diff options
Diffstat (limited to 'clang/lib/CodeGen/CGBuiltin.cpp')
| -rw-r--r-- | clang/lib/CodeGen/CGBuiltin.cpp | 80 |
1 files changed, 78 insertions, 2 deletions
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp index 4d4b742..573be93 100644 --- a/clang/lib/CodeGen/CGBuiltin.cpp +++ b/clang/lib/CodeGen/CGBuiltin.cpp @@ -835,6 +835,38 @@ static Value *emitFrexpBuiltin(CodeGenFunction &CGF, const CallExpr *E, return CGF.Builder.CreateExtractValue(Call, 0); } +static void emitSincosBuiltin(CodeGenFunction &CGF, const CallExpr *E, + llvm::Intrinsic::ID IntrinsicID) { + llvm::Value *Val = CGF.EmitScalarExpr(E->getArg(0)); + llvm::Value *Dest0 = CGF.EmitScalarExpr(E->getArg(1)); + llvm::Value *Dest1 = CGF.EmitScalarExpr(E->getArg(2)); + + llvm::Function *F = CGF.CGM.getIntrinsic(IntrinsicID, {Val->getType()}); + llvm::Value *Call = CGF.Builder.CreateCall(F, Val); + + llvm::Value *SinResult = CGF.Builder.CreateExtractValue(Call, 0); + llvm::Value *CosResult = CGF.Builder.CreateExtractValue(Call, 1); + + QualType DestPtrType = E->getArg(1)->getType()->getPointeeType(); + LValue SinLV = CGF.MakeNaturalAlignAddrLValue(Dest0, DestPtrType); + LValue CosLV = CGF.MakeNaturalAlignAddrLValue(Dest1, DestPtrType); + + llvm::StoreInst *StoreSin = + CGF.Builder.CreateStore(SinResult, SinLV.getAddress()); + llvm::StoreInst *StoreCos = + CGF.Builder.CreateStore(CosResult, CosLV.getAddress()); + + // Mark the two stores as non-aliasing with each other. The order of stores + // emitted by this builtin is arbitrary, enforcing a particular order will + // prevent optimizations later on. + llvm::MDBuilder MDHelper(CGF.getLLVMContext()); + MDNode *Domain = MDHelper.createAnonymousAliasScopeDomain(); + MDNode *AliasScope = MDHelper.createAnonymousAliasScope(Domain); + MDNode *AliasScopeList = MDNode::get(Call->getContext(), AliasScope); + StoreSin->setMetadata(LLVMContext::MD_alias_scope, AliasScopeList); + StoreCos->setMetadata(LLVMContext::MD_noalias, AliasScopeList); +} + /// EmitFAbs - Emit a call to @llvm.fabs(). static Value *EmitFAbs(CodeGenFunction &CGF, Value *V) { Function *F = CGF.CGM.getIntrinsic(Intrinsic::fabs, V->getType()); @@ -3232,6 +3264,14 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID, return RValue::get(emitUnaryMaybeConstrainedFPBuiltin( *this, E, Intrinsic::sinh, Intrinsic::experimental_constrained_sinh)); + case Builtin::BI__builtin_sincos: + case Builtin::BI__builtin_sincosf: + case Builtin::BI__builtin_sincosf16: + case Builtin::BI__builtin_sincosl: + case Builtin::BI__builtin_sincosf128: + emitSincosBuiltin(*this, E, Intrinsic::sincos); + return RValue::get(nullptr); + case Builtin::BIsqrt: case Builtin::BIsqrtf: case Builtin::BIsqrtl: @@ -6757,6 +6797,8 @@ static Value *EmitTargetArchBuiltinExpr(CodeGenFunction *CGF, case llvm::Triple::riscv32: case llvm::Triple::riscv64: return CGF->EmitRISCVBuiltinExpr(BuiltinID, E, ReturnValue); + case llvm::Triple::spirv: + return CGF->EmitSPIRVBuiltinExpr(BuiltinID, E); case llvm::Triple::spirv64: if (CGF->getTarget().getTriple().getOS() != llvm::Triple::OSType::AMDHSA) return nullptr; @@ -11285,6 +11327,19 @@ Value *CodeGenFunction::EmitAArch64SMEBuiltinExpr(unsigned BuiltinID, if (Builtin->LLVMIntrinsic == 0) return nullptr; + if (BuiltinID == SME::BI__builtin_sme___arm_in_streaming_mode) { + // If we already know the streaming mode, don't bother with the intrinsic + // and emit a constant instead + const auto *FD = cast<FunctionDecl>(CurFuncDecl); + if (const auto *FPT = FD->getType()->getAs<FunctionProtoType>()) { + unsigned SMEAttrs = FPT->getAArch64SMEAttributes(); + if (!(SMEAttrs & FunctionType::SME_PStateSMCompatibleMask)) { + bool IsStreaming = SMEAttrs & FunctionType::SME_PStateSMEnabledMask; + return ConstantInt::getBool(Builder.getContext(), IsStreaming); + } + } + } + // Predicates must match the main datatype. for (unsigned i = 0, e = Ops.size(); i != e; ++i) if (auto PredTy = dyn_cast<llvm::VectorType>(Ops[i]->getType())) @@ -19157,8 +19212,9 @@ Value *CodeGenFunction::EmitHLSLBuiltinExpr(unsigned BuiltinID, // TODO: Map to an hlsl_device address space. llvm::Type *RetTy = llvm::PointerType::getUnqual(getLLVMContext()); - return Builder.CreateIntrinsic(RetTy, Intrinsic::dx_resource_getpointer, - ArrayRef<Value *>{HandleOp, IndexOp}); + return Builder.CreateIntrinsic( + RetTy, CGM.getHLSLRuntime().getCreateResourceGetPointerIntrinsic(), + ArrayRef<Value *>{HandleOp, IndexOp}); } case Builtin::BI__builtin_hlsl_all: { Value *Op0 = EmitScalarExpr(E->getArg(0)); @@ -20440,6 +20496,26 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID, } } +Value *CodeGenFunction::EmitSPIRVBuiltinExpr(unsigned BuiltinID, + const CallExpr *E) { + switch (BuiltinID) { + case SPIRV::BI__builtin_spirv_distance: { + Value *X = EmitScalarExpr(E->getArg(0)); + Value *Y = EmitScalarExpr(E->getArg(1)); + assert(E->getArg(0)->getType()->hasFloatingRepresentation() && + E->getArg(1)->getType()->hasFloatingRepresentation() && + "Distance operands must have a float representation"); + assert(E->getArg(0)->getType()->isVectorType() && + E->getArg(1)->getType()->isVectorType() && + "Distance operands must be a vector"); + return Builder.CreateIntrinsic( + /*ReturnType=*/X->getType()->getScalarType(), Intrinsic::spv_distance, + ArrayRef<Value *>{X, Y}, nullptr, "spv.distance"); + } + } + return nullptr; +} + /// Handle a SystemZ function in which the final argument is a pointer /// to an int that receives the post-instruction CC value. At the LLVM level /// this is represented as a function that returns a {result, cc} pair. |