Merged main:213028556419 into origin/amd-gfx:0bc25009f46c

Local branch origin/amd-gfx 0bc2500 Merged main:46236f4c3dbe into origin/amd-gfx:8a9df1520f29
Remote branch main 2130285 [flang][cuda] Make sure allocator id is set for pointer allocate (llvm#129950)

Author: SC llvm team

clang/docs/ReleaseNotes.rst

@@ -248,6 +248,8 @@ Bug Fixes in This Version
   (#GH125500).
 - Fixed clang crash when #embed data does not fit into an array
   (#GH128987).
+- Non-local variable and non-variable declarations in the first clause of a ``for`` loop in C are no longer incorrectly
+  considered an error in C23 mode and are allowed as an extension in earlier language modes.
 
 Bug Fixes to Compiler Builtins
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

clang/include/clang/Basic/Builtins.td

@@ -4765,6 +4765,12 @@ def GetDeviceSideMangledName : LangBuiltin<"CUDA_LANG"> {
 }
 
 // HLSL
+def HLSLAddUint64: LangBuiltin<"HLSL_LANG"> {
+  let Spellings = ["__builtin_hlsl_adduint64"];
+  let Attributes = [NoThrow, Const];
+  let Prototype = "void(...)";
+}
+
 def HLSLResourceGetPointer : LangBuiltin<"HLSL_LANG"> {
   let Spellings = ["__builtin_hlsl_resource_getpointer"];
   let Attributes = [NoThrow];

clang/include/clang/Basic/DiagnosticSemaKinds.td

@@ -10709,6 +10709,11 @@ def err_vector_incorrect_num_elements : Error<
   "%select{too many|too few}0 elements in vector %select{initialization|operand}3 (expected %1 elements, have %2)">;
 def err_altivec_empty_initializer : Error<"expected initializer">;
 
+def err_vector_incorrect_bit_count : Error<
+  "incorrect number of bits in vector operand (expected %select{|a multiple of}0 %1 bits, have %2)">;
+def err_integer_incorrect_bit_count : Error<
+  "incorrect number of bits in integer (expected %0 bits, have %1)">;
+
 def err_invalid_neon_type_code : Error<
   "incompatible constant for this __builtin_neon function">;
 def err_argument_invalid_range : Error<
@@ -10797,6 +10802,23 @@ def err_non_local_variable_decl_in_for : Error<
   "declaration of non-local variable in 'for' loop">;
 def err_non_variable_decl_in_for : Error<
   "non-variable declaration in 'for' loop">;
+
+def ext_c23_non_local_variable_decl_in_for : Extension<
+  "declaration of non-local variable in 'for' loop is a C23 extension">,
+  InGroup<C23>;
+
+def warn_c17_non_local_variable_decl_in_for : Warning<
+  "declaration of non-local variable in 'for' loop is incompatible with C standards before C23">,
+  DefaultIgnore, InGroup<CPre23Compat>;
+
+def ext_c23_non_variable_decl_in_for : Extension<
+  "non-variable declaration in 'for' loop is a C23 extension">,
+  InGroup<C23>;
+
+def warn_c17_non_variable_decl_in_for : Warning<
+  "non-variable declaration in 'for' loop is incompatible with C standards before C23">,
+  DefaultIgnore, InGroup<CPre23Compat>;
+
 def err_toomany_element_decls : Error<
   "only one element declaration is allowed">;
 def err_selector_element_not_lvalue : Error<

clang/include/clang/Basic/TargetInfo.h

@@ -253,6 +253,7 @@ class TargetInfo : public TransferrableTargetInfo,
   const char *MCountName;
   unsigned char RegParmMax, SSERegParmMax;
   TargetCXXABI TheCXXABI;
+  bool UseMicrosoftManglingForC = false;
   const LangASMap *AddrSpaceMap;
 
   mutable StringRef PlatformName;
@@ -1344,6 +1345,11 @@ class TargetInfo : public TransferrableTargetInfo,
     return TheCXXABI;
   }
 
+  /// Should the Microsoft mangling scheme be used for C Calling Convention.
+  bool shouldUseMicrosoftCCforMangling() const {
+    return UseMicrosoftManglingForC;
+  }
+
   /// Target the specified CPU.
   ///
   /// \return  False on error (invalid CPU name).

clang/lib/AST/Mangle.cpp

@@ -74,7 +74,7 @@ static CCMangling getCallingConvMangling(const ASTContext &Context,
       if (FD->isMain() && FD->getNumParams() == 2)
         return CCM_WasmMainArgcArgv;
 
-  if (!Triple.isOSWindows() || !Triple.isX86())
+  if (!TI.shouldUseMicrosoftCCforMangling())
     return CCM_Other;
 
   if (Context.getLangOpts().CPlusPlus && !isExternC(ND) &&

clang/lib/Basic/Targets/OSTargets.h

@@ -817,6 +817,7 @@ class LLVM_LIBRARY_VISIBILITY UEFITargetInfo : public OSTargetInfo<Target> {
       : OSTargetInfo<Target>(Triple, Opts) {
     this->WCharType = TargetInfo::UnsignedShort;
     this->WIntType = TargetInfo::UnsignedShort;
+    this->UseMicrosoftManglingForC = true;
   }
 };
 
@@ -837,6 +838,7 @@ class LLVM_LIBRARY_VISIBILITY WindowsTargetInfo : public OSTargetInfo<Target> {
       : OSTargetInfo<Target>(Triple, Opts) {
     this->WCharType = TargetInfo::UnsignedShort;
     this->WIntType = TargetInfo::UnsignedShort;
+    this->UseMicrosoftManglingForC = true;
   }
 };
 

clang/lib/CodeGen/CGBuiltin.cpp

@@ -19470,6 +19470,62 @@ Value *CodeGenFunction::EmitHLSLBuiltinExpr(unsigned BuiltinID,
     return nullptr;
 
   switch (BuiltinID) {
+  case Builtin::BI__builtin_hlsl_adduint64: {
+    Value *OpA = EmitScalarExpr(E->getArg(0));
+    Value *OpB = EmitScalarExpr(E->getArg(1));
+    QualType Arg0Ty = E->getArg(0)->getType();
+    uint64_t NumElements = Arg0Ty->castAs<VectorType>()->getNumElements();
+    assert(Arg0Ty == E->getArg(1)->getType() &&
+           "AddUint64 operand types must match");
+    assert(Arg0Ty->hasIntegerRepresentation() &&
+           "AddUint64 operands must have an integer representation");
+    assert((NumElements == 2 || NumElements == 4) &&
+           "AddUint64 operands must have 2 or 4 elements");
+
+    llvm::Value *LowA;
+    llvm::Value *HighA;
+    llvm::Value *LowB;
+    llvm::Value *HighB;
+
+    // Obtain low and high words of inputs A and B
+    if (NumElements == 2) {
+      LowA = Builder.CreateExtractElement(OpA, (uint64_t)0, "LowA");
+      HighA = Builder.CreateExtractElement(OpA, (uint64_t)1, "HighA");
+      LowB = Builder.CreateExtractElement(OpB, (uint64_t)0, "LowB");
+      HighB = Builder.CreateExtractElement(OpB, (uint64_t)1, "HighB");
+    } else {
+      LowA = Builder.CreateShuffleVector(OpA, ArrayRef<int>{0, 2}, "LowA");
+      HighA = Builder.CreateShuffleVector(OpA, ArrayRef<int>{1, 3}, "HighA");
+      LowB = Builder.CreateShuffleVector(OpB, ArrayRef<int>{0, 2}, "LowB");
+      HighB = Builder.CreateShuffleVector(OpB, ArrayRef<int>{1, 3}, "HighB");
+    }
+
+    // Use an uadd_with_overflow to compute the sum of low words and obtain a
+    // carry value
+    llvm::Value *Carry;
+    llvm::Value *LowSum = EmitOverflowIntrinsic(
+        *this, llvm::Intrinsic::uadd_with_overflow, LowA, LowB, Carry);
+    llvm::Value *ZExtCarry =
+        Builder.CreateZExt(Carry, HighA->getType(), "CarryZExt");
+
+    // Sum the high words and the carry
+    llvm::Value *HighSum = Builder.CreateAdd(HighA, HighB, "HighSum");
+    llvm::Value *HighSumPlusCarry =
+        Builder.CreateAdd(HighSum, ZExtCarry, "HighSumPlusCarry");
+
+    if (NumElements == 4) {
+      return Builder.CreateShuffleVector(LowSum, HighSumPlusCarry,
+                                         ArrayRef<int>{0, 2, 1, 3},
+                                         "hlsl.AddUint64");
+    }
+
+    llvm::Value *Result = PoisonValue::get(OpA->getType());
+    Result = Builder.CreateInsertElement(Result, LowSum, (uint64_t)0,
+                                         "hlsl.AddUint64.upto0");
+    Result = Builder.CreateInsertElement(Result, HighSumPlusCarry, (uint64_t)1,
+                                         "hlsl.AddUint64");
+    return Result;
+  }
   case Builtin::BI__builtin_hlsl_resource_getpointer: {
     Value *HandleOp = EmitScalarExpr(E->getArg(0));
     Value *IndexOp = EmitScalarExpr(E->getArg(1));

clang/lib/Headers/amdgpuintrin.h

@@ -187,7 +187,7 @@ __gpu_match_any_u64(uint64_t __lane_mask, uint64_t __x) {
   uint64_t __match_mask = 0;
 
   bool __done = 0;
-  while (__gpu_ballot(__lane_mask, __done)) {
+  while (__gpu_ballot(__lane_mask, !__done)) {
     if (!__done) {
       uint64_t __first = __gpu_read_first_lane_u64(__lane_mask, __x);
       if (__first == __x) {

clang/lib/Headers/hlsl/hlsl_alias_intrinsics.h

@@ -174,6 +174,27 @@ float3 acos(float3);
 _HLSL_BUILTIN_ALIAS(__builtin_elementwise_acos)
 float4 acos(float4);
 
+//===----------------------------------------------------------------------===//
+// AddUint64 builtins
+//===----------------------------------------------------------------------===//
+
+/// \fn T AddUint64(T a, T b)
+/// \brief Implements unsigned 64-bit integer addition using pairs of unsigned
+/// 32-bit integers.
+/// \param x [in] The first unsigned 32-bit integer pair(s)
+/// \param y [in] The second unsigned 32-bit integer pair(s)
+///
+/// This function takes one or two pairs (low, high) of unsigned 32-bit integer
+/// values and returns pairs (low, high) of unsigned 32-bit integer
+/// values representing the result of unsigned 64-bit integer addition.
+
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_adduint64)
+uint32_t2 AddUint64(uint32_t2, uint32_t2);
+_HLSL_AVAILABILITY(shadermodel, 6.0)
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_adduint64)
+uint32_t4 AddUint64(uint32_t4, uint32_t4);
+
 //===----------------------------------------------------------------------===//
 // all builtins
 //===----------------------------------------------------------------------===//

clang/lib/Headers/nvptxintrin.h

@@ -207,7 +207,7 @@ __gpu_match_any_u64(uint64_t __lane_mask, uint64_t __x) {
   uint64_t __match_mask = 0;
 
   bool __done = 0;
-  while (__gpu_ballot(__lane_mask, __done)) {
+  while (__gpu_ballot(__lane_mask, !__done)) {
     if (!__done) {
       uint64_t __first = __gpu_read_first_lane_u64(__lane_mask, __x);
       if (__first == __x) {