[clang] [openmp] [Clang][OpenMP] Fix mapping of structs to device (PR #75642)
Gheorghe-Teodor Bercea via cfe-commits
cfe-commits at lists.llvm.org
Fri Dec 15 11:07:59 PST 2023
https://github.com/doru1004 updated https://github.com/llvm/llvm-project/pull/75642
>From e0e1f5e7bb2f95f2568b5dd647b883f4740bcafd Mon Sep 17 00:00:00 2001
From: Doru Bercea <doru.bercea at amd.com>
Date: Fri, 15 Dec 2023 10:22:38 -0500
Subject: [PATCH] Fix mapping of structs to device.
---
clang/lib/CodeGen/CGOpenMPRuntime.cpp | 146 +++++++++++----
clang/test/OpenMP/map_struct_ordering.cpp | 172 ++++++++++++++++++
.../struct_mapping_with_pointers.cpp | 114 ++++++++++++
3 files changed, 399 insertions(+), 33 deletions(-)
create mode 100644 clang/test/OpenMP/map_struct_ordering.cpp
create mode 100644 openmp/libomptarget/test/offloading/struct_mapping_with_pointers.cpp
diff --git a/clang/lib/CodeGen/CGOpenMPRuntime.cpp b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
index 7f7e6f53066644..350e7108b8d5a7 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntime.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
@@ -6811,8 +6811,10 @@ class MappableExprsHandler {
OpenMPMapClauseKind MapType, ArrayRef<OpenMPMapModifierKind> MapModifiers,
ArrayRef<OpenMPMotionModifierKind> MotionModifiers,
OMPClauseMappableExprCommon::MappableExprComponentListRef Components,
- MapCombinedInfoTy &CombinedInfo, StructRangeInfoTy &PartialStruct,
- bool IsFirstComponentList, bool IsImplicit,
+ MapCombinedInfoTy &CombinedInfo,
+ MapCombinedInfoTy &StructBaseCombinedInfo,
+ StructRangeInfoTy &PartialStruct, bool IsFirstComponentList,
+ bool IsImplicit, bool GenerateAllInfoForClauses,
const ValueDecl *Mapper = nullptr, bool ForDeviceAddr = false,
const ValueDecl *BaseDecl = nullptr, const Expr *MapExpr = nullptr,
ArrayRef<OMPClauseMappableExprCommon::MappableExprComponentListRef>
@@ -7098,6 +7100,25 @@ class MappableExprsHandler {
bool IsNonContiguous = CombinedInfo.NonContigInfo.IsNonContiguous;
bool IsPrevMemberReference = false;
+ // We need to check if we will be encountering any MEs. If we do not
+ // encounter any ME expression it means we will be mapping the whole struct.
+ // In that case we need to skip adding an entry for the struct to the
+ // CombinedInfo list and instead add an entry to the StructBaseCombinedInfo
+ // list only when generating all info for clauses.
+ bool IsMappingWholeStruct = true;
+ if (!GenerateAllInfoForClauses) {
+ IsMappingWholeStruct = false;
+ } else {
+ for (auto TempI = I; TempI != CE; ++TempI) {
+ const MemberExpr *PossibleME =
+ dyn_cast<MemberExpr>(TempI->getAssociatedExpression());
+ if (PossibleME) {
+ IsMappingWholeStruct = false;
+ break;
+ }
+ }
+ }
+
for (; I != CE; ++I) {
// If the current component is member of a struct (parent struct) mark it.
if (!EncounteredME) {
@@ -7317,21 +7338,41 @@ class MappableExprsHandler {
break;
}
llvm::Value *Size = getExprTypeSize(I->getAssociatedExpression());
+ // Skip adding an entry in the CurInfo of this combined entry if the
+ // whole struct is currently being mapped. The struct needs to be added
+ // in the first position before any data internal to the struct is being
+ // mapped.
if (!IsMemberPointerOrAddr ||
(Next == CE && MapType != OMPC_MAP_unknown)) {
- CombinedInfo.Exprs.emplace_back(MapDecl, MapExpr);
- CombinedInfo.BasePointers.push_back(BP.getPointer());
- CombinedInfo.DevicePtrDecls.push_back(nullptr);
- CombinedInfo.DevicePointers.push_back(DeviceInfoTy::None);
- CombinedInfo.Pointers.push_back(LB.getPointer());
- CombinedInfo.Sizes.push_back(
- CGF.Builder.CreateIntCast(Size, CGF.Int64Ty, /*isSigned=*/true));
- CombinedInfo.NonContigInfo.Dims.push_back(IsNonContiguous ? DimSize
- : 1);
+ if (!IsMappingWholeStruct) {
+ CombinedInfo.Exprs.emplace_back(MapDecl, MapExpr);
+ CombinedInfo.BasePointers.push_back(BP.getPointer());
+ CombinedInfo.DevicePtrDecls.push_back(nullptr);
+ CombinedInfo.DevicePointers.push_back(DeviceInfoTy::None);
+ CombinedInfo.Pointers.push_back(LB.getPointer());
+ CombinedInfo.Sizes.push_back(CGF.Builder.CreateIntCast(
+ Size, CGF.Int64Ty, /*isSigned=*/true));
+ CombinedInfo.NonContigInfo.Dims.push_back(IsNonContiguous ? DimSize
+ : 1);
+ } else {
+ StructBaseCombinedInfo.Exprs.emplace_back(MapDecl, MapExpr);
+ StructBaseCombinedInfo.BasePointers.push_back(BP.getPointer());
+ StructBaseCombinedInfo.DevicePtrDecls.push_back(nullptr);
+ StructBaseCombinedInfo.DevicePointers.push_back(DeviceInfoTy::None);
+ StructBaseCombinedInfo.Pointers.push_back(LB.getPointer());
+ StructBaseCombinedInfo.Sizes.push_back(CGF.Builder.CreateIntCast(
+ Size, CGF.Int64Ty, /*isSigned=*/true));
+ StructBaseCombinedInfo.NonContigInfo.Dims.push_back(
+ IsNonContiguous ? DimSize : 1);
+ }
// If Mapper is valid, the last component inherits the mapper.
bool HasMapper = Mapper && Next == CE;
- CombinedInfo.Mappers.push_back(HasMapper ? Mapper : nullptr);
+ if (!IsMappingWholeStruct)
+ CombinedInfo.Mappers.push_back(HasMapper ? Mapper : nullptr);
+ else
+ StructBaseCombinedInfo.Mappers.push_back(HasMapper ? Mapper
+ : nullptr);
// We need to add a pointer flag for each map that comes from the
// same expression except for the first one. We also need to signal
@@ -7363,7 +7404,10 @@ class MappableExprsHandler {
}
}
- CombinedInfo.Types.push_back(Flags);
+ if (!IsMappingWholeStruct)
+ CombinedInfo.Types.push_back(Flags);
+ else
+ StructBaseCombinedInfo.Types.push_back(Flags);
}
// If we have encountered a member expression so far, keep track of the
@@ -7954,8 +7998,10 @@ class MappableExprsHandler {
for (const auto &Data : Info) {
StructRangeInfoTy PartialStruct;
- // Temporary generated information.
+ // Current struct information:
MapCombinedInfoTy CurInfo;
+ // Current struct base information:
+ MapCombinedInfoTy StructBaseCurInfo;
const Decl *D = Data.first;
const ValueDecl *VD = cast_or_null<ValueDecl>(D);
for (const auto &M : Data.second) {
@@ -7965,29 +8011,53 @@ class MappableExprsHandler {
// Remember the current base pointer index.
unsigned CurrentBasePointersIdx = CurInfo.BasePointers.size();
+ unsigned StructBasePointersIdx =
+ StructBaseCurInfo.BasePointers.size();
CurInfo.NonContigInfo.IsNonContiguous =
L.Components.back().isNonContiguous();
generateInfoForComponentList(
L.MapType, L.MapModifiers, L.MotionModifiers, L.Components,
- CurInfo, PartialStruct, /*IsFirstComponentList=*/false,
- L.IsImplicit, L.Mapper, L.ForDeviceAddr, VD, L.VarRef);
+ CurInfo, StructBaseCurInfo, PartialStruct,
+ /*IsFirstComponentList=*/false, L.IsImplicit,
+ /*GenerateAllInfoForClauses*/ true, L.Mapper, L.ForDeviceAddr, VD,
+ L.VarRef);
- // If this entry relates with a device pointer, set the relevant
+ // If this entry relates to a device pointer, set the relevant
// declaration and add the 'return pointer' flag.
if (L.ReturnDevicePointer) {
- assert(CurInfo.BasePointers.size() > CurrentBasePointersIdx &&
+ // Check whether a value was added to either CurInfo or
+ // StructBaseCurInfo and error if no value was added to either of
+ // them:
+ assert((CurrentBasePointersIdx < CurInfo.BasePointers.size() ||
+ StructBasePointersIdx <
+ StructBaseCurInfo.BasePointers.size()) &&
"Unexpected number of mapped base pointers.");
+ // Choose a base pointer index which is always valid:
const ValueDecl *RelevantVD =
L.Components.back().getAssociatedDeclaration();
assert(RelevantVD &&
"No relevant declaration related with device pointer??");
- CurInfo.DevicePtrDecls[CurrentBasePointersIdx] = RelevantVD;
- CurInfo.DevicePointers[CurrentBasePointersIdx] =
- L.ForDeviceAddr ? DeviceInfoTy::Address : DeviceInfoTy::Pointer;
- CurInfo.Types[CurrentBasePointersIdx] |=
- OpenMPOffloadMappingFlags::OMP_MAP_RETURN_PARAM;
+ // If StructBaseCurInfo has been updated this iteration then work on the
+ // first new entry added to it i.e. make sure that when multiple values are added to any of the lists, the
+ // first value added is being modified by the assignments below (not the last value added).
+ if (StructBasePointersIdx < StructBaseCurInfo.BasePointers.size()) {
+ StructBaseCurInfo.DevicePtrDecls[StructBasePointersIdx] =
+ RelevantVD;
+ StructBaseCurInfo.DevicePointers[StructBasePointersIdx] =
+ L.ForDeviceAddr ? DeviceInfoTy::Address
+ : DeviceInfoTy::Pointer;
+ StructBaseCurInfo.Types[StructBasePointersIdx] |=
+ OpenMPOffloadMappingFlags::OMP_MAP_RETURN_PARAM;
+ } else {
+ CurInfo.DevicePtrDecls[CurrentBasePointersIdx] = RelevantVD;
+ CurInfo.DevicePointers[CurrentBasePointersIdx] =
+ L.ForDeviceAddr ? DeviceInfoTy::Address
+ : DeviceInfoTy::Pointer;
+ CurInfo.Types[CurrentBasePointersIdx] |=
+ OpenMPOffloadMappingFlags::OMP_MAP_RETURN_PARAM;
+ }
}
}
}
@@ -8034,17 +8104,24 @@ class MappableExprsHandler {
CurInfo.Mappers.push_back(nullptr);
}
}
+
+ // Unify entries in one list making sure the struct mapping precedes the
+ // individual fields:
+ MapCombinedInfoTy UnionCurInfo;
+ UnionCurInfo.append(StructBaseCurInfo);
+ UnionCurInfo.append(CurInfo);
+
// If there is an entry in PartialStruct it means we have a struct with
// individual members mapped. Emit an extra combined entry.
if (PartialStruct.Base.isValid()) {
- CurInfo.NonContigInfo.Dims.push_back(0);
- emitCombinedEntry(CombinedInfo, CurInfo.Types, PartialStruct,
+ UnionCurInfo.NonContigInfo.Dims.push_back(0);
+ // Emit a combined entry:
+ emitCombinedEntry(CombinedInfo, UnionCurInfo.Types, PartialStruct,
/*IsMapThis*/ !VD, OMPBuilder, VD);
}
- // We need to append the results of this capture to what we already
- // have.
- CombinedInfo.append(CurInfo);
+ // We need to append the results of this capture to what we already have.
+ CombinedInfo.append(UnionCurInfo);
}
// Append data for use_device_ptr clauses.
CombinedInfo.append(UseDeviceDataCombinedInfo);
@@ -8554,6 +8631,7 @@ class MappableExprsHandler {
// Associated with a capture, because the mapping flags depend on it.
// Go through all of the elements with the overlapped elements.
bool IsFirstComponentList = true;
+ MapCombinedInfoTy StructBaseCombinedInfo;
for (const auto &Pair : OverlappedData) {
const MapData &L = *Pair.getFirst();
OMPClauseMappableExprCommon::MappableExprComponentListRef Components;
@@ -8568,7 +8646,8 @@ class MappableExprsHandler {
OverlappedComponents = Pair.getSecond();
generateInfoForComponentList(
MapType, MapModifiers, std::nullopt, Components, CombinedInfo,
- PartialStruct, IsFirstComponentList, IsImplicit, Mapper,
+ StructBaseCombinedInfo, PartialStruct, IsFirstComponentList,
+ IsImplicit, /*GenerateAllInfoForClauses*/ false, Mapper,
/*ForDeviceAddr=*/false, VD, VarRef, OverlappedComponents);
IsFirstComponentList = false;
}
@@ -8584,10 +8663,11 @@ class MappableExprsHandler {
L;
auto It = OverlappedData.find(&L);
if (It == OverlappedData.end())
- generateInfoForComponentList(MapType, MapModifiers, std::nullopt,
- Components, CombinedInfo, PartialStruct,
- IsFirstComponentList, IsImplicit, Mapper,
- /*ForDeviceAddr=*/false, VD, VarRef);
+ generateInfoForComponentList(
+ MapType, MapModifiers, std::nullopt, Components, CombinedInfo,
+ StructBaseCombinedInfo, PartialStruct, IsFirstComponentList,
+ IsImplicit, /*GenerateAllInfoForClauses*/ false, Mapper,
+ /*ForDeviceAddr=*/false, VD, VarRef);
IsFirstComponentList = false;
}
}
diff --git a/clang/test/OpenMP/map_struct_ordering.cpp b/clang/test/OpenMP/map_struct_ordering.cpp
new file mode 100644
index 00000000000000..035b39b5b12ab4
--- /dev/null
+++ b/clang/test/OpenMP/map_struct_ordering.cpp
@@ -0,0 +1,172 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --include-generated-funcs --replace-value-regex "__omp_offloading_[0-9a-z]+_[0-9a-z]+" --prefix-filecheck-ir-name _ --version 4
+
+// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - -Wno-openmp-mapping | FileCheck %s --check-prefix=CHECK
+
+// expected-no-diagnostics
+#ifndef HEADER
+#define HEADER
+
+struct Descriptor {
+ int *datum;
+ long int x;
+ int xi;
+ long int arr[1][30];
+};
+
+int map_struct() {
+ Descriptor dat = Descriptor();
+ dat.xi = 3;
+ dat.arr[0][0] = 1;
+
+ #pragma omp target enter data map(to: dat.datum[:10]) map(to: dat)
+
+ #pragma omp target
+ {
+ dat.xi = 4;
+ dat.datum[dat.arr[0][0]] = dat.xi;
+ }
+
+ #pragma omp target exit data map(from: dat)
+
+ return dat.xi;
+}
+
+#endif
+// CHECK-LABEL: define dso_local noundef signext i32 @_Z10map_structv(
+// CHECK-SAME: ) #[[ATTR0:[0-9]+]] {
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[DAT:%.*]] = alloca [[STRUCT_DESCRIPTOR:%.*]], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [3 x ptr], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [3 x ptr], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [3 x ptr], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_SIZES:%.*]] = alloca [3 x i64], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_BASEPTRS4:%.*]] = alloca [1 x ptr], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_PTRS5:%.*]] = alloca [1 x ptr], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_MAPPERS6:%.*]] = alloca [1 x ptr], align 8
+// CHECK-NEXT: [[KERNEL_ARGS:%.*]] = alloca [[STRUCT___TGT_KERNEL_ARGUMENTS:%.*]], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_BASEPTRS7:%.*]] = alloca [1 x ptr], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_PTRS8:%.*]] = alloca [1 x ptr], align 8
+// CHECK-NEXT: [[DOTOFFLOAD_MAPPERS9:%.*]] = alloca [1 x ptr], align 8
+// CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[DAT]], i8 0, i64 264, i1 false)
+// CHECK-NEXT: [[XI:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR]], ptr [[DAT]], i32 0, i32 2
+// CHECK-NEXT: store i32 3, ptr [[XI]], align 8
+// CHECK-NEXT: [[ARR:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR]], ptr [[DAT]], i32 0, i32 3
+// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [1 x [30 x i64]], ptr [[ARR]], i64 0, i64 0
+// CHECK-NEXT: [[ARRAYIDX1:%.*]] = getelementptr inbounds [30 x i64], ptr [[ARRAYIDX]], i64 0, i64 0
+// CHECK-NEXT: store i64 1, ptr [[ARRAYIDX1]], align 8
+// CHECK-NEXT: [[DATUM:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR]], ptr [[DAT]], i32 0, i32 0
+// CHECK-NEXT: [[DATUM2:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR]], ptr [[DAT]], i32 0, i32 0
+// CHECK-NEXT: [[TMP0:%.*]] = load ptr, ptr [[DATUM2]], align 8
+// CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32, ptr [[TMP0]], i64 0
+// CHECK-NEXT: [[TMP1:%.*]] = getelementptr [[STRUCT_DESCRIPTOR]], ptr [[DAT]], i32 1
+// CHECK-NEXT: [[TMP2:%.*]] = ptrtoint ptr [[TMP1]] to i64
+// CHECK-NEXT: [[TMP3:%.*]] = ptrtoint ptr [[DAT]] to i64
+// CHECK-NEXT: [[TMP4:%.*]] = sub i64 [[TMP2]], [[TMP3]]
+// CHECK-NEXT: [[TMP5:%.*]] = sdiv exact i64 [[TMP4]], ptrtoint (ptr getelementptr (i8, ptr null, i32 1) to i64)
+// CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr align 8 [[DOTOFFLOAD_SIZES]], ptr align 8 @.offload_sizes, i64 24, i1 false)
+// CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
+// CHECK-NEXT: store ptr [[DAT]], ptr [[TMP6]], align 8
+// CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_PTRS]], i32 0, i32 0
+// CHECK-NEXT: store ptr [[DAT]], ptr [[TMP7]], align 8
+// CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds [3 x i64], ptr [[DOTOFFLOAD_SIZES]], i32 0, i32 0
+// CHECK-NEXT: store i64 [[TMP5]], ptr [[TMP8]], align 8
+// CHECK-NEXT: [[TMP9:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_MAPPERS]], i64 0, i64 0
+// CHECK-NEXT: store ptr null, ptr [[TMP9]], align 8
+// CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 1
+// CHECK-NEXT: store ptr [[DAT]], ptr [[TMP10]], align 8
+// CHECK-NEXT: [[TMP11:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_PTRS]], i32 0, i32 1
+// CHECK-NEXT: store ptr [[DAT]], ptr [[TMP11]], align 8
+// CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_MAPPERS]], i64 0, i64 1
+// CHECK-NEXT: store ptr null, ptr [[TMP12]], align 8
+// CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 2
+// CHECK-NEXT: store ptr [[DATUM]], ptr [[TMP13]], align 8
+// CHECK-NEXT: [[TMP14:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_PTRS]], i32 0, i32 2
+// CHECK-NEXT: store ptr [[ARRAYIDX3]], ptr [[TMP14]], align 8
+// CHECK-NEXT: [[TMP15:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_MAPPERS]], i64 0, i64 2
+// CHECK-NEXT: store ptr null, ptr [[TMP15]], align 8
+// CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0
+// CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds [3 x ptr], ptr [[DOTOFFLOAD_PTRS]], i32 0, i32 0
+// CHECK-NEXT: [[TMP18:%.*]] = getelementptr inbounds [3 x i64], ptr [[DOTOFFLOAD_SIZES]], i32 0, i32 0
+// CHECK-NEXT: call void @__tgt_target_data_begin_mapper(ptr @[[GLOB1:[0-9]+]], i64 -1, i32 3, ptr [[TMP16]], ptr [[TMP17]], ptr [[TMP18]], ptr @.offload_maptypes, ptr null, ptr null)
+// CHECK-NEXT: [[TMP19:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_BASEPTRS4]], i32 0, i32 0
+// CHECK-NEXT: store ptr [[DAT]], ptr [[TMP19]], align 8
+// CHECK-NEXT: [[TMP20:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_PTRS5]], i32 0, i32 0
+// CHECK-NEXT: store ptr [[DAT]], ptr [[TMP20]], align 8
+// CHECK-NEXT: [[TMP21:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_MAPPERS6]], i64 0, i64 0
+// CHECK-NEXT: store ptr null, ptr [[TMP21]], align 8
+// CHECK-NEXT: [[TMP22:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_BASEPTRS4]], i32 0, i32 0
+// CHECK-NEXT: [[TMP23:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_PTRS5]], i32 0, i32 0
+// CHECK-NEXT: [[TMP24:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 0
+// CHECK-NEXT: store i32 2, ptr [[TMP24]], align 4
+// CHECK-NEXT: [[TMP25:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 1
+// CHECK-NEXT: store i32 1, ptr [[TMP25]], align 4
+// CHECK-NEXT: [[TMP26:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 2
+// CHECK-NEXT: store ptr [[TMP22]], ptr [[TMP26]], align 8
+// CHECK-NEXT: [[TMP27:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 3
+// CHECK-NEXT: store ptr [[TMP23]], ptr [[TMP27]], align 8
+// CHECK-NEXT: [[TMP28:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 4
+// CHECK-NEXT: store ptr @.offload_sizes.1, ptr [[TMP28]], align 8
+// CHECK-NEXT: [[TMP29:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 5
+// CHECK-NEXT: store ptr @.offload_maptypes.2, ptr [[TMP29]], align 8
+// CHECK-NEXT: [[TMP30:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 6
+// CHECK-NEXT: store ptr null, ptr [[TMP30]], align 8
+// CHECK-NEXT: [[TMP31:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 7
+// CHECK-NEXT: store ptr null, ptr [[TMP31]], align 8
+// CHECK-NEXT: [[TMP32:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 8
+// CHECK-NEXT: store i64 0, ptr [[TMP32]], align 8
+// CHECK-NEXT: [[TMP33:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 9
+// CHECK-NEXT: store i64 0, ptr [[TMP33]], align 8
+// CHECK-NEXT: [[TMP34:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 10
+// CHECK-NEXT: store [3 x i32] [i32 -1, i32 0, i32 0], ptr [[TMP34]], align 4
+// CHECK-NEXT: [[TMP35:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 11
+// CHECK-NEXT: store [3 x i32] zeroinitializer, ptr [[TMP35]], align 4
+// CHECK-NEXT: [[TMP36:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 12
+// CHECK-NEXT: store i32 0, ptr [[TMP36]], align 4
+// CHECK-NEXT: [[TMP37:%.*]] = call i32 @__tgt_target_kernel(ptr @[[GLOB1]], i64 -1, i32 -1, i32 0, ptr @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z10map_structv_l23.region_id, ptr [[KERNEL_ARGS]])
+// CHECK-NEXT: [[TMP38:%.*]] = icmp ne i32 [[TMP37]], 0
+// CHECK-NEXT: br i1 [[TMP38]], label [[OMP_OFFLOAD_FAILED:%.*]], label [[OMP_OFFLOAD_CONT:%.*]]
+// CHECK: omp_offload.failed:
+// CHECK-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z10map_structv_l23(ptr [[DAT]]) #[[ATTR3:[0-9]+]]
+// CHECK-NEXT: br label [[OMP_OFFLOAD_CONT]]
+// CHECK: omp_offload.cont:
+// CHECK-NEXT: [[TMP39:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_BASEPTRS7]], i32 0, i32 0
+// CHECK-NEXT: store ptr [[DAT]], ptr [[TMP39]], align 8
+// CHECK-NEXT: [[TMP40:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_PTRS8]], i32 0, i32 0
+// CHECK-NEXT: store ptr [[DAT]], ptr [[TMP40]], align 8
+// CHECK-NEXT: [[TMP41:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_MAPPERS9]], i64 0, i64 0
+// CHECK-NEXT: store ptr null, ptr [[TMP41]], align 8
+// CHECK-NEXT: [[TMP42:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_BASEPTRS7]], i32 0, i32 0
+// CHECK-NEXT: [[TMP43:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_PTRS8]], i32 0, i32 0
+// CHECK-NEXT: call void @__tgt_target_data_end_mapper(ptr @[[GLOB1]], i64 -1, i32 1, ptr [[TMP42]], ptr [[TMP43]], ptr @.offload_sizes.3, ptr @.offload_maptypes.4, ptr null, ptr null)
+// CHECK-NEXT: [[XI10:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR]], ptr [[DAT]], i32 0, i32 2
+// CHECK-NEXT: [[TMP44:%.*]] = load i32, ptr [[XI10]], align 8
+// CHECK-NEXT: ret i32 [[TMP44]]
+//
+//
+// CHECK-LABEL: define internal void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z10map_structv_l23(
+// CHECK-SAME: ptr noundef nonnull align 8 dereferenceable(264) [[DAT:%.*]]) #[[ATTR4:[0-9]+]] {
+// CHECK-NEXT: entry:
+// CHECK-NEXT: [[DAT_ADDR:%.*]] = alloca ptr, align 8
+// CHECK-NEXT: store ptr [[DAT]], ptr [[DAT_ADDR]], align 8
+// CHECK-NEXT: [[TMP0:%.*]] = load ptr, ptr [[DAT_ADDR]], align 8
+// CHECK-NEXT: [[XI:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR:%.*]], ptr [[TMP0]], i32 0, i32 2
+// CHECK-NEXT: store i32 4, ptr [[XI]], align 8
+// CHECK-NEXT: [[XI1:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR]], ptr [[TMP0]], i32 0, i32 2
+// CHECK-NEXT: [[TMP1:%.*]] = load i32, ptr [[XI1]], align 8
+// CHECK-NEXT: [[DATUM:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR]], ptr [[TMP0]], i32 0, i32 0
+// CHECK-NEXT: [[TMP2:%.*]] = load ptr, ptr [[DATUM]], align 8
+// CHECK-NEXT: [[ARR:%.*]] = getelementptr inbounds [[STRUCT_DESCRIPTOR]], ptr [[TMP0]], i32 0, i32 3
+// CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds [1 x [30 x i64]], ptr [[ARR]], i64 0, i64 0
+// CHECK-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds [30 x i64], ptr [[ARRAYIDX]], i64 0, i64 0
+// CHECK-NEXT: [[TMP3:%.*]] = load i64, ptr [[ARRAYIDX2]], align 8
+// CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32, ptr [[TMP2]], i64 [[TMP3]]
+// CHECK-NEXT: store i32 [[TMP1]], ptr [[ARRAYIDX3]], align 4
+// CHECK-NEXT: ret void
+//
+//
+// CHECK-LABEL: define internal void @.omp_offloading.requires_reg(
+// CHECK-SAME: ) #[[ATTR5:[0-9]+]] {
+// CHECK-NEXT: entry:
+// CHECK-NEXT: call void @__tgt_register_requires(i64 1)
+// CHECK-NEXT: ret void
+//
diff --git a/openmp/libomptarget/test/offloading/struct_mapping_with_pointers.cpp b/openmp/libomptarget/test/offloading/struct_mapping_with_pointers.cpp
new file mode 100644
index 00000000000000..cecafe4c584168
--- /dev/null
+++ b/openmp/libomptarget/test/offloading/struct_mapping_with_pointers.cpp
@@ -0,0 +1,114 @@
+// clang-format off
+// RUN: %libomptarget-compilexx-generic && env LIBOMPTARGET_DEBUG=1 %libomptarget-run-generic 2>&1 | %fcheck-generic
+// clang-format on
+
+#include <stdio.h>
+#include <stdlib.h>
+
+struct Descriptor {
+ int *datum;
+ long int x;
+ int *more_datum;
+ int xi;
+ int val_datum, val_more_datum;
+ long int arr[1][30];
+ int val_arr;
+};
+
+int main() {
+ Descriptor dat = Descriptor();
+ dat.datum = (int *)malloc(sizeof(int) * 10);
+ dat.more_datum = (int *)malloc(sizeof(int) * 20);
+ dat.xi = 3;
+ dat.arr[0][0] = 1;
+
+ dat.datum[7] = 7;
+ dat.more_datum[17] = 17;
+
+ /// The struct is mapped with type 0x0 when the pointer fields are mapped.
+ /// The struct is also map explicitely by the user. The second mapping by
+ /// the user must not overwrite the mapping set up for the pointer fields
+ /// when mapping the struct happens after the mapping of the pointers.
+
+ // clang-format off
+ // CHECK: omptarget --> Entry 0: Base=[[DAT_HST_PTR_BASE:0x.*]], Begin=[[DAT_HST_PTR_BASE]], Size=288, Type=0x0, Name=unknown
+ // CHECK: omptarget --> Entry 1: Base=[[DAT_HST_PTR_BASE]], Begin=[[DAT_HST_PTR_BASE]], Size=288, Type=0x1000000000001, Name=unknown
+ // CHECK: omptarget --> Entry 2: Base=[[DAT_HST_PTR_BASE]], Begin=[[DATUM_HST_PTR_BASE:0x.*]], Size=40, Type=0x1000000000011, Name=unknown
+ // CHECK: omptarget --> Entry 3: Base=[[MORE_DATUM_HST_PTR_BASE:0x.*]], Begin=[[MORE_DATUM_HST_PTR_BEGIN:0x.*]], Size=80, Type=0x1000000000011, Name=unknown
+ // clang-format on
+
+ /// The struct will be mapped in the same order as the above entries.
+
+ /// First argument is the struct itself and it will be mapped once.
+
+ // clang-format off
+ // CHECK: omptarget --> Looking up mapping(HstPtrBegin=[[DAT_HST_PTR_BASE]], Size=288)...
+ // CHECK: PluginInterface --> MemoryManagerTy::allocate: size 288 with host pointer [[DAT_HST_PTR_BASE]].
+ // CHECK: omptarget --> Creating new map entry with HstPtrBase=[[DAT_HST_PTR_BASE]], HstPtrBegin=[[DAT_HST_PTR_BASE]], TgtAllocBegin=[[DAT_DEVICE_PTR_BASE:0x.*]], TgtPtrBegin=[[DAT_DEVICE_PTR_BASE]], Size=288, DynRefCount=1, HoldRefCount=0, Name=unknown
+ // CHECK: omptarget --> Moving 288 bytes (hst:[[DAT_HST_PTR_BASE]]) -> (tgt:[[DAT_DEVICE_PTR_BASE]])
+ // clang-format on
+
+ /// Second argument is dat.datum:
+ // clang-format off
+ // CHECK: omptarget --> Looking up mapping(HstPtrBegin=[[DATUM_HST_PTR_BASE]], Size=40)...
+ // CHECK: PluginInterface --> MemoryManagerTy::allocate: size 40 with host pointer [[DATUM_HST_PTR_BASE]].
+ // CHECK: omptarget --> Creating new map entry with HstPtrBase=[[DATUM_HST_PTR_BASE]], HstPtrBegin=[[DATUM_HST_PTR_BASE]], TgtAllocBegin=[[DATUM_DEVICE_PTR_BASE:0x.*]], TgtPtrBegin=[[DATUM_DEVICE_PTR_BASE]], Size=40, DynRefCount=1, HoldRefCount=0, Name=unknown
+ // CHECK: omptarget --> Moving 40 bytes (hst:[[DATUM_HST_PTR_BASE]]) -> (tgt:[[DATUM_DEVICE_PTR_BASE]])
+ // clang-format on
+
+ /// Third argument is dat.more_datum:
+ // clang-format off
+ // CHECK: omptarget --> Looking up mapping(HstPtrBegin=[[MORE_DATUM_HST_PTR_BEGIN]], Size=80)...
+ // CHECK: PluginInterface --> MemoryManagerTy::allocate: size 80 with host pointer [[MORE_DATUM_HST_PTR_BEGIN]].
+ // CHECK: omptarget --> Creating new map entry with HstPtrBase=[[MORE_DATUM_HST_PTR_BEGIN]], HstPtrBegin=[[MORE_DATUM_HST_PTR_BEGIN]], TgtAllocBegin=[[MORE_DATUM_DEVICE_PTR_BEGIN:0x.*]], TgtPtrBegin=[[MORE_DATUM_DEVICE_PTR_BEGIN]], Size=80, DynRefCount=1, HoldRefCount=0, Name=unknown
+ // CHECK: omptarget --> Moving 80 bytes (hst:[[MORE_DATUM_HST_PTR_BEGIN]]) -> (tgt:[[MORE_DATUM_DEVICE_PTR_BEGIN]])
+ // clang-format on
+
+#pragma omp target enter data map(to : dat.datum[ : 10]) \
+ map(to : dat.more_datum[ : 20]) map(to : dat)
+
+ /// Checks induced by having a target region:
+ // clang-format off
+ // CHECK: omptarget --> Entry 0: Base=[[DAT_HST_PTR_BASE]], Begin=[[DAT_HST_PTR_BASE]], Size=288, Type=0x223, Name=unknown
+ // CHECK: omptarget --> Mapping exists (implicit) with HstPtrBegin=[[DAT_HST_PTR_BASE]], TgtPtrBegin=[[DAT_DEVICE_PTR_BASE]], Size=288, DynRefCount=2 (incremented), HoldRefCount=0, Name=unknown
+ // CHECK: omptarget --> Obtained target argument [[DAT_DEVICE_PTR_BASE]] from host pointer [[DAT_HST_PTR_BASE]]
+ // clang-format on
+
+#pragma omp target
+ {
+ dat.xi = 4;
+ dat.datum[7]++;
+ dat.more_datum[17]++;
+ dat.val_datum = dat.datum[7];
+ dat.val_more_datum = dat.more_datum[17];
+ dat.datum[dat.arr[0][0]] = dat.xi;
+ dat.val_arr = dat.datum[dat.arr[0][0]];
+ }
+
+ /// Post-target region checks:
+ // clang-format off
+ // CHECK: omptarget --> Mapping exists with HstPtrBegin=[[DAT_HST_PTR_BASE]], TgtPtrBegin=[[DAT_DEVICE_PTR_BASE]], Size=288, DynRefCount=1 (decremented), HoldRefCount=0
+ // clang-format on
+
+#pragma omp target exit data map(from : dat)
+
+ /// Target data end checks:
+ // clang-format off
+ // CHECK: omptarget --> Mapping exists with HstPtrBegin=[[DAT_HST_PTR_BASE]], TgtPtrBegin=[[DAT_DEVICE_PTR_BASE]], Size=288, DynRefCount=0 (decremented, delayed deletion), HoldRefCount=0
+ // CHECK: omptarget --> Moving 288 bytes (tgt:[[DAT_DEVICE_PTR_BASE]]) -> (hst:[[DAT_HST_PTR_BASE]])
+ // clang-format on
+
+ // CHECK: dat.xi = 4
+ // CHECK: dat.val_datum = 8
+ // CHECK: dat.val_more_datum = 18
+ // CHECK: dat.datum[dat.arr[0][0]] = 0
+ // CHECK: dat.val_arr = 4
+
+ printf("dat.xi = %d\n", dat.xi);
+ printf("dat.val_datum = %d\n", dat.val_datum);
+ printf("dat.val_more_datum = %d\n", dat.val_more_datum);
+ printf("dat.datum[dat.arr[0][0]] = %d\n", dat.datum[dat.arr[0][0]]);
+ printf("dat.val_arr = %d\n", dat.val_arr);
+
+ return 0;
+}
More information about the cfe-commits
mailing list