x86, reboot: cleanup NMI and REBOOT_IRQ

After dealing with pstore conversations about spin locks, I had an idea to simplify the native_smp_stop_other_cpus() path by using both REBOOT_IRQ and NMI instead of using either or.

I also cleaned up a WARN_ON splat from rescheduling.

Tested 10 panics on my core2 quad using 'echo c > /proc/sysrq-trigger' and panic=10 on the commandline. The machine panic'd and rebooted succesfully all 10 times. Though only the first time did I see the WARN_ON splat, the other 9 times I couldn't duplicate it.

Don Zickus (2):
x86, reboot: revert stop_other_cpus to using IRQ with NMI fallback x86, reschedule: check to see if system is shutting down

 arch/x86/kernel/smp.c |  103 +++++++++++++++++++++++--------------------------
 1 files changed, 48 insertions(+), 55 deletions(-)

1.7.7.6

For v3.3, I added code to use the NMI to stop other cpus in the panic case. The idea was to make sure all cpus on the system were definitely halted to help serialize the panic path to execute the rest of the code on a single cpu.

The main problem it was trying to solve was how to stop a cpu that was spinning with its irqs disabled. A IPI irq would be stuck and couldn't get in there, but an NMI could.

Things were great until we had another conversation about some pstore changes. Because some of the backend pstore still uses spinlocks to protect the device access, things could get ugly if a panic happened and we were stuck spinning on a lock.

Now with the NMI shutting down cpus, we could assume no other cpus were running and just bust the spin lock and proceed.

The counter argument was, well if you do that the backend could be in a screwed up state and you might not be able to save anything as a result. If we could have just given the cpu a little more time to finish things, we could have grabbed the spin lock cleanly and everything would have been fine.

Well, how do give a cpu a 'little more time' in the panic case? For the most part you can't without spinning on the lock and even in that case, how long do you spin for?

So instead of making it ugly in the pstore code, I had the idea that most spin locks are held with irqs disabled, naturally blocking other irqs until they are done. We just need an irq to sit there and grab the cpu once the lock is released and irqs are re-enabled again.

I decided to modify stop_other_cpus to go back to the REBOOT_IRQ code and use that IPI irq as the blocking irq. This code has been working for a long time and will give up after one second. To fix the original problem of what happens after one second if a cpu hasn't accepted the IPI, I just added the NMI hammer to clobber the cpu.

The end result of this more complicated-looking-diff-than-it-really-is, is a patch that mostly reverts

3603a25 x86, reboot: Use NMI instead of REBOOT_VECTOR to stop cpus

and instead just sticks another if-case after the REBOOT_IRQ and checks to see if online_cpus are still > 1, and if so, clobber the rest of the cpus with an NMI.

For the most part, the NMI piece will never get hit, thus behaving just like pre-v3.3 code. However, in those rare conditions, we have a fallback plan.

I still wrap the NMI check with the knob 'nonmi_ipi' in case someone still has issues with NMIs in the panic path.

I also reset the original names of the functions.

---
 arch/x86/kernel/smp.c |  100 ++++++++++++++++++++++--------------------------
 1 files changed, 46 insertions(+), 54 deletions(-)

index 66c74f4..48d2b7d 100644
@@ -109,6 +109,9 @@
  *	about nothing of note with C stepping upwards.
  */
 
+static atomic_t stopping_cpu = ATOMIC_INIT(-1);
+static bool smp_no_nmi_ipi = false;
+
 /*
  * this function sends a 'reschedule' IPI to another CPU.
  * it goes straight through and wastes no time serializing

@@ -149,8 +152,6 @@ void native_send_call_func_ipi(const struct cpumask *mask)
 	free_cpumask_var(allbutself);
 }
 
-static atomic_t stopping_cpu = ATOMIC_INIT(-1);
-
 static int smp_stop_nmi_callback(unsigned int val, struct pt_regs *regs)
 {
 	/* We are registered on stopping cpu too, avoid spurious NMI */

@@ -162,7 +163,19 @@ static int smp_stop_nmi_callback(unsigned int val, struct pt_regs *regs)
 	return NMI_HANDLED;
 }
 
-static void native_nmi_stop_other_cpus(int wait)
+/*
+ * this function calls the 'stop' function on all other CPUs in the system.
+ */
+
+asmlinkage void smp_reboot_interrupt(void)
+{
+	ack_APIC_irq();
+	irq_enter();
+	stop_this_cpu(NULL);
+	irq_exit();
+}
+
+static void native_stop_other_cpus(int wait)
 {
 	unsigned long flags;
 	unsigned long timeout;

@@ -174,20 +187,25 @@ static void native_nmi_stop_other_cpus(int wait)
 	 * Use an own vector here because smp_call_function
 	 * does lots of things not suitable in a panic situation.
 	 */
+
+	/*
+	 * We start by using the REBOOT_VECTOR irq.
+	 * The irq is treated as a sync point to allow critical
+	 * regions of code on other cpus to release their spin locks
+	 * and re-enable irqs.  Jumping straight to an NMI might
+	 * accidentally cause deadlocks with further shutdown/panic
+	 * code.  By syncing, we give the cpus up to one second to
+	 * finish their work before we force them off with the NMI.
+	 */
 	if (num_online_cpus() > 1) {
 		/* did someone beat us here? */
 		if (atomic_cmpxchg(&stopping_cpu, -1, safe_smp_processor_id()) != -1)
 			return;
 
-		if (register_nmi_handler(NMI_LOCAL, smp_stop_nmi_callback,
-					 NMI_FLAG_FIRST, "smp_stop"))
-			/* Note: we ignore failures here */
-			return;
-
-		/* sync above data before sending NMI */
+		/* sync above data before sending IRQ */
 		wmb();
 
-		apic->send_IPI_allbutself(NMI_VECTOR);
+		apic->send_IPI_allbutself(REBOOT_VECTOR);
 
 		/*
 		 * Don't wait longer than a second if the caller

@@ -197,63 +215,37 @@ static void native_nmi_stop_other_cpus(int wait)
 		while (num_online_cpus() > 1 && (wait || timeout--))
 			udelay(1);
 	}
+	
+	/* if the REBOOT_VECTOR didn't work, try with the NMI */
+	if ((num_online_cpus() > 1) && (!smp_no_nmi_ipi))  {
+		if (register_nmi_handler(NMI_LOCAL, smp_stop_nmi_callback,
+					 NMI_FLAG_FIRST, "smp_stop"))
+			/* Note: we ignore failures here */
+			/* Hope the REBOOT_IRQ is good enough */
+			goto finish;
 
-	local_irq_save(flags);
-	disable_local_APIC();
-	local_irq_restore(flags);
-}
-
-/*
- * this function calls the 'stop' function on all other CPUs in the system.
- */
-
-asmlinkage void smp_reboot_interrupt(void)
-{
-	ack_APIC_irq();
-	irq_enter();
-	stop_this_cpu(NULL);
-	irq_exit();
-}
-
-static void native_irq_stop_other_cpus(int wait)
-{
-	unsigned long flags;
-	unsigned long timeout;
+		/* sync above data before sending IRQ */
+		wmb();
 
-	if (reboot_force)
-		return;
+		pr_emerg("Shutting down cpus with NMI\n");
 
-	/*
-	 * Use an own vector here because smp_call_function
-	 * does lots of things not suitable in a panic situation.
-	 * On most systems we could also use an NMI here,
-	 * but there are a few systems around where NMI
-	 * is problematic so stay with an non NMI for now
-	 * (this implies we cannot stop CPUs spinning with irq off
-	 * currently)
-	 */
-	if (num_online_cpus() > 1) {
-		apic->send_IPI_allbutself(REBOOT_VECTOR);
+		apic->send_IPI_allbutself(NMI_VECTOR);
 
 		/*
-		 * Don't wait longer than a second if the caller
+		 * Don't wait longer than a 10 ms if the caller
 		 * didn't ask us to wait.
 		 */
-		timeout = USEC_PER_SEC;
+		timeout = USEC_PER_MSEC * 10;
 		while (num_online_cpus() > 1 && (wait || timeout--))
 			udelay(1);
 	}
 
+finish:
 	local_irq_save(flags);
 	disable_local_APIC();
 	local_irq_restore(flags);
 }
 
-static void native_smp_disable_nmi_ipi(void)
-{
-	smp_ops.stop_other_cpus = native_irq_stop_other_cpus;
-}
-
 /*
  * Reschedule call back.
  */

@@ -287,8 +279,8 @@ void smp_call_function_single_interrupt(struct pt_regs *regs)
 
 static int __init nonmi_ipi_setup(char *str)
 {
-        native_smp_disable_nmi_ipi();
-        return 1;
+	smp_no_nmi_ipi = true;
+	return 1;
 }
 
 __setup("nonmi_ipi", nonmi_ipi_setup);

@@ -298,7 +290,7 @@ struct smp_ops smp_ops = {
 	.smp_prepare_cpus	= native_smp_prepare_cpus,
 	.smp_cpus_done		= native_smp_cpus_done,
 
-	.stop_other_cpus	= native_nmi_stop_other_cpus,
+	.stop_other_cpus	= native_stop_other_cpus,
 	.smp_send_reschedule	= native_smp_send_reschedule,
 
 	.cpu_up			= native_cpu_up,

1.7.7.6

I don't think kdump uses the notifier_chain.

Cheers,
Don

Sorry about that. I looked at old kernel. Currently, kdump uses register_nmi_handler() as well.

Seiji

Due to reschedule changes in v3.3, a WARN_ON has popped up because a cpu suddenly became offline without letting anyone know.

This results in the schedule trying to move something to another cpu only to find out it isn't there and getting confused. The splat looks something like:

[   32.448626] ------------[ cut here ]------------
[   32.449160] WARNING: at arch/x86/kernel/smp.c:119 native_smp_send_reschedule+0x25/0x43()
[   32.449621] Pid: 1, comm: init_stage2 Not tainted 3.2.0+ #14
[   32.449621] Call Trace:
[   32.449621]  <IRQ>  [<ffffffff81041a44>] ? native_smp_send_reschedule+0x25/0x43
[   32.449621]  [<ffffffff810735b2>] warn_slowpath_common+0x7b/0x93
[   32.449621]  [<ffffffff810962cc>] ? tick_nohz_handler+0xc9/0xc9
[   32.449621]  [<ffffffff81073675>] warn_slowpath_null+0x15/0x18
[   32.449621]  [<ffffffff81041a44>] native_smp_send_reschedule+0x25/0x43
[   32.449621]  [<ffffffff81067a00>] smp_send_reschedule+0xa/0xc
[   32.449621]  [<ffffffff8106f25e>] scheduler_tick+0x21a/0x242
[   32.449621]  [<ffffffff8107da10>] update_process_times+0x62/0x73
[   32.449621]  [<ffffffff81096336>] tick_sched_timer+0x6a/0x8a
[   32.449621]  [<ffffffff8108c5eb>] __run_hrtimer.clone.26+0x55/0xcb
[   32.449621]  [<ffffffff8108cd77>] hrtimer_interrupt+0xcb/0x19b
[   32.449621]  [<ffffffff810428a8>] smp_apic_timer_interrupt+0x72/0x85
[   32.449621]  [<ffffffff8165a8de>] apic_timer_interrupt+0x6e/0x80
[   32.449621]  <EOI>  [<ffffffff8165928e>] ? _raw_spin_unlock_irqrestore+0x3a/0x3e
[   32.449621]  [<ffffffff81042f4e>] ? arch_local_irq_restore+0x6/0xd
[   32.449621]  [<ffffffff810430c4>] default_send_IPI_mask_allbutself_phys+0x78/0x88
[   32.449621]  [<ffffffff8106c3c4>] ? __migrate_task+0xf1/0xf1
[   32.449621]  [<ffffffff81045445>] physflat_send_IPI_allbutself+0x12/0x14
[   32.449621]  [<ffffffff81041aaf>] native_stop_other_cpus+0x4d/0xa8
[   32.449621]  [<ffffffff810411c6>] native_machine_shutdown+0x56/0x6d
[   32.449621]  [<ffffffff81048499>] kvm_shutdown+0x1a/0x1c
[   32.449621]  [<ffffffff810411f9>] machine_shutdown+0xa/0xc
[   32.449621]  [<ffffffff81041265>] native_machine_restart+0x20/0x32
[   32.449621]  [<ffffffff81041297>] machine_restart+0xa/0xc
[   32.449621]  [<ffffffff81081d53>] kernel_restart+0x49/0x4d

I solved this by re-using the atomic global variable that is set during native_smp_stop_other_cpus(). This is the piece of code that causes the problem and it sets stopping_cpu to reflect the system is going down.

If the variable is set do not yell with the WARN_ON, just return.

v2: use if-condition for WARN-ON instead of if {WARN_ON(1)}

---
 arch/x86/kernel/smp.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

index 48d2b7d..54d570e 100644
@@ -120,7 +120,8 @@ static bool smp_no_nmi_ipi = false;
 static void native_smp_send_reschedule(int cpu)
 {
 	if (unlikely(cpu_is_offline(cpu))) {
-		WARN_ON(1);
+		/* system is not shutting down.. yell */
+		WARN_ON(atomic_read(&stopping_cpu) == -1)
 		return;
 	}
 	apic->send_IPI_mask(cpumask_of(cpu), RESCHEDULE_VECTOR);

1.7.7.6

Ugh, I really need to focus on some stuff at work before I send more crap to the list. Sorry about that.

Cheers,
Don

<snip>

Well maybe it can, it's been like that way for over three years now. I'm surprised no one ran into issues before now.

The only thing I can think that would work is stop_machine(). Pass in a halt function and a cpumask of everyone but smp_processor_id(). That would solve the problem, no?

Cheers,
Don

nope.. same problem, you're not telling anybody you're shooting CPUs down -- this telling is usually done through cpu hotplug notifiers that fix up state.

The only way is to unplug all cpus except the one. Problem with that is that we cannot (as of yet) unplug the boot cpu.

Why? If you have successfully sync'd up the cpus, stopped them and then run a 'stop_cpu' function, you stop all those WARN_ONs I would think. And how much do we care that we fix up the state on a shutdown? We have already shutdown all the processes and unmounted the disks? Most of the drivers have probably been shutdown cleanly. What is left that we have to be polite?

Yeah, well we can migrate to the boot cpu. I think powerpc does that for kdump.

Cheers,
Don

So the one you're running in here is native_smp_send_reschedule(), its send by the scheduler to kick a remote cpu. Doing kstopmachine() flipping online bits and calling hlt with irqs disabled doesn't inform the scheduler those CPUs aren't actually around anymore.

Dunno, probably not much, but things like RCU might just want to try and wait for remote CPUs to complete their grace period, which'll be a while with them being quite dead.

There might also still be work left in workqueues, timers left on timer wheels etc.. timers won't tick since their apic is dead along with the cpu, so that shouldn't be a big problem unless someone is waiting for one to complete. Similar with workqueues, I guess since tasks won't get scheduled.

But the scheduler is still up and thinking its got all these CPUs, poking them and getting WARNed about it.

Right, there's that.

Well we are in luck, migrating to cpu0 is already done in the shutdown path.

arch/x86/kernel/reboot.c::native_machine_shutdown:662

	set_cpus_allowed_ptr(current, cpumask_of(reboot_cpu_id));

Now most of the time it seems to be cpu0, but I guess corner cases can put it elsewhere. I replaced 'stop_other_cpus' with 'for_each_online_cpu(cpu) { cpu_down(cpu) }. And things seemed to work. Output below, patch below that:

=====================
Please stand by while rebooting the system...

[  101.470240] md: stopping all md devices.
[  101.474346] kvm: exiting hardware virtualization
[  101.480071] sd 3:0:0:0: [sdb] Synchronizing SCSI cache
[  101.535977] sd 0:0:0:0: [sda] Synchronizing SCSI cache
[  101.643479] Restarting system.
[  101.646624] machine restart
[  101.649501] DONDON: on cpu2
[  101.652502] DONDON: on new cpu0
[  101.658397] Broke affinity for irq 42
[  101.662168] Broke affinity for irq 47
[  101.669281] bnx2fc: CPU 1 offline: Remove Rx thread
[  101.674381] CPU 1 offline: Remove Rx thread
[  101.680595] Broke affinity for irq 22
[  101.686510] bnx2fc: CPU 2 offline: Remove Rx thread
[  101.691569] CPU 2 offline: Remove Rx thread
[  101.697260] Broke affinity for irq 16
[  101.702057] lockdep: fixing up alternatives.
[  101.706476] SMP alternatives: switching to UP code
[  101.720162] bnx2fc: CPU 3 offline: Remove Rx thread
[  101.725206] CPU 3 offline: Remove Rx thread

====================

Does that work for you?

Cheers,
Don

index d840e69..c3569ac 100644
@@ -644,6 +644,7 @@ void native_machine_shutdown(void)
 
 	/* The boot cpu is always logical cpu 0 */
 	int reboot_cpu_id = 0;
+	int cpu;
 
 #ifdef CONFIG_X86_32
 	/* See if there has been given a command line override */

@@ -652,17 +653,24 @@ void native_machine_shutdown(void)
 		reboot_cpu_id = reboot_cpu;
 #endif
 
+	printk("DONDON: on cpu%d\n", smp_processor_id());
 	/* Make certain the cpu I'm about to reboot on is online */
 	if (!cpu_online(reboot_cpu_id))
 		reboot_cpu_id = smp_processor_id();
 
 	/* Make certain I only run on the appropriate processor */
 	set_cpus_allowed_ptr(current, cpumask_of(reboot_cpu_id));
+	printk("DONDON: on new cpu%d\n", smp_processor_id());
 
 	/* O.K Now that I'm on the appropriate processor,
 	 * stop all of the others.
 	 */
-	stop_other_cpus();
+	//stop_other_cpus();
+	for_each_online_cpu(cpu) {
+		if (cpu == 0)
+			continue;
+		cpu_down(cpu);
+	}
 #endif
 
 	lapic_shutdown();