Cycles: Some more inline policy tweaks for CUDA 8

Makes it so toolkit does exactly the same decision about what to inline,
but unfortunately it has really barely visible difference on GTX-980.

1 files changed, 7 insertions, 7 deletions

diff --git a/intern/cycles/kernel/kernel_path_branched.h b/intern/cycles/kernel/kernel_path_branched.h
index 56516967d8f..64f1468eacf 100644
--- a/intern/cycles/kernel/kernel_path_branched.h
+++ b/intern/cycles/kernel/kernel_path_branched.h
@@ -18,13 +18,13 @@ CCL_NAMESPACE_BEGIN
 
 #ifdef __BRANCHED_PATH__
 
-ccl_device void kernel_branched_path_ao(KernelGlobals *kg,
-                                        ShaderData *sd,
-                                        ShaderData *emission_sd,
-                                        PathRadiance *L,
-                                        PathState *state,
-                                        RNG *rng,
-                                        float3 throughput)
+ccl_device_inline void kernel_branched_path_ao(KernelGlobals *kg,
+                                               ShaderData *sd,
+                                               ShaderData *emission_sd,
+                                               PathRadiance *L,
+                                               PathState *state,
+                                               RNG *rng,
+                                               float3 throughput)
 {
 	int num_samples = kernel_data.integrator.ao_samples;
 	float num_samples_inv = 1.0f/num_samples;