ppsspp/GPU/Vulkan/VertexShaderGeneratorVulkan.cpp
2016-12-21 15:50:20 +01:00

503 lines
18 KiB
C++

// Copyright (c) 2012- PPSSPP Project.
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, version 2.0 or later versions.
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License 2.0 for more details.
// A copy of the GPL 2.0 should have been included with the program.
// If not, see http://www.gnu.org/licenses/
// Official git repository and contact information can be found at
// https://github.com/hrydgard/ppsspp and http://www.ppsspp.org/.
#include <cstdio>
#include <cstdlib>
#include <locale.h>
#include "gfx_es2/gpu_features.h"
#if defined(_WIN32) && defined(_DEBUG)
#include "Common/CommonWindows.h"
#endif
#include "base/stringutil.h"
#include "Common/Vulkan/VulkanLoader.h"
#include "Core/Config.h"
#include "GPU/ge_constants.h"
#include "GPU/GPUState.h"
#include "GPU/Common/ShaderId.h"
#include "GPU/Common/VertexDecoderCommon.h"
#include "GPU/Vulkan/VertexShaderGeneratorVulkan.h"
#include "GPU/Vulkan/PipelineManagerVulkan.h"
#include "GPU/Vulkan/ShaderManagerVulkan.h"
static const char *vulkan_glsl_preamble =
"#version 400\n"
"#extension GL_ARB_separate_shader_objects : enable\n"
"#extension GL_ARB_shading_language_420pack : enable\n\n";
// "Varying" layout - must match fragment shader
// color0 = 0
// color1 = 1
// texcoord = 2
// fog = 3
#undef WRITE
#define WRITE p+=sprintf
static const char * const boneWeightDecl[9] = {
"#ERROR#",
"layout(location = 3) in float w1;\n",
"layout(location = 3) in vec2 w1;\n",
"layout(location = 3) in vec3 w1;\n",
"layout(location = 3) in vec4 w1;\n",
"layout(location = 3) in vec4 w1;\nlayout(location = 4) in float w2;\n",
"layout(location = 3) in vec4 w1;\nlayout(location = 4) in vec2 w2;\n",
"layout(location = 3) in vec4 w1;\nlayout(location = 4) in vec3 w2;\n",
"layout(location = 3) in vec4 w1;\nlayout(location = 4) in vec4 w2;\n",
};
enum DoLightComputation {
LIGHT_OFF,
LIGHT_SHADE,
LIGHT_FULL,
};
// Depth range and viewport
//
// After the multiplication with the projection matrix, we have a 4D vector in clip space.
// In OpenGL, Z is from -1 to 1, while in D3D, Z is from 0 to 1.
// PSP appears to use the OpenGL convention. As Z is from -1 to 1, and the viewport is represented
// by a center and a scale, to find the final Z value, all we need to do is to multiply by ZScale and
// add ZCenter - these are properly scaled to directly give a Z value in [0, 65535].
//
// z = vec.z * ViewportZScale + ViewportZCenter;
//
// That will give us the final value between 0 and 65535, which we can simply floor to simulate
// the limited precision of the PSP's depth buffer. Then we convert it back:
// z = floor(z);
//
// vec.z = (z - ViewportZCenter) / ViewportZScale;
//
// Now, the regular machinery will take over and do the calculation again.
//
// All this above is for full transform mode.
// In through mode, the Z coordinate just goes straight through and there is no perspective division.
// We simulate this of course with pretty much an identity matrix. Rounding Z becomes very easy.
//
// TODO: Skip all this if we can actually get a 16-bit depth buffer along with stencil, which
// is a bit of a rare configuration, although quite common on mobile.
bool GenerateVulkanGLSLVertexShader(const ShaderID &id, char *buffer, bool *usesLighting) {
char *p = buffer;
WRITE(p, "%s", vulkan_glsl_preamble);
bool highpFog = false;
bool highpTexcoord = false;
bool isModeThrough = id.Bit(VS_BIT_IS_THROUGH);
bool lmode = id.Bit(VS_BIT_LMODE) && !isModeThrough; // TODO: Different expression than in shaderIDgen
bool doTexture = id.Bit(VS_BIT_DO_TEXTURE);
bool doTextureProjection = id.Bit(VS_BIT_DO_TEXTURE_PROJ);
GETexMapMode uvGenMode = static_cast<GETexMapMode>(id.Bits(VS_BIT_UVGEN_MODE, 2));
// this is only valid for some settings of uvGenMode
GETexProjMapMode uvProjMode = static_cast<GETexProjMapMode>(id.Bits(VS_BIT_UVPROJ_MODE, 2));
bool doShadeMapping = uvGenMode == GE_TEXMAP_ENVIRONMENT_MAP;
bool doFlatShading = id.Bit(VS_BIT_FLATSHADE);
bool useHWTransform = id.Bit(VS_BIT_USE_HW_TRANSFORM);
bool hasColor = id.Bit(VS_BIT_HAS_COLOR) || !useHWTransform;
bool hasNormal = id.Bit(VS_BIT_HAS_NORMAL) && useHWTransform;
bool hasTexcoord = id.Bit(VS_BIT_HAS_TEXCOORD) || !useHWTransform;
bool enableFog = id.Bit(VS_BIT_ENABLE_FOG);
bool throughmode = id.Bit(VS_BIT_IS_THROUGH);
bool flipNormal = id.Bit(VS_BIT_NORM_REVERSE);
int ls0 = id.Bits(VS_BIT_LS0, 2);
int ls1 = id.Bits(VS_BIT_LS1, 2);
bool enableBones = id.Bit(VS_BIT_ENABLE_BONES);
bool enableLighting = id.Bit(VS_BIT_LIGHTING_ENABLE);
int matUpdate = id.Bits(VS_BIT_MATERIAL_UPDATE, 3);
// The uniforms are passed in as three "clumps" that may or may not be present.
// We will memcpy the parts into place in a big buffer so we can be quite dynamic about what parts
// are present and what parts aren't, but we will not be ultra detailed about it.
*usesLighting = enableLighting || doShadeMapping;
WRITE(p, "\n");
WRITE(p, "layout (std140, set = 0, binding = 2) uniform baseVars {\n%s} base;\n", ub_baseStr);
if (enableLighting || doShadeMapping)
WRITE(p, "layout (std140, set = 0, binding = 3) uniform lightVars {\n%s} light;\n", ub_vs_lightsStr);
if (enableBones)
WRITE(p, "layout (std140, set = 0, binding = 4) uniform boneVars {\n%s} bone;\n", ub_vs_bonesStr);
const char *shading = doFlatShading ? "flat " : "";
DoLightComputation doLight[4] = { LIGHT_OFF, LIGHT_OFF, LIGHT_OFF, LIGHT_OFF };
if (useHWTransform) {
int shadeLight0 = doShadeMapping ? ls0 : -1;
int shadeLight1 = doShadeMapping ? ls1 : -1;
for (int i = 0; i < 4; i++) {
if (i == shadeLight0 || i == shadeLight1)
doLight[i] = LIGHT_SHADE;
if (id.Bit(VS_BIT_LIGHTING_ENABLE) && id.Bit(VS_BIT_LIGHT0_ENABLE + i))
doLight[i] = LIGHT_FULL;
}
}
int numBoneWeights = 0;
int boneWeightScale = id.Bits(VS_BIT_WEIGHT_FMTSCALE, 2);
if (enableBones) {
numBoneWeights = 1 + id.Bits(VS_BIT_BONES, 3);
WRITE(p, "%s", boneWeightDecl[numBoneWeights]);
}
if (useHWTransform)
WRITE(p, "layout (location = %d) in vec3 position;\n", PspAttributeLocation::POSITION);
else
// we pass the fog coord in w
WRITE(p, "layout (location = %d) in vec4 position;\n", PspAttributeLocation::POSITION);
if (useHWTransform && hasNormal)
WRITE(p, "layout (location = %d) in vec3 normal;\n", PspAttributeLocation::NORMAL);
if (doTexture && hasTexcoord) {
if (!useHWTransform && doTextureProjection && !throughmode)
WRITE(p, "layout (location = %d) in vec3 texcoord;\n", PspAttributeLocation::TEXCOORD);
else
WRITE(p, "layout (location = %d) in vec2 texcoord;\n", PspAttributeLocation::TEXCOORD);
}
if (hasColor) {
WRITE(p, "layout (location = %d) in vec4 color0;\n", PspAttributeLocation::COLOR0);
if (lmode && !useHWTransform) // only software transform supplies color1 as vertex data
WRITE(p, "layout (location = %d) in vec3 color1;\n", PspAttributeLocation::COLOR1);
}
WRITE(p, "layout (location = 1) %sout vec4 v_color0;\n", shading);
if (lmode) {
WRITE(p, "layout (location = 2) %sout vec3 v_color1;\n", shading);
}
if (doTexture) {
if (doTextureProjection) {
WRITE(p, "layout (location = 0) out vec3 v_texcoord;\n");
} else {
WRITE(p, "layout (location = 0) out vec2 v_texcoord;\n");
}
}
if (enableFog) {
// See the fragment shader generator
WRITE(p, "layout (location = 3) out float v_fogdepth;\n");
}
// See comment above this function (GenerateVertexShader).
if (!isModeThrough && gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
// Apply the projection and viewport to get the Z buffer value, floor to integer, undo the viewport and projection.
WRITE(p, "\nvec4 depthRoundZVP(vec4 v) {\n");
WRITE(p, " float z = v.z / v.w;\n");
WRITE(p, " z = z * base.depthRange.x + base.depthRange.y;\n");
WRITE(p, " z = floor(z);\n");
WRITE(p, " z = (z - base.depthRange.z) * base.depthRange.w;\n");
WRITE(p, " return vec4(v.x, v.y, z * v.w, v.w);\n");
WRITE(p, "}\n\n");
}
WRITE(p, "out gl_PerVertex { vec4 gl_Position; };\n");
WRITE(p, "void main() {\n");
if (!useHWTransform) {
// Simple pass-through of vertex data to fragment shader
if (doTexture) {
if (throughmode && doTextureProjection) {
WRITE(p, " v_texcoord = vec3(texcoord, 1.0);\n");
} else {
WRITE(p, " v_texcoord = texcoord;\n");
}
}
if (hasColor) {
WRITE(p, " v_color0 = color0;\n");
if (lmode)
WRITE(p, " v_color1 = color1;\n");
} else {
WRITE(p, " v_color0 = base.matambientalpha;\n");
if (lmode)
WRITE(p, " v_color1 = vec3(0.0);\n");
}
if (enableFog) {
WRITE(p, " v_fogdepth = position.w;\n");
}
if (isModeThrough) {
WRITE(p, " gl_Position = base.proj_through_mtx * vec4(position.xyz, 1.0);\n");
} else {
// The viewport is used in this case, so need to compensate for that.
if (gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
WRITE(p, " gl_Position = depthRoundZVP(base.proj_mtx * vec4(position.xyz, 1.0));\n");
} else {
WRITE(p, " gl_Position = base.proj_mtx * vec4(position.xyz, 1.0);\n");
}
}
} else {
// Step 1: World Transform / Skinning
if (!enableBones) {
// No skinning, just standard T&L.
WRITE(p, " vec3 worldpos = (base.world_mtx * vec4(position.xyz, 1.0)).xyz;\n");
if (hasNormal)
WRITE(p, " mediump vec3 worldnormal = normalize((base.world_mtx * vec4(%snormal, 0.0)).xyz);\n", flipNormal ? "-" : "");
else
WRITE(p, " mediump vec3 worldnormal = vec3(0.0, 0.0, 1.0);\n");
} else {
static const char *rescale[4] = { "", " * 1.9921875", " * 1.999969482421875", "" }; // 2*127.5f/128.f, 2*32767.5f/32768.f, 1.0f};
const char *factor = rescale[boneWeightScale];
static const char * const boneWeightAttr[8] = {
"w1.x", "w1.y", "w1.z", "w1.w",
"w2.x", "w2.y", "w2.z", "w2.w",
};
WRITE(p, " mat4 skinMatrix = w1.x * bone.m[0];\n");
if (numBoneWeights > 1) {
for (int i = 1; i < numBoneWeights; i++) {
WRITE(p, " skinMatrix += %s * bone.m[%i];\n", boneWeightAttr[i], i);
}
}
WRITE(p, ";\n");
// Trying to simplify this results in bugs in LBP...
WRITE(p, " vec3 skinnedpos = (skinMatrix * vec4(position, 1.0)).xyz %s;\n", factor);
WRITE(p, " vec3 worldpos = (base.world_mtx * vec4(skinnedpos, 1.0)).xyz;\n");
if (hasNormal) {
WRITE(p, " mediump vec3 skinnednormal = (skinMatrix * vec4(%snormal, 0.0)).xyz %s;\n", flipNormal ? "-" : "", factor);
} else {
WRITE(p, " mediump vec3 skinnednormal = (skinMatrix * vec4(0.0, 0.0, %s1.0, 0.0)).xyz %s;\n", flipNormal ? "-" : "", factor);
}
WRITE(p, " mediump vec3 worldnormal = normalize((base.world_mtx * vec4(skinnednormal, 0.0)).xyz);\n");
}
WRITE(p, " vec4 viewPos = base.view_mtx * vec4(worldpos, 1.0);\n");
// Final view and projection transforms.
if (gstate_c.Supports(GPU_ROUND_DEPTH_TO_16BIT)) {
WRITE(p, " gl_Position = depthRoundZVP(base.proj_mtx * viewPos);\n");
} else {
WRITE(p, " gl_Position = base.proj_mtx * viewPos;\n");
}
// TODO: Declare variables for dots for shade mapping if needed.
const char *ambientStr = ((matUpdate & 1) && hasColor) ? "color0" : "base.matambientalpha";
const char *diffuseStr = ((matUpdate & 2) && hasColor) ? "color0.rgb" : "light.matdiffuse";
const char *specularStr = ((matUpdate & 4) && hasColor) ? "color0.rgb" : "light.matspecular.rgb";
bool diffuseIsZero = true;
bool specularIsZero = true;
bool distanceNeeded = false;
if (enableLighting) {
WRITE(p, " vec4 lightSum0 = light.globalAmbient * %s + vec4(light.matemissive, 0.0);\n", ambientStr);
for (int i = 0; i < 4; i++) {
GELightType type = static_cast<GELightType>(id.Bits(VS_BIT_LIGHT0_TYPE + 4 * i, 2));
GELightComputation comp = static_cast<GELightComputation>(id.Bits(VS_BIT_LIGHT0_COMP + 4 * i, 2));
if (doLight[i] != LIGHT_FULL)
continue;
diffuseIsZero = false;
if (comp != GE_LIGHTCOMP_ONLYDIFFUSE)
specularIsZero = false;
if (type != GE_LIGHTTYPE_DIRECTIONAL)
distanceNeeded = true;
}
if (!specularIsZero) {
WRITE(p, " vec3 lightSum1 = vec3(0.0);\n");
}
if (!diffuseIsZero) {
WRITE(p, " vec3 toLight;\n");
WRITE(p, " vec3 diffuse;\n");
}
if (distanceNeeded) {
WRITE(p, " float distance;\n");
WRITE(p, " float lightScale;\n");
}
}
// Calculate lights if needed. If shade mapping is enabled, lights may need to be
// at least partially calculated.
for (int i = 0; i < 4; i++) {
if (doLight[i] != LIGHT_FULL)
continue;
GELightType type = static_cast<GELightType>(id.Bits(VS_BIT_LIGHT0_TYPE + 4 * i, 2));
GELightComputation comp = static_cast<GELightComputation>(id.Bits(VS_BIT_LIGHT0_COMP + 4 * i, 2));
if (type == GE_LIGHTTYPE_DIRECTIONAL) {
// We prenormalize light positions for directional lights.
WRITE(p, " toLight = light.pos[%i];\n", i);
} else {
WRITE(p, " toLight = light.pos[%i] - worldpos;\n", i);
WRITE(p, " distance = length(toLight);\n");
WRITE(p, " toLight /= distance;\n");
}
bool doSpecular = comp != GE_LIGHTCOMP_ONLYDIFFUSE;
bool poweredDiffuse = comp == GE_LIGHTCOMP_BOTHWITHPOWDIFFUSE;
WRITE(p, " mediump float dot%i = max(dot(toLight, worldnormal), 0.0);\n", i);
if (poweredDiffuse) {
// pow(0.0, 0.0) may be undefined, but the PSP seems to treat it as 1.0.
// Seen in Tales of the World: Radiant Mythology (#2424.)
WRITE(p, " if (dot%i == 0.0 && light.matspecular.a == 0.0) {\n", i);
WRITE(p, " dot%i = 1.0;\n", i);
WRITE(p, " } else {\n");
WRITE(p, " dot%i = pow(dot%i, light.matspecular.a);\n", i, i);
WRITE(p, " }\n");
}
const char *timesLightScale = " * lightScale";
// Attenuation
switch (type) {
case GE_LIGHTTYPE_DIRECTIONAL:
timesLightScale = "";
break;
case GE_LIGHTTYPE_POINT:
WRITE(p, " lightScale = clamp(1.0 / dot(light.att[%i], vec3(1.0, distance, distance*distance)), 0.0, 1.0);\n", i);
break;
case GE_LIGHTTYPE_SPOT:
case GE_LIGHTTYPE_UNKNOWN:
WRITE(p, " float angle%i = dot(normalize(light.dir[%i]), toLight);\n", i, i);
WRITE(p, " if (angle%i >= light.angle[%i]) {\n", i, i);
WRITE(p, " lightScale = clamp(1.0 / dot(light.att[%i], vec3(1.0, distance, distance*distance)), 0.0, 1.0) * pow(angle%i, light.spotCoef[%i]);\n", i, i, i);
WRITE(p, " } else {\n");
WRITE(p, " lightScale = 0.0;\n");
WRITE(p, " }\n");
break;
default:
// ILLEGAL
break;
}
WRITE(p, " diffuse = (light.diffuse[%i] * %s) * dot%i;\n", i, diffuseStr, i);
if (doSpecular) {
WRITE(p, " dot%i = dot(normalize(toLight + vec3(0.0, 0.0, 1.0)), worldnormal);\n", i);
WRITE(p, " if (dot%i > 0.0)\n", i);
WRITE(p, " lightSum1 += light.specular[%i] * %s * (pow(dot%i, light.matspecular.a) %s);\n", i, specularStr, i, timesLightScale);
}
WRITE(p, " lightSum0.rgb += (light.ambient[%i] * %s.rgb + diffuse)%s;\n", i, ambientStr, timesLightScale);
}
if (enableLighting) {
// Sum up ambient, emissive here.
if (lmode) {
WRITE(p, " v_color0 = clamp(lightSum0, 0.0, 1.0);\n");
// v_color1 only exists when lmode = 1.
if (specularIsZero) {
WRITE(p, " v_color1 = vec3(0.0);\n");
} else {
WRITE(p, " v_color1 = clamp(lightSum1, 0.0, 1.0);\n");
}
} else {
if (specularIsZero) {
WRITE(p, " v_color0 = clamp(lightSum0, 0.0, 1.0);\n");
} else {
WRITE(p, " v_color0 = clamp(clamp(lightSum0, 0.0, 1.0) + vec4(lightSum1, 0.0), 0.0, 1.0);\n");
}
}
} else {
// Lighting doesn't affect color.
if (hasColor) {
WRITE(p, " v_color0 = color0;\n");
} else {
WRITE(p, " v_color0 = base.matambientalpha;\n");
}
if (lmode) {
WRITE(p, " v_color1 = vec3(0.0);\n");
}
}
bool scaleUV = !throughmode && (uvGenMode == GE_TEXMAP_TEXTURE_COORDS || uvGenMode == GE_TEXMAP_UNKNOWN);
// Step 3: UV generation
if (doTexture) {
switch (uvGenMode) {
case GE_TEXMAP_TEXTURE_COORDS: // Scale-offset. Easy.
case GE_TEXMAP_UNKNOWN: // Not sure what this is, but Riviera uses it. Treating as coords works.
if (scaleUV) {
if (hasTexcoord) {
WRITE(p, " v_texcoord = texcoord;\n");
} else {
WRITE(p, " v_texcoord = vec2(0.0);\n");
}
} else {
if (hasTexcoord) {
WRITE(p, " v_texcoord = texcoord * base.uvscaleoffset.xy + base.uvscaleoffset.zw;\n");
} else {
WRITE(p, " v_texcoord = base.uvscaleoffset.zw;\n");
}
}
break;
case GE_TEXMAP_TEXTURE_MATRIX: // Projection mapping.
{
std::string temp_tc;
switch (uvProjMode) {
case GE_PROJMAP_POSITION: // Use model space XYZ as source
temp_tc = "vec4(position.xyz, 1.0)";
break;
case GE_PROJMAP_UV: // Use unscaled UV as source
{
// scaleUV is false here.
if (hasTexcoord) {
temp_tc = "vec4(texcoord.xy, 0.0, 1.0)";
} else {
temp_tc = "vec4(0.0, 0.0, 0.0, 1.0)";
}
}
break;
case GE_PROJMAP_NORMALIZED_NORMAL: // Use normalized transformed normal as source
if (hasNormal)
temp_tc = flipNormal ? "vec4(normalize(-normal), 1.0)" : "vec4(normalize(normal), 1.0)";
else
temp_tc = "vec4(0.0, 0.0, 1.0, 1.0)";
break;
case GE_PROJMAP_NORMAL: // Use non-normalized transformed normal as source
if (hasNormal)
temp_tc = flipNormal ? "vec4(-normal, 1.0)" : "vec4(normal, 1.0)";
else
temp_tc = "vec4(0.0, 0.0, 1.0, 1.0)";
break;
}
// Transform by texture matrix. XYZ as we are doing projection mapping.
WRITE(p, " v_texcoord = (base.tex_mtx * %s).xyz * vec3(base.uvscaleoffset.xy, 1.0);\n", temp_tc.c_str());
}
break;
case GE_TEXMAP_ENVIRONMENT_MAP: // Shade mapping - use dots from light sources.
WRITE(p, " v_texcoord = base.uvscaleoffset.xy * vec2(1.0 + dot(normalize(light.pos[%i]), worldnormal), 1.0 + dot(normalize(light.pos[%i]), worldnormal)) * 0.5;\n", ls0, ls1);
break;
default:
// ILLEGAL
break;
}
}
// Compute fogdepth
if (enableFog)
WRITE(p, " v_fogdepth = (viewPos.z + base.fogcoef_stencilreplace.x) * base.fogcoef_stencilreplace.y;\n");
}
WRITE(p, "}\n");
return true;
}