Duckstation/src/core/gpu_sw.h

125 lines
5.1 KiB
C
Raw Normal View History

2019-10-26 02:57:35 +00:00
#pragma once
#include "gpu.h"
#include <array>
#include <memory>
#include <vector>
class HostDisplayTexture;
2019-10-26 02:57:35 +00:00
class GPU_SW final : public GPU
{
public:
GPU_SW();
~GPU_SW() override;
bool IsHardwareRenderer() const override;
JIT optimizations and refactoring (#675) * CPU/Recompiler: Use rel32 call where possible for no-args * JitCodeBuffer: Support using preallocated buffer * CPU/Recompiler/AArch64: Use bl instead of blr for short branches * CPU/CodeCache: Allocate recompiler buffer in program space This means we don't need 64-bit moves for every call out of the recompiler. * GTE: Don't store as u16 and load as u32 * CPU/Recompiler: Add methods to emit global load/stores * GTE: Convert class to namespace * CPU/Recompiler: Call GTE functions directly * Settings: Turn into a global variable * GPU: Replace local pointers with global * InterruptController: Turn into a global pointer * System: Replace local pointers with global * Timers: Turn into a global instance * DMA: Turn into a global instance * SPU: Turn into a global instance * CDROM: Turn into a global instance * MDEC: Turn into a global instance * Pad: Turn into a global instance * SIO: Turn into a global instance * CDROM: Move audio FIFO to the heap * CPU/Recompiler: Drop ASMFunctions No longer needed since we have code in the same 4GB window. * CPUCodeCache: Turn class into namespace * Bus: Local pointer -> global pointers * CPU: Turn class into namespace * Bus: Turn into namespace * GTE: Store registers in CPU state struct Allows relative addressing on ARM. * CPU/Recompiler: Align code storage to page size * CPU/Recompiler: Fix relative branches on A64 * HostInterface: Local references to global * System: Turn into a namespace, move events out * Add guard pages * Android: Fix build
2020-07-31 07:09:18 +00:00
bool Initialize(HostDisplay* host_display) override;
2019-10-26 02:57:35 +00:00
void Reset() override;
ALWAYS_INLINE_RELEASE u16 GetPixel(const u32 x, const u32 y) const { return m_vram[VRAM_WIDTH * y + x]; }
ALWAYS_INLINE_RELEASE const u16* GetPixelPtr(const u32 x, const u32 y) const { return &m_vram[VRAM_WIDTH * y + x]; }
ALWAYS_INLINE_RELEASE u16* GetPixelPtr(const u32 x, const u32 y) { return &m_vram[VRAM_WIDTH * y + x]; }
ALWAYS_INLINE_RELEASE void SetPixel(const u32 x, const u32 y, const u16 value) { m_vram[VRAM_WIDTH * y + x] = value; }
2019-10-26 02:57:35 +00:00
// this is actually (31 * 255) >> 4) == 494, but to simplify addressing we use the next power of two (512)
static constexpr u32 DITHER_LUT_SIZE = 512;
using DitherLUT = std::array<std::array<std::array<u8, 512>, DITHER_MATRIX_SIZE>, DITHER_MATRIX_SIZE>;
static constexpr DitherLUT ComputeDitherLUT();
2019-10-26 02:57:35 +00:00
protected:
struct SWVertex
{
s32 x, y;
u8 r, g, b;
u8 u, v;
ALWAYS_INLINE void SetPosition(VertexPosition p, s32 offset_x, s32 offset_y)
{
x = TruncateVertexPosition(offset_x + p.x);
y = TruncateVertexPosition(offset_y + p.y);
}
ALWAYS_INLINE void SetColorRGB24(u32 color) { std::tie(r, g, b) = UnpackColorRGB24(color); }
ALWAYS_INLINE void SetTexcoord(u16 value) { std::tie(u, v) = UnpackTexcoord(value); }
2019-10-26 02:57:35 +00:00
};
//////////////////////////////////////////////////////////////////////////
// Scanout
//////////////////////////////////////////////////////////////////////////
void CopyOut15Bit(u32 src_x, u32 src_y, u32* dst_ptr, u32 dst_stride, u32 width, u32 height, bool interlaced,
bool interleaved);
void CopyOut24Bit(u32 src_x, u32 src_y, u32* dst_ptr, u32 dst_stride, u32 width, u32 height, bool interlaced,
bool interleaved);
void ClearDisplay() override;
2019-10-26 02:57:35 +00:00
void UpdateDisplay() override;
//////////////////////////////////////////////////////////////////////////
// Rasterization
//////////////////////////////////////////////////////////////////////////
void DispatchRenderCommand() override;
2019-10-26 02:57:35 +00:00
template<bool texture_enable, bool raw_texture_enable, bool transparency_enable, bool dithering_enable>
void ShadePixel(u32 x, u32 y, u8 color_r, u8 color_g, u8 color_b, u8 texcoord_x, u8 texcoord_y);
template<bool texture_enable, bool raw_texture_enable, bool transparency_enable>
void DrawRectangle(s32 origin_x, s32 origin_y, u32 width, u32 height, u8 r, u8 g, u8 b, u8 origin_texcoord_x,
u8 origin_texcoord_y);
using DrawRectangleFunction = void (GPU_SW::*)(s32 origin_x, s32 origin_y, u32 width, u32 height, u8 r, u8 g, u8 b,
u8 origin_texcoord_x, u8 origin_texcoord_y);
DrawRectangleFunction GetDrawRectangleFunction(bool texture_enable, bool raw_texture_enable,
bool transparency_enable);
//////////////////////////////////////////////////////////////////////////
// Polygon and line rasterization ported from Mednafen
//////////////////////////////////////////////////////////////////////////
struct i_deltas
{
u32 du_dx, dv_dx;
u32 dr_dx, dg_dx, db_dx;
u32 du_dy, dv_dy;
u32 dr_dy, dg_dy, db_dy;
};
struct i_group
{
u32 u, v;
u32 r, g, b;
};
template<bool shading_enable, bool texture_enable>
bool CalcIDeltas(i_deltas& idl, const SWVertex* A, const SWVertex* B, const SWVertex* C);
template<bool shading_enable, bool texture_enable>
void AddIDeltas_DX(i_group& ig, const i_deltas& idl, u32 count = 1);
template<bool shading_enable, bool texture_enable>
void AddIDeltas_DY(i_group& ig, const i_deltas& idl, u32 count = 1);
template<bool shading_enable, bool texture_enable, bool raw_texture_enable, bool transparency_enable,
bool dithering_enable>
void DrawSpan(s32 y, s32 x_start, s32 x_bound, i_group ig, const i_deltas& idl);
template<bool shading_enable, bool texture_enable, bool raw_texture_enable, bool transparency_enable,
bool dithering_enable>
void DrawTriangle(const SWVertex* v0, const SWVertex* v1, const SWVertex* v2);
using DrawTriangleFunction = void (GPU_SW::*)(const SWVertex* v0, const SWVertex* v1, const SWVertex* v2);
DrawTriangleFunction GetDrawTriangleFunction(bool shading_enable, bool texture_enable, bool raw_texture_enable,
bool transparency_enable, bool dithering_enable);
template<bool shading_enable, bool transparency_enable, bool dithering_enable>
void DrawLine(const SWVertex* p0, const SWVertex* p1);
using DrawLineFunction = void (GPU_SW::*)(const SWVertex* p0, const SWVertex* p1);
DrawLineFunction GetDrawLineFunction(bool shading_enable, bool transparency_enable, bool dithering_enable);
2019-10-26 02:57:35 +00:00
std::vector<u32> m_display_texture_buffer;
std::unique_ptr<HostDisplayTexture> m_display_texture;
2019-10-26 02:57:35 +00:00
std::array<u16, VRAM_WIDTH * VRAM_HEIGHT> m_vram;
};