dolphin/Source/Core/VideoBackends/Software/TextureSampler.cpp

262 lines
8.3 KiB
C++

// Copyright 2009 Dolphin Emulator Project
// SPDX-License-Identifier: GPL-2.0-or-later
#include "VideoBackends/Software/TextureSampler.h"
#include <algorithm>
#include <cmath>
#include "Common/CommonTypes.h"
#include "Common/MsgHandler.h"
#include "Core/HW/Memmap.h"
#include "VideoCommon/BPMemory.h"
#include "VideoCommon/SamplerCommon.h"
#include "VideoCommon/TextureDecoder.h"
#define ALLOW_MIPMAP 1
namespace TextureSampler
{
static inline void WrapCoord(int* coordp, WrapMode wrap_mode, int image_size)
{
int coord = *coordp;
switch (wrap_mode)
{
case WrapMode::Clamp:
coord = std::clamp(coord, 0, image_size - 1);
break;
case WrapMode::Repeat:
// Per YAGCD's info on TX_SETMODE1_I0 (et al.), mirror "requires the texture size to be a power
// of two. (wrapping is implemented by a logical AND (SIZE-1))". So though this doesn't wrap
// nicely for non-power-of-2 sizes, that's how hardware does it.
coord = coord & (image_size - 1);
break;
case WrapMode::Mirror:
{
// YAGCD doesn't mention this, but this seems to be the check used to implement mirroring.
// With power-of-2 sizes, this correctly checks if it's an even-numbered repeat or an
// odd-numbered one, and thus can decide whether to reflect. It fails in unusual ways
// with non-power-of-2 sizes, but seems to match what happens on actual hardware.
if ((coord & image_size) != 0)
coord = ~coord;
coord = coord & (image_size - 1);
break;
}
default:
// Hardware testing indicates that wrap_mode set to 3 behaves the same as clamp.
PanicAlertFmt("Invalid wrap mode: {}", wrap_mode);
coord = std::clamp(coord, 0, image_size - 1);
break;
}
*coordp = coord;
}
static inline void SetTexel(const u8* inTexel, u32* outTexel, u32 fract)
{
outTexel[0] = inTexel[0] * fract;
outTexel[1] = inTexel[1] * fract;
outTexel[2] = inTexel[2] * fract;
outTexel[3] = inTexel[3] * fract;
}
static inline void AddTexel(const u8* inTexel, u32* outTexel, u32 fract)
{
outTexel[0] += inTexel[0] * fract;
outTexel[1] += inTexel[1] * fract;
outTexel[2] += inTexel[2] * fract;
outTexel[3] += inTexel[3] * fract;
}
void Sample(s32 s, s32 t, s32 lod, bool linear, u8 texmap, u8* sample)
{
int baseMip = 0;
bool mipLinear = false;
#if (ALLOW_MIPMAP)
const FourTexUnits& texUnit = bpmem.tex[(texmap >> 2) & 1];
const TexMode0& tm0 = texUnit.texMode0[texmap & 3];
const s32 lodFract = lod & 0xf;
if (lod > 0 && SamplerCommon::AreBpTexMode0MipmapsEnabled(tm0))
{
// use mipmap
baseMip = lod >> 4;
mipLinear = (lodFract && tm0.mipmap_filter == MipMode::Linear);
// if using nearest mip filter and lodFract >= 0.5 round up to next mip
if (tm0.mipmap_filter == MipMode::Point && lodFract >= 8)
baseMip++;
}
if (mipLinear)
{
u8 sampledTex[4];
u32 texel[4];
SampleMip(s, t, baseMip, linear, texmap, sampledTex);
SetTexel(sampledTex, texel, (16 - lodFract));
SampleMip(s, t, baseMip + 1, linear, texmap, sampledTex);
AddTexel(sampledTex, texel, lodFract);
sample[0] = (u8)(texel[0] >> 4);
sample[1] = (u8)(texel[1] >> 4);
sample[2] = (u8)(texel[2] >> 4);
sample[3] = (u8)(texel[3] >> 4);
}
else
#endif
{
SampleMip(s, t, baseMip, linear, texmap, sample);
}
}
void SampleMip(s32 s, s32 t, s32 mip, bool linear, u8 texmap, u8* sample)
{
const FourTexUnits& texUnit = bpmem.tex[(texmap >> 2) & 1];
const u8 subTexmap = texmap & 3;
const TexMode0& tm0 = texUnit.texMode0[subTexmap];
const TexImage0& ti0 = texUnit.texImage0[subTexmap];
const TexTLUT& texTlut = texUnit.texTlut[subTexmap];
const TextureFormat texfmt = ti0.format;
const TLUTFormat tlutfmt = texTlut.tlut_format;
const u8* imageSrc;
const u8* imageSrcOdd = nullptr;
if (texUnit.texImage1[subTexmap].cache_manually_managed)
{
imageSrc = &texMem[texUnit.texImage1[subTexmap].tmem_even * TMEM_LINE_SIZE];
if (texfmt == TextureFormat::RGBA8)
imageSrcOdd = &texMem[texUnit.texImage2[subTexmap].tmem_odd * TMEM_LINE_SIZE];
}
else
{
const u32 imageBase = texUnit.texImage3[subTexmap].image_base << 5;
imageSrc = Memory::GetPointer(imageBase);
}
int image_width_minus_1 = ti0.width;
int image_height_minus_1 = ti0.height;
const int tlutAddress = texTlut.tmem_offset << 9;
const u8* tlut = &texMem[tlutAddress];
// reduce sample location and texture size to mip level
// move texture pointer to mip location
if (mip)
{
int mipWidth = image_width_minus_1 + 1;
int mipHeight = image_height_minus_1 + 1;
const int fmtWidth = TexDecoder_GetBlockWidthInTexels(texfmt);
const int fmtHeight = TexDecoder_GetBlockHeightInTexels(texfmt);
const int fmtDepth = TexDecoder_GetTexelSizeInNibbles(texfmt);
image_width_minus_1 >>= mip;
image_height_minus_1 >>= mip;
s >>= mip;
t >>= mip;
while (mip)
{
mipWidth = std::max(mipWidth, fmtWidth);
mipHeight = std::max(mipHeight, fmtHeight);
const u32 size = (mipWidth * mipHeight * fmtDepth) >> 1;
imageSrc += size;
mipWidth >>= 1;
mipHeight >>= 1;
mip--;
}
}
if (linear)
{
// offset linear sampling
s -= 64;
t -= 64;
// integer part of sample location
int imageS = s >> 7;
int imageT = t >> 7;
// linear sampling
int imageSPlus1 = imageS + 1;
const int fractS = s & 0x7f;
int imageTPlus1 = imageT + 1;
const int fractT = t & 0x7f;
u8 sampledTex[4];
u32 texel[4];
WrapCoord(&imageS, tm0.wrap_s, image_width_minus_1 + 1);
WrapCoord(&imageT, tm0.wrap_t, image_height_minus_1 + 1);
WrapCoord(&imageSPlus1, tm0.wrap_s, image_width_minus_1 + 1);
WrapCoord(&imageTPlus1, tm0.wrap_t, image_height_minus_1 + 1);
if (!(texfmt == TextureFormat::RGBA8 && texUnit.texImage1[subTexmap].cache_manually_managed))
{
TexDecoder_DecodeTexel(sampledTex, imageSrc, imageS, imageT, image_width_minus_1, texfmt,
tlut, tlutfmt);
SetTexel(sampledTex, texel, (128 - fractS) * (128 - fractT));
TexDecoder_DecodeTexel(sampledTex, imageSrc, imageSPlus1, imageT, image_width_minus_1, texfmt,
tlut, tlutfmt);
AddTexel(sampledTex, texel, (fractS) * (128 - fractT));
TexDecoder_DecodeTexel(sampledTex, imageSrc, imageS, imageTPlus1, image_width_minus_1, texfmt,
tlut, tlutfmt);
AddTexel(sampledTex, texel, (128 - fractS) * (fractT));
TexDecoder_DecodeTexel(sampledTex, imageSrc, imageSPlus1, imageTPlus1, image_width_minus_1,
texfmt, tlut, tlutfmt);
AddTexel(sampledTex, texel, (fractS) * (fractT));
}
else
{
TexDecoder_DecodeTexelRGBA8FromTmem(sampledTex, imageSrc, imageSrcOdd, imageS, imageT,
image_width_minus_1);
SetTexel(sampledTex, texel, (128 - fractS) * (128 - fractT));
TexDecoder_DecodeTexelRGBA8FromTmem(sampledTex, imageSrc, imageSrcOdd, imageSPlus1, imageT,
image_width_minus_1);
AddTexel(sampledTex, texel, (fractS) * (128 - fractT));
TexDecoder_DecodeTexelRGBA8FromTmem(sampledTex, imageSrc, imageSrcOdd, imageS, imageTPlus1,
image_width_minus_1);
AddTexel(sampledTex, texel, (128 - fractS) * (fractT));
TexDecoder_DecodeTexelRGBA8FromTmem(sampledTex, imageSrc, imageSrcOdd, imageSPlus1,
imageTPlus1, image_width_minus_1);
AddTexel(sampledTex, texel, (fractS) * (fractT));
}
sample[0] = (u8)(texel[0] >> 14);
sample[1] = (u8)(texel[1] >> 14);
sample[2] = (u8)(texel[2] >> 14);
sample[3] = (u8)(texel[3] >> 14);
}
else
{
// integer part of sample location
int imageS = s >> 7;
int imageT = t >> 7;
// nearest neighbor sampling
WrapCoord(&imageS, tm0.wrap_s, image_width_minus_1 + 1);
WrapCoord(&imageT, tm0.wrap_t, image_height_minus_1 + 1);
if (!(texfmt == TextureFormat::RGBA8 && texUnit.texImage1[subTexmap].cache_manually_managed))
TexDecoder_DecodeTexel(sample, imageSrc, imageS, imageT, image_width_minus_1, texfmt, tlut,
tlutfmt);
else
TexDecoder_DecodeTexelRGBA8FromTmem(sample, imageSrc, imageSrcOdd, imageS, imageT,
image_width_minus_1);
}
}
} // namespace TextureSampler