Re: [RFC][PATCH] x86, reboot: use NMI instead of REBOOT_VECTOR to stop cpus

[Don Zickus <dzickus@redhat.com>]

On Mon, Oct 10, 2011 at 08:53:33AM +0200, Ingo Molnar wrote:
> 
> * Don Zickus <dzickus@redhat.com> wrote:
> 
> > --- a/arch/x86/kernel/smp.c
> > +++ b/arch/x86/kernel/smp.c
> > @@ -28,6 +28,7 @@
> >  #include <asm/mmu_context.h>
> >  #include <asm/proto.h>
> >  #include <asm/apic.h>
> > +#include <asm/nmi.h>
> >  /*
> >   *	Some notes on x86 processor bugs affecting SMP operation:
> >   *
> > @@ -147,6 +148,57 @@ void native_send_call_func_ipi(const struct cpumask *mask)
> >  	free_cpumask_var(allbutself);
> >  }
> >  
> > +static int stopping_cpu;
> 
> Is access to this variable sufficiently serialized by all callpaths 
> of stop_other_cpus()?

Doesn't seem to be.  I can change this to an atomic_cmpxchg() to do that.

> 
> > +
> > +static int smp_stop_nmi_callback(unsigned int val, struct pt_regs *regs)
> > +{
> > +	/* We are registerd on stopping cpu too, avoid spurious NMI */
> > +	if (raw_smp_processor_id() == stopping_cpu)
> > +		return NMI_HANDLED;
> > +
> > +	stop_this_cpu(NULL);
> > +
> > +	return NMI_HANDLED;
> > +}
> > +
> > +static void native_nmi_stop_other_cpus(int wait)
> > +{
> > +	unsigned long flags;
> > +	unsigned long timeout;
> > +
> > +	if (reboot_force)
> > +		return;
> > +
> > +	/*
> > +	 * Use an own vector here because smp_call_function
> > +	 * does lots of things not suitable in a panic situation.
> > +	 */
> > +	if (num_online_cpus() > 1) {
> > +		stopping_cpu = safe_smp_processor_id();
> > +
> > +		if (register_nmi_handler(NMI_LOCAL, smp_stop_nmi_callback,
> > +					 NMI_FLAG_FIRST, "smp_stop"))
> > +			return;		/* return what? */
> > +
> > +		/* sync above data before sending NMI */
> > +		wmb();
> > +
> > +		apic->send_IPI_allbutself(NMI_VECTOR);
> > +
> > +		/*
> > +		 * Don't wait longer than a second if the caller
> > +		 * didn't ask us to wait.
> > +		 */
> > +		timeout = USEC_PER_SEC;
> > +		while (num_online_cpus() > 1 && (wait || timeout--))
> > +			udelay(1);
> > +	}
> > +
> > +	local_irq_save(flags);
> > +	disable_local_APIC();
> > +	local_irq_restore(flags);
> > +}
> > +
> >  /*
> >   * this function calls the 'stop' function on all other CPUs in the system.
> >   */
> > @@ -159,7 +211,7 @@ asmlinkage void smp_reboot_interrupt(void)
> >  	irq_exit();
> >  }
> >  
> > -static void native_stop_other_cpus(int wait)
> > +static void native_irq_stop_other_cpus(int wait)
> >  {
> >  	unsigned long flags;
> >  	unsigned long timeout;
> > @@ -229,7 +281,7 @@ struct smp_ops smp_ops = {
> >  	.smp_prepare_cpus	= native_smp_prepare_cpus,
> >  	.smp_cpus_done		= native_smp_cpus_done,
> >  
> > -	.stop_other_cpus	= native_stop_other_cpus,
> > +	.stop_other_cpus	= native_nmi_stop_other_cpus,
> >  	.smp_send_reschedule	= native_smp_send_reschedule,
> >  
> >  	.cpu_up			= native_cpu_up,
> 
> I'd be fine about this if you also aded some sort of 
> CONFIG_KERNEL_DEBUG dependent test facility that did a "send NMI to 
> all CPUs and check that they truly arrive" non-destructive test.
> 
> That would at least give people an automatic way to test it without 
> waiting for the first crash of their kernel.

Ok fair enough.  I also wanted to keep the 'old' fucntion around to let
people add something reboot=irq on the command line to sort through those
issues too.

I'll work on the 'DEBUG' idea you suggested and repost.

Thanks,
Don