Let synchronized painting generate correct paint invalidation rects

third_party/WebKit/Source/platform/graphics/paint/PaintController.h

Index: third_party/WebKit/Source/platform/graphics/paint/PaintController.h
diff --git a/third_party/WebKit/Source/platform/graphics/paint/PaintController.h b/third_party/WebKit/Source/platform/graphics/paint/PaintController.h
index 362326283e9e6eef30159a4b5bf83bd36f706ad2..a9d91a2fe116acb50917d17d948be95784d640a7 100644
--- a/third_party/WebKit/Source/platform/graphics/paint/PaintController.h
+++ b/third_party/WebKit/Source/platform/graphics/paint/PaintController.h
@@ -25,7 +25,6 @@
 
 namespace blink {
 
-class GraphicsLayer;
 class GraphicsContext;
 
 static const size_t kInitialDisplayItemListCapacityBytes = 512;
@@ -42,8 +41,11 @@ public:
         return adoptPtr(new PaintController());
     }
 
-    // These methods are called during paint invalidation (or paint if SlimmingPaintSynchronizedPainting is on).
-    void invalidate(const DisplayItemClientWrapper&, PaintInvalidationReason, const IntRect& previousPaintInvalidationRect, const IntRect& newPaintInvalidationRect);
+    // These methods are called during paint invalidation (or paint if SlimmingPaintV2 is on).
+    // |visualContentRect| can be nullptr if we know it's unchanged and PaintController has cached the previous value.
+    // Otherwise we must pass a correct value (bounds of the display item client in coordinates of the

[chrishtr, 2015/10/23 19:57:49]

"bounds" needs to be defined here. I suggest just saying this is a rect in the
space of the GraphicsLayer into which the DisplayItem paints which contains all
pixels within the GraphicsLayer which might be painted into by the DisplayItem.

Also, danakj explained to me that this should be visualLayerRect, not
visualContentRect. "content" in cc means the space that takes into account pinch
zoom and such; "layer" is in the space of the cc::Layer (GraphicsLayer).

[Xianzhu, 2015/10/23 20:56:53]

Done.
I think visualLayerRect looks confusing to blink people and needs more
explanation and reference to cc to be understood. How about avoiding 'content',
just 'visualRect', or 'visualBoundingRect'?

[chrishtr, 2015/10/23 20:59:45]

visualRect is ok then, with comment.

[Xianzhu, 2015/10/23 21:28:22]

Done.

+    // PaintController's GraphicsLayer (spv1) or TODO(pdr): definition of it for spv2.
+    void invalidate(const DisplayItemClientWrapper&, PaintInvalidationReason, const IntRect* visibleContentRect);
     void invalidateUntracked(DisplayItemClient);
     void invalidateAll();
 
@@ -103,9 +105,8 @@ public:
     void endSkippingCache() { ASSERT(m_skippingCacheCount > 0); --m_skippingCacheCount; }
     bool skippingCache() const { return m_skippingCacheCount; }
 
-    // Must be called when a painting is finished. If passed, invalidations are recorded on the given
-    // GraphicsLayer.
-    void commitNewDisplayItems(GraphicsLayer* = nullptr);
+    // Must be called when a painting is finished.
+    void commitNewDisplayItems();
 
     // Returns the approximate memory usage, excluding memory likely to be
     // shared with the embedder after copying to WebPaintController.
@@ -242,14 +243,10 @@ private:
     unsigned m_nextScope;
     Vector<unsigned> m_scopeStack;
 
-    struct Invalidation {
-        IntRect rect;
-        PaintInvalidationReason invalidationReason;
-    };
-
-    Vector<Invalidation> m_invalidations;
-
 #if ENABLE(ASSERT)
+    // Record the debug names of invalidated clients for assertion and debugging.
+    Vector<String> m_invalidations;
+
     // This is used to check duplicated ids during add(). We could also check
     // during commitNewDisplayItems(), but checking during add() helps developer
     // easily find where the duplicated ids are from.