vega: add support to VG_BLEND_SOFTLIGHT_KHR advanced blend mode.

it is a alternative version of softlight blend and it is common used by rendering and svg authoring tools

6 files changed, 125 insertions, 11 deletions

diff --git a/src/gallium/state_trackers/vega/api_params.c b/src/gallium/state_trackers/vega/api_params.c
index 0ddc2514b9..acd450527d 100644
--- a/src/gallium/state_trackers/vega/api_params.c
+++ b/src/gallium/state_trackers/vega/api_params.c
@@ -161,7 +161,7 @@ void vgSeti (VGParamType type, VGint value)
       break;
    case VG_BLEND_MODE:
       if (value < VG_BLEND_SRC ||
-          value > VG_BLEND_SOFTLIGHT_SVG_KHR)
+          value > VG_BLEND_SOFTLIGHT_KHR)
          error = VG_ILLEGAL_ARGUMENT_ERROR;
       else {
          ctx->state.dirty |= BLEND_DIRTY;
diff --git a/src/gallium/state_trackers/vega/asm_fill.h b/src/gallium/state_trackers/vega/asm_fill.h
index acc47733d1..3d7e0ed32b 100644
--- a/src/gallium/state_trackers/vega/asm_fill.h
+++ b/src/gallium/state_trackers/vega/asm_fill.h
@@ -556,6 +556,108 @@ blend_softlight_svg_khr( struct ureg_program *ureg,
 }
 
 static INLINE void
+blend_softlight_khr( struct ureg_program *ureg,
+                     struct ureg_dst     *out,
+                     struct ureg_src     *in,
+                     struct ureg_src     *sampler,
+                     struct ureg_dst     *temp,
+                     struct ureg_src     *constant)
+{
+   unsigned label;
+   struct ureg_src imm, imm2, imm3, imm4, imm5;
+
+   imm  = ureg_imm4f(ureg, 4, 4, 4, 4);
+   imm2 = ureg_imm4f(ureg, 3, 3, 3, 3);
+   imm3 = ureg_imm4f(ureg, 0.5, 0.5, 0.5, 0.5);
+   imm4 = ureg_imm4f(ureg, 16, 16, 16, 16);
+   imm5 = ureg_imm4f(ureg, 12, 12, 12, 12);
+
+   ureg_TEX(ureg, temp[1], TGSI_TEXTURE_2D, in[0], sampler[2]);
+
+   ureg_ADD(ureg, temp[2], ureg_src(temp[0]), ureg_src(temp[0]));
+
+   ureg_SLT(ureg, temp[2], ureg_src(temp[2]),
+            ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W));
+   ureg_MOV(ureg,
+            ureg_writemask(temp[2], TGSI_WRITEMASK_W),
+            ureg_scalar(constant[1], TGSI_SWIZZLE_Y));
+
+   EXTENDED_BLENDER_OVER_FUNC
+
+   label = ureg_get_instruction_number(ureg);
+   label += 2;
+
+   ureg_IF(ureg, ureg_src(temp[2]), &label);
+   ureg_DIV(ureg, temp[2], ureg_src(temp[1]),
+            ureg_scalar(ureg_src(temp[1]), TGSI_SWIZZLE_W));
+   ureg_SUB(ureg, temp[2], ureg_scalar(constant[1], TGSI_SWIZZLE_Y),
+            ureg_src(temp[2]));
+   ureg_ADD(ureg, temp[4], ureg_src(temp[0]), ureg_src(temp[0]));
+   ureg_SUB(ureg, temp[4], ureg_src(temp[4]),
+            ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W));
+   ureg_MUL(ureg, temp[2], ureg_src(temp[2]), ureg_src(temp[4]));
+   ureg_SUB(ureg, temp[2], ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W),
+            ureg_src(temp[2]));
+   ureg_MUL(ureg, temp[2], ureg_src(temp[1]), ureg_src(temp[2]));
+   ureg_ADD(ureg, temp[1], ureg_src(temp[2]), ureg_src(temp[3]));
+
+   label = ureg_get_instruction_number(ureg);
+   label += 2;
+   ureg_ELSE(ureg, &label);
+
+   ureg_MUL(ureg, temp[2], ureg_src(temp[1]), imm);
+   ureg_SLE(ureg, temp[2], ureg_src(temp[2]),
+            ureg_scalar(ureg_src(temp[1]), TGSI_SWIZZLE_W));
+
+   label = ureg_get_instruction_number(ureg);
+   label += 2;
+
+   ureg_IF(ureg, ureg_src(temp[2]), &label);
+   ureg_MUL(ureg, temp[2], imm4, ureg_src(temp[1]));
+   ureg_DIV(ureg, temp[2], ureg_src(temp[2]), ureg_scalar(ureg_src(temp[1]), TGSI_SWIZZLE_W));
+   ureg_SUB(ureg, temp[2], ureg_src(temp[2]), imm5);
+   ureg_MUL(ureg, temp[2], ureg_src(temp[2]), ureg_src(temp[1]));
+   ureg_DIV(ureg, temp[2], ureg_src(temp[2]), ureg_scalar(ureg_src(temp[1]), TGSI_SWIZZLE_W));
+   ureg_ADD(ureg, temp[2], ureg_src(temp[2]), imm2);
+   ureg_ADD(ureg, temp[4], ureg_src(temp[0]), ureg_src(temp[0]));
+   ureg_SUB(ureg, temp[4], ureg_src(temp[4]), ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W));
+   ureg_MUL(ureg, temp[2], ureg_src(temp[2]), ureg_src(temp[4]));
+   ureg_MUL(ureg, temp[2], ureg_src(temp[2]), ureg_src(temp[1]));
+   ureg_MUL(ureg, temp[4], ureg_src(temp[1]), ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W));
+   ureg_ADD(ureg, temp[2], ureg_src(temp[2]), ureg_src(temp[4]));
+   ureg_ADD(ureg, temp[1], ureg_src(temp[2]), ureg_src(temp[3]));
+
+   label = ureg_get_instruction_number(ureg);
+   label += 2;
+
+   ureg_ELSE(ureg, &label);
+   ureg_MUL(ureg, temp[2], ureg_src(temp[1]), ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W));
+   ureg_DIV(ureg, temp[4], ureg_src(temp[1]), ureg_scalar(ureg_src(temp[1]), TGSI_SWIZZLE_W));
+   ureg_POW(ureg, temp[4], ureg_src(temp[4]), imm3);
+   ureg_MUL(ureg, temp[4], ureg_src(temp[4]), ureg_scalar(ureg_src(temp[1]), TGSI_SWIZZLE_W));
+   ureg_SUB(ureg, temp[4], ureg_src(temp[4]), ureg_src(temp[1]));
+   ureg_ADD(ureg, temp[5], ureg_src(temp[0]), ureg_src(temp[0]));
+   ureg_SUB(ureg, temp[5], ureg_src(temp[5]), ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W));
+   ureg_MUL(ureg, temp[4], ureg_src(temp[4]), ureg_src(temp[5]));
+   ureg_ADD(ureg, temp[2], ureg_src(temp[2]), ureg_src(temp[4]));
+   ureg_ADD(ureg, ureg_writemask(temp[1], TGSI_WRITEMASK_XYZ), ureg_src(temp[2]), ureg_src(temp[3]));
+   ureg_ENDIF(ureg);
+
+   ureg_ENDIF(ureg);
+
+   ureg_MUL(ureg, temp[2], ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W),
+            ureg_scalar(ureg_src(temp[1]), TGSI_SWIZZLE_W));
+   ureg_ADD(ureg, temp[3], ureg_scalar(ureg_src(temp[0]), TGSI_SWIZZLE_W),
+            ureg_scalar(ureg_src(temp[1]), TGSI_SWIZZLE_W));
+   ureg_SUB(ureg, ureg_writemask(temp[1], TGSI_WRITEMASK_W),
+            ureg_src(temp[3]), ureg_src(temp[2]));
+
+
+   ureg_MOV(ureg, *out, ureg_src(temp[1]));
+}
+
+
+static INLINE void
 premultiply( struct ureg_program *ureg,
                 struct ureg_dst *out,
                 struct ureg_src *in,
@@ -668,6 +770,8 @@ static const struct shader_asm_info shaders_asm[] = {
     VG_TRUE,  1, 1, 2, 1, 0, 6},
    {VEGA_BLEND_SOFTLIGHT_SVG_KHR_SHADER, blend_softlight_svg_khr,
     VG_TRUE,  1, 1, 2, 1, 0, 6},
+   {VEGA_BLEND_SOFTLIGHT_KHR_SHADER, blend_softlight_khr,
+    VG_TRUE,  1, 1, 2, 1, 0, 6},
 
    /* premultiply */
    {VEGA_PREMULTIPLY_SHADER, premultiply,
diff --git a/src/gallium/state_trackers/vega/shader.c b/src/gallium/state_trackers/vega/shader.c
index 432c2a8e81..66af2cad06 100644
--- a/src/gallium/state_trackers/vega/shader.c
+++ b/src/gallium/state_trackers/vega/shader.c
@@ -129,7 +129,8 @@ static VGint blend_bind_samplers(struct vg_context *ctx,
        bmode == VG_BLEND_LIGHTEN ||
        bmode == VG_BLEND_OVERLAY_KHR ||
        bmode == VG_BLEND_HARDLIGHT_KHR ||
-       bmode == VG_BLEND_SOFTLIGHT_SVG_KHR) {
+       bmode == VG_BLEND_SOFTLIGHT_SVG_KHR ||
+       bmode == VG_BLEND_SOFTLIGHT_KHR) {
       struct st_framebuffer *stfb = ctx->draw_buffer;
 
       vg_prepare_blend_surface(ctx);
@@ -273,6 +274,9 @@ static void setup_shader_program(struct shader *shader)
    case VG_BLEND_SOFTLIGHT_SVG_KHR:
       shader_id |= VEGA_BLEND_SOFTLIGHT_SVG_KHR_SHADER;
       break;
+   case VG_BLEND_SOFTLIGHT_KHR:
+      shader_id |= VEGA_BLEND_SOFTLIGHT_KHR_SHADER;
+      break;
    default:
       /* handled by pipe_blend_state */
       break;
diff --git a/src/gallium/state_trackers/vega/shaders_cache.c b/src/gallium/state_trackers/vega/shaders_cache.c
index 30deeca7dc..d1eaa5bad6 100644
--- a/src/gallium/state_trackers/vega/shaders_cache.c
+++ b/src/gallium/state_trackers/vega/shaders_cache.c
@@ -345,23 +345,27 @@ create_shader(struct pipe_context *pipe,
       debug_assert(shaders_asm[14].id == VEGA_BLEND_SOFTLIGHT_SVG_KHR_SHADER);
       shaders[idx] = &shaders_asm[14];
       ++idx;
+   } else if ((id & VEGA_BLEND_SOFTLIGHT_KHR_SHADER)){
+      debug_assert(shaders_asm[15].id == VEGA_BLEND_SOFTLIGHT_KHR_SHADER);
+      shaders[idx] = &shaders_asm[15];
+      ++idx;
    }
 
    /* fifth stage */
    if ((id & VEGA_PREMULTIPLY_SHADER)) {
-      debug_assert(shaders_asm[15].id == VEGA_PREMULTIPLY_SHADER);
-      shaders[idx] = &shaders_asm[15];
+      debug_assert(shaders_asm[16].id == VEGA_PREMULTIPLY_SHADER);
+      shaders[idx] = &shaders_asm[16];
       ++idx;
    } else if ((id & VEGA_UNPREMULTIPLY_SHADER)) {
-      debug_assert(shaders_asm[16].id == VEGA_UNPREMULTIPLY_SHADER);
-      shaders[idx] = &shaders_asm[16];
+      debug_assert(shaders_asm[17].id == VEGA_UNPREMULTIPLY_SHADER);
+      shaders[idx] = &shaders_asm[17];
       ++idx;
    }
 
    /* sixth stage */
    if ((id & VEGA_BW_SHADER)) {
-      debug_assert(shaders_asm[17].id == VEGA_BW_SHADER);
-      shaders[idx] = &shaders_asm[17];
+      debug_assert(shaders_asm[18].id == VEGA_BW_SHADER);
+      shaders[idx] = &shaders_asm[18];
       ++idx;
    }
 
diff --git a/src/gallium/state_trackers/vega/shaders_cache.h b/src/gallium/state_trackers/vega/shaders_cache.h
index 3495a4c5b9..300e9d0e65 100644
--- a/src/gallium/state_trackers/vega/shaders_cache.h
+++ b/src/gallium/state_trackers/vega/shaders_cache.h
@@ -51,11 +51,12 @@ enum VegaShaderType {
    VEGA_BLEND_OVERLAY_KHR_SHADER       = 1 << 12,
    VEGA_BLEND_HARDLIGHT_KHR_SHADER     = 1 << 13,
    VEGA_BLEND_SOFTLIGHT_SVG_KHR_SHADER = 1 << 14,
+   VEGA_BLEND_SOFTLIGHT_KHR_SHADER     = 1 << 15,
 
-   VEGA_PREMULTIPLY_SHADER             = 1 << 15,
-   VEGA_UNPREMULTIPLY_SHADER           = 1 << 16,
+   VEGA_PREMULTIPLY_SHADER             = 1 << 16,
+   VEGA_UNPREMULTIPLY_SHADER           = 1 << 17,
 
-   VEGA_BW_SHADER                      = 1 << 17
+   VEGA_BW_SHADER                      = 1 << 18
 };
 
 struct vg_shader {
diff --git a/src/gallium/state_trackers/vega/vg_context.c b/src/gallium/state_trackers/vega/vg_context.c
index d6ed464785..779dbf5f3f 100644
--- a/src/gallium/state_trackers/vega/vg_context.c
+++ b/src/gallium/state_trackers/vega/vg_context.c
@@ -342,6 +342,7 @@ void vg_validate_state(struct vg_context *ctx)
       case VG_BLEND_OVERLAY_KHR:
       case VG_BLEND_HARDLIGHT_KHR:
       case VG_BLEND_SOFTLIGHT_SVG_KHR:
+      case VG_BLEND_SOFTLIGHT_KHR:
          blend->rt[0].rgb_src_factor   = PIPE_BLENDFACTOR_ONE;
          blend->rt[0].alpha_src_factor = PIPE_BLENDFACTOR_ONE;
          blend->rt[0].rgb_dst_factor   = PIPE_BLENDFACTOR_ZERO;