f0
> Author: Thomas Gleixner <tglx@linutronix.de>
> AuthorDate: Fri, 26 Feb 2016 18:43:44 +0000
> Committer: Thomas Gleixner <tglx@linutronix.de>
> CommitDate: Tue, 1 Mar 2016 20:36:58 +0100
>
> rcu: Make CPU_DYING_IDLE an explicit call
>
> Make the RCU CPU_DYING_IDLE callback an explicit function call, so it gets
> invoked at the proper place.
>
> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
> Cc: linux-arch@vger.kernel.org
> Cc: Rik van Riel <riel@redhat.com>
> Cc: Rafael Wysocki <rafael.j.wysocki@intel.com>
> Cc: "Srivatsa S. Bhat" <srivatsa@mit.edu>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Arjan van de Ven <arjan@linux.intel.com>
> Cc: Sebastian Siewior <bigeasy@linutronix.de>
> Cc: Rusty Russell <rusty@rustcorp.com.au>
> Cc: Steven Rostedt <rostedt@goodmis.org>
> Cc: Oleg Nesterov <oleg@redhat.com>
> Cc: Tejun Heo <tj@kernel.org>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
> Cc: Linus Torvalds <torvalds@linux-foundation.org>
> Cc: Paul Turner <pjt@google.com>
> Link: http://lkml.kernel.org/r/20160226182341.870167933@linutronix.de
> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
[ . . . ]
> diff --git a/kernel/cpu.c b/kernel/cpu.c
> index 0e8c07f..ff8059b 100644
> --- a/kernel/cpu.c
> +++ b/kernel/cpu.c
> @@ -762,6 +762,7 @@ void cpuhp_report_idle_dead(void)
> BUG_ON(st->state != CPUHP_AP_OFFLINE);
> st->state = CPUHP_AP_IDLE_DEAD;
> complete(&st->done);
Not to be repetitive or anything, but if we delay here, it can break
RCU on a number of architectures. Either the CPU can be killed holding
one of RCU's locks or RCU can wrongly see the CPU as still being alive.
Either can prevent future RCU grace periods from ever completing, thus
OOMing the system.
Thanx, Paul
> + rcu_report_dead(smp_processor_id());
> }
>
> #else
> diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
> index e41dd41..85b4134 100644
> --- a/kernel/rcu/tree.c
> +++ b/kernel/rcu/tree.c
> @@ -2607,28 +2607,6 @@ static void rcu_cleanup_dead_rnp(struct rcu_node *rnp_leaf)
> }
>
> /*
> - * The CPU is exiting the idle loop into the arch_cpu_idle_dead()
> - * function. We now remove it from the rcu_node tree's ->qsmaskinit
> - * bit masks.
> - */
> -static void rcu_cleanup_dying_idle_cpu(int cpu, struct rcu_state *rsp)
> -{
> - unsigned long flags;
> - unsigned long mask;
> - struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu);
> - struct rcu_node *rnp = rdp->mynode; /* Outgoing CPU's rdp & rnp. */
> -
> - if (!IS_ENABLED(CONFIG_HOTPLUG_CPU))
> - return;
> -
> - /* Remove outgoing CPU from mask in the leaf rcu_node structure. */
> - mask = rdp->grpmask;
> - raw_spin_lock_irqsave_rcu_node(rnp, flags); /* Enforce GP memory-order guarantee. */
> - rnp->qsmaskinitnext &= ~mask;
> - raw_spin_unlock_irqrestore(&rnp->lock, flags);
> -}
> -
> -/*
> * The CPU has been completely removed, and some other CPU is reporting
> * this fact from process context. Do the remainder of the cleanup,
> * including orphaning the outgoing CPU's RCU callbacks, and also
> @@ -4247,6 +4225,43 @@ static void rcu_prepare_cpu(int cpu)
> rcu_init_percpu_data(cpu, rsp);
> }
>
> +#ifdef CONFIG_HOTPLUG_CPU
> +/*
> + * The CPU is exiting the idle loop into the arch_cpu_idle_dead()
> + * function. We now remove it from the rcu_node tree's ->qsmaskinit
> + * bit masks.
> + */
> +static void rcu_cleanup_dying_idle_cpu(int cpu, struct rcu_state *rsp)
> +{
> + unsigned long flags;
> + unsigned long mask;
> + struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu);
> + struct rcu_node *rnp = rdp->mynode; /* Outgoing CPU's rdp & rnp. */
> +
> + if (!IS_ENABLED(CONFIG_HOTPLUG_CPU))
> + return;
> +
> + /* Remove outgoing CPU from mask in the leaf rcu_node structure. */
> + mask = rdp->grpmask;
> + raw_spin_lock_irqsave_rcu_node(rnp, flags); /* Enforce GP memory-order guarantee. */
> + rnp->qsmaskinitnext &= ~mask;
> + raw_spin_unlock_irqrestore(&rnp->lock, flags);
> +}
> +
> +void rcu_report_dead(unsigned int cpu)
> +{
> + struct rcu_state *rsp;
> +
> + /* QS for any half-done expedited RCU-sched GP. */
> + preempt_disable();
> + rcu_report_exp_rdp(&rcu_sched_state,
> + this_cpu_ptr(rcu_sched_state.rda), true);
> + preempt_enable();
> + for_each_rcu_flavor(rsp)
> + rcu_cleanup_dying_idle_cpu(cpu, rsp);
> +}
> +#endif
> +
> /*
> * Handle CPU online/offline notification events.
> */
> @@ -4278,17 +4293,6 @@ int rcu_cpu_notify(struct notifier_block *self,
> for_each_rcu_flavor(rsp)
> rcu_cleanup_dying_cpu(rsp);
> break;
> - case CPU_DYING_IDLE:
> - /* QS for any half-done expedited RCU-sched GP. */
> - preempt_disable();
> - rcu_report_exp_rdp(&rcu_sched_state,
> - this_cpu_ptr(rcu_sched_state.rda), true);
> - preempt_enable();
> -
> - for_each_rcu_flavor(rsp) {
> - rcu_cleanup_dying_idle_cpu(cpu, rsp);
> - }
> - break;
> case CPU_DEAD:
> case CPU_DEAD_FROZEN:
> case CPU_UP_CANCELED:
> diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
> index 8abbe89..bd12c6c 100644
> --- a/kernel/sched/idle.c
> +++ b/kernel/sched/idle.c
> @@ -220,8 +220,6 @@ static void cpu_idle_loop(void)
> rmb();
>
> if (cpu_is_offline(smp_processor_id())) {
> - rcu_cpu_notify(NULL, CPU_DYING_IDLE,
> - (void *)(long)smp_processor_id());
> cpuhp_report_idle_dead();
> arch_cpu_idle_dead();
> }
>
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[patch 00/20] cpu/hotplug: Core infrastructure for cpu hotplug rework Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[patch 14/20] cpu/hotplug: Split out the state walk into functions Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Split out the state walk into functions tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 17/20] arch/hotplug: Call into idle with a proper state Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] arch/hotplug: Call into idle with a proper state tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 04/20] cpu/hotplug: Split out cpu down functions Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Split out cpu down functions tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
Re: [tip:smp/hotplug] cpu/hotplug: Split out cpu down functions "Srivatsa S. Bhat" <srivatsa@csail.mit.edu> - 2016-03-03 00:10 +0100
[patch 03/20] cpu/hotplug: Restructure cpu_up code Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Restructure cpu_up code tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
Re: [tip:smp/hotplug] cpu/hotplug: Restructure cpu_up code "Srivatsa S. Bhat" <srivatsa@csail.mit.edu> - 2016-03-03 00:00 +0100
[patch 16/20] cpu/hotplug: Move online calls to hotplugged cpu Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Move online calls to hotplugged cpu tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 02/20] cpu/hotplug: Restructure FROZEN state handling Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Restructure FROZEN state handling tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
Re: [tip:smp/hotplug] cpu/hotplug: Restructure FROZEN state handling "Srivatsa S. Bhat" <srivatsa@csail.mit.edu> - 2016-03-03 00:10 +0100
Re: [tip:smp/hotplug] cpu/hotplug: Restructure FROZEN state handling "Srivatsa S. Bhat" <srivatsa@csail.mit.edu> - 2016-03-03 00:50 +0100
[patch 11/20] cpu/hotplug: Implement setup/removal interface Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Implement setup/removal interface tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 12/20] cpu/hotplug: Move scheduler cpu_online notifier to hotplug core Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Move scheduler cpu_online notifier to hotplug core tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 18/20] cpu/hotplug: Let upcoming cpu bring itself fully up Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Let upcoming cpu bring itself fully up tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
Re: [patch 18/20] cpu/hotplug: Let upcoming cpu bring itself fully up Richard Cochran <richardcochran@gmail.com> - 2016-03-02 18:30 +0100
[patch 15/20] cpu/hotplug: Create hotplug threads Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Create hotplug threads tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 09/20] cpu/hotplug: Add sysfs state interface Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Add sysfs state interface tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
Re: [tip:smp/hotplug] cpu/hotplug: Add sysfs state interface Peter Zijlstra <peterz@infradead.org> - 2016-03-02 13:50 +0100
[patch 06/20] cpu/hotplug: Convert to a state machine for the control processor Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Convert to a state machine for the control processor tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
Re: [tip:smp/hotplug] cpu/hotplug: Convert to a state machine for the control processor Peter Zijlstra <peterz@infradead.org> - 2016-03-02 12:30 +0100
[patch 20/20] rcu: Make CPU_DYING_IDLE an explicit call Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
Re: [patch 20/20] rcu: Make CPU_DYING_IDLE an explicit call "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-02-27 03:20 +0100
Re: [patch 20/20] rcu: Make CPU_DYING_IDLE an explicit call "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-02-27 03:30 +0100
Re: [patch 20/20] rcu: Make CPU_DYING_IDLE an explicit call Thomas Gleixner <tglx@linutronix.de> - 2016-02-27 08:50 +0100
Re: [patch 20/20] rcu: Make CPU_DYING_IDLE an explicit call "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-02-27 12:10 +0100
Re: [patch 20/20] rcu: Make CPU_DYING_IDLE an explicit call Thomas Gleixner <tglx@linutronix.de> - 2016-02-27 12:40 +0100
Re: [patch 20/20] rcu: Make CPU_DYING_IDLE an explicit call "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-02-27 17:40 +0100
[tip:smp/hotplug] rcu: Make CPU_DYING_IDLE an explicit call tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
Re: [tip:smp/hotplug] rcu: Make CPU_DYING_IDLE an explicit call "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-03-02 21:20 +0100
Re: [tip:smp/hotplug] rcu: Make CPU_DYING_IDLE an explicit call Thomas Gleixner <tglx@linutronix.de> - 2016-03-03 08:40 +0100
[tip:smp/hotplug] cpu/hotplug: Plug death reporting race tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-03 11:10 +0100
Re: [tip:smp/hotplug] cpu/hotplug: Plug death reporting race "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-03-03 15:20 +0100
[patch 08/20] cpu/hotplug: Hand in target state to _cpu_up/down Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Hand in target state to _cpu_up/down tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 19/20] cpu/hotplug: Make wait for dead cpu completion based Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 19:50 +0100
[tip:smp/hotplug] cpu/hotplug: Make wait for dead cpu completion based tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 07/20] cpu/hotplug: Convert the hotplugged cpu work to a state machine Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 20:00 +0100
[tip:smp/hotplug] cpu/hotplug: Convert the hotplugged cpu work to a state machine tip-bot for Thomas Gleixner <tipbot@zytor.com> - 2016-03-01 21:00 +0100
[patch 01/20] idle: Move x86ism out of generic code Thomas Gleixner <tglx@linutronix.de> - 2016-02-26 20:00 +0100
Re: [patch 01/20] idle: Move x86ism out of generic code Brian Gerst <brgerst@gmail.com> - 2016-02-27 21:30 +0100
Re: [patch 01/20] idle: Move x86ism out of generic code Thomas Gleixner <tglx@linutronix.de> - 2016-02-29 20:40 +0100
Re: [patch 01/20] idle: Move x86ism out of generic code Will Deacon <will.deacon@arm.com> - 2016-02-29 20:50 +0100
Re: [patch 01/20] idle: Move x86ism out of generic code Thomas Gleixner <tglx@linutronix.de> - 2016-02-29 21:10 +0100
csiph-web