[SurfaceFlinger] Update comments to clarify transform.
Clarify the source coordinate space and destination coordinate space for
transforms.
Bug: N/A
Test: N/A
Change-Id: I93f6770f8a1f03f81e30e42a80c3196ccb1b0fee
diff --git a/services/surfaceflinger/Layer.cpp b/services/surfaceflinger/Layer.cpp
index d9faec4..a79fddf 100644
--- a/services/surfaceflinger/Layer.cpp
+++ b/services/surfaceflinger/Layer.cpp
@@ -2453,6 +2453,7 @@
ySurfaceInset = std::round(ySurfaceInset * yScale);
}
+ // Transform the layer bounds from layer coordinate space to display coordinate space.
Rect transformedLayerBounds = t.transform(layerBounds);
// clamp inset to layer bounds
@@ -2477,7 +2478,7 @@
transformedLayerBounds.bottom = tmp;
}
- // Input coordinate should match the layer bounds.
+ // Input coordinates should be in display coordinate space.
info.frameLeft = transformedLayerBounds.left;
info.frameTop = transformedLayerBounds.top;
info.frameRight = transformedLayerBounds.right;
@@ -2490,7 +2491,7 @@
// the final frame calculated.
// 1. Take the original transform set on the window and get the inverse transform. This is
// used to get the final bounds in display space (ignorning the transform). Apply the
- // inverse transform on the layerBounds to get the untransformed frame (in display space)
+ // inverse transform on the layerBounds to get the untransformed frame (in layer space)
// 2. Take the top and left of the untransformed frame to get the real position on screen.
// Apply the layer transform on top/left so it includes any scale or rotation. These will
// be the new translation values for the transform.