[linux-port] Fix DXASSERT warnings (#1322)

The various DXASSERT macros produced a number of different warnings.
The most obvious was providing an empty string for a macro expecting
a message. Substituting in the _NOMSG macro fixes this.
Fixes 13 clang warnings.

A trickier problem for GCC and Clang is a variadic macro that is
provided no variadic arguments. Both Clang and GCC will avoid
errors by using the ##__VA_ARGS__ non-standard behavior, but it's
not an ideal solution. It turns out there aren't that many DXASSERT
macros that actually take variadic arguments. So I added an _ARGS
variant and replaced those instances with that. An ancillary benefit
is that non variadic asserts can present a message with the assert.
Fixes 583 clang warnings and 582 gcc warnings.

In what I expect was an earlier attempt to overload the DXASSERT
macro, it was defined twice each time with a different number of args.
Sadly, macros can't be overloaded, so this just produced more errors.
Fixes 69 clang warnings and 69 gcc warnings.

Finally, I removed the user declaration of std::error_code because
it is defined in a standard header <system_error>. Because clang was
including that header some places, it saw it as a redefinition. But
the Linux version is a struct and windows is a class. So nothing we
put here can make everyone happy. It doesn't really fit with this
change except for in proximity but doesn't fit anywhere else either.
Fixes 52 clang warnings

include/dxc/Support/Global.h

@@ -23,6 +23,7 @@ typedef _Return_type_success_(return >= 0) long HRESULT;
 #endif // !_HRESULT_DEFINED
 
 #include <stdarg.h>
+#include <system_error>
 #include "dxc/Support/exception.h"
 
 ///////////////////////////////////////////////////////////////////////////////
@@ -72,7 +73,6 @@ struct DxcThreadMalloc {
 
 ///////////////////////////////////////////////////////////////////////////////
 // Error handling support.
-namespace std { class error_code; }
 void CheckLLVMErrorCode(const std::error_code &ec);
 
 
@@ -173,6 +173,8 @@ inline void OutputDebugFormatA(_In_ _Printf_format_string_ _Null_terminated_ con
 
 #ifdef DBG
 
+#ifdef _WIN32
+
 // DXASSERT is used to debug break when 'exp' evaluates to false and is only
 //     intended for internal developer use. It is compiled out in free
 //     builds.  This means that code that should be in the final exe should
@@ -185,25 +187,44 @@ inline void OutputDebugFormatA(_In_ _Printf_format_string_ _Null_terminated_ con
 // This prints 'Hello World (i > 10)' and breaks in the debugger if the
 // assertion doesn't hold.
 //
-#define DXASSERT(exp, fmt, ...)\
+#define DXASSERT_ARGS(exp, fmt, ...)\
   do { _Analysis_assume_(exp); if(!(exp)) {                              \
     OutputDebugFormatA("Error: \t%s\nFile:\n%s(%d)\nFunc:\t%s.\n\t" fmt "\n", "!(" #exp ")", __FILE__, __LINE__, __FUNCTION__, __VA_ARGS__); \
     __debugbreak();\
   } } while(0)
+#define DXASSERT(exp, msg) DXASSERT_ARGS(exp, msg)
 
-#define DXASSERT_LOCALVAR(local, exp, s, ...) DXASSERT(exp, s, __VA_ARGS__)
+#define DXASSERT_LOCALVAR(local, exp, msg) DXASSERT(exp, msg)
 
 #define DXASSERT_NOMSG(exp) DXASSERT(exp, "")
 
 #define DXVERIFY_NOMSG(exp) DXASSERT(exp, "")
 
-#else
+#else // _WIN32
+#include <cassert>
+
+#define DXASSERT_NOMSG assert
+
+#define DXASSERT_LOCALVAR(local, exp, msg) DXASSERT(exp, msg)
+
+#define DXVERIFY_NOMSG assert
+
+#define DXASSERT_ARGS(expr, fmt, ...) do { if (!(expr)) { fprintf(stderr, fmt, __VA_ARGS__); assert(false); } } while (0);
+
+#define DXASSERT(expr, msg) do { if (!(expr)) { fprintf(stderr, msg); assert(false && msg); } } while (0);
+
+#endif // _WIN32
+
+#else // DBG
+
+// DXASSERT_ARGS is disabled in free builds.
+#define DXASSERT_ARGS(exp, s, ...) _Analysis_assume_(exp)
 
 // DXASSERT is disabled in free builds.
-#define DXASSERT(exp, s, ...) _Analysis_assume_(exp)
+#define DXASSERT(exp, msg) _Analysis_assume_(exp)
 
 // DXASSERT_LOCALVAR is disabled in free builds, but we keep the local referenced to avoid a warning.
-#define DXASSERT_LOCALVAR(local, exp, s, ...) do { (local); _Analysis_assume_(exp); } while (0)
+#define DXASSERT_LOCALVAR(local, exp, s, ...) do { (void)(local); _Analysis_assume_(exp); } while (0)
 
 // DXASSERT_NOMSG is disabled in free builds.
 #define DXASSERT_NOMSG(exp) _Analysis_assume_(exp)

lib/HLSL/DxilGenerationPass.cpp

@@ -1127,7 +1127,7 @@ Type *UpdateFieldTypeForLegacyLayout(Type *Ty, bool IsCBuf, DxilFieldAnnotation
       rows = matrix.Rows;
       cols = matrix.Cols;
     } else {
-      DXASSERT(matrix.Orientation == MatrixOrientation::ColumnMajor, "");
+      DXASSERT_NOMSG(matrix.Orientation == MatrixOrientation::ColumnMajor);
       cols = matrix.Rows;
       rows = matrix.Cols;
     }

lib/HLSL/DxilSignatureElement.cpp

@@ -227,7 +227,7 @@ void DxilSignatureElement::SetCompType(CompType CT) {
 
 uint8_t DxilSignatureElement::GetColsAsMask() const {
   unsigned StartCol = IsAllocated() ? m_StartCol : 0;
-  DXASSERT(StartCol + m_Cols <= 4, "else start %u and cols %u exceed limit", StartCol, m_Cols);
+  DXASSERT_ARGS(StartCol + m_Cols <= 4, "else start %u and cols %u exceed limit", StartCol, m_Cols);
   DXASSERT(m_Cols >= 1, "else signature takes no space");
   switch (StartCol) {
   case 0: {

lib/HLSL/HLMatrixLowerPass.cpp

@@ -958,7 +958,7 @@ void HLMatrixLowerPass::TranslateMatVecMul(CallInst *matInst,
 
   unsigned col, row;
   Type *EltTy = GetMatrixInfo(matInst->getType(), col, row);
-  DXASSERT(RVal->getType()->getVectorNumElements() == col, "");
+  DXASSERT_NOMSG(RVal->getType()->getVectorNumElements() == col);
 
   bool isFloat = EltTy->isFloatingPointTy();
 
@@ -1010,7 +1010,7 @@ void HLMatrixLowerPass::TranslateVecMatMul(CallInst *matInst,
 
   unsigned col, row;
   Type *EltTy = GetMatrixInfo(matInst->getType(), col, row);
-  DXASSERT(LVal->getType()->getVectorNumElements() == row, "");
+  DXASSERT_NOMSG(LVal->getType()->getVectorNumElements() == row);
 
   bool isFloat = EltTy->isFloatingPointTy();
 

lib/HLSL/HLModule.cpp

@@ -839,7 +839,7 @@ void HLModule::GetParameterRowsAndCols(Type *Ty, unsigned &rows, unsigned &cols,
       rows = matrix.Rows;
       cols = matrix.Cols;
     } else {
-      DXASSERT(matrix.Orientation == MatrixOrientation::ColumnMajor, "");
+      DXASSERT_NOMSG(matrix.Orientation == MatrixOrientation::ColumnMajor);
       cols = matrix.Rows;
       rows = matrix.Cols;
     }

lib/HLSL/HLOperationLower.cpp

@@ -1668,7 +1668,7 @@ Value *TranslateFUIBinary(CallInst *CI, IntrinsicOp IOP, OP::OpCode opcode,
       break;
     case IntrinsicOp::IOP_min:
     default:
-      DXASSERT(IOP == IntrinsicOp::IOP_min, "");
+      DXASSERT_NOMSG(IOP == IntrinsicOp::IOP_min);
       opcode = OP::OpCode::FMin;
       break;
     }
@@ -1683,7 +1683,7 @@ Value *TranslateFUITrinary(CallInst *CI, IntrinsicOp IOP, OP::OpCode opcode,
     switch (IOP) {
     case IntrinsicOp::IOP_mad:
     default:
-      DXASSERT(IOP == IntrinsicOp::IOP_mad, "");
+      DXASSERT_NOMSG(IOP == IntrinsicOp::IOP_mad);
       opcode = OP::OpCode::FMad;
       break;
     }
@@ -3944,7 +3944,7 @@ Value *TranslateProcessIsolineTessFactors(CallInst *CI, IntrinsicOp IOP, OP::OpC
                               HLOperationLowerHelper &helper,  HLObjectOperationLowerHelper *pObjHelper, bool &Translated) {
   hlsl::OP *hlslOP = &helper.hlslOP;
   // Get partition mode 
-  DXASSERT(helper.functionProps, "");
+  DXASSERT_NOMSG(helper.functionProps);
   DXASSERT(helper.functionProps->shaderKind == ShaderModel::Kind::Hull, "must be hull shader");
   DXIL::TessellatorPartitioning partition = helper.functionProps->ShaderProps.HS.partition;
   
@@ -4124,7 +4124,7 @@ Value *TranslateProcessTessFactors(CallInst *CI, IntrinsicOp IOP, OP::OpCode opc
                               HLOperationLowerHelper &helper,  HLObjectOperationLowerHelper *pObjHelper, bool &Translated) {
   hlsl::OP *hlslOP = &helper.hlslOP;
   // Get partition mode 
-  DXASSERT(helper.functionProps, "");
+  DXASSERT_NOMSG(helper.functionProps);
   DXASSERT(helper.functionProps->shaderKind == ShaderModel::Kind::Hull, "must be hull shader");
   DXIL::TessellatorPartitioning partition = helper.functionProps->ShaderProps.HS.partition;
   
@@ -6262,7 +6262,7 @@ void TranslateDefaultSubscript(CallInst *CI, HLOperationLowerHelper &helper,  HL
     } else if (GetElementPtrInst *GEP = dyn_cast<GetElementPtrInst>(user)) {
       // Must be vector type here.
       unsigned vectorSize = Ty->getVectorNumElements();
-      DXASSERT(GEP->getNumIndices() == 2, "");
+      DXASSERT_NOMSG(GEP->getNumIndices() == 2);
       Use *GEPIdx = GEP->idx_begin();
       GEPIdx++;
       Value *EltIdx = *GEPIdx;

lib/HLSL/HLSignatureLower.cpp

@@ -889,7 +889,7 @@ void GenerateInputOutputUserCall(InputOutputAccessInfo &info, Value *undefVertex
     if (group == HLOpcodeGroup::HLIntrinsic)
       return;
     unsigned opcode = GetHLOpcode(CI);
-    DXASSERT(group == HLOpcodeGroup::HLMatLoadStore, "");
+    DXASSERT_NOMSG(group == HLOpcodeGroup::HLMatLoadStore);
     HLMatLoadStoreOpcode matOp = static_cast<HLMatLoadStoreOpcode>(opcode);
     switch (matOp) {
     case HLMatLoadStoreOpcode::ColMatLoad: {
@@ -1476,7 +1476,7 @@ void HLSignatureLower::GenerateStreamOutputOperation(Value *streamVal, unsigned
     CallInst *CI = cast<CallInst>(user);
     HLOpcodeGroup group = GetHLOpcodeGroupByName(CI->getCalledFunction());
     unsigned opcode = GetHLOpcode(CI);
-    DXASSERT_LOCALVAR(group, group == HLOpcodeGroup::HLIntrinsic, "");
+    DXASSERT_LOCALVAR(group, group == HLOpcodeGroup::HLIntrinsic, "Must be HLIntrinsic here");
     IntrinsicOp IOP = static_cast<IntrinsicOp>(opcode);
     switch (IOP) {
     case IntrinsicOp::MOP_Append:
@@ -1544,4 +1544,4 @@ void HLSignatureLower::Run() {
     GenerateDxilPatchConstantFunctionInputs();
   if (props.IsGS())
     GenerateStreamOutputOperations();
-}
+}

lib/Transforms/Scalar/ScalarReplAggregatesHLSL.cpp

@@ -4530,7 +4530,7 @@ static unsigned AllocateSemanticIndex(
       if (matrix.Orientation == MatrixOrientation::RowMajor) {
         rows = matrix.Rows;
       } else {
-        DXASSERT(matrix.Orientation == MatrixOrientation::ColumnMajor, "");
+        DXASSERT_NOMSG(matrix.Orientation == MatrixOrientation::ColumnMajor);
         rows = matrix.Cols;
       }
     }

tools/clang/lib/CodeGen/CGHLSLMS.cpp

@@ -5033,7 +5033,7 @@ Value *CGMSHLSLRuntime::EmitHLSLInitListExpr(CodeGenFunction &CGF, InitListExpr
   llvm::Type *RetTy = CGF.ConvertType(ResultTy);
   if (DestPtr) {
     SmallVector<Value *, 4> ParamList;
-    DXASSERT(RetTy->isAggregateType(), "");
+    DXASSERT_NOMSG(RetTy->isAggregateType());
     ParamList.emplace_back(DestPtr);
     ParamList.append(EltValList.begin(), EltValList.end());
     idx = 0;

tools/clang/unittests/HLSL/ExecutionTest.cpp

@@ -2461,7 +2461,7 @@ public:
         return &m_table[i];
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -2487,7 +2487,7 @@ public:
         return &(m_table[i].m_int32Table);
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -2498,7 +2498,7 @@ public:
         return &(m_table[i].m_int8Table);
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -2509,7 +2509,7 @@ public:
         return &(m_table[i].m_int16Table);
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -2520,7 +2520,7 @@ public:
         return &(m_table[i].m_uint32Table);
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -2531,7 +2531,7 @@ public:
         return &(m_table[i].m_floatTable);
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -2543,7 +2543,7 @@ public:
         return &(m_table[i].m_halfTable);
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -2554,7 +2554,7 @@ public:
         return &(m_table[i].m_doubleTable);
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -2565,7 +2565,7 @@ public:
         return &(m_table[i].m_boolTable);
       }
     }
-    DXASSERT(false, "Invalid Table Parameter Name %s", name);
+    DXASSERT_ARGS(false, "Invalid Table Parameter Name %s", name);
     return nullptr;
   }
 
@@ -5030,7 +5030,7 @@ ShaderOpKind GetShaderOpKind(LPCWSTR str) {
       return ShaderOpKindTable[i].kind;
     }
   }
-  DXASSERT(false, "Invalid ShaderOp name: %s", str);
+  DXASSERT_ARGS(false, "Invalid ShaderOp name: %s", str);
   return ShaderOpKind::ShaderOpInvalid;
 }
 
@@ -5259,7 +5259,7 @@ static OutType computeExpectedWithShaderOp(const std::vector<InType> &inputs,
   case ShaderOpKind::WaveActiveAllEqual:
     return computeExpected<InType, OutType, ShaderOpKind::WaveActiveAllEqual>()(inputs, masks, maskValue, index);
   default:
-    DXASSERT(false, "Invalid ShaderOp Name: %s", str);
+    DXASSERT_ARGS(false, "Invalid ShaderOp Name: %s", str);
     return (OutType) 0;
   }
 };

tools/clang/unittests/HLSL/ShaderOpTest.cpp

@@ -2005,7 +2005,7 @@ void ParseDataFromText(LPCWSTR pText, LPCWSTR pEnd, DXIL::ComponentType compType
     V.insert(V.end(), pB, pB + sizeof(int));
   }
   else {
-    DXASSERT(false, "Unsupported stream component type : %u", compType);
+    DXASSERT_ARGS(false, "Unsupported stream component type : %u", compType);
   }
 }