From: Erik Gilling <konkers@android.com>
Date: Fri, 1 Mar 2013 00:43:18 +0000 (-0800)
Subject: staging: sync: Protect unlocked access to fence status
X-Git-Url: http://git.cdn.openwrt.org/?a=commitdiff_plain;h=c679212dbfd060513e156133326122bf9f496579;p=openwrt%2Fstaging%2Fblogic.git

staging: sync: Protect unlocked access to fence status

Fence status is checked outside of locks in both sync_fence_wait and
sync_fence_poll.  This patch adds propper barrier protection in these
cases to avoid seeing stale status.

Cc: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Cc: Erik Gilling <konkers@android.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Rob Clark <robclark@gmail.com>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: dri-devel@lists.freedesktop.org
Cc: Android Kernel Team <kernel-team@android.com>
Signed-off-by: Erik Gilling <konkers@android.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---

diff --git a/drivers/staging/android/sync.c b/drivers/staging/android/sync.c
index 36ffa2024875..2394189c5958 100644
--- a/drivers/staging/android/sync.c
+++ b/drivers/staging/android/sync.c
@@ -556,6 +556,16 @@ int sync_fence_cancel_async(struct sync_fence *fence,
 }
 EXPORT_SYMBOL(sync_fence_cancel_async);
 
+static bool sync_fence_check(struct sync_fence *fence)
+{
+	/*
+	 * Make sure that reads to fence->status are ordered with the
+	 * wait queue event triggering
+	 */
+	smp_rmb();
+	return fence->status != 0;
+}
+
 int sync_fence_wait(struct sync_fence *fence, long timeout)
 {
 	int err = 0;
@@ -563,7 +573,7 @@ int sync_fence_wait(struct sync_fence *fence, long timeout)
 	if (timeout > 0) {
 		timeout = msecs_to_jiffies(timeout);
 		err = wait_event_interruptible_timeout(fence->wq,
-						       fence->status != 0,
+						       sync_fence_check(fence),
 						       timeout);
 	} else if (timeout < 0) {
 		err = wait_event_interruptible(fence->wq, fence->status != 0);
@@ -630,6 +640,12 @@ static unsigned int sync_fence_poll(struct file *file, poll_table *wait)
 
 	poll_wait(file, &fence->wq, wait);
 
+	/*
+	 * Make sure that reads to fence->status are ordered with the
+	 * wait queue event triggering
+	 */
+	smp_rmb();
+
 	if (fence->status == 1)
 		return POLLIN;
 	else if (fence->status < 0)