EEVEE: Light: 2.5D Culling: Initial implementation

This follows closely the implementation of 2.5D tiled light
culling described in the presentation:
"Improved Culling for Tiled and Clustered Rendering"
from Michal Drobot
http://advances.realtimerendering.com/s2017/2017_Sig_Improved_Culling_final.pdf

I chose the tile + Z binning approach for its high depth range support
and low CPU overhead & low memory consumption compared to the cluster
based culling. The cons is that the culling is a bit less precise in
some aspect but it is quite balanced.

The culling is done by the `Culling` object which is templated to easily
be reused for light probes cullg.

The Z-binning process is described starting from slide 20 in the
reference pdf.

I also implemented a debug pass to visualize false negative (light
culled when they shouldn't) and light evaluation density.
This is useful to detect failure case and hotspot. This could be exposed
as a developper only render pass in the future.

Some optimization of the reference implementation requires extensions
not yet added to GPU module and will be added later.

1 files changed, 12 insertions, 3 deletions

diff --git a/source/blender/draw/engines/eevee/eevee_shader.hh b/source/blender/draw/engines/eevee/eevee_shader.hh
index a492da9bf27..d3f41dae5d1 100644
--- a/source/blender/draw/engines/eevee/eevee_shader.hh
+++ b/source/blender/draw/engines/eevee/eevee_shader.hh
@@ -40,6 +40,10 @@ extern char datatoc_common_view_lib_glsl[];
 
 extern char datatoc_eevee_camera_lib_glsl[];
 extern char datatoc_eevee_camera_velocity_frag_glsl[];
+extern char datatoc_eevee_culling_debug_frag_glsl[];
+extern char datatoc_eevee_culling_iter_lib_glsl[];
+extern char datatoc_eevee_culling_lib_glsl[];
+extern char datatoc_eevee_culling_light_frag_glsl[];
 extern char datatoc_eevee_depth_of_field_accumulator_lib_glsl[];
 extern char datatoc_eevee_depth_of_field_bokeh_lut_frag_glsl[];
 extern char datatoc_eevee_depth_of_field_downsample_frag_glsl[];
@@ -60,7 +64,6 @@ extern char datatoc_eevee_depth_of_field_tiles_flatten_frag_glsl[];
 extern char datatoc_eevee_film_filter_frag_glsl[];
 extern char datatoc_eevee_film_lib_glsl[];
 extern char datatoc_eevee_film_resolve_frag_glsl[];
-extern char datatoc_eevee_light_lib_glsl[];
 extern char datatoc_eevee_motion_blur_gather_frag_glsl[];
 extern char datatoc_eevee_motion_blur_lib_glsl[];
 extern char datatoc_eevee_motion_blur_tiles_dilate_frag_glsl[];
@@ -80,7 +83,10 @@ namespace blender::eevee {
 
 /* Keep alphabetical order and clean prefix. */
 enum eShaderType {
-  DOF_BOKEH_LUT = 0,
+  CULLING_DEBUG = 0,
+  CULLING_LIGHT,
+
+  DOF_BOKEH_LUT,
   DOF_GATHER_BACKGROUND_LUT,
   DOF_GATHER_BACKGROUND,
   DOF_FILTER,
@@ -154,7 +160,8 @@ class ShaderModule {
     DRW_SHADER_LIB_ADD(shader_lib_, common_view_lib);
     DRW_SHADER_LIB_ADD(shader_lib_, eevee_sampling_lib);
     DRW_SHADER_LIB_ADD(shader_lib_, eevee_camera_lib);
-    DRW_SHADER_LIB_ADD(shader_lib_, eevee_light_lib);
+    DRW_SHADER_LIB_ADD(shader_lib_, eevee_culling_lib);
+    DRW_SHADER_LIB_ADD(shader_lib_, eevee_culling_iter_lib);
     DRW_SHADER_LIB_ADD(shader_lib_, eevee_velocity_lib);
     DRW_SHADER_LIB_ADD(shader_lib_, eevee_depth_of_field_lib);
     DRW_SHADER_LIB_ADD(shader_lib_, eevee_depth_of_field_accumulator_lib);
@@ -178,6 +185,8 @@ class ShaderModule {
   SHADER(enum_, common_fullscreen_vert, nullptr, frag_, defs_)
 #define SHADER_FULLSCREEN(enum_, frag_) SHADER_FULLSCREEN_DEFINES(enum_, frag_, nullptr)
 
+    SHADER_FULLSCREEN(CULLING_DEBUG, eevee_culling_debug_frag);
+    SHADER_FULLSCREEN(CULLING_LIGHT, eevee_culling_light_frag);
     SHADER_FULLSCREEN(FILM_FILTER, eevee_film_filter_frag);
     SHADER_FULLSCREEN(FILM_RESOLVE, eevee_film_resolve_frag);
     SHADER_FULLSCREEN(DOF_BOKEH_LUT, eevee_depth_of_field_bokeh_lut_frag);