diff options
Diffstat (limited to 'test/OpenMP/for_reduction_codegen.cpp')
| -rw-r--r-- | test/OpenMP/for_reduction_codegen.cpp | 336 |
1 files changed, 286 insertions, 50 deletions
diff --git a/test/OpenMP/for_reduction_codegen.cpp b/test/OpenMP/for_reduction_codegen.cpp index d5afae990bf4..cef457333062 100644 --- a/test/OpenMP/for_reduction_codegen.cpp +++ b/test/OpenMP/for_reduction_codegen.cpp @@ -27,7 +27,7 @@ struct S { // CHECK-DAG: [[REDUCTION_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 18, i32 0, i32 0, i8* // CHECK-DAG: [[REDUCTION_LOCK:@.+]] = common global [8 x i32] zeroinitializer -template <typename T> +template <typename T, int length> T tmain() { T t; S<T> test; @@ -36,6 +36,7 @@ T tmain() { S<T> s_arr[] = {1, 2}; S<T> &var = test; S<T> var1; + S<T> arr[length]; #pragma omp parallel #pragma omp for reduction(+:t_var) reduction(&:var) reduction(&& : var1) reduction(min: t_var1) nowait for (int i = 0; i < 2; ++i) { @@ -48,6 +49,12 @@ T tmain() { vec[i] = t_var; s_arr[i] = var; } +#pragma omp parallel +#pragma omp for reduction(+ : arr[1:length-2]) + for (int i = 0; i < 2; ++i) { + vec[i] = t_var; + s_arr[i] = var; + } return T(); } @@ -180,12 +187,12 @@ int main() { S<float> test; float t_var = 0, t_var1; int vec[] = {1, 2}; - S<float> s_arr[] = {1, 2}; + S<float> s_arr[] = {1, 2, 3, 4}; S<float> &var = test; S<float> var1, arrs[10][4]; S<float> **var2 = foo(); - S<float> vvar2[2]; - S<float> (&var3)[2] = s_arr; + S<float> vvar2[5]; + S<float> (&var3)[4] = s_arr; #pragma omp parallel #pragma omp for reduction(+:t_var) reduction(&:var) reduction(&& : var1) reduction(min: t_var1) for (int i = 0; i < 2; ++i) { @@ -200,11 +207,28 @@ int main() { #pragma omp for reduction(+:arr) reduction(&:arrs) for (int i = 0; i < 10; ++i) ++arr[1][i]; + // arr is a VLA, but the array section has constant length so we can generate a constant sized array! +#pragma omp parallel +#pragma omp for reduction(+:arr[1][0:2]) + for (int i = 0; i < 10; ++i) + ++arr[1][i]; #pragma omp parallel #pragma omp for reduction(& : var2[0 : 5][1 : 6]) for (int i = 0; i < 10; ++i) ; #pragma omp parallel +#pragma omp for reduction(& : var2[1][1 : 6]) + for (int i = 0; i < 10; ++i) + ; +#pragma omp parallel +#pragma omp for reduction(& : var2[1 : 1][1 : 6]) + for (int i = 0; i < 10; ++i) + ; +#pragma omp parallel +#pragma omp for reduction(& : var2[1 : 1][1]) + for (int i = 0; i < 10; ++i) + ; +#pragma omp parallel #pragma omp for reduction(& : vvar2[0 : 5]) for (int i = 0; i < 10; ++i) ; @@ -213,28 +237,43 @@ int main() { for (int i = 0; i < 10; ++i) ; #pragma omp parallel +#pragma omp for reduction(& : var3[ : 2]) + for (int i = 0; i < 10; ++i) + ; + // TODO: The compiler should also be able to generate a constant sized array in this case! +#pragma omp parallel +#pragma omp for reduction(& : var3[2 : ]) + for (int i = 0; i < 10; ++i) + ; +#pragma omp parallel #pragma omp for reduction(& : var3) for (int i = 0; i < 10; ++i) ; - return tmain<int>(); + return tmain<int, 42>(); #endif } // CHECK: define {{.*}}i{{[0-9]+}} @main() // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]], // CHECK: call {{.*}} [[S_FLOAT_TY_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]]) -// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, float*, [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]*, float*, [2 x i32]*, [2 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, float*, [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]*, float*, [2 x i32]*, [4 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK:@.+]] to void // CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 5, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i64, i64, i32*, [2 x i32]*, [10 x [4 x [[S_FLOAT_TY]]]]*)* [[MAIN_MICROTASK1:@.+]] to void // CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i64, i64, i32*, [10 x [4 x [[S_FLOAT_TY]]]]*)* [[MAIN_MICROTASK2:@.+]] to void -// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[S_FLOAT_TY]]***)* [[MAIN_MICROTASK3:@.+]] to void -// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [2 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK4:@.+]] to void -// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [2 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK5:@.+]] to void -// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [2 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK6:@.+]] to void -// CHECK: = call {{.*}}i{{.+}} [[TMAIN_INT:@.+]]() +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 3, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i64, i64, i32*)* [[MAIN_MICROTASK3:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[S_FLOAT_TY]]***)* [[MAIN_MICROTASK4:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[S_FLOAT_TY]]***)* [[MAIN_MICROTASK5:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[S_FLOAT_TY]]***)* [[MAIN_MICROTASK6:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[S_FLOAT_TY]]***)* [[MAIN_MICROTASK7:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [5 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK8:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [4 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK9:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [4 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK10:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [4 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK11:@.+]] to void +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [4 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK12:@.+]] to void +// CHECK: = call {{.*}}i{{.+}} [[TMAIN_INT_42:@.+]]() // CHECK: call {{.*}} [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]* // CHECK: ret // -// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, float* dereferenceable(4) %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}}, float* dereferenceable(4) %{{.+}}, [2 x i32]* dereferenceable(8) %vec, [2 x [[S_FLOAT_TY]]]* dereferenceable(8) %{{.+}}) +// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, float* dereferenceable(4) %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}}, float* dereferenceable(4) %{{.+}}, [2 x i32]* dereferenceable(8) %vec, [4 x [[S_FLOAT_TY]]]* dereferenceable(16) %{{.+}}) // CHECK: [[T_VAR_PRIV:%.+]] = alloca float, // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]], // CHECK: [[VAR1_PRIV:%.+]] = alloca [[S_FLOAT_TY]], @@ -491,7 +530,7 @@ int main() { // CHECK: store float [[UP]], float* [[T_VAR1_LHS]], // CHECK: ret void -// CHECK: define internal void [[MAIN_MICROTASK1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i64 %{{.+}}, i64 %{{.+}}, i32* %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, [10 x [4 x [[S_FLOAT_TY]]]]* dereferenceable(160) %{{.+}}) +// CHECK: define internal void [[MAIN_MICROTASK1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i64 %{{.+}}, i64 %{{.+}}, i32* {{.+}} %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, [10 x [4 x [[S_FLOAT_TY]]]]* dereferenceable(160) %{{.+}}) // Reduction list for runtime. // CHECK: [[RED_LIST:%.+]] = alloca [4 x i8*], @@ -692,7 +731,7 @@ int main() { // CHECK: ret void -// CHECK: define internal void [[MAIN_MICROTASK2]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i64 %{{.+}}, i64 %{{.+}}, i32* %{{.+}}, [10 x [4 x [[S_FLOAT_TY]]]]* dereferenceable(160) %{{.+}}) +// CHECK: define internal void [[MAIN_MICROTASK2]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i64 %{{.+}}, i64 %{{.+}}, i32* {{.+}} %{{.+}}, [10 x [4 x [[S_FLOAT_TY]]]]* dereferenceable(160) %{{.+}}) // CHECK: [[ARRS_PRIV:%.+]] = alloca [10 x [4 x [[S_FLOAT_TY]]]], @@ -882,7 +921,35 @@ int main() { // CHECK: ret void -// CHECK: define internal void [[MAIN_MICROTASK3]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[S_FLOAT_TY]]*** dereferenceable(8) %{{.+}}) +// CHECK: define internal void [[MAIN_MICROTASK3]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i32* {{.*}} %{{.+}}) + +// CHECK: [[VLA1_ORIG_ADDR:%.+]] = alloca i64 +// CHECK: [[VLA2_ORIG_ADDR:%.+]] = alloca i64 +// CHECK: [[ARR_ORIG_ADDR:%.+]] = alloca i32*, +// CHECK: [[ARR_PRIV:%.+]] = alloca [1 x [2 x i32]], + +// Reduction list for runtime. +// CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], +// CHECK: [[VLA1:%.+]] = load i64, i64* [[VLA1_ORIG_ADDR]] +// CHECK: [[VLA2:%.+]] = load i64, i64* [[VLA2_ORIG_ADDR]] +// CHECK: [[ARR_ORIG:%.+]] = load i32*, i32** [[ARR_ORIG_ADDR]], + +// CHECK: [[LOW_OFFSET:%.+]] = mul nsw i64 1, [[VLA2]] +// CHECK: [[ARRIDX:%.+]] = getelementptr inbounds i32, i32* [[ARR_ORIG]], i64 [[LOW_OFFSET]] +// CHECK: [[LOW:%.+]] = getelementptr inbounds i32, i32* [[ARRIDX]], i64 0 + +// CHECK: [[START:%.+]] = ptrtoint i32* [[ARR_ORIG]] to i64 +// CHECK: [[LOW_BOUND:%.+]] = ptrtoint i32* [[LOW]] to i64 +// CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] +// CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (i32* getelementptr (i32, i32* null, i32 1) to i64) +// CHECK: [[ARR_PRIV_PTR:%.+]] = bitcast [1 x [2 x i32]]* [[ARR_PRIV]] to i32* +// CHECK: [[ARR_PRIV:%.+]] = getelementptr i32, i32* [[ARR_PRIV_PTR]], i64 [[OFFSET]] + +// CHECK: ret void + +// CHECK: define internal void [[MAIN_MICROTASK4]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[S_FLOAT_TY]]*** dereferenceable(8) %{{.+}}) // CHECK: [[VAR2_ORIG_ADDR:%.+]] = alloca [[S_FLOAT_TY]]***, @@ -910,44 +977,186 @@ int main() { // CHECK: store [[S_FLOAT_TY]]* [[PSEUDO_VAR2_PRIV]], [[S_FLOAT_TY]]** [[REF]] // CHECK: ret void -// CHECK: define internal void [[MAIN_MICROTASK4]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [2 x [[S_FLOAT_TY]]]* dereferenceable(8) %{{.+}}) +// CHECK: define internal void [[MAIN_MICROTASK5]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[S_FLOAT_TY]]*** dereferenceable(8) %{{.+}}) -// CHECK: [[VVAR2_ORIG_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR2_ORIG_ADDR:%.+]] = alloca [[S_FLOAT_TY]]***, +// CHECK: [[VAR2_PRIV:%.+]] = alloca [1 x [6 x [[S_FLOAT_TY]]]], // Reduction list for runtime. -// CHECK: [[RED_LIST:%.+]] = alloca [2 x i8*], +// CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], -// CHECK: [[VVAR2_ORIG:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[VVAR2_ORIG_ADDR]], +// CHECK: [[VAR2_ORIG:%.+]] = load [[S_FLOAT_TY]]***, [[S_FLOAT_TY]]**** [[VAR2_ORIG_ADDR]], -// CHECK: [[LOW:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[VVAR2_ORIG]], i64 0, i64 0 -// CHECK: [[LAST:%.+]] = ptrtoint [[S_FLOAT_TY]]* %{{.+}} to i64 -// CHECK: [[FIRST:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW]] to i64 -// CHECK: [[BYTE_DIF:%.+]] = sub i64 [[LAST]], [[FIRST]] -// CHECK: [[DIF:%.+]] = sdiv exact i64 [[BYTE_DIF]], ptrtoint (float* getelementptr (float, float* null, i32 1) to i64) -// CHECK: [[SIZE:%.+]] = add nuw i64 [[DIF]], 1 -// CHECK: call i8* @llvm.stacksave() -// CHECK: [[VVAR2_PRIV:%.+]] = alloca [[S_FLOAT_TY]], i64 [[SIZE]], -// CHECK: [[ORIG_START:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[VVAR2_ORIG]] to [[S_FLOAT_TY]]* +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], +// CHECK: [[ARRIDX:%.+]] = getelementptr inbounds [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[LD]], i64 1 +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[ARRIDX]], +// CHECK: [[LOW:%.+]] = getelementptr inbounds [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[LD]], i64 1 +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], + +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], +// CHECK: [[ORIG_START:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[LD]], // CHECK: [[START:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[ORIG_START]] to i64 // CHECK: [[LOW_BOUND:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW]] to i64 // CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] // CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (float* getelementptr (float, float* null, i32 1) to i64) -// CHECK: [[PSEUDO_VVAR2_PRIV:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[VVAR2_PRIV]], i64 [[OFFSET]] -// CHECK: [[VVAR2_PRIV:%.+]] = bitcast [[S_FLOAT_TY]]* [[PSEUDO_VVAR2_PRIV]] to [2 x [[S_FLOAT_TY]]]* +// CHECK: [[VAR2_PRIV_PTR:%.+]] = bitcast [1 x [6 x [[S_FLOAT_TY]]]]* [[VAR2_PRIV]] to [[S_FLOAT_TY]]* +// CHECK: [[VAR2_PRIV:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[VAR2_PRIV_PTR]], i64 [[OFFSET]] +// CHECK: store [[S_FLOAT_TY]]** [[REF:.+]], [[S_FLOAT_TY]]*** % +// CHECK: store [[S_FLOAT_TY]]* [[VAR2_PRIV]], [[S_FLOAT_TY]]** [[REF]] // CHECK: ret void -// CHECK: define internal void [[MAIN_MICROTASK5]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [2 x [[S_FLOAT_TY]]]* dereferenceable(8) %{{.+}}) +// CHECK: define internal void [[MAIN_MICROTASK6]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[S_FLOAT_TY]]*** dereferenceable(8) %{{.+}}) -// CHECK: [[VAR3_ORIG_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR2_ORIG_ADDR:%.+]] = alloca [[S_FLOAT_TY]]***, +// CHECK: [[VAR2_PRIV:%.+]] = alloca [1 x [6 x [[S_FLOAT_TY]]]], // Reduction list for runtime. -// CHECK: [[RED_LIST:%.+]] = alloca [2 x i8*], +// CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], +// CHECK: [[VAR2_ORIG:%.+]] = load [[S_FLOAT_TY]]***, [[S_FLOAT_TY]]**** [[VAR2_ORIG_ADDR]], + +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], +// CHECK: [[ARRIDX:%.+]] = getelementptr inbounds [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[LD]], i64 1 +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[ARRIDX]], +// CHECK: [[LOW:%.+]] = getelementptr inbounds [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[LD]], i64 1 +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], + +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], +// CHECK: [[ORIG_START:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[LD]], +// CHECK: [[START:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[ORIG_START]] to i64 +// CHECK: [[LOW_BOUND:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW]] to i64 +// CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] +// CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (float* getelementptr (float, float* null, i32 1) to i64) +// CHECK: [[VAR2_PRIV_PTR:%.+]] = bitcast [1 x [6 x [[S_FLOAT_TY]]]]* [[VAR2_PRIV]] to [[S_FLOAT_TY]]* +// CHECK: [[VAR2_PRIV:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[VAR2_PRIV_PTR]], i64 [[OFFSET]] +// CHECK: store [[S_FLOAT_TY]]** [[REF:.+]], [[S_FLOAT_TY]]*** % +// CHECK: store [[S_FLOAT_TY]]* [[VAR2_PRIV]], [[S_FLOAT_TY]]** [[REF]] +// CHECK: ret void + +// CHECK: define internal void [[MAIN_MICROTASK7]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[S_FLOAT_TY]]*** dereferenceable(8) %{{.+}}) + +// CHECK: [[VAR2_ORIG_ADDR:%.+]] = alloca [[S_FLOAT_TY]]***, +// CHECK: [[VAR2_PRIV:%.+]] = alloca [[S_FLOAT_TY]], + +// Reduction list for runtime. +// CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], +// CHECK: [[VAR2_ORIG:%.+]] = load [[S_FLOAT_TY]]***, [[S_FLOAT_TY]]**** [[VAR2_ORIG_ADDR]], + +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], +// CHECK: [[ARRIDX:%.+]] = getelementptr inbounds [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[LD]], i64 1 +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[ARRIDX]], +// CHECK: [[LOW:%.+]] = getelementptr inbounds [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[LD]], i64 1 +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], + +// CHECK: [[LD:%.+]] = load [[S_FLOAT_TY]]**, [[S_FLOAT_TY]]*** [[VAR2_ORIG]], +// CHECK: [[ORIG_START:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[LD]], +// CHECK: [[START:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[ORIG_START]] to i64 +// CHECK: [[LOW_BOUND:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW]] to i64 +// CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] +// CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (float* getelementptr (float, float* null, i32 1) to i64) +// CHECK: [[PSEUDO_VAR2_PRIV:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[VAR2_PRIV]], i64 [[OFFSET]] +// CHECK: store [[S_FLOAT_TY]]** [[REF:.+]], [[S_FLOAT_TY]]*** % +// CHECK: store [[S_FLOAT_TY]]* [[PSEUDO_VAR2_PRIV]], [[S_FLOAT_TY]]** [[REF]] +// CHECK: ret void + +// CHECK: define internal void [[MAIN_MICROTASK8]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [5 x [[S_FLOAT_TY]]]* dereferenceable(20) %{{.+}}) + +// CHECK: [[VVAR2_ORIG_ADDR:%.+]] = alloca [5 x [[S_FLOAT_TY]]]*, +// CHECK: [[VVAR2_PRIV:%.+]] = alloca [5 x [[S_FLOAT_TY]]], + +// Reduction list for runtime. +// CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], +// CHECK: [[VVAR2_ORIG:%.+]] = load [5 x [[S_FLOAT_TY]]]*, [5 x [[S_FLOAT_TY]]]** [[VVAR2_ORIG_ADDR]], + +// CHECK: [[LOW:%.+]] = getelementptr inbounds [5 x [[S_FLOAT_TY]]], [5 x [[S_FLOAT_TY]]]* [[VVAR2_ORIG]], i64 0, i64 0 +// CHECK: [[ORIG_START:%.+]] = bitcast [5 x [[S_FLOAT_TY]]]* [[VVAR2_ORIG]] to [[S_FLOAT_TY]]* +// CHECK: [[START:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[ORIG_START]] to i64 +// CHECK: [[LOW_BOUND:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW]] to i64 +// CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] +// CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (float* getelementptr (float, float* null, i32 1) to i64) +// CHECK: [[VVAR2_PRIV_PTR:%.+]] = bitcast [5 x [[S_FLOAT_TY]]]* [[VVAR2_PRIV]] to [[S_FLOAT_TY]]* +// CHECK: [[VVAR2_PRIV:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[VVAR2_PRIV_PTR]], i64 [[OFFSET]] +// CHECK: ret void + +// CHECK: define internal void [[MAIN_MICROTASK9]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [4 x [[S_FLOAT_TY]]]* dereferenceable(16) %{{.+}}) + +// CHECK: [[VAR3_ORIG_ADDR:%.+]] = alloca [4 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR3_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]], + +// Reduction list for runtime. +// CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], + +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], +// CHECK: store [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]], [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR:%.+]], +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], + +// CHECK: [[LOW:%.+]] = getelementptr inbounds [4 x [[S_FLOAT_TY]]], [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]], i64 0, i64 1 +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], + +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], +// CHECK: [[ORIG_START:%.+]] = bitcast [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]] to [[S_FLOAT_TY]]* +// CHECK: [[START:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[ORIG_START]] to i64 +// CHECK: [[LOW_BOUND:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW]] to i64 +// CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] +// CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (float* getelementptr (float, float* null, i32 1) to i64) +// CHECK: [[VAR3_PRIV_PTR:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]] to [[S_FLOAT_TY]]* +// CHECK: [[PSEUDO_VAR3_PRIV:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[VAR3_PRIV_PTR]], i64 [[OFFSET]] +// CHECK: [[VAR3_PRIV:%.+]] = bitcast [[S_FLOAT_TY]]* [[PSEUDO_VAR3_PRIV]] to [4 x [[S_FLOAT_TY]]]* + +// CHECK: store [4 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]], [4 x [[S_FLOAT_TY]]]** % + +// CHECK: ret void + +// CHECK: define internal void [[MAIN_MICROTASK10]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [4 x [[S_FLOAT_TY]]]* dereferenceable(16) %{{.+}}) + +// CHECK: [[VAR3_ORIG_ADDR:%.+]] = alloca [4 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR3_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]], + +// Reduction list for runtime. +// CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], + +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], +// CHECK: store [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]], [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR:%.+]], +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], + +// CHECK: [[LOW:%.+]] = getelementptr inbounds [4 x [[S_FLOAT_TY]]], [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]], i64 0, i64 0 +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], + +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], +// CHECK: [[ORIG_START:%.+]] = bitcast [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]] to [[S_FLOAT_TY]]* +// CHECK: [[START:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[ORIG_START]] to i64 +// CHECK: [[LOW_BOUND:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW]] to i64 +// CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] +// CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (float* getelementptr (float, float* null, i32 1) to i64) +// CHECK: [[VAR3_PRIV_PTR:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]] to [[S_FLOAT_TY]]* +// CHECK: [[PSEUDO_VAR3_PRIV:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[VAR3_PRIV_PTR]], i64 [[OFFSET]] +// CHECK: [[VAR3_PRIV:%.+]] = bitcast [[S_FLOAT_TY]]* [[PSEUDO_VAR3_PRIV]] to [4 x [[S_FLOAT_TY]]]* + +// CHECK: store [4 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]], [4 x [[S_FLOAT_TY]]]** % + +// CHECK: ret void + +// CHECK: define internal void [[MAIN_MICROTASK11]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [4 x [[S_FLOAT_TY]]]* dereferenceable(16) %{{.+}}) + +// CHECK: [[VAR3_ORIG_ADDR:%.+]] = alloca [4 x [[S_FLOAT_TY]]]*, + +// Reduction list for runtime. +// CHECK: [[RED_LIST:%.+]] = alloca [2 x i8*], + +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], +// CHECK: store [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]], [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR:%.+]], +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], -// CHECK: [[VAR3_ORIG:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], -// CHECK: store [2 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]], [2 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR:%.+]], // CHECK: [[LAST:%.+]] = ptrtoint [[S_FLOAT_TY]]* %{{.+}} to i64 // CHECK: [[FIRST:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW:%.+]] to i64 // CHECK: [[BYTE_DIF:%.+]] = sub i64 [[LAST]], [[FIRST]] @@ -955,44 +1164,47 @@ int main() { // CHECK: [[SIZE:%.+]] = add nuw i64 [[DIF]], 1 // CHECK: call i8* @llvm.stacksave() // CHECK: [[VAR3_PRIV:%.+]] = alloca [[S_FLOAT_TY]], i64 [[SIZE]], -// CHECK: [[VAR3_ORIG:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], -// CHECK: [[ORIG_START:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]] to [[S_FLOAT_TY]]* +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], +// CHECK: [[ORIG_START:%.+]] = bitcast [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]] to [[S_FLOAT_TY]]* // CHECK: [[START:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[ORIG_START]] to i64 // CHECK: [[LOW_BOUND:%.+]] = ptrtoint [[S_FLOAT_TY]]* [[LOW]] to i64 // CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] // CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (float* getelementptr (float, float* null, i32 1) to i64) // CHECK: [[PSEUDO_VAR3_PRIV:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[VAR3_PRIV]], i64 [[OFFSET]] -// CHECK: [[VAR3_PRIV:%.+]] = bitcast [[S_FLOAT_TY]]* [[PSEUDO_VAR3_PRIV]] to [2 x [[S_FLOAT_TY]]]* +// CHECK: [[VAR3_PRIV:%.+]] = bitcast [[S_FLOAT_TY]]* [[PSEUDO_VAR3_PRIV]] to [4 x [[S_FLOAT_TY]]]* -// CHECK: store [2 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]], [2 x [[S_FLOAT_TY]]]** % +// CHECK: store [4 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]], [4 x [[S_FLOAT_TY]]]** % // CHECK: ret void -// CHECK: define internal void [[MAIN_MICROTASK6]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [2 x [[S_FLOAT_TY]]]* dereferenceable(8) %{{.+}}) +// CHECK: define internal void [[MAIN_MICROTASK12]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [4 x [[S_FLOAT_TY]]]* dereferenceable(16) %{{.+}}) -// CHECK: [[VAR3_ORIG_ADDR:%.+]] = alloca [2 x [[S_FLOAT_TY]]]*, -// CHECK: [[VAR3_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]], +// CHECK: [[VAR3_ORIG_ADDR:%.+]] = alloca [4 x [[S_FLOAT_TY]]]*, +// CHECK: [[VAR3_PRIV:%.+]] = alloca [4 x [[S_FLOAT_TY]]], // Reduction list for runtime. // CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], -// CHECK: [[VAR3_ORIG:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], -// CHECK: store [2 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]], [2 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR:%.+]], -// CHECK: [[VAR3_ORIG:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], -// CHECK: getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]], i32 0, i32 0 -// CHECK: getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* %{{.+}}, i64 2 +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], +// CHECK: store [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]], [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR:%.+]], +// CHECK: [[VAR3_ORIG:%.+]] = load [4 x [[S_FLOAT_TY]]]*, [4 x [[S_FLOAT_TY]]]** [[VAR3_ORIG_ADDR]], +// CHECK: getelementptr inbounds [4 x [[S_FLOAT_TY]]], [4 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]], i32 0, i32 0 +// CHECK: getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* %{{.+}}, i64 4 -// CHECK: store [2 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]], [2 x [[S_FLOAT_TY]]]** % -// CHECK: bitcast [2 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]] to [[S_FLOAT_TY]]* +// CHECK: store [4 x [[S_FLOAT_TY]]]* [[VAR3_PRIV]], [4 x [[S_FLOAT_TY]]]** % +// CHECK: bitcast [4 x [[S_FLOAT_TY]]]* [[VAR3_ORIG]] to [[S_FLOAT_TY]]* // CHECK: ret void -// CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]() +// CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT_42]]() // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]], // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]]) // CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [[S_INT_TY]]*, [[S_INT_TY]]*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*)* [[TMAIN_MICROTASK:@.+]] to void +// Not interested in this one: +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, +// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 5, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [42 x [[S_INT_TY]]]*, [2 x i32]*, i32*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[TMAIN_MICROTASK2:@.+]] to void // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]* // CHECK: ret // @@ -1002,6 +1214,7 @@ int main() { // CHECK: alloca i{{[0-9]+}}, // CHECK: alloca i{{[0-9]+}}, // CHECK: alloca i{{[0-9]+}}, +// CHECK: alloca i{{[0-9]+}}, // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]], // CHECK: [[VAR1_PRIV:%.+]] = alloca [[S_INT_TY]], @@ -1226,5 +1439,28 @@ int main() { // CHECK: store i{{[0-9]+}} [[UP]], i{{[0-9]+}}* [[T_VAR1_LHS]], // CHECK: ret void +// CHECK: define internal void [[TMAIN_MICROTASK2]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [42 x [[S_INT_TY]]]* dereferenceable(168) %{{.*}}, [2 x i32]* dereferenceable(8) %{{.*}}, i32* dereferenceable(4) %{{.*}}, [2 x [[S_INT_TY]]]* dereferenceable(8) %{{.*}}, [[S_INT_TY]]* dereferenceable(4) %{{.*}}) + +// CHECK: [[ARR_ORIG_ADDR:%.+]] = alloca [42 x [[S_INT_TY]]]*, +// CHECK: [[ARR_PRIV:%.+]] = alloca [40 x [[S_INT_TY]]], + +// Reduction list for runtime. +// CHECK: [[RED_LIST:%.+]] = alloca [1 x i8*], + +// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]], + +// CHECK: [[ARR_ORIG:%.+]] = load [42 x [[S_INT_TY]]]*, [42 x [[S_INT_TY]]]** [[ARR_ORIG_ADDR]], +// CHECK: [[LOW:%.+]] = getelementptr inbounds [42 x [[S_INT_TY]]], [42 x [[S_INT_TY]]]* [[ARR_ORIG]], i64 0, i64 1 +// CHECK: [[ORIG_START:%.+]] = bitcast [42 x [[S_INT_TY]]]* [[ARR_ORIG]] to [[S_INT_TY]]* +// CHECK: [[START:%.+]] = ptrtoint [[S_INT_TY]]* [[ORIG_START]] to i64 +// CHECK: [[LOW_BOUND:%.+]] = ptrtoint [[S_INT_TY]]* [[LOW]] to i64 +// CHECK: [[OFFSET_BYTES:%.+]] = sub i64 [[START]], [[LOW_BOUND]] +// CHECK: [[OFFSET:%.+]] = sdiv exact i64 [[OFFSET_BYTES]], ptrtoint (i32* getelementptr (i32, i32* null, i32 1) to i64) +// CHECK: [[ARR_PRIV_PTR:%.+]] = bitcast [40 x [[S_INT_TY]]]* [[ARR_PRIV]] to [[S_INT_TY]]* +// CHECK: [[PSEUDO_ARR_PRIV:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[ARR_PRIV_PTR]], i64 [[OFFSET]] +// CHECK: [[ARR_PRIV:%.+]] = bitcast [[S_INT_TY]]* [[PSEUDO_ARR_PRIV]] to [42 x [[S_INT_TY]]]* + +// CHECK: ret void + #endif |