Merge from 'main' to 'sycl-web' (44 commits)

  CONFLICT (content): Merge conflict in clang/docs/LanguageExtensions.rst

bolt/test/runtime/instrument-wrong-target.s

@@ -1,7 +1,8 @@
 # Test that BOLT errs when trying to instrument a binary with a different
 # architecture than the one BOLT is built for.
 
-# REQUIRES: x86_64-linux,bolt-runtime,target=x86_64{{.*}}
+# REQUIRES: x86_64-linux,bolt-runtime
+# REQUIRES: target-x86_64 && aarch64-registered-target
 
 # RUN: llvm-mc -triple aarch64 -filetype=obj %s -o %t.o
 # RUN: ld.lld -q -pie -o %t.exe %t.o

clang/docs/LanguageExtensions.rst

@@ -3494,6 +3494,37 @@ internal linkage.
   // Computes a unique stable name for a given variable.
   constexpr const char * __builtin_sycl_unique_stable_id( expr );
 
+``__builtin_popcountg``
+-----------------------
+
+``__builtin_popcountg`` returns the number of 1 bits in the argument. The
+argument can be of any integer type.
+
+**Syntax**:
+
+.. code-block:: c++
+
+  int __builtin_popcountg(type x)
+
+**Examples**:
+
+.. code-block:: c++
+
+  int x = 1;
+  int x_pop = __builtin_popcountg(x);
+
+  unsigned long y = 3;
+  int y_pop = __builtin_popcountg(y);
+
+  _BitInt(128) z = 7;
+  int z_pop = __builtin_popcountg(z);
+
+**Description**:
+
+``__builtin_popcountg`` is meant to be a type-generic alternative to the
+``__builtin_popcount{,l,ll}`` builtins, with support for other integer types,
+such as ``__int128`` and C23 ``_BitInt(N)``.
+
 Multiprecision Arithmetic Builtins
 ----------------------------------
 

clang/docs/ReleaseNotes.rst

@@ -283,6 +283,8 @@ Bug Fixes to C++ Support
   (`#78524 <https://github.com/llvm/llvm-project/issues/78524>`_)
 - Clang no longer instantiates the exception specification of discarded candidate function
   templates when determining the primary template of an explicit specialization.
+- Fixed a crash in Microsoft compatibility mode where unqualified dependent base class
+  lookup searches the bases of an incomplete class.
 
 Bug Fixes to AST Handling
 ^^^^^^^^^^^^^^^^^^^^^^^^^

clang/include/clang/Analysis/Analyses/ThreadSafetyCommon.h

@@ -527,8 +527,10 @@ class SExprBuilder {
   BlockInfo *CurrentBlockInfo = nullptr;
 };
 
+#ifndef NDEBUG
 // Dump an SCFG to llvm::errs().
 void printSCFG(CFGWalker &Walker);
+#endif // NDEBUG
 
 } // namespace threadSafety
 } // namespace clang

clang/include/clang/Basic/Builtins.td

@@ -688,6 +688,12 @@ def Popcount : Builtin, BitInt_Long_LongLongTemplate {
   let Prototype = "int(unsigned T)";
 }
 
+def Popcountg : Builtin {
+  let Spellings = ["__builtin_popcountg"];
+  let Attributes = [NoThrow, Const];
+  let Prototype = "int(...)";
+}
+
 def Clrsb : Builtin, BitInt_Long_LongLongTemplate {
   let Spellings = ["__builtin_clrsb"];
   let Attributes = [NoThrow, Const, Constexpr];
@@ -4524,6 +4530,12 @@ def HLSLCreateHandle : LangBuiltin<"HLSL_LANG"> {
   let Prototype = "void*(unsigned char)";
 }
 
+def HLSLDotProduct : LangBuiltin<"HLSL_LANG"> {
+  let Spellings = ["__builtin_hlsl_dot"];
+  let Attributes = [NoThrow, Const];
+  let Prototype = "void(...)";
+}
+
 // Builtins for XRay.
 def XRayCustomEvent : Builtin {
   let Spellings = ["__xray_customevent"];

clang/include/clang/Basic/DiagnosticSemaKinds.td

@@ -12065,7 +12065,8 @@ def err_builtin_invalid_arg_type: Error <
   "pointer to a valid matrix element type|"
   "signed integer or floating point type|vector type|"
   "floating point type|"
-  "vector of integers}1 (was %2)">;
+  "vector of integers|"
+  "type of integer}1 (was %2)">;
 
 def err_builtin_matrix_disabled: Error<
   "matrix types extension is disabled. Pass -fenable-matrix to enable it">;

clang/include/clang/Sema/Sema.h

@@ -14502,6 +14502,7 @@ class Sema final {
   bool CheckPPCBuiltinFunctionCall(const TargetInfo &TI, unsigned BuiltinID,
                                    CallExpr *TheCall);
   bool CheckAMDGCNBuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall);
+  bool CheckHLSLBuiltinFunctionCall(unsigned BuiltinID, CallExpr *TheCall);
   bool CheckRISCVLMUL(CallExpr *TheCall, unsigned ArgNum);
   bool CheckRISCVBuiltinFunctionCall(const TargetInfo &TI, unsigned BuiltinID,
                                      CallExpr *TheCall);
@@ -14590,6 +14591,8 @@ class Sema final {
 
   bool CheckPPCMMAType(QualType Type, SourceLocation TypeLoc);
 
+  bool SemaBuiltinVectorMath(CallExpr *TheCall, QualType &Res);
+  bool SemaBuiltinVectorToScalarMath(CallExpr *TheCall);
   bool SemaBuiltinElementwiseMath(CallExpr *TheCall);
   bool SemaBuiltinElementwiseTernaryMath(CallExpr *TheCall);
   bool PrepareBuiltinElementwiseMathOneArgCall(CallExpr *TheCall);

clang/lib/AST/Interp/InterpBuiltin.cpp

@@ -85,7 +85,8 @@ static void pushInteger(InterpState &S, T Val, QualType QT) {
     pushInteger(S, APSInt(Val, !std::is_signed_v<T>), QT);
   else
     pushInteger(S,
-                APSInt(APInt(sizeof(T) * 8, Val, std::is_signed_v<T>),
+                APSInt(APInt(sizeof(T) * 8, static_cast<uint64_t>(Val),
+                             std::is_signed_v<T>),
                        !std::is_signed_v<T>),
                 QT);
 }
@@ -464,7 +465,7 @@ static bool interp__builtin_popcount(InterpState &S, CodePtr OpPC,
 
   PrimType ArgT = *S.getContext().classify(Call->getArg(0)->getType());
   APSInt Val = peekToAPSInt(S.Stk, ArgT);
-  pushInteger(S, APSInt(APInt(32, Val.popcount())), Call->getType());
+  pushInteger(S, Val.popcount(), Call->getType());
   return true;
 }
 
@@ -805,7 +806,7 @@ static bool interp__builtin_clz(InterpState &S, CodePtr OpPC,
   if (ZeroIsUndefined && Val == 0)
     return false;
 
-  pushInteger(S, APSInt(APInt(32, Val.countl_zero())), Call->getType());
+  pushInteger(S, Val.countl_zero(), Call->getType());
   return true;
 }
 

clang/lib/Analysis/FlowSensitive/Transfer.cpp

@@ -671,9 +671,9 @@ class TransferVisitor : public ConstStmtVisitor<TransferVisitor> {
     }
 
     if (!Type->isStructureOrClassType()) {
-      // Until array initialization is implemented, we don't need to care about
-      // cases where `getNumInits() > 1`.
-      if (S->getNumInits() == 1)
+      // Until array initialization is implemented, we skip arrays and don't
+      // need to care about cases where `getNumInits() > 1`.
+      if (!Type->isArrayType() && S->getNumInits() == 1)
         propagateValueOrStorageLocation(*S->getInit(0), *S, Env);
       return;
     }

clang/lib/Analysis/ThreadSafetyCommon.cpp

@@ -995,7 +995,7 @@ void SExprBuilder::exitCFG(const CFGBlock *Last) {
   IncompleteArgs.clear();
 }
 
-/*
+#ifndef NDEBUG
 namespace {
 
 class TILPrinter :
@@ -1016,4 +1016,4 @@ void printSCFG(CFGWalker &Walker) {
 
 } // namespace threadSafety
 } // namespace clang
-*/
+#endif // NDEBUG

clang/lib/CodeGen/CGBuiltin.cpp

@@ -45,6 +45,7 @@
 #include "llvm/IR/IntrinsicsAMDGPU.h"
 #include "llvm/IR/IntrinsicsARM.h"
 #include "llvm/IR/IntrinsicsBPF.h"
+#include "llvm/IR/IntrinsicsDirectX.h"
 #include "llvm/IR/IntrinsicsHexagon.h"
 #include "llvm/IR/IntrinsicsNVPTX.h"
 #include "llvm/IR/IntrinsicsPowerPC.h"
@@ -3296,7 +3297,8 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
   case Builtin::BI__popcnt64:
   case Builtin::BI__builtin_popcount:
   case Builtin::BI__builtin_popcountl:
-  case Builtin::BI__builtin_popcountll: {
+  case Builtin::BI__builtin_popcountll:
+  case Builtin::BI__builtin_popcountg: {
     Value *ArgValue = EmitScalarExpr(E->getArg(0));
 
     llvm::Type *ArgType = ArgValue->getType();
@@ -6069,6 +6071,10 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
     llvm_unreachable("Bad evaluation kind in EmitBuiltinExpr");
   }
 
+  // EmitHLSLBuiltinExpr will check getLangOpts().HLSL
+  if (Value *V = EmitHLSLBuiltinExpr(BuiltinID, E))
+    return RValue::get(V);
+
   if (getLangOpts().HIPStdPar && getLangOpts().CUDAIsDevice)
     return EmitHipStdParUnsupportedBuiltin(this, FD);
 
@@ -18056,6 +18062,52 @@ llvm::Value *CodeGenFunction::EmitScalarOrConstFoldImmArg(unsigned ICEArguments,
   return Arg;
 }
 
+Value *CodeGenFunction::EmitHLSLBuiltinExpr(unsigned BuiltinID,
+                                            const CallExpr *E) {
+  if (!getLangOpts().HLSL)
+    return nullptr;
+
+  switch (BuiltinID) {
+  case Builtin::BI__builtin_hlsl_dot: {
+    Value *Op0 = EmitScalarExpr(E->getArg(0));
+    Value *Op1 = EmitScalarExpr(E->getArg(1));
+    llvm::Type *T0 = Op0->getType();
+    llvm::Type *T1 = Op1->getType();
+    if (!T0->isVectorTy() && !T1->isVectorTy()) {
+      if (T0->isFloatingPointTy())
+        return Builder.CreateFMul(Op0, Op1, "dx.dot");
+
+      if (T0->isIntegerTy())
+        return Builder.CreateMul(Op0, Op1, "dx.dot");
+
+      // Bools should have been promoted
+      llvm_unreachable(
+          "Scalar dot product is only supported on ints and floats.");
+    }
+    // A VectorSplat should have happened
+    assert(T0->isVectorTy() && T1->isVectorTy() &&
+           "Dot product of vector and scalar is not supported.");
+
+    // A vector sext or sitofp should have happened
+    assert(T0->getScalarType() == T1->getScalarType() &&
+           "Dot product of vectors need the same element types.");
+
+    [[maybe_unused]] auto *VecTy0 =
+        E->getArg(0)->getType()->getAs<VectorType>();
+    [[maybe_unused]] auto *VecTy1 =
+        E->getArg(1)->getType()->getAs<VectorType>();
+    // A HLSLVectorTruncation should have happend
+    assert(VecTy0->getNumElements() == VecTy1->getNumElements() &&
+           "Dot product requires vectors to be of the same size.");
+
+    return Builder.CreateIntrinsic(
+        /*ReturnType*/ T0->getScalarType(), Intrinsic::dx_dot,
+        ArrayRef<Value *>{Op0, Op1}, nullptr, "dx.dot");
+  } break;
+  }
+  return nullptr;
+}
+
 Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
                                               const CallExpr *E) {
   llvm::AtomicOrdering AO = llvm::AtomicOrdering::SequentiallyConsistent;

clang/lib/CodeGen/CodeGenFunction.h

@@ -4427,6 +4427,7 @@ class CodeGenFunction : public CodeGenTypeCache {
   llvm::Value *EmitX86BuiltinExpr(unsigned BuiltinID, const CallExpr *E);
   llvm::Value *EmitPPCBuiltinExpr(unsigned BuiltinID, const CallExpr *E);
   llvm::Value *EmitAMDGPUBuiltinExpr(unsigned BuiltinID, const CallExpr *E);
+  llvm::Value *EmitHLSLBuiltinExpr(unsigned BuiltinID, const CallExpr *E);
   llvm::Value *EmitScalarOrConstFoldImmArg(unsigned ICEArguments, unsigned Idx,
                                            const CallExpr *E);
   llvm::Value *EmitSystemZBuiltinExpr(unsigned BuiltinID, const CallExpr *E);

clang/lib/Driver/ToolChain.cpp

@@ -759,19 +759,29 @@ std::string ToolChain::getCompilerRT(const ArgList &Args, StringRef Component,
   // Check for runtime files in the new layout without the architecture first.
   std::string CRTBasename =
       buildCompilerRTBasename(Args, Component, Type, /*AddArch=*/false);
+  SmallString<128> Path;
   for (const auto &LibPath : getLibraryPaths()) {
     SmallString<128> P(LibPath);
     llvm::sys::path::append(P, CRTBasename);
     if (getVFS().exists(P))
       return std::string(P);
+    if (Path.empty())
+      Path = P;
   }
+  if (getTriple().isOSAIX())
+    Path.clear();
 
-  // Fall back to the old expected compiler-rt name if the new one does not
-  // exist.
+  // Check the filename for the old layout if the new one does not exist.
   CRTBasename =
       buildCompilerRTBasename(Args, Component, Type, /*AddArch=*/true);
-  SmallString<128> Path(getCompilerRTPath());
-  llvm::sys::path::append(Path, CRTBasename);
+  SmallString<128> OldPath(getCompilerRTPath());
+  llvm::sys::path::append(OldPath, CRTBasename);
+  if (Path.empty() || getVFS().exists(OldPath))
+    return std::string(OldPath);
+
+  // If none is found, use a file name from the new layout, which may get
+  // printed in an error message, aiding users in knowing what Clang is
+  // looking for.
   return std::string(Path);
 }
 

clang/lib/Driver/ToolChains/CommonArgs.cpp

@@ -1322,13 +1322,16 @@ void tools::addFortranRuntimeLibs(const ToolChain &TC, const ArgList &Args,
   // add the correct libraries to link against as dependents in the object
   // file.
   if (!TC.getTriple().isKnownWindowsMSVCEnvironment()) {
-    StringRef f128LibName = TC.getDriver().getFlangF128MathLibrary();
-    f128LibName.consume_front_insensitive("lib");
-    if (!f128LibName.empty()) {
+    StringRef F128LibName = TC.getDriver().getFlangF128MathLibrary();
+    F128LibName.consume_front_insensitive("lib");
+    if (!F128LibName.empty()) {
+      bool AsNeeded = !TC.getTriple().isOSAIX();
       CmdArgs.push_back("-lFortranFloat128Math");
-      addAsNeededOption(TC, Args, CmdArgs, /*as_needed=*/true);
-      CmdArgs.push_back(Args.MakeArgString("-l" + f128LibName));
-      addAsNeededOption(TC, Args, CmdArgs, /*as_needed=*/false);
+      if (AsNeeded)
+        addAsNeededOption(TC, Args, CmdArgs, /*as_needed=*/true);
+      CmdArgs.push_back(Args.MakeArgString("-l" + F128LibName));
+      if (AsNeeded)
+        addAsNeededOption(TC, Args, CmdArgs, /*as_needed=*/false);
     }
     CmdArgs.push_back("-lFortranRuntime");
     CmdArgs.push_back("-lFortranDecimal");