shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/clang/test/OpenMP/parallel_private_codegen.cpp
Alexey Bataev 738bab743b [OPENMP]Add support for allocate vars in untied tasks. 
Local vars, marked with pragma allocate, mustbe allocate by the call of
the runtime function and cannot be allocated as other local variables.
Instead, we allocate a space for the pointer in private record and store
the address, returned by kmpc_alloc call in this pointer.
So, for untied tasks

```
 #pragma omp task untied
 {
   S s;
    #pragma omp allocate(s) allocator(allocator)
   s = x;
 }
```
compiler generates something like this:
```
struct task_with_privates {
  S *ptr;
};

void entry(task_with_privates *p) {
  S *s = p->s;
  switch(partid) {
  case 1:
    p->s = (S*)kmpc_alloc();
    kmpc_omp_task();
    br exit;
  case 2:
    *s = x;
    kmpc_omp_task();
    br exit;
  case 2:
    ~S(s);
    kmpc_free((void*)s);
    br exit;
  }
exit:
}
```

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D86558
2020-09-15 13:39:14 -04:00

376 lines
17 KiB
C++
Raw Blame History

// RUN: %clang_cc1 -verify -fopenmp -x c++ -triple x86_64-unknown-unknown -emit-llvm %s -o - | FileCheck %s
// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple x86_64-unknown-unknown -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp -x c++ -triple x86_64-unknown-unknown -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s
// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -DLAMBDA -triple %itanium_abi_triple -emit-llvm %s -o - | FileCheck -check-prefix=LAMBDA %s
// RUN: %clang_cc1 -verify -fopenmp -x c++ -fblocks -DBLOCKS -triple %itanium_abi_triple -emit-llvm %s -o - | FileCheck -check-prefix=BLOCKS %s
// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -triple x86_64-unknown-unknown -emit-llvm %s -o - | FileCheck --check-prefix SIMD-ONLY0 %s
// RUN: %clang_cc1 -fopenmp-simd -x c++ -std=c++11 -triple x86_64-unknown-unknown -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp-simd -x c++ -triple x86_64-unknown-unknown -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck --check-prefix SIMD-ONLY0 %s
// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -std=c++11 -DLAMBDA -triple %itanium_abi_triple -emit-llvm %s -o - | FileCheck --check-prefix SIMD-ONLY0 %s
// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -fblocks -DBLOCKS -triple %itanium_abi_triple -emit-llvm %s -o - | FileCheck --check-prefix SIMD-ONLY0 %s
// SIMD-ONLY0-NOT: {{__kmpc|__tgt}}
// expected-no-diagnostics
#ifndef HEADER
#define HEADER
enum omp_allocator_handle_t {
omp_null_allocator = 0,
omp_default_mem_alloc = 1,
omp_large_cap_mem_alloc = 2,
omp_const_mem_alloc = 3,
omp_high_bw_mem_alloc = 4,
omp_low_lat_mem_alloc = 5,
omp_cgroup_mem_alloc = 6,
omp_pteam_mem_alloc = 7,
omp_thread_mem_alloc = 8,
KMP_ALLOCATOR_MAX_HANDLE = __UINTPTR_MAX__
};
template <class T>
struct S {
T f;
S(T a) : f(a) {}
S() : f() {}
operator T() { return T(); }
~S() {}
};
volatile int g __attribute__((aligned(128))) = 1212;
struct SS {
int a;
int b : 4;
int &c;
SS(int &d) : a(0), b(0), c(d) {
#pragma omp parallel private(a, b, c)
#ifdef LAMBDA
[&]() {
++this->a, --b, (this)->c /= 1;
#pragma omp parallel private(a, b, c)
++(this)->a, --b, this->c /= 1;
}();
#elif defined(BLOCKS)
^{
++a;
--this->b;
(this)->c /= 1;
#pragma omp parallel private(a, b, c)
++(this)->a, --b, this->c /= 1;
}();
#else
++this->a, --b, c /= 1;
#endif
}
};
template<typename T>
struct SST {
T a;
SST() : a(T()) {
#pragma omp parallel private(a) allocate(omp_large_cap_mem_alloc:a)
#ifdef LAMBDA
[&]() {
[&]() {
++this->a;
#pragma omp parallel private(a)
++(this)->a;
}();
}();
#elif defined(BLOCKS)
^{
^{
++a;
#pragma omp parallel private(a)
++(this)->a;
}();
}();
#else
++(this)->a;
#endif
}
};
// CHECK: [[SS_TY:%.+]] = type { i{{[0-9]+}}, i8
// LAMBDA: [[SS_TY:%.+]] = type { i{{[0-9]+}}, i8
// BLOCKS: [[SS_TY:%.+]] = type { i{{[0-9]+}}, i8
// CHECK: [[S_FLOAT_TY:%.+]] = type { float }
// CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
// CHECK: [[SST_TY:%.+]] = type { i{{[0-9]+}} }
template <typename T>
T tmain() {
S<T> test;
SST<T> sst;
T t_var __attribute__((aligned(128))) = T();
T vec[] __attribute__((aligned(128))) = {1, 2};
S<T> s_arr[] __attribute__((aligned(128))) = {1, 2};
S<T> var __attribute__((aligned(128))) (3);
#pragma omp parallel private(t_var, vec, s_arr, var)
{
vec[0] = t_var;
s_arr[0] = var;
}
return T();
}
int main() {
static int sivar;
SS ss(sivar);
#ifdef LAMBDA
// LAMBDA: [[G:@.+]] = {{(dso_local )?}}global i{{[0-9]+}} 1212,
// LAMBDA-LABEL: @main
// LAMBDA: alloca [[SS_TY]],
// LAMBDA: alloca [[CAP_TY:%.+]],
// LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@[^(]+]]([[CAP_TY]]*
[&]() {
// LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
// LAMBDA-NOT: = getelementptr inbounds %{{.+}},
// LAMBDA: call{{.*}} void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
#pragma omp parallel private(g, sivar)
{
// LAMBDA: define {{.+}} @{{.+}}([[SS_TY]]*
// LAMBDA: store i{{[0-9]+}} 0, i{{[0-9]+}}* %
// LAMBDA: store i8
// LAMBDA: 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]+}}*, [[SS_TY]]*)* [[SS_MICROTASK:@.+]] to void
// LAMBDA: ret
// LAMBDA: define internal void [[SS_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[SS_TY]]* %{{.+}})
// LAMBDA-NOT: getelementptr {{.*}}[[SS_TY]], [[SS_TY]]* %
// LAMBDA: call{{.*}} void
// LAMBDA: ret void
// LAMBDA: define internal void @{{.+}}(i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[SS_TY]]* %{{.+}})
// LAMBDA: [[A_PRIV:%.+]] = alloca i{{[0-9]+}},
// LAMBDA: [[B_PRIV:%.+]] = alloca i{{[0-9]+}},
// LAMBDA: [[C_PRIV:%.+]] = alloca i{{[0-9]+}},
// LAMBDA: store i{{[0-9]+}}* [[A_PRIV]], i{{[0-9]+}}** [[REFA:%.+]],
// LAMBDA: store i{{[0-9]+}}* [[C_PRIV]], i{{[0-9]+}}** [[REFC:%.+]],
// LAMBDA-NEXT: [[A_PRIV:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[REFA]],
// LAMBDA-NEXT: [[A_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[A_PRIV]],
// LAMBDA-NEXT: [[INC:%.+]] = add nsw i{{[0-9]+}} [[A_VAL]], 1
// LAMBDA-NEXT: store i{{[0-9]+}} [[INC]], i{{[0-9]+}}* [[A_PRIV]],
// LAMBDA-NEXT: [[B_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[B_PRIV]],
// LAMBDA-NEXT: [[DEC:%.+]] = add nsw i{{[0-9]+}} [[B_VAL]], -1
// LAMBDA-NEXT: store i{{[0-9]+}} [[DEC]], i{{[0-9]+}}* [[B_PRIV]],
// LAMBDA-NEXT: [[C_PRIV:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[REFC]],
// LAMBDA-NEXT: [[C_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[C_PRIV]],
// LAMBDA-NEXT: [[DIV:%.+]] = sdiv i{{[0-9]+}} [[C_VAL]], 1
// LAMBDA-NEXT: store i{{[0-9]+}} [[DIV]], i{{[0-9]+}}* [[C_PRIV]],
// LAMBDA-NEXT: ret void
// LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
// LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
// LAMBDA: [[SIVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
g = 1;
sivar = 2;
// LAMBDA: store i{{[0-9]+}} 1, i{{[0-9]+}}* [[G_PRIVATE_ADDR]],
// LAMBDA: store i{{[0-9]+}} 2, i{{[0-9]+}}* [[SIVAR_PRIVATE_ADDR]],
// LAMBDA: [[G_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
// LAMBDA: store i{{[0-9]+}}* [[G_PRIVATE_ADDR]], i{{[0-9]+}}** [[G_PRIVATE_ADDR_REF]]
// LAMBDA: [[SIVAR_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
// LAMBDA: store i{{[0-9]+}}* [[SIVAR_PRIVATE_ADDR]], i{{[0-9]+}}** [[SIVAR_PRIVATE_ADDR_REF]]
// LAMBDA: call{{.*}} void [[INNER_LAMBDA:@.+]](%{{.+}}* [[ARG]])
[&]() {
// LAMBDA: define {{.+}} void [[INNER_LAMBDA]](%{{.+}}* [[ARG_PTR:%.+]])
// LAMBDA: store %{{.+}}* [[ARG_PTR]], %{{.+}}** [[ARG_PTR_REF:%.+]],
g = 2;
sivar = 4;
// LAMBDA: [[ARG_PTR:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_PTR_REF]]
// LAMBDA: [[G_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
// LAMBDA: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_PTR_REF]]
// LAMBDA: store i{{[0-9]+}} 2, i{{[0-9]+}}* [[G_REF]]
// LAMBDA: [[SIVAR_PTR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG_PTR]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
// LAMBDA: [[SIVAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[SIVAR_PTR_REF]]
// LAMBDA: store i{{[0-9]+}} 4, i{{[0-9]+}}* [[SIVAR_REF]]
}();
}
}();
return 0;
#elif defined(BLOCKS)
// BLOCKS: [[G:@.+]] = {{(dso_local )?}}global i{{[0-9]+}} 1212,
// BLOCKS-LABEL: @main
// BLOCKS: call
// BLOCKS: call{{.*}} void {{%.+}}(i8
^{
// BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
// BLOCKS-NOT: = getelementptr inbounds %{{.+}},
// BLOCKS: call{{.*}} void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
#pragma omp parallel private(g, sivar)
{
// BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
// BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
// BLOCKS: [[SIVAR_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
g = 1;
sivar = 20;
// BLOCKS: store i{{[0-9]+}} 1, i{{[0-9]+}}* [[G_PRIVATE_ADDR]],
// BLOCKS: store i{{[0-9]+}} 20, i{{[0-9]+}}* [[SIVAR_PRIVATE_ADDR]],
// BLOCKS-NOT: [[G]]{{[[^:word:]]}}
// BLOCKS: i{{[0-9]+}}* [[G_PRIVATE_ADDR]]
// BLOCKS-NOT: [[G]]{{[[^:word:]]}}
// BLOCKS-NOT: [[SIVAR]]{{[[^:word:]]}}
// BLOCKS: i{{[0-9]+}}* [[SIVAR_PRIVATE_ADDR]]
// BLOCKS-NOT: [[SIVAR]]{{[[^:word:]]}}
// BLOCKS: call{{.*}} void {{%.+}}(i8
^{
// BLOCKS: define {{.+}} void {{@.+}}(i8*
g = 2;
sivar = 40;
// BLOCKS-NOT: [[G]]{{[[^:word:]]}}
// BLOCKS: store i{{[0-9]+}} 2, i{{[0-9]+}}*
// BLOCKS-NOT: [[G]]{{[[^:word:]]}}
// BLOCKS-NOT: [[SIVAR]]{{[[^:word:]]}}
// BLOCKS: store i{{[0-9]+}} 40, i{{[0-9]+}}*
// BLOCKS-NOT: [[SIVAR]]{{[[^:word:]]}}
// BLOCKS: ret
}();
}
}();
return 0;
// BLOCKS: define {{.+}} @{{.+}}([[SS_TY]]*
// BLOCKS: store i{{[0-9]+}} 0, i{{[0-9]+}}* %
// BLOCKS: store i8
// BLOCKS: 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]+}}*, [[SS_TY]]*)* [[SS_MICROTASK:@.+]] to void
// BLOCKS: ret
// BLOCKS: define internal void [[SS_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[SS_TY]]* %{{.+}})
// BLOCKS-NOT: getelementptr {{.*}}[[SS_TY]], [[SS_TY]]* %
// BLOCKS: call{{.*}} void
// BLOCKS: ret void
// BLOCKS: define internal void @{{.+}}(i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[SS_TY]]* %{{.+}})
// BLOCKS: [[A_PRIV:%.+]] = alloca i{{[0-9]+}},
// BLOCKS: [[B_PRIV:%.+]] = alloca i{{[0-9]+}},
// BLOCKS: [[C_PRIV:%.+]] = alloca i{{[0-9]+}},
// BLOCKS: store i{{[0-9]+}}* [[A_PRIV]], i{{[0-9]+}}** [[REFA:%.+]],
// BLOCKS: store i{{[0-9]+}}* [[C_PRIV]], i{{[0-9]+}}** [[REFC:%.+]],
// BLOCKS-NEXT: [[A_PRIV:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[REFA]],
// BLOCKS-NEXT: [[A_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[A_PRIV]],
// BLOCKS-NEXT: [[INC:%.+]] = add nsw i{{[0-9]+}} [[A_VAL]], 1
// BLOCKS-NEXT: store i{{[0-9]+}} [[INC]], i{{[0-9]+}}* [[A_PRIV]],
// BLOCKS-NEXT: [[B_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[B_PRIV]],
// BLOCKS-NEXT: [[DEC:%.+]] = add nsw i{{[0-9]+}} [[B_VAL]], -1
// BLOCKS-NEXT: store i{{[0-9]+}} [[DEC]], i{{[0-9]+}}* [[B_PRIV]],
// BLOCKS-NEXT: [[C_PRIV:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[REFC]],
// BLOCKS-NEXT: [[C_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[C_PRIV]],
// BLOCKS-NEXT: [[DIV:%.+]] = sdiv i{{[0-9]+}} [[C_VAL]], 1
// BLOCKS-NEXT: store i{{[0-9]+}} [[DIV]], i{{[0-9]+}}* [[C_PRIV]],
// BLOCKS-NEXT: ret void
#else
S<float> test;
int t_var = 0;
int vec[] = {1, 2};
S<float> s_arr[] = {1, 2};
S<float> var(3);
#pragma omp parallel private(t_var, vec, s_arr, var, sivar)
{
vec[0] = t_var;
s_arr[0] = var;
sivar = 3;
}
return tmain<int>();
#endif
}
// CHECK: define i{{[0-9]+}} @main()
// CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK:@.+]] to void
// CHECK: = call i{{.+}} [[TMAIN_INT:@.+]]()
// CHECK: call void [[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 %{{.+}})
// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
// CHECK: [[SIVAR_PRIV:%.+]] = alloca i{{[0-9]+}},
// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
// CHECK-NOT: [[T_VAR_PRIV]]
// CHECK-NOT: [[VEC_PRIV]]
// CHECK: {{.+}}:
// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_FLOAT_TY]]*
// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]])
// CHECK-NOT: [[T_VAR_PRIV]]
// CHECK-NOT: [[VEC_PRIV]]
// CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
// CHECK-DAG: call void [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
// CHECK-DAG: call void [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]*
// CHECK: ret void
// CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
// CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[TMAIN_MICROTASK:@.+]] to void
// CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
// CHECK: ret
//
// CHECK: define {{.+}} @{{.+}}([[SS_TY]]*
// CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* %
// CHECK: store i8
// 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]+}}*, [[SS_TY]]*)* [[SS_MICROTASK:@.+]] to void
// CHECK: ret
// CHECK: define internal void [[SS_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[SS_TY]]* %{{.+}})
// CHECK: [[A_PRIV:%.+]] = alloca i{{[0-9]+}},
// CHECK: [[B_PRIV:%.+]] = alloca i{{[0-9]+}},
// CHECK: [[C_PRIV:%.+]] = alloca i{{[0-9]+}},
// CHECK: store i{{[0-9]+}}* [[A_PRIV]], i{{[0-9]+}}** [[REFA:%.+]],
// CHECK: store i{{[0-9]+}}* [[C_PRIV]], i{{[0-9]+}}** [[REFC:%.+]],
// CHECK-NEXT: [[A_PRIV:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[REFA]],
// CHECK-NEXT: [[A_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[A_PRIV]],
// CHECK-NEXT: [[INC:%.+]] = add nsw i{{[0-9]+}} [[A_VAL]], 1
// CHECK-NEXT: store i{{[0-9]+}} [[INC]], i{{[0-9]+}}* [[A_PRIV]],
// CHECK-NEXT: [[B_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[B_PRIV]],
// CHECK-NEXT: [[DEC:%.+]] = add nsw i{{[0-9]+}} [[B_VAL]], -1
// CHECK-NEXT: store i{{[0-9]+}} [[DEC]], i{{[0-9]+}}* [[B_PRIV]],
// CHECK-NEXT: [[C_PRIV:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[REFC]],
// CHECK-NEXT: [[C_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[C_PRIV]],
// CHECK-NEXT: [[DIV:%.+]] = sdiv i{{[0-9]+}} [[C_VAL]], 1
// CHECK-NEXT: store i{{[0-9]+}} [[DIV]], i{{[0-9]+}}* [[C_PRIV]],
// CHECK-NEXT: ret void
// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
// CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}}, align 128
// CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}], align 128
// CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]], align 128
// CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]], align 128
// CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
// CHECK-NOT: [[T_VAR_PRIV]]
// CHECK-NOT: [[VEC_PRIV]]
// CHECK-NOT: [[SIVAR_PRIV]]
// CHECK: {{.+}}:
// CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_INT_TY]]*
// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]])
// CHECK-NOT: [[T_VAR_PRIV]]
// CHECK-NOT: [[VEC_PRIV]]
// CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]])
// CHECK-DAG: call void [[S_INT_TY_DESTR]]([[S_INT_TY]]* [[VAR_PRIV]])
// CHECK-DAG: call void [[S_INT_TY_DESTR]]([[S_INT_TY]]*
// CHECK: ret void
// CHECK: define {{.+}} @{{.+}}([[SST_TY]]* %
// CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* %
// 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]+}}*, [[SST_TY]]*)* [[SST_MICROTASK:@.+]] to void
// CHECK: ret
// CHECK: define internal void [[SST_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [[SST_TY]]* %{{.+}})
// CHECK: [[GTID_ADDR_PTR:%.+]] = alloca i32*,
// CHECK: [[GTID_ADDR:%.+]] = load i32*, i32** [[GTID_ADDR_PTR]],
// CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_ADDR]],
// CHECK: [[A_VOID_PTR:%.+]] = call i8* @__kmpc_alloc(i32 [[GTID]], i64 4, i8* inttoptr (i64 2 to i8*))
// CHECK: [[A_PRIV_ADDR:%.+]] = bitcast i8* [[A_VOID_PTR]] to i32*
// CHECK: store i{{[0-9]+}}* [[A_PRIV_ADDR]], i{{[0-9]+}}** [[REF:%.+]],
// CHECK-NEXT: [[A_PRIV:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[REF]],
// CHECK-NEXT: [[A_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[A_PRIV]],
// CHECK-NEXT: [[INC:%.+]] = add nsw i{{[0-9]+}} [[A_VAL]], 1
// CHECK-NEXT: store i{{[0-9]+}} [[INC]], i{{[0-9]+}}* [[A_PRIV]],
// CHECK-NEXT: [[A_VOID_PTR:%.+]] = bitcast i32* [[A_PRIV_ADDR]] to i8*
// CHECK-NEXT: call void @__kmpc_free(i32 [[GTID]], i8* [[A_VOID_PTR]], i8* inttoptr (i64 2 to i8*))
// CHECK-NEXT: ret void
#endif
Reference in New Issue View Git Blame Copy Permalink