public void PrepareForReturn()
{
if (_config.Stage == ShaderStage.Fragment)
{
if (_config.OmapDepth)
{
Operand dest = Attribute(AttributeConsts.FragmentOutputDepth);
Operand src = Register(_config.GetDepthRegister(), RegisterType.Gpr);
this.Copy(dest, src);
}
int regIndex = 0;
for (int attachment = 0; attachment < 8; attachment++)
{
OmapTarget target = _config.OmapTargets[attachment];
for (int component = 0; component < 4; component++)
{
if (target.ComponentEnabled(component))
{
Operand dest = Attribute(AttributeConsts.FragmentOutputColorBase + attachment * 16 + component * 4);
Operand src = Register(regIndex, RegisterType.Gpr);
this.Copy(dest, src);
regIndex++;
}
}
}
}
}
public void PrepareForReturn()
{
if (Config.Stage == ShaderStage.Fragment)
{
if (Config.OmapDepth)
{
Operand dest = Attribute(AttributeConsts.FragmentOutputDepth);
Operand src = Register(Config.GetDepthRegister(), RegisterType.Gpr);
this.Copy(dest, src);
}
int regIndex = 0;
for (int rtIndex = 0; rtIndex < 8; rtIndex++)
{
OmapTarget target = Config.OmapTargets[rtIndex];
for (int component = 0; component < 4; component++)
{
if (!target.ComponentEnabled(component))
{
continue;
}
int fragmentOutputColorAttr = AttributeConsts.FragmentOutputColorBase + rtIndex * 16;
Operand src = Register(regIndex, RegisterType.Gpr);
// Perform B <-> R swap if needed, for BGRA formats (not supported on OpenGL).
if (component == 0 || component == 2)
{
Operand isBgra = Attribute(AttributeConsts.FragmentOutputIsBgraBase + rtIndex * 4);
Operand lblIsBgra = Label();
Operand lblEnd = Label();
this.BranchIfTrue(lblIsBgra, isBgra);
this.Copy(Attribute(fragmentOutputColorAttr + component * 4), src);
this.Branch(lblEnd);
MarkLabel(lblIsBgra);
this.Copy(Attribute(fragmentOutputColorAttr + (2 - component) * 4), src);
MarkLabel(lblEnd);
}
else
{
this.Copy(Attribute(fragmentOutputColorAttr + component * 4), src);
}
regIndex++;
}
}
}
}
public ShaderHeader(IGpuAccessor gpuAccessor, ulong address)
{
int commonWord0 = gpuAccessor.MemoryRead <int>(address + 0);
int commonWord1 = gpuAccessor.MemoryRead <int>(address + 4);
int commonWord2 = gpuAccessor.MemoryRead <int>(address + 8);
int commonWord3 = gpuAccessor.MemoryRead <int>(address + 12);
int commonWord4 = gpuAccessor.MemoryRead <int>(address + 16);
SphType = commonWord0.Extract(0, 5);
Version = commonWord0.Extract(5, 5);
Stage = (ShaderStage)commonWord0.Extract(10, 4);
// Invalid.
if (Stage == ShaderStage.Compute)
{
Stage = ShaderStage.Vertex;
}
MrtEnable = commonWord0.Extract(14);
KillsPixels = commonWord0.Extract(15);
DoesGlobalStore = commonWord0.Extract(16);
SassVersion = commonWord0.Extract(17, 4);
GpPassthrough = commonWord0.Extract(24);
DoesLoadOrStore = commonWord0.Extract(26);
DoesFp64 = commonWord0.Extract(27);
StreamOutMask = commonWord0.Extract(28, 4);
ShaderLocalMemoryLowSize = commonWord1.Extract(0, 24);
PerPatchAttributeCount = commonWord1.Extract(24, 8);
ShaderLocalMemoryHighSize = commonWord2.Extract(0, 24);
ThreadsPerInputPrimitive = commonWord2.Extract(24, 8);
ShaderLocalMemoryCrsSize = commonWord3.Extract(0, 24);
OutputTopology = (OutputTopology)commonWord3.Extract(24, 4);
MaxOutputVertexCount = commonWord4.Extract(0, 12);
StoreReqStart = commonWord4.Extract(12, 8);
StoreReqEnd = commonWord4.Extract(24, 8);
ImapTypes = new ImapPixelType[32];
for (ulong i = 0; i < 32; i++)
{
byte imap = gpuAccessor.MemoryRead <byte>(address + 0x18 + i);
ImapTypes[i] = new ImapPixelType(
(PixelImap)((imap >> 0) & 3),
(PixelImap)((imap >> 2) & 3),
(PixelImap)((imap >> 4) & 3),
(PixelImap)((imap >> 6) & 3));
}
int type2OmapTarget = gpuAccessor.MemoryRead <int>(address + 0x48);
int type2Omap = gpuAccessor.MemoryRead <int>(address + 0x4c);
OmapTargets = new OmapTarget[8];
for (int offset = 0; offset < OmapTargets.Length * 4; offset += 4)
{
OmapTargets[offset >> 2] = new OmapTarget(
type2OmapTarget.Extract(offset + 0),
type2OmapTarget.Extract(offset + 1),
type2OmapTarget.Extract(offset + 2),
type2OmapTarget.Extract(offset + 3));
}
OmapSampleMask = type2Omap.Extract(0);
OmapDepth = type2Omap.Extract(1);
}
public void PrepareForReturn()
{
if (Config.Stage == ShaderStage.Vertex && (Config.Flags & TranslationFlags.VertexA) == 0)
{
// Here we attempt to implement viewport swizzle on the vertex shader.
// Perform permutation and negation of the output gl_Position components.
// Note that per-viewport swizzling can't be supported using thhis approach.
int swizzleX = Config.GpuAccessor.QueryViewportSwizzle(0);
int swizzleY = Config.GpuAccessor.QueryViewportSwizzle(1);
int swizzleZ = Config.GpuAccessor.QueryViewportSwizzle(2);
int swizzleW = Config.GpuAccessor.QueryViewportSwizzle(3);
bool nonStandardSwizzle = swizzleX != 0 || swizzleY != 2 || swizzleZ != 4 || swizzleW != 6;
if (!Config.GpuAccessor.QuerySupportsViewportSwizzle() && nonStandardSwizzle)
{
Operand[] temp = new Operand[4];
temp[0] = this.Copy(Attribute(AttributeConsts.PositionX));
temp[1] = this.Copy(Attribute(AttributeConsts.PositionY));
temp[2] = this.Copy(Attribute(AttributeConsts.PositionZ));
temp[3] = this.Copy(Attribute(AttributeConsts.PositionW));
this.Copy(Attribute(AttributeConsts.PositionX), this.FPNegate(temp[(swizzleX >> 1) & 3], (swizzleX & 1) != 0));
this.Copy(Attribute(AttributeConsts.PositionY), this.FPNegate(temp[(swizzleY >> 1) & 3], (swizzleY & 1) != 0));
this.Copy(Attribute(AttributeConsts.PositionZ), this.FPNegate(temp[(swizzleZ >> 1) & 3], (swizzleZ & 1) != 0));
this.Copy(Attribute(AttributeConsts.PositionW), this.FPNegate(temp[(swizzleW >> 1) & 3], (swizzleW & 1) != 0));
}
}
else if (Config.Stage == ShaderStage.Fragment)
{
if (Config.OmapDepth)
{
Operand dest = Attribute(AttributeConsts.FragmentOutputDepth);
Operand src = Register(Config.GetDepthRegister(), RegisterType.Gpr);
this.Copy(dest, src);
}
int regIndex = 0;
for (int attachment = 0; attachment < 8; attachment++)
{
OmapTarget target = Config.OmapTargets[attachment];
for (int component = 0; component < 4; component++)
{
if (target.ComponentEnabled(component))
{
Operand dest = Attribute(AttributeConsts.FragmentOutputColorBase + attachment * 16 + component * 4);
Operand src = Register(regIndex, RegisterType.Gpr);
this.Copy(dest, src);
regIndex++;
}
}
}
}
}
public void PrepareForReturn()
{
if (Config.Stage == ShaderStage.Vertex && (Config.Flags & TranslationFlags.VertexA) == 0)
{
// Here we attempt to implement viewport swizzle on the vertex shader.
// Perform permutation and negation of the output gl_Position components.
// Note that per-viewport swizzling can't be supported using this approach.
int swizzleX = Config.GpuAccessor.QueryViewportSwizzle(0);
int swizzleY = Config.GpuAccessor.QueryViewportSwizzle(1);
int swizzleZ = Config.GpuAccessor.QueryViewportSwizzle(2);
int swizzleW = Config.GpuAccessor.QueryViewportSwizzle(3);
bool nonStandardSwizzle = swizzleX != 0 || swizzleY != 2 || swizzleZ != 4 || swizzleW != 6;
if (!Config.GpuAccessor.QuerySupportsViewportSwizzle() && nonStandardSwizzle)
{
Operand[] temp = new Operand[4];
temp[0] = this.Copy(Attribute(AttributeConsts.PositionX));
temp[1] = this.Copy(Attribute(AttributeConsts.PositionY));
temp[2] = this.Copy(Attribute(AttributeConsts.PositionZ));
temp[3] = this.Copy(Attribute(AttributeConsts.PositionW));
this.Copy(Attribute(AttributeConsts.PositionX), this.FPNegate(temp[(swizzleX >> 1) & 3], (swizzleX & 1) != 0));
this.Copy(Attribute(AttributeConsts.PositionY), this.FPNegate(temp[(swizzleY >> 1) & 3], (swizzleY & 1) != 0));
this.Copy(Attribute(AttributeConsts.PositionZ), this.FPNegate(temp[(swizzleZ >> 1) & 3], (swizzleZ & 1) != 0));
this.Copy(Attribute(AttributeConsts.PositionW), this.FPNegate(temp[(swizzleW >> 1) & 3], (swizzleW & 1) != 0));
}
}
else if (Config.Stage == ShaderStage.Fragment)
{
if (Config.OmapDepth)
{
Operand dest = Attribute(AttributeConsts.FragmentOutputDepth);
Operand src = Register(Config.GetDepthRegister(), RegisterType.Gpr);
this.Copy(dest, src);
}
int regIndex = 0;
for (int rtIndex = 0; rtIndex < 8; rtIndex++)
{
OmapTarget target = Config.OmapTargets[rtIndex];
for (int component = 0; component < 4; component++)
{
if (!target.ComponentEnabled(component))
{
continue;
}
int fragmentOutputColorAttr = AttributeConsts.FragmentOutputColorBase + rtIndex * 16;
Operand src = Register(regIndex, RegisterType.Gpr);
// Perform B <-> R swap if needed, for BGRA formats (not supported on OpenGL).
if (component == 0 || component == 2)
{
Operand isBgra = Attribute(AttributeConsts.FragmentOutputIsBgraBase + rtIndex * 4);
Operand lblIsBgra = Label();
Operand lblEnd = Label();
this.BranchIfTrue(lblIsBgra, isBgra);
this.Copy(Attribute(fragmentOutputColorAttr + component * 4), src);
this.Branch(lblEnd);
MarkLabel(lblIsBgra);
this.Copy(Attribute(fragmentOutputColorAttr + (2 - component) * 4), src);
MarkLabel(lblEnd);
}
else
{
this.Copy(Attribute(fragmentOutputColorAttr + component * 4), src);
}
regIndex++;
}
}
}
}
public ShaderHeader(IGpuAccessor gpuAccessor, ulong address)
{
ReadOnlySpan <int> header = MemoryMarshal.Cast <ulong, int>(gpuAccessor.GetCode(address, 0x50));
int commonWord0 = header[0];
int commonWord1 = header[1];
int commonWord2 = header[2];
int commonWord3 = header[3];
int commonWord4 = header[4];
SphType = commonWord0.Extract(0, 5);
Version = commonWord0.Extract(5, 5);
Stage = (ShaderStage)commonWord0.Extract(10, 4);
// Invalid.
if (Stage == ShaderStage.Compute)
{
Stage = ShaderStage.Vertex;
}
MrtEnable = commonWord0.Extract(14);
KillsPixels = commonWord0.Extract(15);
DoesGlobalStore = commonWord0.Extract(16);
SassVersion = commonWord0.Extract(17, 4);
GpPassthrough = commonWord0.Extract(24);
DoesLoadOrStore = commonWord0.Extract(26);
DoesFp64 = commonWord0.Extract(27);
StreamOutMask = commonWord0.Extract(28, 4);
ShaderLocalMemoryLowSize = commonWord1.Extract(0, 24);
PerPatchAttributeCount = commonWord1.Extract(24, 8);
ShaderLocalMemoryHighSize = commonWord2.Extract(0, 24);
ThreadsPerInputPrimitive = commonWord2.Extract(24, 8);
ShaderLocalMemoryCrsSize = commonWord3.Extract(0, 24);
OutputTopology = (OutputTopology)commonWord3.Extract(24, 4);
MaxOutputVertexCount = commonWord4.Extract(0, 12);
StoreReqStart = commonWord4.Extract(12, 8);
StoreReqEnd = commonWord4.Extract(24, 8);
ImapTypes = new ImapPixelType[32];
for (int i = 0; i < 32; i++)
{
byte imap = (byte)(header[6 + (i >> 2)] >> ((i & 3) * 8));
ImapTypes[i] = new ImapPixelType(
(PixelImap)((imap >> 0) & 3),
(PixelImap)((imap >> 2) & 3),
(PixelImap)((imap >> 4) & 3),
(PixelImap)((imap >> 6) & 3));
}
int type2OmapTarget = header[18];
int type2Omap = header[19];
OmapTargets = new OmapTarget[8];
for (int offset = 0; offset < OmapTargets.Length * 4; offset += 4)
{
OmapTargets[offset >> 2] = new OmapTarget(
type2OmapTarget.Extract(offset + 0),
type2OmapTarget.Extract(offset + 1),
type2OmapTarget.Extract(offset + 2),
type2OmapTarget.Extract(offset + 3));
}
OmapSampleMask = type2Omap.Extract(0);
OmapDepth = type2Omap.Extract(1);
}
