r321816 - [OPENMP] Add debug info for generated functions.
Jonas Hahnfeld via cfe-commits
cfe-commits at lists.llvm.org
Thu Jan 4 12:54:51 PST 2018
Hi Alexey,
should this change be backported to 6.0?
Regards,
Jonas
Am 2018-01-04 20:45, schrieb Alexey Bataev via cfe-commits:
> Author: abataev
> Date: Thu Jan 4 11:45:16 2018
> New Revision: 321816
>
> URL: http://llvm.org/viewvc/llvm-project?rev=321816&view=rev
> Log:
> [OPENMP] Add debug info for generated functions.
>
> Most of the generated functions for the OpenMP were generated with
> disabled debug info. Patch fixes this for better user experience.
>
> Modified:
> cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp
> cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h
> cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
> cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.h
> cfe/trunk/test/OpenMP/target_parallel_debug_codegen.cpp
>
> Modified: cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp
> URL:
> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp?rev=321816&r1=321815&r2=321816&view=diff
> ==============================================================================
> --- cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp (original)
> +++ cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp Thu Jan 4 11:45:16 2018
> @@ -1216,7 +1216,8 @@ emitCombinerOrInitializer(CodeGenModule
> CodeGenFunction CGF(CGM);
> // Map "T omp_in;" variable to "*omp_in_parm" value in all
> expressions.
> // Map "T omp_out;" variable to "*omp_out_parm" value in all
> expressions.
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FnInfo, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FnInfo, Args,
> In->getLocation(),
> + Out->getLocation());
> CodeGenFunction::OMPPrivateScope Scope(CGF);
> Address AddrIn = CGF.GetAddrOfLocalVar(&OmpInParm);
> Scope.addPrivate(In, [&CGF, AddrIn, PtrTy]() -> Address {
> @@ -2383,7 +2384,8 @@ llvm::Function *CGOpenMPRuntime::emitThr
> // threadprivate copy of the variable VD
> CodeGenFunction CtorCGF(CGM);
> FunctionArgList Args;
> - ImplicitParamDecl Dst(CGM.getContext(),
> CGM.getContext().VoidPtrTy,
> + ImplicitParamDecl Dst(CGM.getContext(), /*DC=*/nullptr, Loc,
> + /*Id=*/nullptr,
> CGM.getContext().VoidPtrTy,
> ImplicitParamDecl::Other);
> Args.push_back(&Dst);
>
> @@ -2393,13 +2395,13 @@ llvm::Function *CGOpenMPRuntime::emitThr
> auto Fn = CGM.CreateGlobalInitOrDestructFunction(
> FTy, ".__kmpc_global_ctor_.", FI, Loc);
> CtorCGF.StartFunction(GlobalDecl(), CGM.getContext().VoidPtrTy,
> Fn, FI,
> - Args, SourceLocation());
> + Args, Loc, Loc);
> auto ArgVal = CtorCGF.EmitLoadOfScalar(
> CtorCGF.GetAddrOfLocalVar(&Dst), /*Volatile=*/false,
> CGM.getContext().VoidPtrTy, Dst.getLocation());
> Address Arg = Address(ArgVal, VDAddr.getAlignment());
> - Arg = CtorCGF.Builder.CreateElementBitCast(Arg,
> -
> CtorCGF.ConvertTypeForMem(ASTTy));
> + Arg = CtorCGF.Builder.CreateElementBitCast(
> + Arg, CtorCGF.ConvertTypeForMem(ASTTy));
> CtorCGF.EmitAnyExprToMem(Init, Arg,
> Init->getType().getQualifiers(),
> /*IsInitializer=*/true);
> ArgVal = CtorCGF.EmitLoadOfScalar(
> @@ -2414,7 +2416,8 @@ llvm::Function *CGOpenMPRuntime::emitThr
> // of the variable VD
> CodeGenFunction DtorCGF(CGM);
> FunctionArgList Args;
> - ImplicitParamDecl Dst(CGM.getContext(),
> CGM.getContext().VoidPtrTy,
> + ImplicitParamDecl Dst(CGM.getContext(), /*DC=*/nullptr, Loc,
> + /*Id=*/nullptr,
> CGM.getContext().VoidPtrTy,
> ImplicitParamDecl::Other);
> Args.push_back(&Dst);
>
> @@ -2425,7 +2428,7 @@ llvm::Function *CGOpenMPRuntime::emitThr
> FTy, ".__kmpc_global_dtor_.", FI, Loc);
> auto NL = ApplyDebugLocation::CreateEmpty(DtorCGF);
> DtorCGF.StartFunction(GlobalDecl(), CGM.getContext().VoidTy,
> Fn, FI, Args,
> - SourceLocation());
> + Loc, Loc);
> // Create a scope with an artificial location for the body of
> this function.
> auto AL = ApplyDebugLocation::CreateArtificial(DtorCGF);
> auto ArgVal = DtorCGF.EmitLoadOfScalar(
> @@ -2469,7 +2472,7 @@ llvm::Function *CGOpenMPRuntime::emitThr
> FunctionArgList ArgList;
> InitCGF.StartFunction(GlobalDecl(), CGM.getContext().VoidTy,
> InitFunction,
> CGM.getTypes().arrangeNullaryFunction(),
> ArgList,
> - Loc);
> + Loc, Loc);
> emitThreadPrivateVarInit(InitCGF, VDAddr, Ctor, CopyCtor, Dtor,
> Loc);
> InitCGF.FinishFunction();
> return InitFunction;
> @@ -2783,12 +2786,15 @@ static Address emitAddrOfVarFromArray(Co
> static llvm::Value *emitCopyprivateCopyFunction(
> CodeGenModule &CGM, llvm::Type *ArgsType,
> ArrayRef<const Expr *> CopyprivateVars, ArrayRef<const Expr *>
> DestExprs,
> - ArrayRef<const Expr *> SrcExprs, ArrayRef<const Expr *>
> AssignmentOps) {
> + ArrayRef<const Expr *> SrcExprs, ArrayRef<const Expr *>
> AssignmentOps,
> + SourceLocation Loc) {
> auto &C = CGM.getContext();
> // void copy_func(void *LHSArg, void *RHSArg);
> FunctionArgList Args;
> - ImplicitParamDecl LHSArg(C, C.VoidPtrTy, ImplicitParamDecl::Other);
> - ImplicitParamDecl RHSArg(C, C.VoidPtrTy, ImplicitParamDecl::Other);
> + ImplicitParamDecl LHSArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> C.VoidPtrTy,
> + ImplicitParamDecl::Other);
> + ImplicitParamDecl RHSArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> C.VoidPtrTy,
> + ImplicitParamDecl::Other);
> Args.push_back(&LHSArg);
> Args.push_back(&RHSArg);
> auto &CGFI =
> CGM.getTypes().arrangeBuiltinFunctionDeclaration(C.VoidTy, Args);
> @@ -2797,7 +2803,7 @@ static llvm::Value *emitCopyprivateCopyF
> ".omp.copyprivate.copy_func", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, Fn, CGFI);
> CodeGenFunction CGF(CGM);
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args, Loc, Loc);
> // Dest = (void*[n])(LHSArg);
> // Src = (void*[n])(RHSArg);
> Address LHS(CGF.Builder.CreatePointerBitCastOrAddrSpaceCast(
> @@ -2888,7 +2894,7 @@ void CGOpenMPRuntime::emitSingleRegion(C
> // threads in the corresponding parallel region.
> auto *CpyFn = emitCopyprivateCopyFunction(
> CGM,
> CGF.ConvertTypeForMem(CopyprivateArrayTy)->getPointerTo(),
> - CopyprivateVars, SrcExprs, DstExprs, AssignmentOps);
> + CopyprivateVars, SrcExprs, DstExprs, AssignmentOps, Loc);
> auto *BufSize = CGF.getTypeSize(CopyprivateArrayTy);
> Address CL =
> CGF.Builder.CreatePointerBitCastOrAddrSpaceCast(CopyprivateList,
> @@ -3399,11 +3405,13 @@ createOffloadingBinaryDescriptorFunction
> Args.push_back(&DummyPtr);
>
> CodeGenFunction CGF(CGM);
> + // Disable debug info for global (de-)initializer because they are
> not part of
> + // some particular construct.
> + CGF.disableDebugInfo();
> auto &FI =
> CGM.getTypes().arrangeBuiltinFunctionDeclaration(C.VoidTy, Args);
> auto FTy = CGM.getTypes().GetFunctionType(FI);
> - auto *Fn =
> - CGM.CreateGlobalInitOrDestructFunction(FTy, Name, FI,
> SourceLocation());
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FI, Args,
> SourceLocation());
> + auto *Fn = CGM.CreateGlobalInitOrDestructFunction(FTy, Name, FI);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FI, Args);
> Codegen(CGF);
> CGF.FinishFunction();
> return Fn;
> @@ -3411,7 +3419,6 @@ createOffloadingBinaryDescriptorFunction
>
> llvm::Function *
> CGOpenMPRuntime::createOffloadingBinaryDescriptorRegistration() {
> -
> // If we don't have entries or if we are emitting code for the
> device, we
> // don't need to do anything.
> if (CGM.getLangOpts().OpenMPIsDevice ||
> OffloadEntriesInfoManager.empty())
> @@ -3945,7 +3952,8 @@ emitProxyTaskFunction(CodeGenModule &CGM
> ".omp_task_entry.", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, TaskEntry,
> TaskEntryFnInfo);
> CodeGenFunction CGF(CGM);
> - CGF.StartFunction(GlobalDecl(), KmpInt32Ty, TaskEntry,
> TaskEntryFnInfo, Args);
> + CGF.StartFunction(GlobalDecl(), KmpInt32Ty, TaskEntry,
> TaskEntryFnInfo, Args,
> + Loc, Loc);
>
> // TaskFunction(gtid, tt->task_data.part_id, &tt->privates,
> task_privates_map,
> // tt,
> @@ -4045,9 +4053,8 @@ static llvm::Value *emitDestructorsFunct
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, DestructorFn,
> DestructorFnInfo);
> CodeGenFunction CGF(CGM);
> - CGF.disableDebugInfo();
> CGF.StartFunction(GlobalDecl(), KmpInt32Ty, DestructorFn,
> DestructorFnInfo,
> - Args);
> + Args, Loc, Loc);
>
> LValue Base = CGF.EmitLoadOfPointerLValue(
> CGF.GetAddrOfLocalVar(&TaskTypeArg),
> @@ -4139,9 +4146,8 @@ emitTaskPrivateMappingFunction(CodeGenMo
> TaskPrivatesMap->removeFnAttr(llvm::Attribute::OptimizeNone);
> TaskPrivatesMap->addFnAttr(llvm::Attribute::AlwaysInline);
> CodeGenFunction CGF(CGM);
> - CGF.disableDebugInfo();
> CGF.StartFunction(GlobalDecl(), C.VoidTy, TaskPrivatesMap,
> - TaskPrivatesMapFnInfo, Args);
> + TaskPrivatesMapFnInfo, Args, Loc, Loc);
>
> // *privi = &.privates.privi;
> LValue Base = CGF.EmitLoadOfPointerLValue(
> @@ -4322,8 +4328,8 @@ emitTaskDupFunction(CodeGenModule &CGM,
> ".omp_task_dup.", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, TaskDup,
> TaskDupFnInfo);
> CodeGenFunction CGF(CGM);
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, TaskDup, TaskDupFnInfo,
> Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, TaskDup, TaskDupFnInfo,
> Args, Loc,
> + Loc);
>
> LValue TDBase = CGF.EmitLoadOfPointerLValue(
> CGF.GetAddrOfLocalVar(&DstArg),
> @@ -4943,15 +4949,17 @@ static void emitReductionCombiner(CodeGe
> }
>
> llvm::Value *CGOpenMPRuntime::emitReductionFunction(
> - CodeGenModule &CGM, llvm::Type *ArgsType, ArrayRef<const Expr *>
> Privates,
> - ArrayRef<const Expr *> LHSExprs, ArrayRef<const Expr *> RHSExprs,
> - ArrayRef<const Expr *> ReductionOps) {
> + CodeGenModule &CGM, SourceLocation Loc, llvm::Type *ArgsType,
> + ArrayRef<const Expr *> Privates, ArrayRef<const Expr *> LHSExprs,
> + ArrayRef<const Expr *> RHSExprs, ArrayRef<const Expr *>
> ReductionOps) {
> auto &C = CGM.getContext();
>
> // void reduction_func(void *LHSArg, void *RHSArg);
> FunctionArgList Args;
> - ImplicitParamDecl LHSArg(C, C.VoidPtrTy, ImplicitParamDecl::Other);
> - ImplicitParamDecl RHSArg(C, C.VoidPtrTy, ImplicitParamDecl::Other);
> + ImplicitParamDecl LHSArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> C.VoidPtrTy,
> + ImplicitParamDecl::Other);
> + ImplicitParamDecl RHSArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> C.VoidPtrTy,
> + ImplicitParamDecl::Other);
> Args.push_back(&LHSArg);
> Args.push_back(&RHSArg);
> auto &CGFI =
> CGM.getTypes().arrangeBuiltinFunctionDeclaration(C.VoidTy, Args);
> @@ -4960,7 +4968,7 @@ llvm::Value *CGOpenMPRuntime::emitReduct
> ".omp.reduction.reduction_func", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, Fn, CGFI);
> CodeGenFunction CGF(CGM);
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args, Loc, Loc);
>
> // Dst = (void*[n])(LHSArg);
> // Src = (void*[n])(RHSArg);
> @@ -5149,8 +5157,8 @@ void CGOpenMPRuntime::emitReduction(Code
>
> // 2. Emit reduce_func().
> auto *ReductionFn = emitReductionFunction(
> - CGM, CGF.ConvertTypeForMem(ReductionArrayTy)->getPointerTo(),
> Privates,
> - LHSExprs, RHSExprs, ReductionOps);
> + CGM, Loc,
> CGF.ConvertTypeForMem(ReductionArrayTy)->getPointerTo(),
> + Privates, LHSExprs, RHSExprs, ReductionOps);
>
> // 3. Create static kmp_critical_name lock = { 0 };
> auto *Lock = getCriticalRegionLock(".reduction");
> @@ -5365,7 +5373,8 @@ static llvm::Value *emitReduceInitFuncti
> ReductionCodeGen &RCG,
> unsigned N) {
> auto &C = CGM.getContext();
> FunctionArgList Args;
> - ImplicitParamDecl Param(C, C.VoidPtrTy, ImplicitParamDecl::Other);
> + ImplicitParamDecl Param(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> C.VoidPtrTy,
> + ImplicitParamDecl::Other);
> Args.emplace_back(&Param);
> auto &FnInfo =
> CGM.getTypes().arrangeBuiltinFunctionDeclaration(C.VoidTy,
> Args);
> @@ -5374,8 +5383,7 @@ static llvm::Value *emitReduceInitFuncti
> ".red_init.", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, Fn, FnInfo);
> CodeGenFunction CGF(CGM);
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FnInfo, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FnInfo, Args, Loc,
> Loc);
> Address PrivateAddr = CGF.EmitLoadOfPointer(
> CGF.GetAddrOfLocalVar(&Param),
> C.getPointerType(C.VoidPtrTy).castAs<PointerType>());
> @@ -5435,8 +5443,10 @@ static llvm::Value *emitReduceCombFuncti
> auto *LHSVD = cast<VarDecl>(cast<DeclRefExpr>(LHS)->getDecl());
> auto *RHSVD = cast<VarDecl>(cast<DeclRefExpr>(RHS)->getDecl());
> FunctionArgList Args;
> - ImplicitParamDecl ParamInOut(C, C.VoidPtrTy,
> ImplicitParamDecl::Other);
> - ImplicitParamDecl ParamIn(C, C.VoidPtrTy, ImplicitParamDecl::Other);
> + ImplicitParamDecl ParamInOut(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> + C.VoidPtrTy, ImplicitParamDecl::Other);
> + ImplicitParamDecl ParamIn(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> C.VoidPtrTy,
> + ImplicitParamDecl::Other);
> Args.emplace_back(&ParamInOut);
> Args.emplace_back(&ParamIn);
> auto &FnInfo =
> @@ -5446,8 +5456,7 @@ static llvm::Value *emitReduceCombFuncti
> ".red_comb.", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, Fn, FnInfo);
> CodeGenFunction CGF(CGM);
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FnInfo, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FnInfo, Args, Loc,
> Loc);
> llvm::Value *Size = nullptr;
> // If the size of the reduction item is non-constant, load it from
> global
> // threadprivate variable.
> @@ -5506,7 +5515,8 @@ static llvm::Value *emitReduceFiniFuncti
> return nullptr;
> auto &C = CGM.getContext();
> FunctionArgList Args;
> - ImplicitParamDecl Param(C, C.VoidPtrTy, ImplicitParamDecl::Other);
> + ImplicitParamDecl Param(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> C.VoidPtrTy,
> + ImplicitParamDecl::Other);
> Args.emplace_back(&Param);
> auto &FnInfo =
> CGM.getTypes().arrangeBuiltinFunctionDeclaration(C.VoidTy,
> Args);
> @@ -5515,8 +5525,7 @@ static llvm::Value *emitReduceFiniFuncti
> ".red_fini.", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, Fn, FnInfo);
> CodeGenFunction CGF(CGM);
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FnInfo, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, FnInfo, Args, Loc,
> Loc);
> Address PrivateAddr = CGF.EmitLoadOfPointer(
> CGF.GetAddrOfLocalVar(&Param),
> C.getPointerType(C.VoidPtrTy).castAs<PointerType>());
>
> Modified: cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h
> URL:
> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h?rev=321816&r1=321815&r2=321816&view=diff
> ==============================================================================
> --- cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h (original)
> +++ cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h Thu Jan 4 11:45:16 2018
> @@ -1048,7 +1048,8 @@ public:
> /// \param RHSExprs List of RHS in \a ReductionOps reduction
> operations.
> /// \param ReductionOps List of reduction operations in form 'LHS
> binop RHS'
> /// or 'operator binop(LHS, RHS)'.
> - llvm::Value *emitReductionFunction(CodeGenModule &CGM, llvm::Type
> *ArgsType,
> + llvm::Value *emitReductionFunction(CodeGenModule &CGM,
> SourceLocation Loc,
> + llvm::Type *ArgsType,
> ArrayRef<const Expr *> Privates,
> ArrayRef<const Expr *> LHSExprs,
> ArrayRef<const Expr *> RHSExprs,
>
> Modified: cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
> URL:
> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp?rev=321816&r1=321815&r2=321816&view=diff
> ==============================================================================
> --- cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp (original)
> +++ cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp Thu Jan 4 11:45:16
> 2018
> @@ -250,8 +250,8 @@ static llvm::Value *getMasterThreadID(Co
> }
>
> CGOpenMPRuntimeNVPTX::WorkerFunctionState::WorkerFunctionState(
> - CodeGenModule &CGM)
> - : WorkerFn(nullptr), CGFI(nullptr) {
> + CodeGenModule &CGM, SourceLocation Loc)
> + : WorkerFn(nullptr), CGFI(nullptr), Loc(Loc) {
> createWorkerFunction(CGM);
> }
>
> @@ -297,28 +297,28 @@ void CGOpenMPRuntimeNVPTX::emitGenericKe
> ExecutionModeRAII ModeRAII(CurrentExecutionMode,
>
> CGOpenMPRuntimeNVPTX::ExecutionMode::Generic);
> EntryFunctionState EST;
> - WorkerFunctionState WST(CGM);
> + WorkerFunctionState WST(CGM, D.getLocStart());
> Work.clear();
> WrapperFunctionsMap.clear();
>
> // Emit target region as a standalone region.
> class NVPTXPrePostActionTy : public PrePostActionTy {
> - CGOpenMPRuntimeNVPTX &RT;
> CGOpenMPRuntimeNVPTX::EntryFunctionState &EST;
> CGOpenMPRuntimeNVPTX::WorkerFunctionState &WST;
>
> public:
> - NVPTXPrePostActionTy(CGOpenMPRuntimeNVPTX &RT,
> - CGOpenMPRuntimeNVPTX::EntryFunctionState
> &EST,
> + NVPTXPrePostActionTy(CGOpenMPRuntimeNVPTX::EntryFunctionState
> &EST,
> CGOpenMPRuntimeNVPTX::WorkerFunctionState
> &WST)
> - : RT(RT), EST(EST), WST(WST) {}
> + : EST(EST), WST(WST) {}
> void Enter(CodeGenFunction &CGF) override {
> - RT.emitGenericEntryHeader(CGF, EST, WST);
> + static_cast<CGOpenMPRuntimeNVPTX &>(CGF.CGM.getOpenMPRuntime())
> + .emitGenericEntryHeader(CGF, EST, WST);
> }
> void Exit(CodeGenFunction &CGF) override {
> - RT.emitGenericEntryFooter(CGF, EST);
> + static_cast<CGOpenMPRuntimeNVPTX &>(CGF.CGM.getOpenMPRuntime())
> + .emitGenericEntryFooter(CGF, EST);
> }
> - } Action(*this, EST, WST);
> + } Action(EST, WST);
> CodeGen.setAction(Action);
> emitTargetOutlinedFunctionHelper(D, ParentName, OutlinedFn,
> OutlinedFnID,
> IsOffloadEntry, CodeGen);
> @@ -347,7 +347,7 @@ void CGOpenMPRuntimeNVPTX::emitGenericEn
> Bld.CreateCondBr(IsWorker, WorkerBB, MasterCheckBB);
>
> CGF.EmitBlock(WorkerBB);
> - emitCall(CGF, WST.WorkerFn);
> + emitOutlinedFunctionCall(CGF, WST.Loc, WST.WorkerFn);
> CGF.EmitBranch(EST.ExitBB);
>
> CGF.EmitBlock(MasterCheckBB);
> @@ -479,8 +479,8 @@ void CGOpenMPRuntimeNVPTX::emitWorkerFun
> ASTContext &Ctx = CGM.getContext();
>
> CodeGenFunction CGF(CGM, /*suppressNewContext=*/true);
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), Ctx.VoidTy, WST.WorkerFn, *WST.CGFI,
> {});
> + CGF.StartFunction(GlobalDecl(), Ctx.VoidTy, WST.WorkerFn, *WST.CGFI,
> {},
> + WST.Loc, WST.Loc);
> emitWorkerLoop(CGF, WST);
> CGF.FinishFunction();
> }
> @@ -571,8 +571,9 @@ void CGOpenMPRuntimeNVPTX::emitWorkerLoo
> Address Capture = CGF.EmitLoadOfPointer(SharedArgs,
> Ctx.getPointerType(
> Ctx.getPointerType(Ctx.VoidPtrTy)).castAs<PointerType>());
> - emitCall(CGF, W, {Bld.getInt16(/*ParallelLevel=*/0),
> - getMasterThreadID(CGF), Capture.getPointer()});
> + emitOutlinedFunctionCall(CGF, WST.Loc, W,
> + {Bld.getInt16(/*ParallelLevel=*/0),
> + getMasterThreadID(CGF),
> Capture.getPointer()});
>
> // Go to end of parallel region.
> CGF.EmitBranch(TerminateBB);
> @@ -1319,27 +1320,31 @@ static void emitReductionListCopy(
> /// local = local @ remote
> /// else
> /// local = remote
> -static llvm::Value *
> -emitReduceScratchpadFunction(CodeGenModule &CGM,
> - ArrayRef<const Expr *> Privates,
> - QualType ReductionArrayTy, llvm::Value
> *ReduceFn) {
> +static llvm::Value *emitReduceScratchpadFunction(
> + CodeGenModule &CGM, ArrayRef<const Expr *> Privates,
> + QualType ReductionArrayTy, llvm::Value *ReduceFn, SourceLocation
> Loc) {
> auto &C = CGM.getContext();
> auto Int32Ty = C.getIntTypeForBitwidth(32, /* Signed */ true);
>
> // Destination of the copy.
> - ImplicitParamDecl ReduceListArg(C, C.VoidPtrTy,
> ImplicitParamDecl::Other);
> + ImplicitParamDecl ReduceListArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> + C.VoidPtrTy,
> ImplicitParamDecl::Other);
> // Base address of the scratchpad array, with each element storing a
> // Reduce list per team.
> - ImplicitParamDecl ScratchPadArg(C, C.VoidPtrTy,
> ImplicitParamDecl::Other);
> + ImplicitParamDecl ScratchPadArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> + C.VoidPtrTy,
> ImplicitParamDecl::Other);
> // A source index into the scratchpad array.
> - ImplicitParamDecl IndexArg(C, Int32Ty, ImplicitParamDecl::Other);
> + ImplicitParamDecl IndexArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> Int32Ty,
> + ImplicitParamDecl::Other);
> // Row width of an element in the scratchpad array, typically
> // the number of teams.
> - ImplicitParamDecl WidthArg(C, Int32Ty, ImplicitParamDecl::Other);
> + ImplicitParamDecl WidthArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> Int32Ty,
> + ImplicitParamDecl::Other);
> // If should_reduce == 1, then it's load AND reduce,
> // If should_reduce == 0 (or otherwise), then it only loads (+
> copy).
> // The latter case is used for initialization.
> - ImplicitParamDecl ShouldReduceArg(C, Int32Ty,
> ImplicitParamDecl::Other);
> + ImplicitParamDecl ShouldReduceArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> + Int32Ty,
> ImplicitParamDecl::Other);
>
> FunctionArgList Args;
> Args.push_back(&ReduceListArg);
> @@ -1354,10 +1359,7 @@ emitReduceScratchpadFunction(CodeGenModu
> "_omp_reduction_load_and_reduce", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*DC=*/nullptr, Fn, CGFI);
> CodeGenFunction CGF(CGM);
> - // We don't need debug information in this function as nothing here
> refers to
> - // user code.
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args, Loc, Loc);
>
> auto &Bld = CGF.Builder;
>
> @@ -1421,7 +1423,8 @@ emitReduceScratchpadFunction(CodeGenModu
> ReduceListAddr.getPointer(), CGF.VoidPtrTy);
> llvm::Value *RemoteDataPtr =
> Bld.CreatePointerBitCastOrAddrSpaceCast(
> RemoteReduceList.getPointer(), CGF.VoidPtrTy);
> - CGF.EmitCallOrInvoke(ReduceFn, {LocalDataPtr, RemoteDataPtr});
> + CGM.getOpenMPRuntime().emitOutlinedFunctionCall(
> + CGF, Loc, ReduceFn, {LocalDataPtr, RemoteDataPtr});
> Bld.CreateBr(MergeBB);
>
> CGF.EmitBlock(ElseBB);
> @@ -1445,22 +1448,27 @@ emitReduceScratchpadFunction(CodeGenModu
> ///
> static llvm::Value *emitCopyToScratchpad(CodeGenModule &CGM,
> ArrayRef<const Expr *>
> Privates,
> - QualType ReductionArrayTy) {
> + QualType ReductionArrayTy,
> + SourceLocation Loc) {
>
> auto &C = CGM.getContext();
> auto Int32Ty = C.getIntTypeForBitwidth(32, /* Signed */ true);
>
> // Source of the copy.
> - ImplicitParamDecl ReduceListArg(C, C.VoidPtrTy,
> ImplicitParamDecl::Other);
> + ImplicitParamDecl ReduceListArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> + C.VoidPtrTy,
> ImplicitParamDecl::Other);
> // Base address of the scratchpad array, with each element storing a
> // Reduce list per team.
> - ImplicitParamDecl ScratchPadArg(C, C.VoidPtrTy,
> ImplicitParamDecl::Other);
> + ImplicitParamDecl ScratchPadArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> + C.VoidPtrTy,
> ImplicitParamDecl::Other);
> // A destination index into the scratchpad array, typically the team
> // identifier.
> - ImplicitParamDecl IndexArg(C, Int32Ty, ImplicitParamDecl::Other);
> + ImplicitParamDecl IndexArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> Int32Ty,
> + ImplicitParamDecl::Other);
> // Row width of an element in the scratchpad array, typically
> // the number of teams.
> - ImplicitParamDecl WidthArg(C, Int32Ty, ImplicitParamDecl::Other);
> + ImplicitParamDecl WidthArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> Int32Ty,
> + ImplicitParamDecl::Other);
>
> FunctionArgList Args;
> Args.push_back(&ReduceListArg);
> @@ -1474,10 +1482,7 @@ static llvm::Value *emitCopyToScratchpad
> "_omp_reduction_copy_to_scratchpad", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*DC=*/nullptr, Fn, CGFI);
> CodeGenFunction CGF(CGM);
> - // We don't need debug information in this function as nothing here
> refers to
> - // user code.
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args, Loc, Loc);
>
> auto &Bld = CGF.Builder;
>
> @@ -1534,17 +1539,19 @@ static llvm::Value *emitCopyToScratchpad
> /// sync
> static llvm::Value *emitInterWarpCopyFunction(CodeGenModule &CGM,
> ArrayRef<const Expr *>
> Privates,
> - QualType
> ReductionArrayTy) {
> + QualType
> ReductionArrayTy,
> + SourceLocation Loc) {
> auto &C = CGM.getContext();
> auto &M = CGM.getModule();
>
> // ReduceList: thread local Reduce list.
> // At the stage of the computation when this function is called,
> partially
> // aggregated values reside in the first lane of every active warp.
> - ImplicitParamDecl ReduceListArg(C, C.VoidPtrTy,
> ImplicitParamDecl::Other);
> + ImplicitParamDecl ReduceListArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> + C.VoidPtrTy,
> ImplicitParamDecl::Other);
> // NumWarps: number of warps active in the parallel region. This
> could
> // be smaller than 32 (max warps in a CTA) for partial block
> reduction.
> - ImplicitParamDecl NumWarpsArg(C,
> + ImplicitParamDecl NumWarpsArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> C.getIntTypeForBitwidth(32, /* Signed
> */ true),
> ImplicitParamDecl::Other);
> FunctionArgList Args;
> @@ -1557,10 +1564,7 @@ static llvm::Value *emitInterWarpCopyFun
> "_omp_reduction_inter_warp_copy_func", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*DC=*/nullptr, Fn, CGFI);
> CodeGenFunction CGF(CGM);
> - // We don't need debug information in this function as nothing here
> refers to
> - // user code.
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args, Loc, Loc);
>
> auto &Bld = CGF.Builder;
>
> @@ -1781,21 +1785,23 @@ static llvm::Value *emitInterWarpCopyFun
> /// (2k+1)th thread is ignored in the value aggregation. Therefore
> /// we copy the Reduce list from the (2k+1)th lane to (k+1)th lane
> so
> /// that the contiguity assumption still holds.
> -static llvm::Value *
> -emitShuffleAndReduceFunction(CodeGenModule &CGM,
> - ArrayRef<const Expr *> Privates,
> - QualType ReductionArrayTy, llvm::Value
> *ReduceFn) {
> +static llvm::Value *emitShuffleAndReduceFunction(
> + CodeGenModule &CGM, ArrayRef<const Expr *> Privates,
> + QualType ReductionArrayTy, llvm::Value *ReduceFn, SourceLocation
> Loc) {
> auto &C = CGM.getContext();
>
> // Thread local Reduce list used to host the values of data to be
> reduced.
> - ImplicitParamDecl ReduceListArg(C, C.VoidPtrTy,
> ImplicitParamDecl::Other);
> + ImplicitParamDecl ReduceListArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> + C.VoidPtrTy,
> ImplicitParamDecl::Other);
> // Current lane id; could be logical.
> - ImplicitParamDecl LaneIDArg(C, C.ShortTy, ImplicitParamDecl::Other);
> + ImplicitParamDecl LaneIDArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> C.ShortTy,
> + ImplicitParamDecl::Other);
> // Offset of the remote source lane relative to the current lane.
> - ImplicitParamDecl RemoteLaneOffsetArg(C, C.ShortTy,
> - ImplicitParamDecl::Other);
> + ImplicitParamDecl RemoteLaneOffsetArg(C, /*DC=*/nullptr, Loc,
> /*Id=*/nullptr,
> + C.ShortTy,
> ImplicitParamDecl::Other);
> // Algorithm version. This is expected to be known at compile time.
> - ImplicitParamDecl AlgoVerArg(C, C.ShortTy,
> ImplicitParamDecl::Other);
> + ImplicitParamDecl AlgoVerArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr,
> + C.ShortTy, ImplicitParamDecl::Other);
> FunctionArgList Args;
> Args.push_back(&ReduceListArg);
> Args.push_back(&LaneIDArg);
> @@ -1808,10 +1814,7 @@ emitShuffleAndReduceFunction(CodeGenModu
> "_omp_reduction_shuffle_and_reduce_func", &CGM.getModule());
> CGM.SetInternalFunctionAttributes(/*D=*/nullptr, Fn, CGFI);
> CodeGenFunction CGF(CGM);
> - // We don't need debug information in this function as nothing here
> refers to
> - // user code.
> - CGF.disableDebugInfo();
> - CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args);
> + CGF.StartFunction(GlobalDecl(), C.VoidTy, Fn, CGFI, Args, Loc, Loc);
>
> auto &Bld = CGF.Builder;
>
> @@ -1898,7 +1901,8 @@ emitShuffleAndReduceFunction(CodeGenModu
> LocalReduceList.getPointer(), CGF.VoidPtrTy);
> llvm::Value *RemoteReduceListPtr =
> Bld.CreatePointerBitCastOrAddrSpaceCast(
> RemoteReduceList.getPointer(), CGF.VoidPtrTy);
> - CGF.EmitCallOrInvoke(ReduceFn, {LocalReduceListPtr,
> RemoteReduceListPtr});
> + CGM.getOpenMPRuntime().emitOutlinedFunctionCall(
> + CGF, Loc, ReduceFn, {LocalReduceListPtr, RemoteReduceListPtr});
> Bld.CreateBr(MergeBB);
>
> CGF.EmitBlock(ElseBB);
> @@ -2228,8 +2232,8 @@ void CGOpenMPRuntimeNVPTX::emitReduction
>
> // 2. Emit reduce_func().
> auto *ReductionFn = emitReductionFunction(
> - CGM, CGF.ConvertTypeForMem(ReductionArrayTy)->getPointerTo(),
> Privates,
> - LHSExprs, RHSExprs, ReductionOps);
> + CGM, Loc,
> CGF.ConvertTypeForMem(ReductionArrayTy)->getPointerTo(),
> + Privates, LHSExprs, RHSExprs, ReductionOps);
>
> // 4. Build res = __kmpc_reduce{_nowait}(<gtid>, <n>,
> sizeof(RedList),
> // RedList, shuffle_reduce_func, interwarp_copy_func);
> @@ -2239,9 +2243,9 @@ void CGOpenMPRuntimeNVPTX::emitReduction
> ReductionList.getPointer(), CGF.VoidPtrTy);
>
> auto *ShuffleAndReduceFn = emitShuffleAndReduceFunction(
> - CGM, Privates, ReductionArrayTy, ReductionFn);
> + CGM, Privates, ReductionArrayTy, ReductionFn, Loc);
> auto *InterWarpCopyFn =
> - emitInterWarpCopyFunction(CGM, Privates, ReductionArrayTy);
> + emitInterWarpCopyFunction(CGM, Privates, ReductionArrayTy, Loc);
>
> llvm::Value *Res = nullptr;
> if (ParallelReduction) {
> @@ -2259,9 +2263,9 @@ void CGOpenMPRuntimeNVPTX::emitReduction
>
> if (TeamsReduction) {
> auto *ScratchPadCopyFn =
> - emitCopyToScratchpad(CGM, Privates, ReductionArrayTy);
> + emitCopyToScratchpad(CGM, Privates, ReductionArrayTy, Loc);
> auto *LoadAndReduceFn = emitReduceScratchpadFunction(
> - CGM, Privates, ReductionArrayTy, ReductionFn);
> + CGM, Privates, ReductionArrayTy, ReductionFn, Loc);
>
> llvm::Value *Args[] = {ThreadId,
> CGF.Builder.getInt32(RHSExprs.size()),
> @@ -2422,10 +2426,15 @@ llvm::Function *CGOpenMPRuntimeNVPTX::cr
> Ctx.getIntTypeForBitwidth(/*DestWidth=*/32, /*Signed=*/false);
> QualType Int32PtrQTy = Ctx.getPointerType(Int32QTy);
> QualType VoidPtrPtrQTy = Ctx.getPointerType(Ctx.VoidPtrTy);
> - ImplicitParamDecl ParallelLevelArg(Ctx, Int16QTy,
> ImplicitParamDecl::Other);
> - ImplicitParamDecl WrapperArg(Ctx, Int32QTy,
> ImplicitParamDecl::Other);
> - ImplicitParamDecl SharedArgsList(Ctx, VoidPtrPtrQTy,
> - ImplicitParamDecl::Other);
> + ImplicitParamDecl ParallelLevelArg(Ctx, /*DC=*/nullptr,
> D.getLocStart(),
> + /*Id=*/nullptr, Int16QTy,
> + ImplicitParamDecl::Other);
> + ImplicitParamDecl WrapperArg(Ctx, /*DC=*/nullptr, D.getLocStart(),
> + /*Id=*/nullptr, Int32QTy,
> + ImplicitParamDecl::Other);
> + ImplicitParamDecl SharedArgsList(Ctx, /*DC=*/nullptr,
> D.getLocStart(),
> + /*Id=*/nullptr, VoidPtrPtrQTy,
> + ImplicitParamDecl::Other);
> WrapperArgs.emplace_back(&ParallelLevelArg);
> WrapperArgs.emplace_back(&WrapperArg);
> WrapperArgs.emplace_back(&SharedArgsList);
> @@ -2440,7 +2449,8 @@ llvm::Function *CGOpenMPRuntimeNVPTX::cr
> Fn->setLinkage(llvm::GlobalValue::InternalLinkage);
>
> CodeGenFunction CGF(CGM, /*suppressNewContext=*/true);
> - CGF.StartFunction(GlobalDecl(), Ctx.VoidTy, Fn, CGFI, WrapperArgs);
> + CGF.StartFunction(GlobalDecl(), Ctx.VoidTy, Fn, CGFI, WrapperArgs,
> + D.getLocStart(), D.getLocStart());
>
> const auto *RD = CS.getCapturedRecordDecl();
> auto CurField = RD->field_begin();
> @@ -2489,7 +2499,7 @@ llvm::Function *CGOpenMPRuntimeNVPTX::cr
> Args.emplace_back(Arg);
> }
>
> - emitCall(CGF, OutlinedParallelFn, Args);
> + emitOutlinedFunctionCall(CGF, D.getLocStart(), OutlinedParallelFn,
> Args);
> CGF.FinishFunction();
> return Fn;
> }
>
> Modified: cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.h
> URL:
> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.h?rev=321816&r1=321815&r2=321816&view=diff
> ==============================================================================
> --- cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.h (original)
> +++ cfe/trunk/lib/CodeGen/CGOpenMPRuntimeNVPTX.h Thu Jan 4 11:45:16
> 2018
> @@ -36,8 +36,9 @@ private:
> public:
> llvm::Function *WorkerFn;
> const CGFunctionInfo *CGFI;
> + SourceLocation Loc;
>
> - WorkerFunctionState(CodeGenModule &CGM);
> + WorkerFunctionState(CodeGenModule &CGM, SourceLocation Loc);
>
> private:
> void createWorkerFunction(CodeGenModule &CGM);
>
> Modified: cfe/trunk/test/OpenMP/target_parallel_debug_codegen.cpp
> URL:
> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/OpenMP/target_parallel_debug_codegen.cpp?rev=321816&r1=321815&r2=321816&view=diff
> ==============================================================================
> --- cfe/trunk/test/OpenMP/target_parallel_debug_codegen.cpp (original)
> +++ cfe/trunk/test/OpenMP/target_parallel_debug_codegen.cpp Thu Jan 4
> 11:45:16 2018
> @@ -116,8 +116,8 @@ int main() {
> // CHECK: !DILocalVariable(name: ".bound_tid.",
> // CHECK-SAME: DIFlagArtificial
> // CHECK: !DILocalVariable(name: "c",
> -// CHECK-SAMEi-NOT: DIFlagArtificial
> +// CHECK-SAME: line: 11
> // CHECK: !DILocalVariable(name: "a",
> -// CHECK-SAMEi-NOT: DIFlagArtificial
> +// CHECK-SAME: line: 9
> // CHECK: !DILocalVariable(name: "b",
> -// CHECK-SAMEi-NOT: DIFlagArtificial
> +// CHECK-SAME: line: 10
>
>
> _______________________________________________
> cfe-commits mailing list
> cfe-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
More information about the cfe-commits
mailing list