Skip to content

[Clang][OpenMP] Switch to __kmpc_parallel_60 with strict parameter #171082

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
kevinsala merged 5 commits into from
Dec 8, 2025
Merged

[Clang][OpenMP] Switch to __kmpc_parallel_60 with strict parameter #171082

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
446fa26
[Clang][OpenMP] Switch to __kmpc_parallel_60 with nthreads strict par…
kevinsala Dec 8, 2025
4b3c8b9
Fix tests
kevinsala Dec 8, 2025
02f1770
Add comment
kevinsala Dec 8, 2025
760f0b8
Merge remote-tracking branch 'upstream/main' into parallel_60_with_st…
kevinsala Dec 8, 2025
5732e8a
Fix llvm unit test
kevinsala Dec 8, 2025
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
9 changes: 7 additions & 2 deletions clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1291,6 +1291,9 @@ void CGOpenMPRuntimeGPU::emitParallelCall(
else
NumThreadsVal = Bld.CreateZExtOrTrunc(NumThreadsVal, CGF.Int32Ty);

// No strict prescriptiveness for the number of threads.
llvm::Value *StrictNumThreadsVal = llvm::ConstantInt::get(CGF.Int32Ty, 0);

assert(IfCondVal && "Expected a value");
llvm::Value *RTLoc = emitUpdateLocation(CGF, Loc);
llvm::Value *Args[] = {
Expand All @@ -1303,9 +1306,11 @@ void CGOpenMPRuntimeGPU::emitParallelCall(
ID,
Bld.CreateBitOrPointerCast(CapturedVarsAddrs.emitRawPointer(CGF),
CGF.VoidPtrPtrTy),
llvm::ConstantInt::get(CGM.SizeTy, CapturedVars.size())};
llvm::ConstantInt::get(CGM.SizeTy, CapturedVars.size()),
StrictNumThreadsVal};

CGF.EmitRuntimeCall(OMPBuilder.getOrCreateRuntimeFunction(
CGM.getModule(), OMPRTL___kmpc_parallel_51),
CGM.getModule(), OMPRTL___kmpc_parallel_60),
Args);
};

Expand Down
6 changes: 3 additions & 3 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 @@ -327,7 +327,7 @@ int main() {
// CHECK-NEXT: store ptr [[TMP1]], ptr [[TMP29]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP31:%.*]] = load i32, ptr [[TMP30]], align 4
// CHECK-NEXT: call void @__kmpc_parallel_51(ptr addrspacecast (ptr addrspace(1) @[[GLOB1]] to ptr), i32 [[TMP31]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo2v_l30_omp_outlined_omp_outlined, ptr null, ptr [[CAPTURED_VARS_ADDRS_ASCAST]], i64 5)
// CHECK-NEXT: call void @__kmpc_parallel_60(ptr addrspacecast (ptr addrspace(1) @[[GLOB1]] to ptr), i32 [[TMP31]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo2v_l30_omp_outlined_omp_outlined, ptr null, ptr [[CAPTURED_VARS_ADDRS_ASCAST]], i64 5, i32 0)
// CHECK-NEXT: br label [[OMP_INNER_FOR_INC:%.*]]
// CHECK: omp.inner.for.inc:
// CHECK-NEXT: [[TMP32:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
Expand Down Expand Up @@ -701,7 +701,7 @@ int main() {
// CHECK-NEXT: store ptr [[A]], ptr [[TMP27]], align 8
// CHECK-NEXT: [[TMP28:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP29:%.*]] = load i32, ptr [[TMP28]], align 4
// CHECK-NEXT: call void @__kmpc_parallel_51(ptr addrspacecast (ptr addrspace(1) @[[GLOB1]] to ptr), i32 [[TMP29]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo3v_l52_omp_outlined_omp_outlined, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo3v_l52_omp_outlined_omp_outlined_wrapper, ptr [[CAPTURED_VARS_ADDRS_ASCAST]], i64 3)
// CHECK-NEXT: call void @__kmpc_parallel_60(ptr addrspacecast (ptr addrspace(1) @[[GLOB1]] to ptr), i32 [[TMP29]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo3v_l52_omp_outlined_omp_outlined, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo3v_l52_omp_outlined_omp_outlined_wrapper, ptr [[CAPTURED_VARS_ADDRS_ASCAST]], i64 3, i32 0)
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
Expand Down Expand Up @@ -1070,7 +1070,7 @@ int main() {
// CHECK-NEXT: store ptr [[A]], ptr [[TMP28]], align 8
// CHECK-NEXT: [[TMP29:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP30:%.*]] = load i32, ptr [[TMP29]], align 4
// CHECK-NEXT: call void @__kmpc_parallel_51(ptr addrspacecast (ptr addrspace(1) @[[GLOB1]] to ptr), i32 [[TMP30]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo4v_l76_omp_outlined_omp_outlined, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo4v_l76_omp_outlined_omp_outlined_wrapper, ptr [[CAPTURED_VARS_ADDRS_ASCAST]], i64 3)
// CHECK-NEXT: call void @__kmpc_parallel_60(ptr addrspacecast (ptr addrspace(1) @[[GLOB1]] to ptr), i32 [[TMP30]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo4v_l76_omp_outlined_omp_outlined, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z4foo4v_l76_omp_outlined_omp_outlined_wrapper, ptr [[CAPTURED_VARS_ADDRS_ASCAST]], i64 3, i32 0)
// CHECK-NEXT: store i32 0, ptr [[J_ASCAST]], align 4
// CHECK-NEXT: br label [[FOR_COND:%.*]]
// CHECK: for.cond:
Expand Down
2 changes: 1 addition & 1 deletion clang/test/OpenMP/amdgpu_target_with_aligned_attribute.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -158,7 +158,7 @@ void write_to_aligned_array(int *a, int N) {
// CHECK-AMD-NEXT: store ptr [[TMP19]], ptr [[TMP26]], align 8
// CHECK-AMD-NEXT: [[TMP27:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR_ASCAST]], align 8
// CHECK-AMD-NEXT: [[TMP28:%.*]] = load i32, ptr [[TMP27]], align 4
// CHECK-AMD-NEXT: call void @__kmpc_parallel_51(ptr addrspacecast (ptr addrspace(1) @[[GLOB1]] to ptr), i32 [[TMP28]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_write_to_aligned_array_l14_omp_outlined_omp_outlined, ptr null, ptr [[CAPTURED_VARS_ADDRS_ASCAST]], i64 4)
// CHECK-AMD-NEXT: call void @__kmpc_parallel_60(ptr addrspacecast (ptr addrspace(1) @[[GLOB1]] to ptr), i32 [[TMP28]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}_write_to_aligned_array_l14_omp_outlined_omp_outlined, ptr null, ptr [[CAPTURED_VARS_ADDRS_ASCAST]], i64 4, i32 0)
// CHECK-AMD-NEXT: br label [[OMP_INNER_FOR_INC:%.*]]
// CHECK-AMD: omp.inner.for.inc:
// CHECK-AMD-NEXT: [[TMP29:%.*]] = load i32, ptr [[DOTOMP_IV_ASCAST]], align 4
Expand Down
2 changes: 1 addition & 1 deletion clang/test/OpenMP/declare_target_codegen_globalization.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,7 @@ int maini1() {
// CHECK1-NEXT: [[TMP2:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1:[0-9]+]])
// CHECK1-NEXT: [[TMP3:%.*]] = getelementptr inbounds [1 x ptr], ptr [[CAPTURED_VARS_ADDRS]], i64 0, i64 0
// CHECK1-NEXT: store ptr [[TMP0]], ptr [[TMP3]], align 8
// CHECK1-NEXT: call void @__kmpc_parallel_51(ptr @[[GLOB1]], i32 [[TMP2]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z6maini1v_l16_omp_outlined, ptr null, ptr [[CAPTURED_VARS_ADDRS]], i64 1)
// CHECK1-NEXT: call void @__kmpc_parallel_60(ptr @[[GLOB1]], i32 [[TMP2]], i32 1, i32 -1, i32 -1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z6maini1v_l16_omp_outlined, ptr null, ptr [[CAPTURED_VARS_ADDRS]], i64 1, i32 0)
// CHECK1-NEXT: call void @__kmpc_target_deinit()
// CHECK1-NEXT: ret void
// CHECK1: worker.exit:
Expand Down
2 changes: 1 addition & 1 deletion clang/test/OpenMP/metadirective_device_arch_codegen.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,7 @@ int metadirective1() {
// CHECK: %{{[0-9]}} = call{{.*}} i32 @__kmpc_target_init
// CHECK: user_code.entry:
// CHECK: call{{.*}} void @[[METADIRECTIVE]]_omp_outlined
// CHECK-NOT: call{{.*}} void @__kmpc_parallel_51
// CHECK-NOT: call{{.*}} void @__kmpc_parallel_60
// CHECK: ret void


Expand Down
4 changes: 2 additions & 2 deletions clang/test/OpenMP/metadirective_device_isa_codegen_amdgcn.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,7 +24,7 @@ int amdgcn_device_isa_selected() {

// CHECK: define weak_odr protected amdgpu_kernel void @__omp_offloading_{{.*}}amdgcn_device_isa_selected
// CHECK: user_code.entry:
// CHECK: call void @__kmpc_parallel_51
// CHECK: call void @__kmpc_parallel_60
// CHECK-NOT: call i32 @__kmpc_single
// CHECK: ret void

Expand All @@ -47,7 +47,7 @@ int amdgcn_device_isa_not_selected() {
// CHECK: define weak_odr protected amdgpu_kernel void @__omp_offloading_{{.*}}amdgcn_device_isa_not_selected
// CHECK: user_code.entry:
// CHECK: call i32 @__kmpc_single
// CHECK-NOT: call void @__kmpc_parallel_51
// CHECK-NOT: call void @__kmpc_parallel_60
// CHECK: ret void

#endif
Loading
Loading