Cycles: Split BVH nodes storage into inner and leaf nodes

This way we can get rid of inefficient memory usage caused by BVH boundbox
part being unused by leaf nodes but still being allocated for them. Doing
such split allows to save 6 of float4 values for QBVH per leaf node and 3
of float4 values for regular BVH per leaf node.

This translates into following memory save using 01.01.01.G rendered
without hair:

                   Device memory size   Device memory peak   Global memory peak
Before the patch:  4957                 5051                 7668
With the patch:    4467                 4562                 7332

The measurements are done against current master. Still need to run speed tests
and it's hard to predict if it's faster or not: on the one hand leaf nodes are
now much more coherent in cache, on the other hand they're not so much coherent
with regular nodes anymore.

Reviewers: brecht, juicyfruit

Subscribers: venomgfx, eyecandy

Differential Revision: https://developer.blender.org/D1236

1 files changed, 1 insertions, 1 deletions

diff --git a/intern/cycles/kernel/geom/geom_bvh_subsurface.h b/intern/cycles/kernel/geom/geom_bvh_subsurface.h
index 73d392b2ea6..a73139f9c88 100644
--- a/intern/cycles/kernel/geom/geom_bvh_subsurface.h
+++ b/intern/cycles/kernel/geom/geom_bvh_subsurface.h
@@ -187,7 +187,7 @@ ccl_device uint BVH_FUNCTION_FULL_NAME(BVH)(KernelGlobals *kg,
 
 			/* if node is leaf, fetch triangle list */
 			if(nodeAddr < 0) {
-				float4 leaf = kernel_tex_fetch(__bvh_nodes, (-nodeAddr-1)*BVH_NODE_SIZE+3);
+				float4 leaf = kernel_tex_fetch(__bvh_leaf_nodes, (-nodeAddr-1)*BVH_NODE_LEAF_SIZE);
 				int primAddr = __float_as_int(leaf.x);
 
 #if BVH_FEATURE(BVH_INSTANCING)