Cycles: support for custom shader AOVs

Custom render passes are added in the Shader AOVs panel in the view layer
settings, with a name and data type. In shader nodes, an AOV Output node
is then used to output either a value or color to the pass.

Arbitrary names can be used for these passes, as long as they don't conflict
with built-in passes that are enabled. The AOV Output node can be used in both
material and world shader nodes.

Implemented by Lukas, with tweaks by Brecht.

Differential Revision: https://developer.blender.org/D4837

2 files changed, 6 insertions, 2 deletions

diff --git a/intern/cycles/kernel/split/kernel_indirect_background.h b/intern/cycles/kernel/split/kernel_indirect_background.h
index b1c65f61e2c..6d500650cc0 100644
--- a/intern/cycles/kernel/split/kernel_indirect_background.h
+++ b/intern/cycles/kernel/split/kernel_indirect_background.h
@@ -58,8 +58,10 @@ ccl_device void kernel_indirect_background(KernelGlobals *kg)
     ccl_global Ray *ray = &kernel_split_state.ray[ray_index];
     float3 throughput = kernel_split_state.throughput[ray_index];
     ShaderData *sd = kernel_split_sd(sd, ray_index);
+    uint buffer_offset = kernel_split_state.buffer_offset[ray_index];
+    ccl_global float *buffer = kernel_split_params.tile.buffer + buffer_offset;
 
-    kernel_path_background(kg, state, ray, throughput, sd, L);
+    kernel_path_background(kg, state, ray, throughput, sd, buffer, L);
     kernel_split_path_end(kg, ray_index);
   }
 }
diff --git a/intern/cycles/kernel/split/kernel_shader_eval.h b/intern/cycles/kernel/split/kernel_shader_eval.h
index 8e39c9797e5..c760a2b2049 100644
--- a/intern/cycles/kernel/split/kernel_shader_eval.h
+++ b/intern/cycles/kernel/split/kernel_shader_eval.h
@@ -50,8 +50,10 @@ ccl_device void kernel_shader_eval(KernelGlobals *kg)
   ccl_global char *ray_state = kernel_split_state.ray_state;
   if (IS_STATE(ray_state, ray_index, RAY_ACTIVE)) {
     ccl_global PathState *state = &kernel_split_state.path_state[ray_index];
+    uint buffer_offset = kernel_split_state.buffer_offset[ray_index];
+    ccl_global float *buffer = kernel_split_params.tile.buffer + buffer_offset;
 
-    shader_eval_surface(kg, kernel_split_sd(sd, ray_index), state, state->flag);
+    shader_eval_surface(kg, kernel_split_sd(sd, ray_index), state, buffer, state->flag);
 #ifdef __BRANCHED_PATH__
     if (kernel_data.integrator.branched) {
       shader_merge_closures(kernel_split_sd(sd, ray_index));