X-Git-Url: http://pilppa.org/gitweb/gitweb.cgi?a=blobdiff_plain;f=kernel%2Fsmp.c;h=782e2b93e4652214ad5b6f21cd89935178e3b9f9;hb=7946612de2087e163308e26034286fc2dc9dacf1;hp=462c785ca1eee80dea958541edf19786259c88e6;hpb=acee709cab689ec7703770e8b8cb5cc3a4abcb31;p=linux-2.6-omap-h63xx.git

diff --git a/kernel/smp.c b/kernel/smp.c
index 462c785ca1e..782e2b93e46 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -33,7 +33,7 @@ struct call_single_queue {
 	spinlock_t lock;
 };
 
-void __cpuinit init_call_single_data(void)
+static int __cpuinit init_call_single_data(void)
 {
 	int i;
 
@@ -43,7 +43,9 @@ void __cpuinit init_call_single_data(void)
 		spin_lock_init(&q->lock);
 		INIT_LIST_HEAD(&q->list);
 	}
+	return 0;
 }
+early_initcall(init_call_single_data);
 
 static void csd_flag_wait(struct call_single_data *data)
 {
@@ -133,7 +135,8 @@ void generic_smp_call_function_interrupt(void)
 			 */
 			smp_wmb();
 			data->csd.flags &= ~CSD_FLAG_WAIT;
-		} else
+		}
+		if (data->csd.flags & CSD_FLAG_ALLOC)
 			call_rcu(&data->rcu_head, rcu_free_call_data);
 	}
 	rcu_read_unlock();
@@ -258,6 +261,42 @@ void __smp_call_function_single(int cpu, struct call_single_data *data)
 	generic_exec_single(cpu, data);
 }
 
+/* Dummy function */
+static void quiesce_dummy(void *unused)
+{
+}
+
+/*
+ * Ensure stack based data used in call function mask is safe to free.
+ *
+ * This is needed by smp_call_function_mask when using on-stack data, because
+ * a single call function queue is shared by all CPUs, and any CPU may pick up
+ * the data item on the queue at any time before it is deleted. So we need to
+ * ensure that all CPUs have transitioned through a quiescent state after
+ * this call.
+ *
+ * This is a very slow function, implemented by sending synchronous IPIs to
+ * all possible CPUs. For this reason, we have to alloc data rather than use
+ * stack based data even in the case of synchronous calls. The stack based
+ * data is then just used for deadlock/oom fallback which will be very rare.
+ *
+ * If a faster scheme can be made, we could go back to preferring stack based
+ * data -- the data allocation/free is non-zero cost.
+ */
+static void smp_call_function_mask_quiesce_stack(cpumask_t mask)
+{
+	struct call_single_data data;
+	int cpu;
+
+	data.func = quiesce_dummy;
+	data.info = NULL;
+
+	for_each_cpu_mask(cpu, mask) {
+		data.flags = CSD_FLAG_WAIT;
+		generic_exec_single(cpu, &data);
+	}
+}
+
 /**
  * smp_call_function_mask(): Run a function on a set of other CPUs.
  * @mask: The set of cpus to run on.
@@ -283,6 +322,7 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
 	cpumask_t allbutself;
 	unsigned long flags;
 	int cpu, num_cpus;
+	int slowpath = 0;
 
 	/* Can deadlock when called with interrupts disabled */
 	WARN_ON(irqs_disabled());
@@ -304,15 +344,16 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
 		return smp_call_function_single(cpu, func, info, wait);
 	}
 
-	if (!wait) {
-		data = kmalloc(sizeof(*data), GFP_ATOMIC);
-		if (data)
-			data->csd.flags = CSD_FLAG_ALLOC;
-	}
-	if (!data) {
+	data = kmalloc(sizeof(*data), GFP_ATOMIC);
+	if (data) {
+		data->csd.flags = CSD_FLAG_ALLOC;
+		if (wait)
+			data->csd.flags |= CSD_FLAG_WAIT;
+	} else {
 		data = &d;
 		data->csd.flags = CSD_FLAG_WAIT;
 		wait = 1;
+		slowpath = 1;
 	}
 
 	spin_lock_init(&data->lock);
@@ -329,8 +370,11 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info,
 	arch_send_call_function_ipi(mask);
 
 	/* optionally wait for the CPUs to complete */
-	if (wait)
+	if (wait) {
 		csd_flag_wait(&data->csd);
+		if (unlikely(slowpath))
+			smp_call_function_mask_quiesce_stack(mask);
+	}
 
 	return 0;
 }