lee/renderdoc
1
0
Fork 0
mirror of https://github.com/baldurk/renderdoc.git synced 2026-08-23 23:16:31 +00:00
Code Issues Packages Projects Releases Wiki Activity

DXIL debugger handle all float types for DXOp::Saturate

Browse Source
This commit is contained in:
Jake Turner
2025-10-25 10:16:56 +01:00
parent eebb66eea1
commit 8a725233d6
1 changed files with 18 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
renderdoc/driver/shaders/dxil/dxil_debug.cpp
+18 -4
View File
@@ -3125,16 +3125,30 @@ bool ThreadState::ExecuteInstruction(DebugAPIWrapper *apiWrapper,
{
ShaderVariable arg;
RDCASSERT(GetShaderVariable(inst.args[1], opCode, dxOpCode, arg));
RDCASSERTEQUAL(arg.type, VarType::Float);
RDCASSERTEQUAL(result.type, VarType::Float);
result.value.f32v[0] = dxbc_min(1.0f, dxbc_max(0.0f, arg.value.f32v[0]));
RDCASSERT(IsFloatingPointType(arg.type));
RDCASSERTEQUAL(result.type, arg.type);
if(result.type == VarType::Double)
{
result.value.f64v[0] = dxbc_min(1.0, dxbc_max(0.0, arg.value.f64v[0]));
break;
}
else if(result.type == VarType::Float)
{
result.value.f32v[0] = dxbc_min(1.0f, dxbc_max(0.0f, arg.value.f32v[0]));
break;
}
else if(result.type == VarType::Half)
{
result.value.u16v[0] =
ConvertToHalf(dxbc_min(1.0f, dxbc_max(0.0f, ConvertFromHalf(arg.value.u16v[0]))));
break;
}
break;
}
case DXOp::Dot2:
case DXOp::Dot3:
case DXOp::Dot4:
{
// Float or Int
// 2/3/4 Vector
// Result type must match input types
uint32_t numComps = 4;