Outline SkSpinlock::acquire().

The proportion of time spent doing useful work is well over 99% in acquire(),
so outlining it doesn't hurt speed at all, and makes it much easier to pick out
on a profile.

It'd be about 50/50 work/overhead if we outlined the extremely-cheap release().
I also tried outlining some SkRefCnt methods with similar mixed results.

BUG=skia:

No public API changes.
TBR=reed@google.com

Review URL: https://codereview.chromium.org/1212253013

1 files changed, 13 insertions, 0 deletions

diff --git a/src/core/SkSpinlock.cpp b/src/core/SkSpinlock.cpp
new file mode 100644
index 0000000000..0f764278df
--- /dev/null
+++ b/src/core/SkSpinlock.cpp
@@ -0,0 +1,13 @@
+/*
+ * Copyright 2015 Google Inc.
+ *
+ * Use of this source code is governed by a BSD-style license that can be
+ * found in the LICENSE file.
+ */
+
+#include "SkSpinlock.h"
+
+void SkPODSpinlock::contendedAcquire() {
+    // To act as a mutex, we need an acquire barrier when we take the lock.
+    while(sk_atomic_exchange(&fLocked, true, sk_memory_order_acquire)) { /*spin*/ }
+}