msm: kgsl: Prevent deadlock in cmdbatch timer

Dispatcher can acquire drawctxt->lock if context is pending
and the fence it is waiting on just got signalled.
Dispatcher acquires drawctxt->lock and tries to delete the
cmdbatch timer using delete_timer_sync(). Delete_timer_sync()
waits till timer and its pending handlers are deleted.
But if the timer expires at the same time, timer handler
could be waiting on drawctxt->lock leading to a
deadlock. To prevent this use spin_trylock_bh() instead of
spin_lock_bh(). spin_trylock_bh() does not wait for the lock
if it does not get it and allows the timer handler to finish.
This prevents the deadlock.

Change-Id: Ic2344fed5fccb581b58ec0b66b45ba68af9f1459
Signed-off-by: Tarun Karra <tkarra@codeaurora.org>

drivers/gpu/msm/adreno_drawctxt.c

@@ -64,8 +64,22 @@ void adreno_drawctxt_dump(struct kgsl_device *device,
 	 * We may have cmdbatch timer running, which also uses same
 	 * lock, take a lock with software interrupt disabled (bh)
 	 * to avoid spin lock recursion.
+	 *
+	 * Use Spin trylock because dispatcher can acquire drawctxt->lock
+	 * if context is pending and the fence it is waiting on just got
+	 * signalled. Dispatcher acquires drawctxt->lock and tries to
+	 * delete the cmdbatch timer using del_timer_sync().
+	 * del_timer_sync() waits till timer and its pending handlers
+	 * are deleted. But if the timer expires at the same time,
+	 * timer handler could be waiting on drawctxt->lock leading to a
+	 * deadlock. To prevent this use spin_trylock_bh.
 	 */
-	spin_lock_bh(&drawctxt->lock);
+	if (!spin_trylock_bh(&drawctxt->lock)) {
+		dev_err(device->dev, "  context[%d]: could not get lock\n",
+			context->id);
+		return;
+	}
+
 	dev_err(device->dev,
 		"  context[%d]: queue=%d, submit=%d, start=%d, retire=%d\n",
 		context->id, queue, drawctxt->submitted_timestamp,