Clarify a comment in kvm/eventfd.c: we do
rcu_assign_pointer without explicit synchronize_rcu
afterwards, and the reason this works is because we
use a spinlock to synch against another place
which does call synchronize_rcu.
Signed-off-by: Michael S. Tsirkin <mst@xxxxxxxxxx>
---
diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c
index 2ca4535..e347b53 100644
--- a/virt/kvm/eventfd.c
+++ b/virt/kvm/eventfd.c
@@ -313,8 +313,9 @@ kvm_irqfd_deassign(struct kvm *kvm, int fd, int gsi)
if (irqfd->eventfd == eventfd && irqfd->gsi == gsi) {
/*
* This rcu_assign_pointer is needed for when
- * another thread calls kvm_irqfd_update before
- * we flush workqueue below.
+ * another thread calls kvm_irq_routing_update before
+ * we flush workqueue below (we synchronize with
+ * kvm_irq_routing_update using irqfds.lock).
* It is paired with synchronize_rcu done by caller
* of that function.
*/
--
MST
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
