Magic-Mirror/llvm-capstone
1
0
Fork 0
mirror of https://github.com/capstone-engine/llvm-capstone.git synced 2024-11-27 07:31:28 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[clang][OpenMP] Add interop support for multiple depend clauses

Browse Source 
This patch removes the constraint of the `interop` directive where only a single
`depend` clause was allowed.

Differential Revision: https://reviews.llvm.org/D155692
This commit is contained in:
Michael Halkenhaeuser 2023-07-18 12:56:12 -04:00
parent 2c38740ca6
commit 7d4e14c76b
2 changed files with 103 additions and 34 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
clang/lib/CodeGen/CGStmtOpenMP.cpp
View File

@ -6968,27 +6968,27 @@ void CodeGenFunction::EmitOMPTeamsDistributeParallelForSimdDirective(
void CodeGenFunction::EmitOMPInteropDirective(const OMPInteropDirective &S) {
llvm::OpenMPIRBuilder &OMPBuilder = CGM.getOpenMPRuntime().getOMPBuilder();
llvm::Value *Device = nullptr;
llvm::Value *NumDependences = nullptr;
llvm::Value *DependenceList = nullptr;
if (const auto *C = S.getSingleClause<OMPDeviceClause>())
Device = EmitScalarExpr(C->getDevice());
llvm::Value *NumDependences = nullptr;
llvm::Value *DependenceAddress = nullptr;
if (const auto *DC = S.getSingleClause<OMPDependClause>()) {
OMPTaskDataTy::DependData Dependencies(DC->getDependencyKind(),
DC->getModifier());
Dependencies.DepExprs.append(DC->varlist_begin(), DC->varlist_end());
std::pair<llvm::Value *, Address> DependencePair =
CGM.getOpenMPRuntime().emitDependClause(*this, Dependencies,
DC->getBeginLoc());
NumDependences = DependencePair.first;
DependenceAddress = Builder.CreatePointerCast(
DependencePair.second.getPointer(), CGM.Int8PtrTy);
// Build list and emit dependences
OMPTaskDataTy Data;
buildDependences(S, Data);
if (!Data.Dependences.empty()) {
Address DependenciesArray = Address::invalid();
std::tie(NumDependences, DependenciesArray) =
CGM.getOpenMPRuntime().emitDependClause(*this, Data.Dependences,
S.getBeginLoc());
DependenceList = DependenciesArray.getPointer();
}
Data.HasNowaitClause = S.hasClausesOfKind<OMPNowaitClause>();
assert(!(S.hasClausesOfKind<OMPNowaitClause>() &&
!(S.getSingleClause<OMPInitClause>() ||
S.getSingleClause<OMPDestroyClause>() ||
S.getSingleClause<OMPUseClause>())) &&
assert(!(Data.HasNowaitClause && !(S.getSingleClause<OMPInitClause>() ||
S.getSingleClause<OMPDestroyClause>() ||
S.getSingleClause<OMPUseClause>())) &&
"OMPNowaitClause clause is used separately in OMPInteropDirective.");
if (const auto *C = S.getSingleClause<OMPInitClause>()) {
@ -7002,20 +7002,20 @@ void CodeGenFunction::EmitOMPInteropDirective(const OMPInteropDirective &S) {
InteropType = llvm::omp::OMPInteropType::TargetSync;
}
OMPBuilder.createOMPInteropInit(Builder, InteropvarPtr, InteropType, Device,
NumDependences, DependenceAddress,
S.hasClausesOfKind<OMPNowaitClause>());
NumDependences, DependenceList,
Data.HasNowaitClause);
} else if (const auto *C = S.getSingleClause<OMPDestroyClause>()) {
llvm::Value *InteropvarPtr =
EmitLValue(C->getInteropVar()).getPointer(*this);
OMPBuilder.createOMPInteropDestroy(Builder, InteropvarPtr, Device,
NumDependences, DependenceAddress,
S.hasClausesOfKind<OMPNowaitClause>());
NumDependences, DependenceList,
Data.HasNowaitClause);
} else if (const auto *C = S.getSingleClause<OMPUseClause>()) {
llvm::Value *InteropvarPtr =
EmitLValue(C->getInteropVar()).getPointer(*this);
OMPBuilder.createOMPInteropUse(Builder, InteropvarPtr, Device,
NumDependences, DependenceAddress,
S.hasClausesOfKind<OMPNowaitClause>());
NumDependences, DependenceList,
Data.HasNowaitClause);
}
}

93
clang/test/OpenMP/interop_irbuilder.cpp
View File

@ -7,7 +7,7 @@ typedef void *omp_interop_t;
void test1() {
int device_id = 4;
int D0, D1;
int D0, D1, D2;
omp_interop_t interop;
#pragma omp interop init(target : interop)
@ -20,6 +20,9 @@ void test1() {
#pragma omp interop use(interop) depend(in : D0, D1) nowait
#pragma omp interop use(interop) depend(in : D0) depend(inout : D1) \
depend(out : D2) nowait
#pragma omp interop destroy(interop) depend(in : D0, D1)
}
@ -31,7 +34,7 @@ struct S {
void S::member_test() {
int device_id = 4;
int D0, D1;
int D0, D1, D2;
#pragma omp interop init(target : interop)
@ -43,6 +46,9 @@ void S::member_test() {
#pragma omp interop use(interop) depend(in : D0, D1) nowait
#pragma omp interop use(interop) depend(in : D0) depend(inout : D1) \
depend(out : D2) nowait
#pragma omp interop destroy(interop) depend(in : D0, D1)
}
// CHECK-LABEL: @_Z5test1v(
@ -50,11 +56,14 @@ void S::member_test() {
// CHECK-NEXT: [[DEVICE_ID:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[D0:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[D1:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[D2:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[INTEROP:%.*]] = alloca ptr, align 8
// CHECK-NEXT: [[DOTDEP_ARR_ADDR:%.*]] = alloca [2 x %struct.kmp_depend_info], align 8
// CHECK-NEXT: [[DEP_COUNTER_ADDR:%.*]] = alloca i64, align 8
// CHECK-NEXT: [[DOTDEP_ARR_ADDR5:%.*]] = alloca [2 x %struct.kmp_depend_info], align 8
// CHECK-NEXT: [[DOTDEP_ARR_ADDR5:%.*]] = alloca [3 x %struct.kmp_depend_info], align 8
// CHECK-NEXT: [[DEP_COUNTER_ADDR6:%.*]] = alloca i64, align 8
// CHECK-NEXT: [[DOTDEP_ARR_ADDR8:%.*]] = alloca [2 x %struct.kmp_depend_info], align 8
// CHECK-NEXT: [[DEP_COUNTER_ADDR9:%.*]] = alloca i64, align 8
// CHECK-NEXT: store i32 4, ptr [[DEVICE_ID]], align 4
// CHECK-NEXT: [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1:[0-9]+]])
// CHECK-NEXT: call void @__tgt_interop_init(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM]], ptr [[INTEROP]], i32 1, i32 -1, i32 0, ptr null, i32 0)
@ -86,7 +95,7 @@ void S::member_test() {
// CHECK-NEXT: store i64 2, ptr [[DEP_COUNTER_ADDR]], align 8
// CHECK-NEXT: [[OMP_GLOBAL_THREAD_NUM4:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1]])
// CHECK-NEXT: call void @__tgt_interop_use(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM4]], ptr [[INTEROP]], i32 -1, i32 2, ptr [[TMP2]], i32 1)
// CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds [2 x %struct.kmp_depend_info], ptr [[DOTDEP_ARR_ADDR5]], i64 0, i64 0
// CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds [3 x %struct.kmp_depend_info], ptr [[DOTDEP_ARR_ADDR5]], i64 0, i64 0
// CHECK-NEXT: [[TMP14:%.*]] = ptrtoint ptr [[D0]] to i64
// CHECK-NEXT: [[TMP15:%.*]] = getelementptr [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP13]], i64 0
// CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP15]], i32 0, i32 0
@ -102,10 +111,38 @@ void S::member_test() {
// CHECK-NEXT: [[TMP22:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP20]], i32 0, i32 1
// CHECK-NEXT: store i64 4, ptr [[TMP22]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP20]], i32 0, i32 2
// CHECK-NEXT: store i8 1, ptr [[TMP23]], align 8
// CHECK-NEXT: store i64 2, ptr [[DEP_COUNTER_ADDR6]], align 8
// CHECK-NEXT: store i8 3, ptr [[TMP23]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = ptrtoint ptr [[D2]] to i64
// CHECK-NEXT: [[TMP25:%.*]] = getelementptr [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP13]], i64 2
// CHECK-NEXT: [[TMP26:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP25]], i32 0, i32 0
// CHECK-NEXT: store i64 [[TMP24]], ptr [[TMP26]], align 8
// CHECK-NEXT: [[TMP27:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP25]], i32 0, i32 1
// CHECK-NEXT: store i64 4, ptr [[TMP27]], align 8
// CHECK-NEXT: [[TMP28:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP25]], i32 0, i32 2
// CHECK-NEXT: store i8 3, ptr [[TMP28]], align 8
// CHECK-NEXT: store i64 3, ptr [[DEP_COUNTER_ADDR6]], align 8
// CHECK-NEXT: [[OMP_GLOBAL_THREAD_NUM7:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1]])
// CHECK-NEXT: call void @__tgt_interop_destroy(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM7]], ptr [[INTEROP]], i32 -1, i32 2, ptr [[TMP13]], i32 0)
// CHECK-NEXT: call void @__tgt_interop_use(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM7]], ptr [[INTEROP]], i32 -1, i32 3, ptr [[TMP13]], i32 1)
// CHECK-NEXT: [[TMP29:%.*]] = getelementptr inbounds [2 x %struct.kmp_depend_info], ptr [[DOTDEP_ARR_ADDR8]], i64 0, i64 0
// CHECK-NEXT: [[TMP30:%.*]] = ptrtoint ptr [[D0]] to i64
// CHECK-NEXT: [[TMP31:%.*]] = getelementptr [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP29]], i64 0
// CHECK-NEXT: [[TMP32:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP31]], i32 0, i32 0
// CHECK-NEXT: store i64 [[TMP30]], ptr [[TMP32]], align 8
// CHECK-NEXT: [[TMP33:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP31]], i32 0, i32 1
// CHECK-NEXT: store i64 4, ptr [[TMP33]], align 8
// CHECK-NEXT: [[TMP34:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP31]], i32 0, i32 2
// CHECK-NEXT: store i8 1, ptr [[TMP34]], align 8
// CHECK-NEXT: [[TMP35:%.*]] = ptrtoint ptr [[D1]] to i64
// CHECK-NEXT: [[TMP36:%.*]] = getelementptr [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP29]], i64 1
// CHECK-NEXT: [[TMP37:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP36]], i32 0, i32 0
// CHECK-NEXT: store i64 [[TMP35]], ptr [[TMP37]], align 8
// CHECK-NEXT: [[TMP38:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP36]], i32 0, i32 1
// CHECK-NEXT: store i64 4, ptr [[TMP38]], align 8
// CHECK-NEXT: [[TMP39:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP36]], i32 0, i32 2
// CHECK-NEXT: store i8 1, ptr [[TMP39]], align 8
// CHECK-NEXT: store i64 2, ptr [[DEP_COUNTER_ADDR9]], align 8
// CHECK-NEXT: [[OMP_GLOBAL_THREAD_NUM10:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1]])
// CHECK-NEXT: call void @__tgt_interop_destroy(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM10]], ptr [[INTEROP]], i32 -1, i32 2, ptr [[TMP29]], i32 0)
// CHECK-NEXT: ret void
//
//
@ -115,10 +152,13 @@ void S::member_test() {
// CHECK-NEXT: [[DEVICE_ID:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[D0:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[D1:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[D2:%.*]] = alloca i32, align 4
// CHECK-NEXT: [[DOTDEP_ARR_ADDR:%.*]] = alloca [2 x %struct.kmp_depend_info], align 8
// CHECK-NEXT: [[DEP_COUNTER_ADDR:%.*]] = alloca i64, align 8
// CHECK-NEXT: [[DOTDEP_ARR_ADDR10:%.*]] = alloca [2 x %struct.kmp_depend_info], align 8
// CHECK-NEXT: [[DOTDEP_ARR_ADDR10:%.*]] = alloca [3 x %struct.kmp_depend_info], align 8
// CHECK-NEXT: [[DEP_COUNTER_ADDR11:%.*]] = alloca i64, align 8
// CHECK-NEXT: [[DOTDEP_ARR_ADDR14:%.*]] = alloca [2 x %struct.kmp_depend_info], align 8
// CHECK-NEXT: [[DEP_COUNTER_ADDR15:%.*]] = alloca i64, align 8
// CHECK-NEXT: store ptr [[THIS:%.*]], ptr [[THIS_ADDR]], align 8
// CHECK-NEXT: [[THIS1:%.*]] = load ptr, ptr [[THIS_ADDR]], align 8
// CHECK-NEXT: store i32 4, ptr [[DEVICE_ID]], align 4
@ -157,7 +197,7 @@ void S::member_test() {
// CHECK-NEXT: [[INTEROP8:%.*]] = getelementptr inbounds [[STRUCT_S]], ptr [[THIS1]], i32 0, i32 0
// CHECK-NEXT: [[OMP_GLOBAL_THREAD_NUM9:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1]])
// CHECK-NEXT: call void @__tgt_interop_use(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM9]], ptr [[INTEROP8]], i32 -1, i32 2, ptr [[TMP2]], i32 1)
// CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds [2 x %struct.kmp_depend_info], ptr [[DOTDEP_ARR_ADDR10]], i64 0, i64 0
// CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds [3 x %struct.kmp_depend_info], ptr [[DOTDEP_ARR_ADDR10]], i64 0, i64 0
// CHECK-NEXT: [[TMP14:%.*]] = ptrtoint ptr [[D0]] to i64
// CHECK-NEXT: [[TMP15:%.*]] = getelementptr [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP13]], i64 0
// CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP15]], i32 0, i32 0
@ -173,10 +213,39 @@ void S::member_test() {
// CHECK-NEXT: [[TMP22:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP20]], i32 0, i32 1
// CHECK-NEXT: store i64 4, ptr [[TMP22]], align 8
// CHECK-NEXT: [[TMP23:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP20]], i32 0, i32 2
// CHECK-NEXT: store i8 1, ptr [[TMP23]], align 8
// CHECK-NEXT: store i64 2, ptr [[DEP_COUNTER_ADDR11]], align 8
// CHECK-NEXT: store i8 3, ptr [[TMP23]], align 8
// CHECK-NEXT: [[TMP24:%.*]] = ptrtoint ptr [[D2]] to i64
// CHECK-NEXT: [[TMP25:%.*]] = getelementptr [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP13]], i64 2
// CHECK-NEXT: [[TMP26:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP25]], i32 0, i32 0
// CHECK-NEXT: store i64 [[TMP24]], ptr [[TMP26]], align 8
// CHECK-NEXT: [[TMP27:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP25]], i32 0, i32 1
// CHECK-NEXT: store i64 4, ptr [[TMP27]], align 8
// CHECK-NEXT: [[TMP28:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP25]], i32 0, i32 2
// CHECK-NEXT: store i8 3, ptr [[TMP28]], align 8
// CHECK-NEXT: store i64 3, ptr [[DEP_COUNTER_ADDR11]], align 8
// CHECK-NEXT: [[INTEROP12:%.*]] = getelementptr inbounds [[STRUCT_S]], ptr [[THIS1]], i32 0, i32 0
// CHECK-NEXT: [[OMP_GLOBAL_THREAD_NUM13:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1]])
// CHECK-NEXT: call void @__tgt_interop_destroy(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM13]], ptr [[INTEROP12]], i32 -1, i32 2, ptr [[TMP13]], i32 0)
// CHECK-NEXT: call void @__tgt_interop_use(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM13]], ptr [[INTEROP12]], i32 -1, i32 3, ptr [[TMP13]], i32 1)
// CHECK-NEXT: [[TMP29:%.*]] = getelementptr inbounds [2 x %struct.kmp_depend_info], ptr [[DOTDEP_ARR_ADDR14]], i64 0, i64 0
// CHECK-NEXT: [[TMP30:%.*]] = ptrtoint ptr [[D0]] to i64
// CHECK-NEXT: [[TMP31:%.*]] = getelementptr [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP29]], i64 0
// CHECK-NEXT: [[TMP32:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP31]], i32 0, i32 0
// CHECK-NEXT: store i64 [[TMP30]], ptr [[TMP32]], align 8
// CHECK-NEXT: [[TMP33:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP31]], i32 0, i32 1
// CHECK-NEXT: store i64 4, ptr [[TMP33]], align 8
// CHECK-NEXT: [[TMP34:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP31]], i32 0, i32 2
// CHECK-NEXT: store i8 1, ptr [[TMP34]], align 8
// CHECK-NEXT: [[TMP35:%.*]] = ptrtoint ptr [[D1]] to i64
// CHECK-NEXT: [[TMP36:%.*]] = getelementptr [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP29]], i64 1
// CHECK-NEXT: [[TMP37:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP36]], i32 0, i32 0
// CHECK-NEXT: store i64 [[TMP35]], ptr [[TMP37]], align 8
// CHECK-NEXT: [[TMP38:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP36]], i32 0, i32 1
// CHECK-NEXT: store i64 4, ptr [[TMP38]], align 8
// CHECK-NEXT: [[TMP39:%.*]] = getelementptr inbounds [[STRUCT_KMP_DEPEND_INFO]], ptr [[TMP36]], i32 0, i32 2
// CHECK-NEXT: store i8 1, ptr [[TMP39]], align 8
// CHECK-NEXT: store i64 2, ptr [[DEP_COUNTER_ADDR15]], align 8
// CHECK-NEXT: [[INTEROP16:%.*]] = getelementptr inbounds [[STRUCT_S]], ptr [[THIS1]], i32 0, i32 0
// CHECK-NEXT: [[OMP_GLOBAL_THREAD_NUM17:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1]])
// CHECK-NEXT: call void @__tgt_interop_destroy(ptr @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM17]], ptr [[INTEROP16]], i32 -1, i32 2, ptr [[TMP29]], i32 0)
// CHECK-NEXT: ret void
//