While the generic callback functions have an 'int' return and thus
appear to be allowed to return error, this is not true for all states.

Specifically, what used to be STARTING/DYING are ran with IRQs
disabled from critical parts of CPU bringup/teardown and are not
allowed to fail. Add WARNs to enforce this rule.

But since some callbacks are indeed allowed to fail, we have the
situation where a state-machine rollback encounters a failure, in this
case we're stuck, we can't go forward and we can't go back. Also add a
WARN for that case.

AFAICT this is a fundamental 'problem' with no real obvious solution.
We want the 'prepare' callbacks to allow failure on either up or down.
Typically on prepare-up this would be things like -ENOMEM from
resource allocations, and the typical usage in prepare-down would be
something like -EBUSY to avoid CPUs being taken away.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 kernel/cpu.c |   26 ++++++++++++++++++++++----
 1 file changed, 22 insertions(+), 4 deletions(-)

--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -202,7 +202,14 @@ static int cpuhp_invoke_callback(unsigne
 	hlist_for_each(node, &step->list) {
 		if (!cnt--)
 			break;
-		cbm(cpu, node);
+
+		trace_cpuhp_multi_enter(cpu, st->target, state, cbm, node);
+		ret = cbm(cpu, node);
+		trace_cpuhp_exit(cpu, st->state, state, ret);
+		/*
+		 * Rollback must not fail,
+		 */
+		WARN_ON_ONCE(ret);
 	}
 	return ret;
 }
@@ -695,6 +702,7 @@ static int take_cpu_down(void *_param)
 	struct cpuhp_cpu_state *st = this_cpu_ptr(&cpuhp_state);
 	enum cpuhp_state target = max((int)st->target, CPUHP_AP_OFFLINE);
 	int err, cpu = smp_processor_id();
+	int ret;
 
 	/* Ensure this CPU doesn't handle any more interrupts. */
 	err = __cpu_disable();
@@ -708,8 +716,13 @@ static int take_cpu_down(void *_param)
 	WARN_ON(st->state != CPUHP_TEARDOWN_CPU);
 	st->state--;
 	/* Invoke the former CPU_DYING callbacks */
-	for (; st->state > target; st->state--)
-		cpuhp_invoke_callback(cpu, st->state, false, NULL, NULL);
+	for (; st->state > target; st->state--) {
+		ret = cpuhp_invoke_callback(cpu, st->state, false, NULL, NULL);
+		/*
+		 * DYING must not fail!
+		 */
+		WARN_ON_ONCE(ret);
+	}
 
 	/* Give up timekeeping duties */
 	tick_handover_do_timer();
@@ -884,11 +897,16 @@ void notify_cpu_starting(unsigned int cp
 {
 	struct cpuhp_cpu_state *st = per_cpu_ptr(&cpuhp_state, cpu);
 	enum cpuhp_state target = min((int)st->target, CPUHP_AP_ONLINE);
+	int ret;
 
 	rcu_cpu_starting(cpu);	/* Enables RCU usage on this CPU. */
 	while (st->state < target) {
 		st->state++;
-		cpuhp_invoke_callback(cpu, st->state, true, NULL, NULL);
+		ret = cpuhp_invoke_callback(cpu, st->state, true, NULL, NULL);
+		/*
+		 * STARTING must not fail!
+		 */
+		WARN_ON_ONCE(ret);
 	}
 }