From dc81a906402b0a39c3ff1964cdf935b46b062b5d Mon Sep 17 00:00:00 2001
From: ReinUsesLisp <reinuseslisp@airmail.cc>
Date: Wed, 30 Dec 2020 01:47:08 -0300
Subject: [PATCH] host_shaders: Add compute shader to copy BC4 as RG32UI to
 RGBA8

---
 src/video_core/host_shaders/CMakeLists.txt    |  1 +
 .../host_shaders/opengl_copy_bc4.comp         | 70 +++++++++++++++++++
 2 files changed, 71 insertions(+)
 create mode 100644 src/video_core/host_shaders/opengl_copy_bc4.comp

diff --git a/src/video_core/host_shaders/CMakeLists.txt b/src/video_core/host_shaders/CMakeLists.txt
index 5770c47617..338bf9eec0 100644
--- a/src/video_core/host_shaders/CMakeLists.txt
+++ b/src/video_core/host_shaders/CMakeLists.txt
@@ -2,6 +2,7 @@ set(SHADER_FILES
     block_linear_unswizzle_2d.comp
     block_linear_unswizzle_3d.comp
     full_screen_triangle.vert
+    opengl_copy_bc4.comp
     opengl_present.frag
     opengl_present.vert
     pitch_unswizzle.comp
diff --git a/src/video_core/host_shaders/opengl_copy_bc4.comp b/src/video_core/host_shaders/opengl_copy_bc4.comp
new file mode 100644
index 0000000000..7b8e20fbe0
--- /dev/null
+++ b/src/video_core/host_shaders/opengl_copy_bc4.comp
@@ -0,0 +1,70 @@
+// Copyright 2020 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#version 430 core
+#extension GL_ARB_gpu_shader_int64 : require
+
+layout (local_size_x = 4, local_size_y = 4) in;
+
+layout(binding = 0, rg32ui) readonly uniform uimage3D bc4_input;
+layout(binding = 1, rgba8ui) writeonly uniform uimage3D bc4_output;
+
+layout(location = 0) uniform uvec3 src_offset;
+layout(location = 1) uniform uvec3 dst_offset;
+
+// https://www.khronos.org/registry/OpenGL/extensions/ARB/ARB_texture_compression_rgtc.txt
+uint DecompressBlock(uint64_t bits, uvec2 coord) {
+    const uint code_offset = 16 + 3 * (4 * coord.y + coord.x);
+    const uint code = uint(bits >> code_offset) & 7;
+    const uint red0 = uint(bits >> 0) & 0xff;
+    const uint red1 = uint(bits >> 8) & 0xff;
+    if (red0 > red1) {
+        switch (code) {
+        case 0:
+            return red0;
+        case 1:
+            return red1;
+        case 2:
+            return (6 * red0 + 1 * red1) / 7;
+        case 3:
+            return (5 * red0 + 2 * red1) / 7;
+        case 4:
+            return (4 * red0 + 3 * red1) / 7;
+        case 5:
+            return (3 * red0 + 4 * red1) / 7;
+        case 6:
+            return (2 * red0 + 5 * red1) / 7;
+        case 7:
+            return (1 * red0 + 6 * red1) / 7;
+        }
+    } else {
+        switch (code) {
+        case 0:
+            return red0;
+        case 1:
+            return red1;
+        case 2:
+            return (4 * red0 + 1 * red1) / 5;
+        case 3:
+            return (3 * red0 + 2 * red1) / 5;
+        case 4:
+            return (2 * red0 + 3 * red1) / 5;
+        case 5:
+            return (1 * red0 + 4 * red1) / 5;
+        case 6:
+            return 0;
+        case 7:
+            return 0xff;
+        }
+    }
+    return 0;
+}
+
+void main() {
+    uvec2 packed_bits = imageLoad(bc4_input, ivec3(gl_WorkGroupID + src_offset)).rg;
+    uint64_t bits = packUint2x32(packed_bits);
+    uint red = DecompressBlock(bits, gl_LocalInvocationID.xy);
+    uvec4 color = uvec4(red & 0xff, 0, 0, 0xff);
+    imageStore(bc4_output, ivec3(gl_GlobalInvocationID + dst_offset), color);
+}