Implement HLE macro for DrawElementsIndirect (#3748)

* Implement HLE macro for DrawElementsIndirect

* Shader cache version bump

* Use GL_ARB_shader_draw_parameters extension on OpenGL

* Fix DrawIndexedIndirectCount on Vulkan when extension is not supported

* Implement DrawIndex

* Alignment

* Fix some validation errors

* Rename BaseIds to DrawParameters

* Fix incorrect index buffer and vertex buffer size in some cases

* Add HLE macros for DrawArraysInstanced and DrawElementsInstanced

* Perform a regular draw when indirect data is not modified

* Use non-indirect draw methods if indirect buffer was not GPU modified

* Only check if draw parameters match if the shader actually uses them

* Expose Macro HLE setting on GUI

* Reset FirstVertex and FirstInstance after draw

* Update shader cache version again since some people already tested this

* PR feedback

Co-authored-by: riperiperi <rhy3756547@hotmail.com>

1 files changed, 58 insertions, 0 deletions

diff --git a/Ryujinx.Graphics.Vulkan/Shaders/ConvertIndexBufferShaderSource.comp b/Ryujinx.Graphics.Vulkan/Shaders/ConvertIndexBufferShaderSource.comp
new file mode 100644
index 00000000..d56d6cfd
--- /dev/null
+++ b/Ryujinx.Graphics.Vulkan/Shaders/ConvertIndexBufferShaderSource.comp
@@ -0,0 +1,58 @@
+#version 450 core
+
+#extension GL_EXT_scalar_block_layout : require
+#extension GL_EXT_shader_8bit_storage : require
+
+layout (local_size_x = 16, local_size_y = 1, local_size_z = 1) in;
+
+layout (std430, set = 0, binding = 0) uniform index_buffer_pattern
+{
+    int ibp_pattern[8];
+    int ibp_primitive_vertices;
+    int ibp_primitive_vertices_out;
+    int ibp_index_size;
+    int ibp_index_size_out;
+    int ibp_base_index;
+    int ibp_index_stride;
+    int src_offset;
+    int total_primitives;
+};
+
+layout (std430, set = 1, binding = 1) buffer in_s
+{
+    uint8_t[] in_data;
+};
+
+layout (std430, set = 1, binding = 2) buffer out_s
+{
+    uint8_t[] out_data;
+};
+
+void main()
+{
+    int primitiveIndex = int(gl_GlobalInvocationID.x);
+    if (primitiveIndex >= total_primitives)
+    {
+        return;
+    }
+
+    int inOffset = primitiveIndex * ibp_index_stride;
+    int outOffset = primitiveIndex * ibp_primitive_vertices_out;
+
+    for (int i = 0; i < ibp_primitive_vertices_out; i++)
+    {
+        int j;
+        int io = max(0, inOffset + ibp_base_index + ibp_pattern[i]) * ibp_index_size;
+        int oo = (outOffset + i) * ibp_index_size_out;
+
+        for (j = 0; j < ibp_index_size; j++)
+        {
+            out_data[oo + j] = in_data[src_offset + io + j];
+        }
+
+        for (; j < ibp_index_size_out; j++)
+        {
+            out_data[oo + j] = uint8_t(0);
+        }
+    }
+}