xen: use stronger barrier after unlocking lock

We need to have a stronger barrier between releasing the lock and
checking for any waiting spinners.  A compiler barrier is not sufficient
because the CPU's ordering rules do not prevent the read xl->spinners
from happening before the unlock assignment, as they are different
memory locations.

We need to have an explicit barrier to enforce the write-read ordering
to different memory locations.

Because of it, I can't bring up > 4 HVM guests on one SMP machine.

[ Code and commit comments expanded -J ]

[ Impact: avoid deadlock when using Xen PV spinlocks ]

Signed-off-by: Yang Xiaowei <xiaowei.yang@intel.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

1 files changed, 7 insertions, 2 deletions

diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index 2f91e56..36a5141 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -326,8 +326,13 @@ static void xen_spin_unlock(struct raw_spinlock *lock)
 	smp_wmb();		/* make sure no writes get moved after unlock */
 	xl->lock = 0;		/* release lock */
 
-	/* make sure unlock happens before kick */
-	barrier();
+	/*
+	 * Make sure unlock happens before checking for waiting
+	 * spinners.  We need a strong barrier to enforce the
+	 * write-read ordering to different memory locations, as the
+	 * CPU makes no implied guarantees about their ordering.
+	 */
+	mb();
 
 	if (unlikely(xl->spinners))
 		xen_spin_unlock_slow(xl);