1 files changed, 102 insertions, 0 deletions
diff --git a/archive/src/ral/ral_impl.h b/archive/src/ral/ral_impl.h
new file mode 100644
index 0000000..16c9767
--- /dev/null
+++ b/archive/src/ral/ral_impl.h
@@ -0,0 +1,102 @@
+/**
+ * @brief
+ */
+#pragma once
+#include "buf.h"
+#include "defines.h"
+#include "ral_types.h"
+
+struct GLFWwindow;
+struct ResourcePools;
+
+// Forward declare structs - these must be defined in the backend implementation
+typedef struct GPU_Swapchain GPU_Swapchain;
+typedef struct GPU_Device GPU_Device;
+typedef struct GPU_PipelineLayout GPU_PipelineLayout;
+typedef struct GPU_Pipeline GPU_Pipeline;
+typedef struct GPU_Renderpass GPU_Renderpass;
+typedef struct GPU_CmdEncoder GPU_CmdEncoder;  // Recording
+typedef struct GPU_CmdBuffer GPU_CmdBuffer;    // Ready for submission
+typedef struct GPU_Buffer GPU_Buffer;
+typedef struct GPU_Texture GPU_Texture;
+
+bool GPU_Backend_Init(const char* window_name, struct GLFWwindow* window,
+                      struct ResourcePools* res_pools);
+void GPU_Backend_Shutdown();
+
+bool GPU_Device_Create(GPU_Device* out_device);
+void GPU_Device_Destroy(GPU_Device* device);
+
+bool GPU_Swapchain_Create(GPU_Swapchain* out_swapchain);
+void GPU_Swapchain_Destroy(GPU_Swapchain* swapchain);
+void GPU_Swapchain_Resize(i32 new_width, i32 new_height);
+u32x2 GPU_Swapchain_GetDimensions();
+
+PUB GPU_Renderpass* GPU_Renderpass_Create(GPU_RenderpassDesc description);
+PUB void GPU_Renderpass_Destroy(GPU_Renderpass* pass);
+
+PUB GPU_Pipeline* GPU_GraphicsPipeline_Create(GraphicsPipelineDesc description,
+                                              GPU_Renderpass* renderpass);
+PUB void GraphicsPipeline_Destroy(GPU_Pipeline* pipeline);
+
+// --- Command buffer
+PUB GPU_CmdEncoder GPU_CmdEncoder_Create();
+PUB void GPU_CmdEncoder_Destroy(GPU_CmdEncoder* encoder);
+PUB void GPU_CmdEncoder_Begin(GPU_CmdEncoder* encoder);
+PUB void GPU_CmdEncoder_Finish(GPU_CmdEncoder* encoder);
+PUB void GPU_CmdEncoder_BeginRender(GPU_CmdEncoder* encoder, GPU_Renderpass* renderpass);
+PUB void GPU_CmdEncoder_EndRender(GPU_CmdEncoder* encoder);
+PUB GPU_CmdEncoder* GPU_GetDefaultEncoder();
+PUB void GPU_QueueSubmit(GPU_CmdBuffer* cmd_buffer);
+
+// --- Buffers
+PUB BufferHandle GPU_BufferCreate(u64 size, GPU_BufferType buf_type, GPU_BufferFlags flags,
+                                  const void* data);
+PUB void GPU_BufferDestroy(BufferHandle handle);
+PUB void GPU_BufferUpload(BufferHandle buffer, size_t n_bytes, const void* data);
+
+// --- Textures
+PUB TextureHandle GPU_TextureCreate(TextureDesc desc, bool create_view, const void* data);
+PUB GPU_Texture* GPU_TextureAlloc(TextureHandle* out_handle);
+PUB void GPU_TextureDestroy(TextureHandle handle);
+PUB void GPU_TextureUpload(TextureHandle handle, size_t n_bytes, const void* data);
+
+// --- Data copy commands
+PUB void GPU_EncodeCopyBufToBuf(GPU_CmdEncoder* encoder, BufferHandle src, u64 src_offset,
+                                BufferHandle dst, u64 dst_offset, u64 copy_size);
+
+// PUB void GPU_EncodeCopyBufToTex(GPU_CmdEncoder* encoder, BufferHandle src, TextureHandle dst,
+//     u32 x_offset, u32 y_offset, u32 width, u32 height, const void* data);
+/** @brief Convenience method for writing data directly into a texture. Staging memory is handled
+ * internally. */
+PUB void GPU_WriteTextureRegion(GPU_CmdEncoder* encoder, TextureHandle dst, u32 x_offset,
+                                u32 y_offset, u32 width, u32 height, const void* data);
+PUB void GPU_WriteBuffer(GPU_CmdEncoder* encoder, BufferHandle buf, u64 offset, u64 size,
+                         const void* data);
+
+// --- Render commands
+PUB void GPU_EncodeBindPipeline(GPU_CmdEncoder* encoder, GPU_Pipeline* pipeline);
+PUB void GPU_EncodeBindShaderData(GPU_CmdEncoder* encoder, u32 group, ShaderDataLayout layout);
+void GPU_EncodeSetDefaults(GPU_CmdEncoder* encoder);
+PUB void GPU_EncodeSetVertexBuffer(GPU_CmdEncoder* encoder, BufferHandle buf);
+PUB void GPU_EncodeSetIndexBuffer(GPU_CmdEncoder* encoder, BufferHandle buf);
+
+PUB void GPU_EncodeDraw(GPU_CmdEncoder* encoder, PrimitiveTopology topology, u64 count);
+PUB void GPU_EncodeDrawIndexed(GPU_CmdEncoder* encoder, PrimitiveTopology topology,
+                               u64 index_count);
+// convenience versions of the above
+PUB void GPU_EncodeDrawTris(GPU_CmdEncoder* encoder, u64 count);
+PUB void GPU_EncodeDrawIndexedTris(GPU_CmdEncoder* encoder, u64 index_count);
+PUB void GPU_EncodeDrawInstanced(GPU_CmdEncoder* encoder, u64 index_count,
+                                 u64 instance_count);  // TODO: implement instanced rendering
+
+// --- Frame cycle
+PUB bool GPU_Backend_BeginFrame();
+PUB void GPU_Backend_EndFrame();
+
+// Concrete implementation
+#if defined(CEL_REND_BACKEND_OPENGL)
+#include "backend_opengl.h"
+#elif defined(CEL_REND_BACKEND_VULKAN)
+#include "backend_vulkan.h"
+#endif