Workbench: Cleanups & Simplifications

* Move the curvature computation to the cavity pass: One can argue it's not
  the best performance wise (it gets a tiny perf pernalty if it is done
  alone without the ssao), but it make the code cleaner and reduce
  considerably the number of shader variation possible.
* Lower shader variation to 2^8 instead of 2^12

1 files changed, 63 insertions, 34 deletions

diff --git a/source/blender/draw/engines/workbench/workbench_deferred.c b/source/blender/draw/engines/workbench/workbench_deferred.c
index b801181e1fc..1ca76da672a 100644
--- a/source/blender/draw/engines/workbench/workbench_deferred.c
+++ b/source/blender/draw/engines/workbench/workbench_deferred.c
@@ -59,7 +59,7 @@
 static struct {
 	struct GPUShader *prepass_sh_cache[MAX_SHADERS];
 	struct GPUShader *composite_sh_cache[MAX_SHADERS];
-	struct GPUShader *cavity_sh;
+	struct GPUShader *cavity_sh[MAX_CAVITY_SHADERS];
 	struct GPUShader *ghost_resolve_sh;
 	struct GPUShader *shadow_fail_sh;
 	struct GPUShader *shadow_fail_manifold_sh;
@@ -167,13 +167,23 @@ static char *workbench_build_prepass_vert(bool is_hair)
 	return str;
 }
 
-static char *workbench_build_cavity_frag(void)
+static char *workbench_build_cavity_frag(bool cavity, bool curvature, bool high_dpi)
 {
 	char *str = NULL;
 
 	DynStr *ds = BLI_dynstr_new();
 
+	if (cavity) {
+		BLI_dynstr_append(ds, "#define USE_CAVITY\n");
+	}
+	if (curvature) {
+		BLI_dynstr_append(ds, "#define USE_CURVATURE\n");
+	}
+	if (high_dpi) {
+		BLI_dynstr_append(ds, "#define CURVATURE_OFFSET 2\n");
+	}
 	BLI_dynstr_append(ds, datatoc_workbench_common_lib_glsl);
+	BLI_dynstr_append(ds, datatoc_workbench_curvature_lib_glsl);
 	BLI_dynstr_append(ds, datatoc_workbench_cavity_frag_glsl);
 	BLI_dynstr_append(ds, datatoc_workbench_cavity_lib_glsl);
 
@@ -182,6 +192,23 @@ static char *workbench_build_cavity_frag(void)
 	return str;
 }
 
+static GPUShader *workbench_cavity_shader_get(bool cavity, bool curvature)
+{
+	const bool high_dpi = (U.pixelsize > 1.5f);
+	int index = 0;
+	SET_FLAG_FROM_TEST(index, cavity, 1 << 0);
+	SET_FLAG_FROM_TEST(index, curvature, 1 << 1);
+	SET_FLAG_FROM_TEST(index, high_dpi, 1 << 2);
+
+	GPUShader **sh = &e_data.cavity_sh[index];
+	if (*sh == NULL) {
+		char *cavity_frag = workbench_build_cavity_frag(cavity, curvature, high_dpi);
+		*sh = DRW_shader_create_fullscreen(cavity_frag, NULL);
+		MEM_freeN(cavity_frag);
+	}
+	return *sh;
+}
+
 static void ensure_deferred_shaders(WORKBENCH_PrivateData *wpd, int index, bool use_textures, bool is_hair)
 {
 	if (e_data.prepass_sh_cache[index] == NULL) {
@@ -303,6 +330,7 @@ void workbench_deferred_engine_init(WORKBENCH_Data *vedata)
 #else
 		const char *shadow_frag = datatoc_gpu_shader_depth_only_frag_glsl;
 #endif
+		/* TODO only compile on demand */
 		e_data.shadow_pass_sh = DRW_shader_create(
 		        datatoc_workbench_shadow_vert_glsl,
 		        datatoc_workbench_shadow_geom_glsl,
@@ -337,10 +365,6 @@ void workbench_deferred_engine_init(WORKBENCH_Data *vedata)
 		        shadow_frag,
 		        "#define SHADOW_FAIL\n");
 
-		char *cavity_frag = workbench_build_cavity_frag();
-		e_data.cavity_sh = DRW_shader_create_fullscreen(cavity_frag, NULL);
-		MEM_freeN(cavity_frag);
-
 		e_data.ghost_resolve_sh = DRW_shader_create_fullscreen(datatoc_workbench_ghost_resolve_frag_glsl, NULL);
 	}
 	workbench_volume_engine_init();
@@ -353,21 +377,13 @@ void workbench_deferred_engine_init(WORKBENCH_Data *vedata)
 	{
 		const float *viewport_size = DRW_viewport_size_get();
 		const int size[2] = {(int)viewport_size[0], (int)viewport_size[1]};
+		const GPUTextureFormat nor_tex_format = NORMAL_ENCODING_ENABLED() ? GPU_RG16 : GPU_RGBA32F;
 		e_data.object_id_tx = DRW_texture_pool_query_2D(size[0], size[1], GPU_R32UI, &draw_engine_workbench_solid);
 		e_data.color_buffer_tx = DRW_texture_pool_query_2D(size[0], size[1], GPU_RGBA8, &draw_engine_workbench_solid);
-		e_data.cavity_buffer_tx = DRW_texture_pool_query_2D(size[0], size[1], GPU_RG16, &draw_engine_workbench_solid);
+		e_data.normal_buffer_tx = DRW_texture_pool_query_2D(size[0], size[1], nor_tex_format, &draw_engine_workbench_solid);
+		e_data.cavity_buffer_tx = DRW_texture_pool_query_2D(size[0], size[1], GPU_R16, &draw_engine_workbench_solid);
 		e_data.specular_buffer_tx = DRW_texture_pool_query_2D(size[0], size[1], GPU_RGBA8, &draw_engine_workbench_solid);
-		e_data.composite_buffer_tx = DRW_texture_pool_query_2D(
-		        size[0], size[1], GPU_RGBA16F, &draw_engine_workbench_solid);
-
-		if (NORMAL_ENCODING_ENABLED()) {
-			e_data.normal_buffer_tx = DRW_texture_pool_query_2D(
-			        size[0], size[1], GPU_RG16, &draw_engine_workbench_solid);
-		}
-		else {
-			e_data.normal_buffer_tx = DRW_texture_pool_query_2D(
-			        size[0], size[1], GPU_RGBA32F, &draw_engine_workbench_solid);
-		}
+		e_data.composite_buffer_tx = DRW_texture_pool_query_2D(size[0], size[1], GPU_RGBA16F, &draw_engine_workbench_solid);
 
 		GPU_framebuffer_ensure_config(&fbl->prepass_fb, {
 			GPU_ATTACHMENT_TEXTURE(dtxl->depth),
@@ -438,21 +454,30 @@ void workbench_deferred_engine_init(WORKBENCH_Data *vedata)
 		workbench_aa_create_pass(vedata, &e_data.color_buffer_tx);
 	}
 
-	{
+	if (SSAO_ENABLED(wpd) || CURVATURE_ENABLED(wpd)) {
 		int state = DRW_STATE_WRITE_COLOR;
+		GPUShader *shader = workbench_cavity_shader_get(SSAO_ENABLED(wpd), CURVATURE_ENABLED(wpd));
 		psl->cavity_pass = DRW_pass_create("Cavity", state);
-		DRWShadingGroup *grp = DRW_shgroup_create(e_data.cavity_sh, psl->cavity_pass);
-		DRW_shgroup_uniform_texture_ref(grp, "depthBuffer", &dtxl->depth);
-		DRW_shgroup_uniform_texture_ref(grp, "colorBuffer", &e_data.color_buffer_tx);
+		DRWShadingGroup *grp = DRW_shgroup_create(shader, psl->cavity_pass);
 		DRW_shgroup_uniform_texture_ref(grp, "normalBuffer", &e_data.normal_buffer_tx);
-
-		DRW_shgroup_uniform_vec2(grp, "invertedViewportSize", DRW_viewport_invert_size_get(), 1);
-		DRW_shgroup_uniform_vec4(grp, "viewvecs[0]", (float *)wpd->viewvecs, 3);
-		DRW_shgroup_uniform_vec4(grp, "ssao_params", wpd->ssao_params, 1);
-		DRW_shgroup_uniform_vec4(grp, "ssao_settings", wpd->ssao_settings, 1);
-		DRW_shgroup_uniform_mat4(grp, "WinMatrix", wpd->winmat);
-		DRW_shgroup_uniform_texture(grp, "ssao_jitter", e_data.jitter_tx);
 		DRW_shgroup_uniform_block(grp, "samples_block", e_data.sampling_ubo);
+
+		if (SSAO_ENABLED(wpd)) {
+			DRW_shgroup_uniform_texture_ref(grp, "depthBuffer", &dtxl->depth);
+			DRW_shgroup_uniform_texture_ref(grp, "colorBuffer", &e_data.color_buffer_tx);
+			DRW_shgroup_uniform_vec2(grp, "invertedViewportSize", DRW_viewport_invert_size_get(), 1);
+			DRW_shgroup_uniform_vec4(grp, "viewvecs[0]", (float *)wpd->viewvecs, 3);
+			DRW_shgroup_uniform_vec4(grp, "ssao_params", wpd->ssao_params, 1);
+			DRW_shgroup_uniform_vec4(grp, "ssao_settings", wpd->ssao_settings, 1);
+			DRW_shgroup_uniform_mat4(grp, "WinMatrix", wpd->winmat);
+			DRW_shgroup_uniform_texture(grp, "ssao_jitter", e_data.jitter_tx);
+		}
+
+		if (CURVATURE_ENABLED(wpd)) {
+			DRW_shgroup_uniform_texture_ref(grp, "objectId", &e_data.object_id_tx);
+			DRW_shgroup_uniform_vec2(grp, "curvature_settings", &wpd->world_data.curvature_ridge, 1);
+		}
+
 		DRW_shgroup_call_add(grp, DRW_cache_fullscreen_quad_get(), NULL);
 	}
 }
@@ -478,7 +503,9 @@ void workbench_deferred_engine_free(void)
 		DRW_SHADER_FREE_SAFE(e_data.prepass_sh_cache[index]);
 		DRW_SHADER_FREE_SAFE(e_data.composite_sh_cache[index]);
 	}
-	DRW_SHADER_FREE_SAFE(e_data.cavity_sh);
+	for (int index = 0; index < MAX_CAVITY_SHADERS; ++index) {
+		DRW_SHADER_FREE_SAFE(e_data.cavity_sh[index]);
+	}
 	DRW_SHADER_FREE_SAFE(e_data.ghost_resolve_sh);
 	DRW_UBO_FREE_SAFE(e_data.sampling_ubo);
 	DRW_TEXTURE_FREE_SAFE(e_data.jitter_tx);
@@ -502,11 +529,13 @@ static void workbench_composite_uniforms(WORKBENCH_PrivateData *wpd, DRWShadingG
 	if (NORMAL_VIEWPORT_COMP_PASS_ENABLED(wpd)) {
 		DRW_shgroup_uniform_texture_ref(grp, "normalBuffer", &e_data.normal_buffer_tx);
 	}
-	if (SSAO_ENABLED(wpd)) {
+	if (SSAO_ENABLED(wpd) || CURVATURE_ENABLED(wpd)) {
 		DRW_shgroup_uniform_texture_ref(grp, "cavityBuffer", &e_data.cavity_buffer_tx);
 	}
-	if (SPECULAR_HIGHLIGHT_ENABLED(wpd) || MATCAP_ENABLED(wpd)) {
+	if (SPECULAR_HIGHLIGHT_ENABLED(wpd)) {
 		DRW_shgroup_uniform_texture_ref(grp, "specularBuffer", &e_data.specular_buffer_tx);
+	}
+	if (SPECULAR_HIGHLIGHT_ENABLED(wpd) || STUDIOLIGHT_TYPE_MATCAP_ENABLED(wpd)) {
 		DRW_shgroup_uniform_vec4(grp, "viewvecs[0]", (float *)wpd->viewvecs, 3);
 	}
 	DRW_shgroup_uniform_block(grp, "world_block", wpd->world_ubo);
@@ -514,7 +543,7 @@ static void workbench_composite_uniforms(WORKBENCH_PrivateData *wpd, DRWShadingG
 
 	if (STUDIOLIGHT_TYPE_MATCAP_ENABLED(wpd)) {
 		BKE_studiolight_ensure_flag(wpd->studio_light, STUDIOLIGHT_EQUIRECT_RADIANCE_GPUTEXTURE);
-		DRW_shgroup_uniform_texture(grp, "matcapImage", wpd->studio_light->equirect_radiance_gputexture );
+		DRW_shgroup_uniform_texture(grp, "matcapImage", wpd->studio_light->equirect_radiance_gputexture);
 	}
 }
 
@@ -902,7 +931,7 @@ void workbench_deferred_draw_scene(WORKBENCH_Data *vedata)
 		DRW_draw_pass(psl->ghost_resolve_pass);
 	}
 
-	if (SSAO_ENABLED(wpd)) {
+	if (SSAO_ENABLED(wpd) || CURVATURE_ENABLED(wpd)) {
 		GPU_framebuffer_bind(fbl->cavity_fb);
 		DRW_draw_pass(psl->cavity_pass);
 	}