Skip to content

[OpenMP] Introduce the KernelLaunchEnvironment as implicit argument #70401

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 1, 2023
Merged

[OpenMP] Introduce the KernelLaunchEnvironment as implicit argument #70401

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 18 additions & 8 deletions clang/lib/Sema/SemaOpenMP.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4249,12 +4249,15 @@ void Sema::ActOnOpenMPRegionStart(OpenMPDirectiveKind DKind, Scope *CurScope) {
getCurCapturedRegion()->TheCapturedDecl->addAttr(
AlwaysInlineAttr::CreateImplicit(
Context, {}, AlwaysInlineAttr::Keyword_forceinline));
Sema::CapturedParamNameType ParamsTarget[] = {
std::make_pair(StringRef(), QualType()) // __context with shared vars
};
SmallVector<Sema::CapturedParamNameType, 2> ParamsTarget;
if (getLangOpts().OpenMPIsTargetDevice)
ParamsTarget.push_back(std::make_pair(StringRef("dyn_ptr"), VoidPtrTy));
ParamsTarget.push_back(
std::make_pair(StringRef(), QualType())); // __context with shared vars;
// Start a captured region for 'target' with no implicit parameters.
ActOnCapturedRegionStart(DSAStack->getConstructLoc(), CurScope, CR_OpenMP,
ParamsTarget, /*OpenMPCaptureLevel=*/1);
ParamsTarget,
/*OpenMPCaptureLevel=*/1);
Sema::CapturedParamNameType ParamsTeamsOrParallel[] = {
std::make_pair(".global_tid.", KmpInt32PtrTy),
std::make_pair(".bound_tid.", KmpInt32PtrTy),
Expand Down Expand Up @@ -4293,8 +4296,13 @@ void Sema::ActOnOpenMPRegionStart(OpenMPDirectiveKind DKind, Scope *CurScope) {
getCurCapturedRegion()->TheCapturedDecl->addAttr(
AlwaysInlineAttr::CreateImplicit(
Context, {}, AlwaysInlineAttr::Keyword_forceinline));
SmallVector<Sema::CapturedParamNameType, 2> ParamsTarget;
if (getLangOpts().OpenMPIsTargetDevice)
ParamsTarget.push_back(std::make_pair(StringRef("dyn_ptr"), VoidPtrTy));
ParamsTarget.push_back(
std::make_pair(StringRef(), QualType())); // __context with shared vars;
ActOnCapturedRegionStart(DSAStack->getConstructLoc(), CurScope, CR_OpenMP,
std::make_pair(StringRef(), QualType()),
ParamsTarget,
/*OpenMPCaptureLevel=*/1);
break;
}
Expand Down Expand Up @@ -4499,9 +4507,11 @@ void Sema::ActOnOpenMPRegionStart(OpenMPDirectiveKind DKind, Scope *CurScope) {
getCurCapturedRegion()->TheCapturedDecl->addAttr(
AlwaysInlineAttr::CreateImplicit(
Context, {}, AlwaysInlineAttr::Keyword_forceinline));
Sema::CapturedParamNameType ParamsTarget[] = {
std::make_pair(StringRef(), QualType()) // __context with shared vars
};
SmallVector<Sema::CapturedParamNameType, 2> ParamsTarget;
if (getLangOpts().OpenMPIsTargetDevice)
ParamsTarget.push_back(std::make_pair(StringRef("dyn_ptr"), VoidPtrTy));
ParamsTarget.push_back(
std::make_pair(StringRef(), QualType())); // __context with shared vars;
// Start a captured region for 'target' with no implicit parameters.
ActOnCapturedRegionStart(DSAStack->getConstructLoc(), CurScope, CR_OpenMP,
ParamsTarget, /*OpenMPCaptureLevel=*/1);
Expand Down
14 changes: 10 additions & 4 deletions clang/test/OpenMP/amdgcn_target_codegen.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,15 +29,18 @@ int test_amdgcn_target_tid_threads_simd() {

#endif
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z30test_amdgcn_target_tid_threadsv_l14
// CHECK-SAME: (ptr noundef nonnull align 4 dereferenceable(4000) [[ARR:%.*]]) #[[ATTR0:[0-9]+]] {
// CHECK-SAME: (ptr noalias noundef [[DYN_PTR:%.*]], ptr noundef nonnull align 4 dereferenceable(4000) [[ARR:%.*]]) #[[ATTR0:[0-9]+]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[ARR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[I:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[ARR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[ARR_ADDR]] to ptr
// CHECK-NEXT: [[I_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[I]] to ptr
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[ARR]], ptr [[ARR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load ptr, ptr [[ARR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z30test_amdgcn_target_tid_threadsv_l14_kernel_environment to ptr))
// CHECK-NEXT: [[TMP1:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z30test_amdgcn_target_tid_threadsv_l14_kernel_environment to ptr), ptr [[DYN_PTR]])
// CHECK-NEXT: [[EXEC_USER_CODE:%.*]] = icmp eq i32 [[TMP1]], -1
// CHECK-NEXT: br i1 [[EXEC_USER_CODE]], label [[USER_CODE_ENTRY:%.*]], label [[WORKER_EXIT:%.*]]
// CHECK: user_code.entry:
Expand Down Expand Up @@ -66,19 +69,22 @@ int test_amdgcn_target_tid_threads_simd() {
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z35test_amdgcn_target_tid_threads_simdv_l23
// CHECK-SAME: (ptr noundef nonnull align 4 dereferenceable(4000) [[ARR:%.*]]) #[[ATTR1:[0-9]+]] {
// CHECK-SAME: (ptr noalias noundef [[DYN_PTR:%.*]], ptr noundef nonnull align 4 dereferenceable(4000) [[ARR:%.*]]) #[[ATTR1:[0-9]+]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[ARR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[TMP:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[I:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[ARR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[ARR_ADDR]] to ptr
// CHECK-NEXT: [[TMP_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[TMP]] to ptr
// CHECK-NEXT: [[DOTOMP_IV_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTOMP_IV]] to ptr
// CHECK-NEXT: [[I_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[I]] to ptr
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[ARR]], ptr [[ARR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load ptr, ptr [[ARR_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z35test_amdgcn_target_tid_threads_simdv_l23_kernel_environment to ptr))
// CHECK-NEXT: [[TMP1:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z35test_amdgcn_target_tid_threads_simdv_l23_kernel_environment to ptr), ptr [[DYN_PTR]])
// CHECK-NEXT: [[EXEC_USER_CODE:%.*]] = icmp eq i32 [[TMP1]], -1
// CHECK-NEXT: br i1 [[EXEC_USER_CODE]], label [[USER_CODE_ENTRY:%.*]], label [[WORKER_EXIT:%.*]]
// CHECK: user_code.entry:
Expand Down
28 changes: 20 additions & 8 deletions clang/test/OpenMP/amdgcn_target_device_vla.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -97,21 +97,24 @@ int main() {

#endif
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo1v_l12
// CHECK-SAME: (ptr noundef nonnull align 4 dereferenceable(4) [[SUM:%.*]]) #[[ATTR0:[0-9]+]] {
// CHECK-SAME: (ptr noalias noundef [[DYN_PTR:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[SUM:%.*]]) #[[ATTR0:[0-9]+]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[SUM_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[N:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[__VLA_EXPR0:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[I:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[I1:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[SUM_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[SUM_ADDR]] to ptr
// CHECK-NEXT: [[N_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N]] to ptr
// CHECK-NEXT: [[__VLA_EXPR0_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[__VLA_EXPR0]] to ptr
// CHECK-NEXT: [[I_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[I]] to ptr
// CHECK-NEXT: [[I1_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[I1]] to ptr
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[SUM]], ptr [[SUM_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load ptr, ptr [[SUM_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo1v_l12_kernel_environment to ptr))
// CHECK-NEXT: [[TMP1:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo1v_l12_kernel_environment to ptr), ptr [[DYN_PTR]])
// CHECK-NEXT: [[EXEC_USER_CODE:%.*]] = icmp eq i32 [[TMP1]], -1
// CHECK-NEXT: br i1 [[EXEC_USER_CODE]], label [[USER_CODE_ENTRY:%.*]], label [[WORKER_EXIT:%.*]]
// CHECK: user_code.entry:
Expand Down Expand Up @@ -174,26 +177,29 @@ int main() {
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo2v_l30
// CHECK-SAME: (i64 noundef [[M:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[RESULT:%.*]]) #[[ATTR0]] {
// CHECK-SAME: (ptr noalias noundef [[DYN_PTR:%.*]], i64 noundef [[M:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[RESULT:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[M_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[RESULT_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[M_CASTED:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[DOTZERO_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTTHREADID_TEMP_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[M_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[M_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[RESULT_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RESULT_ADDR]] to ptr
// CHECK-NEXT: [[M_CASTED_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[M_CASTED]] to ptr
// CHECK-NEXT: [[DOTZERO_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTZERO_ADDR]] to ptr
// CHECK-NEXT: [[DOTTHREADID_TEMP__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTTHREADID_TEMP_]] to ptr
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[M]], ptr [[M_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[RESULT]], ptr [[RESULT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = load ptr, ptr [[RESULT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo2v_l30_kernel_environment to ptr))
// CHECK-NEXT: [[TMP2:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo2v_l30_kernel_environment to ptr), ptr [[DYN_PTR]])
// CHECK-NEXT: [[EXEC_USER_CODE:%.*]] = icmp eq i32 [[TMP2]], -1
// CHECK-NEXT: br i1 [[EXEC_USER_CODE]], label [[USER_CODE_ENTRY:%.*]], label [[WORKER_EXIT:%.*]]
// CHECK: user_code.entry:
Expand Down Expand Up @@ -540,26 +546,29 @@ int main() {
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo3v_l52
// CHECK-SAME: (i64 noundef [[M:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[RESULT:%.*]]) #[[ATTR0]] {
// CHECK-SAME: (ptr noalias noundef [[DYN_PTR:%.*]], i64 noundef [[M:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[RESULT:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[M_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[RESULT_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[M_CASTED:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[DOTZERO_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTTHREADID_TEMP_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[M_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[M_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
// CHECK-NEXT: [[RESULT_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[RESULT_ADDR]] to ptr
// CHECK-NEXT: [[M_CASTED_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[M_CASTED]] to ptr
// CHECK-NEXT: [[DOTZERO_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTZERO_ADDR]] to ptr
// CHECK-NEXT: [[DOTTHREADID_TEMP__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTTHREADID_TEMP_]] to ptr
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[M]], ptr [[M_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[RESULT]], ptr [[RESULT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = load ptr, ptr [[RESULT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo3v_l52_kernel_environment to ptr))
// CHECK-NEXT: [[TMP2:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo3v_l52_kernel_environment to ptr), ptr [[DYN_PTR]])
// CHECK-NEXT: [[EXEC_USER_CODE:%.*]] = icmp eq i32 [[TMP2]], -1
// CHECK-NEXT: br i1 [[EXEC_USER_CODE]], label [[USER_CODE_ENTRY:%.*]], label [[WORKER_EXIT:%.*]]
// CHECK: user_code.entry:
Expand Down Expand Up @@ -894,8 +903,9 @@ int main() {
//
//
// CHECK-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo4v_l76
// CHECK-SAME: (i64 noundef [[M:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[RESULT:%.*]]) #[[ATTR0]] {
// CHECK-SAME: (ptr noalias noundef [[DYN_PTR:%.*]], i64 noundef [[M:%.*]], i64 noundef [[N:%.*]], i64 noundef [[VLA:%.*]], ptr noundef nonnull align 4 dereferenceable(4) [[RESULT:%.*]]) #[[ATTR0]] {
// CHECK-NEXT: entry:
// CHECK-NEXT: [[DYN_PTR_ADDR:%.*]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[M_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[N_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[VLA_ADDR:%.*]] = alloca i64, align 8, addrspace(5)
Expand All @@ -904,6 +914,7 @@ int main() {
// CHECK-NEXT: [[N_CASTED:%.*]] = alloca i64, align 8, addrspace(5)
// CHECK-NEXT: [[DOTZERO_ADDR:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DOTTHREADID_TEMP_:%.*]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[M_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[M_ADDR]] to ptr
// CHECK-NEXT: [[N_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_ADDR]] to ptr
// CHECK-NEXT: [[VLA_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[VLA_ADDR]] to ptr
Expand All @@ -912,13 +923,14 @@ int main() {
// CHECK-NEXT: [[N_CASTED_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[N_CASTED]] to ptr
// CHECK-NEXT: [[DOTZERO_ADDR_ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTZERO_ADDR]] to ptr
// CHECK-NEXT: [[DOTTHREADID_TEMP__ASCAST:%.*]] = addrspacecast ptr addrspace(5) [[DOTTHREADID_TEMP_]] to ptr
// CHECK-NEXT: store ptr [[DYN_PTR]], ptr [[DYN_PTR_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[M]], ptr [[M_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[N]], ptr [[N_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[VLA]], ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: store ptr [[RESULT]], ptr [[RESULT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load i64, ptr [[VLA_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = load ptr, ptr [[RESULT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo4v_l76_kernel_environment to ptr))
// CHECK-NEXT: [[TMP2:%.*]] = call i32 @__kmpc_target_init(ptr addrspacecast (ptr addrspace(1) @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo4v_l76_kernel_environment to ptr), ptr [[DYN_PTR]])
// CHECK-NEXT: [[EXEC_USER_CODE:%.*]] = icmp eq i32 [[TMP2]], -1
// CHECK-NEXT: br i1 [[EXEC_USER_CODE]], label [[USER_CODE_ENTRY:%.*]], label [[WORKER_EXIT:%.*]]
// CHECK: user_code.entry:
Expand Down
2 changes: 2 additions & 0 deletions clang/test/OpenMP/amdgcn_target_init_temp_alloca.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,8 +11,10 @@ int test_amdgcn_target_temp_alloca() {

int arr[N];

// CHECK: [[DYN_PTR_ADDR:%.+]] = alloca ptr, align 8, addrspace(5)
// CHECK: [[VAR_ADDR:%.+]] = alloca ptr, align 8, addrspace(5)
// CHECK-NEXT: [[VAR2_ADDR:%.+]] = alloca i32, align 4, addrspace(5)
// CHECK-NEXT: [[DYN_PTR_ADDR_CAST:%.+]] = addrspacecast ptr addrspace(5) [[DYN_PTR_ADDR]] to ptr
// CHECK-NEXT: [[VAR_ADDR_CAST:%.+]] = addrspacecast ptr addrspace(5) [[VAR_ADDR]] to ptr
// CHECK-NEXT: [[VAR2_ADDR_CAST:%.+]] = addrspacecast ptr addrspace(5) [[VAR2_ADDR]] to ptr
// CHECK: store ptr [[VAR:%.+]], ptr [[VAR_ADDR_CAST]], align 8
Expand Down
Loading