1c298c9274
These transforms all perform a variant of (gep (gep p, x), y) to (gep p, (x + y)). We can preserve both inbounds and nuw during such transforms (https://alive2.llvm.org/ce/z/Stu4cN), but not nusw, which would require proving that the new add is nsw. For the constant offset case, I've conservatively retained the logic that checks for negative intermediate offsets, though I'm not sure it's still reachable nowadays.
56 lines
3.2 KiB
C++
56 lines
3.2 KiB
C++
// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --function-signature --include-generated-funcs --prefix-filecheck-ir-name _
|
|
// RUN: %clang_cc1 -fopenmp -O1 -x c++ -triple x86_64-apple-darwin10 -emit-llvm %s -o - | FileCheck %s --check-prefix=CHECK
|
|
|
|
template <class Function, class... Args>
|
|
void run_task(Function function, Args... args) {
|
|
#pragma omp task untied
|
|
{ function(args...); }
|
|
}
|
|
|
|
void bar(int a, float b);
|
|
|
|
void foo() {
|
|
int a;
|
|
float b;
|
|
run_task(bar, a, b);
|
|
}
|
|
// CHECK-LABEL: define {{[^@]+}}@_Z3foov
|
|
// CHECK-SAME: () local_unnamed_addr #[[ATTR0:[0-9]+]] {
|
|
// CHECK-NEXT: entry:
|
|
// CHECK-NEXT: [[TMP0:%.*]] = tail call i32 @__kmpc_global_thread_num(ptr nonnull @[[GLOB1:[0-9]+]])
|
|
// CHECK-NEXT: [[TMP1:%.*]] = tail call ptr @__kmpc_omp_task_alloc(ptr nonnull @[[GLOB1]], i32 [[TMP0]], i32 0, i64 56, i64 1, ptr nonnull @.omp_task_entry.)
|
|
// CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP1]], i64 40
|
|
// CHECK-NEXT: store ptr @_Z3barif, ptr [[TMP2]], align 8, !tbaa [[TBAA3:![0-9]+]]
|
|
// CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP1]], i64 16
|
|
// CHECK-NEXT: store i32 0, ptr [[TMP3]], align 8, !tbaa [[TBAA12:![0-9]+]]
|
|
// CHECK-NEXT: [[TMP4:%.*]] = tail call i32 @__kmpc_omp_task(ptr nonnull @[[GLOB1]], i32 [[TMP0]], ptr [[TMP1]])
|
|
// CHECK-NEXT: ret void
|
|
//
|
|
//
|
|
// CHECK-LABEL: define {{[^@]+}}@.omp_task_entry.
|
|
// CHECK-SAME: (i32 noundef [[TMP0:%.*]], ptr noalias noundef [[TMP1:%.*]]) #[[ATTR3:[0-9]+]] {
|
|
// CHECK-NEXT: entry:
|
|
// CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP1]], i64 16
|
|
// CHECK-NEXT: tail call void @llvm.experimental.noalias.scope.decl(metadata [[META13:![0-9]+]])
|
|
// CHECK-NEXT: [[TMP3:%.*]] = load i32, ptr [[TMP2]], align 4, !tbaa [[TBAA16:![0-9]+]], !alias.scope [[META13]], !noalias [[META17:![0-9]+]]
|
|
// CHECK-NEXT: switch i32 [[TMP3]], label [[DOTOMP_OUTLINED__EXIT:%.*]] [
|
|
// CHECK-NEXT: i32 0, label [[DOTUNTIED_JMP__I:%.*]]
|
|
// CHECK-NEXT: i32 1, label [[DOTUNTIED_NEXT__I:%.*]]
|
|
// CHECK-NEXT: ]
|
|
// CHECK: .untied.jmp..i:
|
|
// CHECK-NEXT: store i32 1, ptr [[TMP2]], align 4, !tbaa [[TBAA16]], !alias.scope [[META13]], !noalias [[META17]]
|
|
// CHECK-NEXT: [[TMP4:%.*]] = tail call i32 @__kmpc_omp_task(ptr nonnull @[[GLOB1]], i32 [[TMP0]], ptr [[TMP1]]), !noalias [[META13]]
|
|
// CHECK-NEXT: br label [[DOTOMP_OUTLINED__EXIT]]
|
|
// CHECK: .untied.next..i:
|
|
// CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP1]], i64 40
|
|
// CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP1]], i64 52
|
|
// CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds nuw i8, ptr [[TMP1]], i64 48
|
|
// CHECK-NEXT: [[TMP8:%.*]] = load ptr, ptr [[TMP5]], align 8, !tbaa [[TBAA19:![0-9]+]], !noalias [[META13]]
|
|
// CHECK-NEXT: [[TMP9:%.*]] = load i32, ptr [[TMP7]], align 4, !tbaa [[TBAA16]], !noalias [[META13]]
|
|
// CHECK-NEXT: [[TMP10:%.*]] = load float, ptr [[TMP6]], align 4, !tbaa [[TBAA20:![0-9]+]], !noalias [[META13]]
|
|
// CHECK-NEXT: tail call void [[TMP8]](i32 noundef [[TMP9]], float noundef [[TMP10]]) #[[ATTR2:[0-9]+]], !noalias [[META13]]
|
|
// CHECK-NEXT: br label [[DOTOMP_OUTLINED__EXIT]]
|
|
// CHECK: .omp_outlined..exit:
|
|
// CHECK-NEXT: ret i32 0
|
|
//
|