ShuriZma
/
xenia-canary
mirror of https://github.com/xenia-canary/xenia-canary.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Further improvements to FetchTexture2D.

This commit is contained in:
gibbed 2014-01-21 18:17:13 -08:00
parent 58db76c42c
commit a9dfe13799
3 changed files with 92 additions and 54 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

139
src/xenia/gpu/d3d11/d3d11_graphics_driver.cc
View File

@ -1101,11 +1101,14 @@ D3D11GraphicsDriver::TextureInfo D3D11GraphicsDriver::GetTextureInfo(
TextureInfo info;
info.format = DXGI_FORMAT_UNKNOWN;
info.block_size = 0;
info.pitch = 0;
info.needs_power_of_two = false;
info.texel_pitch = 0;
info.is_compressed = false;
switch (fetch.format) {
case FMT_8:
switch (fetch.swizzle) {
case XE_GPU_SWIZZLE_RRR1:
info.format = DXGI_FORMAT_R8_UNORM;
break;
case XE_GPU_SWIZZLE_000R:
info.format = DXGI_FORMAT_A8_UNORM;
break;
@ -1115,7 +1118,7 @@ D3D11GraphicsDriver::TextureInfo D3D11GraphicsDriver::GetTextureInfo(
break;
}
info.block_size = 1;
info.pitch = 1;
info.texel_pitch = 1;
break;
case FMT_8_8_8_8:
switch (fetch.swizzle) {
@ -1137,7 +1140,7 @@ D3D11GraphicsDriver::TextureInfo D3D11GraphicsDriver::GetTextureInfo(
break;
}
info.block_size = 1;
info.pitch = 4;
info.texel_pitch = 4;
break;
case FMT_4_4_4_4:
switch (fetch.swizzle) {
@ -1150,30 +1153,46 @@ D3D11GraphicsDriver::TextureInfo D3D11GraphicsDriver::GetTextureInfo(
break;
}
info.block_size = 1;
info.pitch = 2;
info.texel_pitch = 2;
break;
case FMT_16_16_16_16_FLOAT:
info.format = DXGI_FORMAT_R16G16B16A16_FLOAT;
switch (fetch.swizzle) {
case XE_GPU_SWIZZLE_RGBA:
info.format = DXGI_FORMAT_R16G16B16A16_FLOAT;
break;
default:
XELOGW("D3D11: unhandled swizzle for FMT_16_16_16_16_FLOAT");
info.format = DXGI_FORMAT_R16G16B16A16_FLOAT;
break;
}
info.block_size = 1;
info.pitch = 8;
info.texel_pitch = 8;
break;
case FMT_32_FLOAT:
info.format = DXGI_FORMAT_R32_FLOAT;
switch (fetch.swizzle) {
case XE_GPU_SWIZZLE_R111:
info.format = DXGI_FORMAT_R32_FLOAT;
break;
default:
XELOGW("D3D11: unhandled swizzle for FMT_32_FLOAT");
info.format = DXGI_FORMAT_R32_FLOAT;
break;
}
info.block_size = 1;
info.pitch = 4;
info.texel_pitch = 4;
break;
case FMT_DXT1:
info.format = DXGI_FORMAT_BC1_UNORM;
info.block_size = 4;
info.pitch = 8;
info.needs_power_of_two = true;
info.texel_pitch = 8;
info.is_compressed = true;
break;
case FMT_DXT2_3:
case FMT_DXT4_5:
info.format = (fetch.format == FMT_DXT4_5 ? DXGI_FORMAT_BC3_UNORM : DXGI_FORMAT_BC2_UNORM);
info.block_size = 4;
info.pitch = 16;
info.needs_power_of_two = true;
info.texel_pitch = 16;
info.is_compressed = true;
break;
case FMT_1_REVERSE:
case FMT_1:
@ -1265,24 +1284,24 @@ XEFORCEINLINE void TextureSwap(uint8_t* dest, const uint8_t* src, uint32_t pitch
switch (endianness) {
case XE_GPU_ENDIAN_8IN16:
if (pitch < 4) {
for (uint32_t i = 0; i < pitch; i += 2, dest += 2) {
*(uint16_t*)dest = XESWAP16(*(uint16_t*)dest);
for (uint32_t i = 0; i < pitch; i += 2, src += 2, dest += 2) {
*(uint16_t*)dest = XESWAP16(*(uint16_t*)src);
}
}
else {
for (uint32_t i = 0; i < pitch; i += 4, dest += 4, src += 4) {
for (uint32_t i = 0; i < pitch; i += 4, src += 4, dest += 4) {
uint32_t value = *(uint32_t*)src;
*(uint32_t*)dest = ((value >> 16) & 0xFFFF) | (value << 16);
}
}
break;
case XE_GPU_ENDIAN_8IN32: // Swap bytes.
for (uint32_t i = 0; i < pitch; i += 4, dest += 4, src += 4) {
*(uint32_t*)dest = XESWAP32(*(uint32_t*)src);
for (uint32_t i = 0; i < pitch; i += 4, src += 4, dest += 4) {
*(uint32_t*)dest = *(uint32_t*)src;
}
break;
case XE_GPU_ENDIAN_16IN32: // Swap half words.
for (uint32_t i = 0; i < pitch; i += 4, dest += 4, src += 4) {
for (uint32_t i = 0; i < pitch; i += 4, src += 4, dest += 4) {
uint32_t value = *(uint32_t*)src;
*(uint32_t*)dest = ((value >> 16) & 0xFFFF) | (value << 16);
}
@ -1295,24 +1314,20 @@ XEFORCEINLINE void TextureSwap(uint8_t* dest, const uint8_t* src, uint32_t pitch
}
// https://code.google.com/p/crunch/source/browse/trunk/inc/crn_decomp.h#4104
XEFORCEINLINE uint32_t TiledOffset2DOuter(uint32_t y, uint32_t aligned_width, uint32_t log_bpp)
XEFORCEINLINE uint32_t TiledOffset2DOuter(uint32_t y, uint32_t width, uint32_t log_bpp)
{
uint32_t macro = ((y >> 5) * (aligned_width >> 5)) << (log_bpp + 7);
uint32_t macro = ((y >> 5) * (width >> 5)) << (log_bpp + 7);
uint32_t micro = ((y & 6) << 2) << log_bpp;
return macro +
((micro & ~15) << 1) +
(micro & 15) +
((y & 8) << (3 + log_bpp)) + ((y & 1) << 4);
return macro + ((micro & ~15) << 1) + (micro & 15) + ((y & 8) << (3 + log_bpp)) + ((y & 1) << 4);
}
XEFORCEINLINE uint32_t TiledOffset2DInner(uint32_t x, uint32_t y, uint32_t log_bpp, uint32_t base_offset)
XEFORCEINLINE uint32_t TiledOffset2DInner(uint32_t x, uint32_t y, uint32_t bpp, uint32_t base_offset)
{
uint32_t macro = (x >> 5) << (log_bpp + 7);
uint32_t micro = (x & 7) << log_bpp;
uint32_t offset = base_offset + macro + ((micro & ~15) << 1) + (micro & 15);
uint32_t macro = (x >> 5) << (bpp + 7);
uint32_t micro = (x & 7) << bpp;
uint32_t offset = base_offset + (macro + ((micro & ~15) << 1) + (micro & 15));
return ((offset & ~511) << 3) + ((offset & 448) << 2) + (offset & 63) +
((y & 16) << 7) +
(((((y & 8) >> 2) + (x >> 3)) & 3) << 6);
((y & 16) << 7) + (((((y & 8) >> 2) + (x >> 3)) & 3) << 6);
}
int D3D11GraphicsDriver::FetchTexture2D(
@ -1321,15 +1336,31 @@ int D3D11GraphicsDriver::FetchTexture2D(
ID3D11Resource** out_texture) {
uint32_t address = (fetch.address << 12) + address_translation_;
uint32_t width = 1 + fetch.size_2d.width;
uint32_t height = 1 + fetch.size_2d.height;
uint32_t padded_width = info.needs_power_of_two == false ? width : MAX(256, XENEXTPOW2(width));
uint32_t padded_height = info.needs_power_of_two == false ? height : XENEXTPOW2(height);
uint32_t logical_width = 1 + fetch.size_2d.width;
uint32_t logical_height = 1 + fetch.size_2d.height;
uint32_t input_width, input_height;
uint32_t output_width, output_height;
if (!info.is_compressed) {
// must be 32x32
input_width = XEROUNDUP((logical_width + 31) & ~31, 256);
input_height = (logical_height + 31) & ~31;
output_width = input_width;
output_height = input_height;
}
else {
// must be 128x128
input_width = (logical_width + 127) & ~127;
input_height = (logical_height + 127) & ~127;
output_width = input_width;
output_height = input_height;
}
D3D11_TEXTURE2D_DESC texture_desc;
xe_zero_struct(&texture_desc, sizeof(texture_desc));
texture_desc.Width = padded_width;
texture_desc.Height = padded_height;
texture_desc.Width = output_width;
texture_desc.Height = output_height;
texture_desc.MipLevels = 1;
texture_desc.ArraySize = 1;
texture_desc.Format = info.format;
@ -1354,32 +1385,36 @@ int D3D11GraphicsDriver::FetchTexture2D(
return 1;
}
auto row_pitch = (width / info.block_size) * info.pitch;
auto padded_row_pitch = (padded_width / info.block_size) * info.pitch;
auto logical_pitch = (logical_width / info.block_size) * info.texel_pitch;
auto input_pitch = (input_width / info.block_size) * info.texel_pitch;
auto output_pitch = res.RowPitch; // (output_width / info.block_size) * info.texel_pitch;
const uint8_t* src = memory_->Translate(address);
uint8_t* dest = (uint8_t*)res.pData;
//memset(dest, 0, res.RowPitch * (padded_height / info.block_size)); // TODO(gibbed): remove me later
memset(dest, 0, output_pitch * (output_height / info.block_size)); // TODO(gibbed): remove me later
if (!fetch.tiled) {
dest = (uint8_t*)res.pData;
for (uint32_t y = 0; y < height / info.block_size; y++) {
for (uint32_t x = 0; x < row_pitch; x += info.pitch) {
TextureSwap(dest + x, src + x, info.pitch, (XE_GPU_ENDIAN)fetch.endianness);
for (uint32_t y = 0; y < input_height / info.block_size; y++) {
for (uint32_t x = 0; x < logical_pitch; x += info.texel_pitch) {
TextureSwap(dest + x, src + x, info.texel_pitch, (XE_GPU_ENDIAN)fetch.endianness);
}
dest += res.RowPitch;
src += row_pitch;
src += input_pitch;
dest += output_pitch;
}
}
else {
auto aligned_width = (width + 31) & ~31; // 32x32
auto log_bpp = (info.pitch >> 2) + ((info.pitch >> 1) >> (info.pitch >> 2));
for (uint32_t y = 0; y < height; y++) {
auto base_offset = TiledOffset2DOuter(y, aligned_width, log_bpp);
for (uint32_t x = 0; x < width; x++) {
auto offset = TiledOffset2DInner(x, y, log_bpp, base_offset);
TextureSwap(dest + (y * res.RowPitch) + (x * info.pitch), src + offset, info.pitch, (XE_GPU_ENDIAN)fetch.endianness);
uint32_t block_width = logical_width / info.block_size;
uint32_t block_height = logical_height / info.block_size;
auto bpp = (info.texel_pitch >> 2) + ((info.texel_pitch >> 1) >> (info.texel_pitch >> 2));
for (uint32_t y = 0, output_base_offset = 0; y < block_height; y++, output_base_offset += output_pitch) {
auto input_base_offset = TiledOffset2DOuter(y, (input_width / info.block_size), bpp);
for (uint32_t x = 0, output_offset = output_base_offset; x < block_width; x++, output_offset += info.texel_pitch) {
auto input_offset = TiledOffset2DInner(x, y, bpp, input_base_offset) >> bpp;
TextureSwap(dest + output_offset,
src + input_offset * info.texel_pitch,
info.texel_pitch, (XE_GPU_ENDIAN)fetch.endianness);
}
}
}

4
src/xenia/gpu/d3d11/d3d11_graphics_driver.h
View File

@ -69,8 +69,8 @@ private:
typedef struct {
DXGI_FORMAT format;
uint32_t block_size;
uint32_t pitch;
bool needs_power_of_two;
uint32_t texel_pitch;
bool is_compressed;
} TextureInfo;
TextureInfo GetTextureInfo(xenos::xe_gpu_texture_fetch_t& fetch);
int FetchTexture1D(xenos::xe_gpu_texture_fetch_t& fetch,

3
src/xenia/gpu/xenos/xenos.h
View File

@ -68,6 +68,9 @@ typedef enum {
XE_GPU_SWIZZLE_RGB1 = XE_GPU_MAKE_SWIZZLE(R, G, B, 1),
XE_GPU_SWIZZLE_BGR1 = XE_GPU_MAKE_SWIZZLE(B, G, R, 1),
XE_GPU_SWIZZLE_000R = XE_GPU_MAKE_SWIZZLE(0, 0, 0, R),
XE_GPU_SWIZZLE_RRR1 = XE_GPU_MAKE_SWIZZLE(R, R, R, 1),
XE_GPU_SWIZZLE_R111 = XE_GPU_MAKE_SWIZZLE(R, 1, 1, 1),
XE_GPU_SWIZZLE_R000 = XE_GPU_MAKE_SWIZZLE(R, 0, 0, 0),
} XE_GPU_SWIZZLE;
XEFORCEINLINE uint32_t GpuSwap(uint32_t value, XE_GPU_ENDIAN endianness) {