Cycles: Speedup of Christensen-Burley SSS falloff function

The idea is simply to pre-compute fitting and parameterization
in the bssrdf_setup() function and re-use the values in both
sample() and eval().

The only trick is where to store the pre-calculated values and
the answer is inside of ShaderClosure->custom{1,2,3}. There's
no memory bump here because we now simply re-use padding fields
for the pre-calculated values. Similar trick we can do for other
BSDFs.

Seems to give nice speedup up to 7% here on my desktop with
Core i7 CPU, SSE4.1 kernel.

1 files changed, 8 insertions, 1 deletions

diff --git a/intern/cycles/kernel/kernel_types.h b/intern/cycles/kernel/kernel_types.h
index 93ff732ca2f..42d40e1daee 100644
--- a/intern/cycles/kernel/kernel_types.h
+++ b/intern/cycles/kernel/kernel_types.h
@@ -642,7 +642,14 @@ typedef ccl_addr_space struct ShaderClosure {
 	float data0;
 	float data1;
 	float data2;
-	int pad1, pad2, pad3;
+
+	/* Following fields could be used to store pre-calculated
+	 * values by various BSDF closures for more effective sampling
+	 * and evaluation.
+	 */
+	float custom1;
+	float custom2;
+	float custom3;
 
 #ifdef __OSL__
 	void *prim, *pad4;