diff --git a/clang/lib/Sema/SemaOpenMP.cpp b/clang/lib/Sema/SemaOpenMP.cpp --- a/clang/lib/Sema/SemaOpenMP.cpp +++ b/clang/lib/Sema/SemaOpenMP.cpp @@ -6634,6 +6634,7 @@ case OMPD_target_teams_loop: Res = ActOnOpenMPTargetTeamsGenericLoopDirective( ClausesWithImplicit, AStmt, StartLoc, EndLoc, VarsWithInheritedDSA); + AllowedNameModifiers.push_back(OMPD_target); break; case OMPD_parallel_loop: Res = ActOnOpenMPParallelGenericLoopDirective( diff --git a/clang/test/OpenMP/target_teams_generic_loop_if_codegen.cpp b/clang/test/OpenMP/target_teams_generic_loop_if_codegen.cpp --- a/clang/test/OpenMP/target_teams_generic_loop_if_codegen.cpp +++ b/clang/test/OpenMP/target_teams_generic_loop_if_codegen.cpp @@ -48,7 +48,7 @@ #pragma omp target teams loop for(int i = 0 ; i < 100; i++) {} -#pragma omp target teams loop if (parallel: false) +#pragma omp target teams loop if (target: false) for(int i = 0 ; i < 100; i++) { gtid_test(); } @@ -65,7 +65,7 @@ for(int i = 0 ; i < 100; i++) { fn2(); } -#pragma omp target teams loop if (parallel: Arg) +#pragma omp target teams loop if (target: Arg) for(int i = 0 ; i < 100; i++) { fn3(); } @@ -110,8 +110,6 @@ // CHECK1-NEXT: entry: // CHECK1-NEXT: [[TMP:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[KERNEL_ARGS:%.*]] = alloca [[STRUCT___TGT_KERNEL_ARGUMENTS:%.*]], align 8 -// CHECK1-NEXT: [[_TMP1:%.*]] = alloca i32, align 4 -// CHECK1-NEXT: [[KERNEL_ARGS2:%.*]] = alloca [[STRUCT___TGT_KERNEL_ARGUMENTS]], align 8 // CHECK1-NEXT: [[TMP0:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 0 // CHECK1-NEXT: store i32 2, ptr [[TMP0]], align 4 // CHECK1-NEXT: [[TMP1:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 1 @@ -145,39 +143,7 @@ // CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9gtid_testv_l48() #[[ATTR2:[0-9]+]] // CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT]] // CHECK1: omp_offload.cont: -// CHECK1-NEXT: [[TMP15:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 0 -// CHECK1-NEXT: store i32 2, ptr [[TMP15]], align 4 -// CHECK1-NEXT: [[TMP16:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 1 -// CHECK1-NEXT: store i32 0, ptr [[TMP16]], align 4 -// CHECK1-NEXT: [[TMP17:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 2 -// CHECK1-NEXT: store ptr null, ptr [[TMP17]], align 8 -// CHECK1-NEXT: [[TMP18:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 3 -// CHECK1-NEXT: store ptr null, ptr [[TMP18]], align 8 -// CHECK1-NEXT: [[TMP19:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 4 -// CHECK1-NEXT: store ptr null, ptr [[TMP19]], align 8 -// CHECK1-NEXT: [[TMP20:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 5 -// CHECK1-NEXT: store ptr null, ptr [[TMP20]], align 8 -// CHECK1-NEXT: [[TMP21:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 6 -// CHECK1-NEXT: store ptr null, ptr [[TMP21]], align 8 -// CHECK1-NEXT: [[TMP22:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 7 -// CHECK1-NEXT: store ptr null, ptr [[TMP22]], align 8 -// CHECK1-NEXT: [[TMP23:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 8 -// CHECK1-NEXT: store i64 100, ptr [[TMP23]], align 8 -// CHECK1-NEXT: [[TMP24:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 9 -// CHECK1-NEXT: store i64 0, ptr [[TMP24]], align 8 -// CHECK1-NEXT: [[TMP25:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 10 -// CHECK1-NEXT: store [3 x i32] zeroinitializer, ptr [[TMP25]], align 4 -// CHECK1-NEXT: [[TMP26:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 11 -// CHECK1-NEXT: store [3 x i32] [i32 1, i32 0, i32 0], ptr [[TMP26]], align 4 -// CHECK1-NEXT: [[TMP27:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 12 -// CHECK1-NEXT: store i32 0, ptr [[TMP27]], align 4 -// CHECK1-NEXT: [[TMP28:%.*]] = call i32 @__tgt_target_kernel(ptr @[[GLOB3]], i64 -1, i32 0, i32 1, ptr @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9gtid_testv_l51.region_id, ptr [[KERNEL_ARGS2]]) -// CHECK1-NEXT: [[TMP29:%.*]] = icmp ne i32 [[TMP28]], 0 -// CHECK1-NEXT: br i1 [[TMP29]], label [[OMP_OFFLOAD_FAILED3:%.*]], label [[OMP_OFFLOAD_CONT4:%.*]] -// CHECK1: omp_offload.failed3: // CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9gtid_testv_l51() #[[ATTR2]] -// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT4]] -// CHECK1: omp_offload.cont4: // CHECK1-NEXT: ret void // // @@ -337,7 +303,6 @@ // CHECK1-NEXT: [[DOTOMP_STRIDE:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[DOTOMP_IS_LAST:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[I:%.*]] = alloca i32, align 4 -// CHECK1-NEXT: [[DOTBOUND_ZERO_ADDR:%.*]] = alloca i32, align 4 // CHECK1-NEXT: store ptr [[DOTGLOBAL_TID_]], ptr [[DOTGLOBAL_TID__ADDR]], align 8 // CHECK1-NEXT: store ptr [[DOTBOUND_TID_]], ptr [[DOTBOUND_TID__ADDR]], align 8 // CHECK1-NEXT: store i32 0, ptr [[DOTOMP_COMB_LB]], align 4 @@ -371,16 +336,12 @@ // CHECK1-NEXT: [[TMP8:%.*]] = zext i32 [[TMP7]] to i64 // CHECK1-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_COMB_UB]], align 4 // CHECK1-NEXT: [[TMP10:%.*]] = zext i32 [[TMP9]] to i64 -// CHECK1-NEXT: call void @__kmpc_serialized_parallel(ptr @[[GLOB3]], i32 [[TMP1]]) -// CHECK1-NEXT: [[TMP11:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR]], align 8 -// CHECK1-NEXT: store i32 0, ptr [[DOTBOUND_ZERO_ADDR]], align 4 -// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9gtid_testv_l51.omp_outlined.omp_outlined(ptr [[TMP11]], ptr [[DOTBOUND_ZERO_ADDR]], i64 [[TMP8]], i64 [[TMP10]]) #[[ATTR2]] -// CHECK1-NEXT: call void @__kmpc_end_serialized_parallel(ptr @[[GLOB3]], i32 [[TMP1]]) +// CHECK1-NEXT: call void (ptr, i32, ptr, ...) @__kmpc_fork_call(ptr @[[GLOB3]], i32 2, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z9gtid_testv_l51.omp_outlined.omp_outlined, i64 [[TMP8]], i64 [[TMP10]]) // CHECK1-NEXT: br label [[OMP_INNER_FOR_INC:%.*]] // CHECK1: omp.inner.for.inc: -// CHECK1-NEXT: [[TMP12:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4 -// CHECK1-NEXT: [[TMP13:%.*]] = load i32, ptr [[DOTOMP_STRIDE]], align 4 -// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP12]], [[TMP13]] +// CHECK1-NEXT: [[TMP11:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4 +// CHECK1-NEXT: [[TMP12:%.*]] = load i32, ptr [[DOTOMP_STRIDE]], align 4 +// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP11]], [[TMP12]] // CHECK1-NEXT: store i32 [[ADD]], ptr [[DOTOMP_IV]], align 4 // CHECK1-NEXT: br label [[OMP_INNER_FOR_COND]] // CHECK1: omp.inner.for.end: @@ -1026,13 +987,8 @@ // CHECK1-NEXT: [[ARG_ADDR:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[TMP:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[KERNEL_ARGS:%.*]] = alloca [[STRUCT___TGT_KERNEL_ARGUMENTS:%.*]], align 8 -// CHECK1-NEXT: [[DOTCAPTURE_EXPR_:%.*]] = alloca i8, align 1 -// CHECK1-NEXT: [[DOTCAPTURE_EXPR__CASTED:%.*]] = alloca i64, align 8 -// CHECK1-NEXT: [[DOTOFFLOAD_BASEPTRS:%.*]] = alloca [1 x ptr], align 8 -// CHECK1-NEXT: [[DOTOFFLOAD_PTRS:%.*]] = alloca [1 x ptr], align 8 -// CHECK1-NEXT: [[DOTOFFLOAD_MAPPERS:%.*]] = alloca [1 x ptr], align 8 -// CHECK1-NEXT: [[_TMP4:%.*]] = alloca i32, align 4 -// CHECK1-NEXT: [[KERNEL_ARGS5:%.*]] = alloca [[STRUCT___TGT_KERNEL_ARGUMENTS]], align 8 +// CHECK1-NEXT: [[_TMP1:%.*]] = alloca i32, align 4 +// CHECK1-NEXT: [[KERNEL_ARGS2:%.*]] = alloca [[STRUCT___TGT_KERNEL_ARGUMENTS]], align 8 // CHECK1-NEXT: store i32 [[ARG]], ptr [[ARG_ADDR]], align 4 // CHECK1-NEXT: [[TMP0:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS]], i32 0, i32 0 // CHECK1-NEXT: store i32 2, ptr [[TMP0]], align 4 @@ -1070,58 +1026,46 @@ // CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l64() #[[ATTR2]] // CHECK1-NEXT: [[TMP15:%.*]] = load i32, ptr [[ARG_ADDR]], align 4 // CHECK1-NEXT: [[TOBOOL:%.*]] = icmp ne i32 [[TMP15]], 0 -// CHECK1-NEXT: [[FROMBOOL:%.*]] = zext i1 [[TOBOOL]] to i8 -// CHECK1-NEXT: store i8 [[FROMBOOL]], ptr [[DOTCAPTURE_EXPR_]], align 1 -// CHECK1-NEXT: [[TMP16:%.*]] = load i8, ptr [[DOTCAPTURE_EXPR_]], align 1 -// CHECK1-NEXT: [[TOBOOL1:%.*]] = trunc i8 [[TMP16]] to i1 -// CHECK1-NEXT: [[FROMBOOL2:%.*]] = zext i1 [[TOBOOL1]] to i8 -// CHECK1-NEXT: store i8 [[FROMBOOL2]], ptr [[DOTCAPTURE_EXPR__CASTED]], align 1 -// CHECK1-NEXT: [[TMP17:%.*]] = load i64, ptr [[DOTCAPTURE_EXPR__CASTED]], align 8 -// CHECK1-NEXT: [[TMP18:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0 -// CHECK1-NEXT: store i64 [[TMP17]], ptr [[TMP18]], align 8 -// CHECK1-NEXT: [[TMP19:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_PTRS]], i32 0, i32 0 -// CHECK1-NEXT: store i64 [[TMP17]], ptr [[TMP19]], align 8 -// CHECK1-NEXT: [[TMP20:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_MAPPERS]], i64 0, i64 0 +// CHECK1-NEXT: br i1 [[TOBOOL]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]] +// CHECK1: omp_if.then: +// CHECK1-NEXT: [[TMP16:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 0 +// CHECK1-NEXT: store i32 2, ptr [[TMP16]], align 4 +// CHECK1-NEXT: [[TMP17:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 1 +// CHECK1-NEXT: store i32 0, ptr [[TMP17]], align 4 +// CHECK1-NEXT: [[TMP18:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 2 +// CHECK1-NEXT: store ptr null, ptr [[TMP18]], align 8 +// CHECK1-NEXT: [[TMP19:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 3 +// CHECK1-NEXT: store ptr null, ptr [[TMP19]], align 8 +// CHECK1-NEXT: [[TMP20:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 4 // CHECK1-NEXT: store ptr null, ptr [[TMP20]], align 8 -// CHECK1-NEXT: [[TMP21:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_BASEPTRS]], i32 0, i32 0 -// CHECK1-NEXT: [[TMP22:%.*]] = getelementptr inbounds [1 x ptr], ptr [[DOTOFFLOAD_PTRS]], i32 0, i32 0 -// CHECK1-NEXT: [[TMP23:%.*]] = load i8, ptr [[DOTCAPTURE_EXPR_]], align 1 -// CHECK1-NEXT: [[TOBOOL3:%.*]] = trunc i8 [[TMP23]] to i1 -// CHECK1-NEXT: [[TMP24:%.*]] = select i1 [[TOBOOL3]], i32 0, i32 1 -// CHECK1-NEXT: [[TMP25:%.*]] = insertvalue [3 x i32] zeroinitializer, i32 [[TMP24]], 0 -// CHECK1-NEXT: [[TMP26:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 0 -// CHECK1-NEXT: store i32 2, ptr [[TMP26]], align 4 -// CHECK1-NEXT: [[TMP27:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 1 -// CHECK1-NEXT: store i32 1, ptr [[TMP27]], align 4 -// CHECK1-NEXT: [[TMP28:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 2 -// CHECK1-NEXT: store ptr [[TMP21]], ptr [[TMP28]], align 8 -// CHECK1-NEXT: [[TMP29:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 3 -// CHECK1-NEXT: store ptr [[TMP22]], ptr [[TMP29]], align 8 -// CHECK1-NEXT: [[TMP30:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 4 -// CHECK1-NEXT: store ptr @.offload_sizes.1, ptr [[TMP30]], align 8 -// CHECK1-NEXT: [[TMP31:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 5 -// CHECK1-NEXT: store ptr @.offload_maptypes.2, ptr [[TMP31]], align 8 -// CHECK1-NEXT: [[TMP32:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 6 -// CHECK1-NEXT: store ptr null, ptr [[TMP32]], align 8 -// CHECK1-NEXT: [[TMP33:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 7 -// CHECK1-NEXT: store ptr null, ptr [[TMP33]], align 8 -// CHECK1-NEXT: [[TMP34:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 8 -// CHECK1-NEXT: store i64 100, ptr [[TMP34]], align 8 -// CHECK1-NEXT: [[TMP35:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 9 -// CHECK1-NEXT: store i64 0, ptr [[TMP35]], align 8 -// CHECK1-NEXT: [[TMP36:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 10 -// CHECK1-NEXT: store [3 x i32] zeroinitializer, ptr [[TMP36]], align 4 -// CHECK1-NEXT: [[TMP37:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 11 -// CHECK1-NEXT: store [3 x i32] [[TMP25]], ptr [[TMP37]], align 4 -// CHECK1-NEXT: [[TMP38:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS5]], i32 0, i32 12 -// CHECK1-NEXT: store i32 0, ptr [[TMP38]], align 4 -// CHECK1-NEXT: [[TMP39:%.*]] = call i32 @__tgt_target_kernel(ptr @[[GLOB3]], i64 -1, i32 0, i32 [[TMP24]], ptr @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68.region_id, ptr [[KERNEL_ARGS5]]) -// CHECK1-NEXT: [[TMP40:%.*]] = icmp ne i32 [[TMP39]], 0 -// CHECK1-NEXT: br i1 [[TMP40]], label [[OMP_OFFLOAD_FAILED6:%.*]], label [[OMP_OFFLOAD_CONT7:%.*]] -// CHECK1: omp_offload.failed6: -// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68(i64 [[TMP17]]) #[[ATTR2]] -// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT7]] -// CHECK1: omp_offload.cont7: +// CHECK1-NEXT: [[TMP21:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 5 +// CHECK1-NEXT: store ptr null, ptr [[TMP21]], align 8 +// CHECK1-NEXT: [[TMP22:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 6 +// CHECK1-NEXT: store ptr null, ptr [[TMP22]], align 8 +// CHECK1-NEXT: [[TMP23:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 7 +// CHECK1-NEXT: store ptr null, ptr [[TMP23]], align 8 +// CHECK1-NEXT: [[TMP24:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 8 +// CHECK1-NEXT: store i64 100, ptr [[TMP24]], align 8 +// CHECK1-NEXT: [[TMP25:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 9 +// CHECK1-NEXT: store i64 0, ptr [[TMP25]], align 8 +// CHECK1-NEXT: [[TMP26:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 10 +// CHECK1-NEXT: store [3 x i32] zeroinitializer, ptr [[TMP26]], align 4 +// CHECK1-NEXT: [[TMP27:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 11 +// CHECK1-NEXT: store [3 x i32] zeroinitializer, ptr [[TMP27]], align 4 +// CHECK1-NEXT: [[TMP28:%.*]] = getelementptr inbounds [[STRUCT___TGT_KERNEL_ARGUMENTS]], ptr [[KERNEL_ARGS2]], i32 0, i32 12 +// CHECK1-NEXT: store i32 0, ptr [[TMP28]], align 4 +// CHECK1-NEXT: [[TMP29:%.*]] = call i32 @__tgt_target_kernel(ptr @[[GLOB3]], i64 -1, i32 0, i32 0, ptr @.{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68.region_id, ptr [[KERNEL_ARGS2]]) +// CHECK1-NEXT: [[TMP30:%.*]] = icmp ne i32 [[TMP29]], 0 +// CHECK1-NEXT: br i1 [[TMP30]], label [[OMP_OFFLOAD_FAILED3:%.*]], label [[OMP_OFFLOAD_CONT4:%.*]] +// CHECK1: omp_offload.failed3: +// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68() #[[ATTR2]] +// CHECK1-NEXT: br label [[OMP_OFFLOAD_CONT4]] +// CHECK1: omp_offload.cont4: +// CHECK1-NEXT: br label [[OMP_IF_END:%.*]] +// CHECK1: omp_if.else: +// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68() #[[ATTR2]] +// CHECK1-NEXT: br label [[OMP_IF_END]] +// CHECK1: omp_if.end: // CHECK1-NEXT: ret i32 0 // // @@ -1407,26 +1351,17 @@ // // // CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68 -// CHECK1-SAME: (i64 noundef [[DOTCAPTURE_EXPR_:%.*]]) #[[ATTR1]] { +// CHECK1-SAME: () #[[ATTR1]] { // CHECK1-NEXT: entry: -// CHECK1-NEXT: [[DOTCAPTURE_EXPR__ADDR:%.*]] = alloca i64, align 8 -// CHECK1-NEXT: [[DOTCAPTURE_EXPR__CASTED:%.*]] = alloca i64, align 8 -// CHECK1-NEXT: store i64 [[DOTCAPTURE_EXPR_]], ptr [[DOTCAPTURE_EXPR__ADDR]], align 8 -// CHECK1-NEXT: [[TMP0:%.*]] = load i8, ptr [[DOTCAPTURE_EXPR__ADDR]], align 1 -// CHECK1-NEXT: [[TOBOOL:%.*]] = trunc i8 [[TMP0]] to i1 -// CHECK1-NEXT: [[FROMBOOL:%.*]] = zext i1 [[TOBOOL]] to i8 -// CHECK1-NEXT: store i8 [[FROMBOOL]], ptr [[DOTCAPTURE_EXPR__CASTED]], align 1 -// CHECK1-NEXT: [[TMP1:%.*]] = load i64, ptr [[DOTCAPTURE_EXPR__CASTED]], align 8 -// CHECK1-NEXT: call void (ptr, i32, ptr, ...) @__kmpc_fork_teams(ptr @[[GLOB3]], i32 1, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68.omp_outlined, i64 [[TMP1]]) +// CHECK1-NEXT: call void (ptr, i32, ptr, ...) @__kmpc_fork_teams(ptr @[[GLOB3]], i32 0, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68.omp_outlined) // CHECK1-NEXT: ret void // // // CHECK1-LABEL: define {{[^@]+}}@{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68.omp_outlined -// CHECK1-SAME: (ptr noalias noundef [[DOTGLOBAL_TID_:%.*]], ptr noalias noundef [[DOTBOUND_TID_:%.*]], i64 noundef [[DOTCAPTURE_EXPR_:%.*]]) #[[ATTR1]] { +// CHECK1-SAME: (ptr noalias noundef [[DOTGLOBAL_TID_:%.*]], ptr noalias noundef [[DOTBOUND_TID_:%.*]]) #[[ATTR1]] { // CHECK1-NEXT: entry: // CHECK1-NEXT: [[DOTGLOBAL_TID__ADDR:%.*]] = alloca ptr, align 8 // CHECK1-NEXT: [[DOTBOUND_TID__ADDR:%.*]] = alloca ptr, align 8 -// CHECK1-NEXT: [[DOTCAPTURE_EXPR__ADDR:%.*]] = alloca i64, align 8 // CHECK1-NEXT: [[DOTOMP_IV:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[TMP:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[DOTOMP_COMB_LB:%.*]] = alloca i32, align 4 @@ -1434,10 +1369,8 @@ // CHECK1-NEXT: [[DOTOMP_STRIDE:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[DOTOMP_IS_LAST:%.*]] = alloca i32, align 4 // CHECK1-NEXT: [[I:%.*]] = alloca i32, align 4 -// CHECK1-NEXT: [[DOTBOUND_ZERO_ADDR:%.*]] = alloca i32, align 4 // CHECK1-NEXT: store ptr [[DOTGLOBAL_TID_]], ptr [[DOTGLOBAL_TID__ADDR]], align 8 // CHECK1-NEXT: store ptr [[DOTBOUND_TID_]], ptr [[DOTBOUND_TID__ADDR]], align 8 -// CHECK1-NEXT: store i64 [[DOTCAPTURE_EXPR_]], ptr [[DOTCAPTURE_EXPR__ADDR]], align 8 // CHECK1-NEXT: store i32 0, ptr [[DOTOMP_COMB_LB]], align 4 // CHECK1-NEXT: store i32 99, ptr [[DOTOMP_COMB_UB]], align 4 // CHECK1-NEXT: store i32 1, ptr [[DOTOMP_STRIDE]], align 4 @@ -1469,25 +1402,12 @@ // CHECK1-NEXT: [[TMP8:%.*]] = zext i32 [[TMP7]] to i64 // CHECK1-NEXT: [[TMP9:%.*]] = load i32, ptr [[DOTOMP_COMB_UB]], align 4 // CHECK1-NEXT: [[TMP10:%.*]] = zext i32 [[TMP9]] to i64 -// CHECK1-NEXT: [[TMP11:%.*]] = load i8, ptr [[DOTCAPTURE_EXPR__ADDR]], align 1 -// CHECK1-NEXT: [[TOBOOL:%.*]] = trunc i8 [[TMP11]] to i1 -// CHECK1-NEXT: br i1 [[TOBOOL]], label [[OMP_IF_THEN:%.*]], label [[OMP_IF_ELSE:%.*]] -// CHECK1: omp_if.then: // CHECK1-NEXT: call void (ptr, i32, ptr, ...) @__kmpc_fork_call(ptr @[[GLOB3]], i32 2, ptr @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68.omp_outlined.omp_outlined, i64 [[TMP8]], i64 [[TMP10]]) -// CHECK1-NEXT: br label [[OMP_IF_END:%.*]] -// CHECK1: omp_if.else: -// CHECK1-NEXT: call void @__kmpc_serialized_parallel(ptr @[[GLOB3]], i32 [[TMP1]]) -// CHECK1-NEXT: [[TMP12:%.*]] = load ptr, ptr [[DOTGLOBAL_TID__ADDR]], align 8 -// CHECK1-NEXT: store i32 0, ptr [[DOTBOUND_ZERO_ADDR]], align 4 -// CHECK1-NEXT: call void @{{__omp_offloading_[0-9a-z]+_[0-9a-z]+}}__Z5tmainIiEiT__l68.omp_outlined.omp_outlined(ptr [[TMP12]], ptr [[DOTBOUND_ZERO_ADDR]], i64 [[TMP8]], i64 [[TMP10]]) #[[ATTR2]] -// CHECK1-NEXT: call void @__kmpc_end_serialized_parallel(ptr @[[GLOB3]], i32 [[TMP1]]) -// CHECK1-NEXT: br label [[OMP_IF_END]] -// CHECK1: omp_if.end: // CHECK1-NEXT: br label [[OMP_INNER_FOR_INC:%.*]] // CHECK1: omp.inner.for.inc: -// CHECK1-NEXT: [[TMP13:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4 -// CHECK1-NEXT: [[TMP14:%.*]] = load i32, ptr [[DOTOMP_STRIDE]], align 4 -// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP13]], [[TMP14]] +// CHECK1-NEXT: [[TMP11:%.*]] = load i32, ptr [[DOTOMP_IV]], align 4 +// CHECK1-NEXT: [[TMP12:%.*]] = load i32, ptr [[DOTOMP_STRIDE]], align 4 +// CHECK1-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP11]], [[TMP12]] // CHECK1-NEXT: store i32 [[ADD]], ptr [[DOTOMP_IV]], align 4 // CHECK1-NEXT: br label [[OMP_INNER_FOR_COND]] // CHECK1: omp.inner.for.end: diff --git a/clang/test/OpenMP/target_teams_generic_loop_messages.cpp b/clang/test/OpenMP/target_teams_generic_loop_messages.cpp --- a/clang/test/OpenMP/target_teams_generic_loop_messages.cpp +++ b/clang/test/OpenMP/target_teams_generic_loop_messages.cpp @@ -127,6 +127,16 @@ for (i=0; i<1000; ++i) z = i+11; + // expected-error@+1 {{directive name modifier 'teams' is not allowed for '#pragma omp target teams loop'}} + #pragma omp target teams loop if(teams:1) + for (i=0; i<1000; ++i) + z = i+11; + + // expected-error@+1 {{directive name modifier 'parallel' is not allowed for '#pragma omp target teams loop'}} + #pragma omp target teams loop if(parallel:0) + for (i=0; i<1000; ++i) + z = i+11; + } template