1 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
2 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-enable-irbuilder -x c++ -emit-llvm %s -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -o - | FileCheck --check-prefixes=CHECK %s
3 // RUN: %clang_cc1 -fopenmp -fopenmp-enable-irbuilder -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -debug-info-kind=limited -std=c++11 -verify %s -emit-llvm -o - | FileCheck --check-prefixes=CHECK-DEBUG %s
4 
5 // expected-no-diagnostics
6 
7 // TODO: Teach the update script to check new functions too.
8 
9 #ifndef HEADER
10 #define HEADER
11 
12 // CHECK-LABEL: @_Z14parallel_for_0v(
13 // CHECK-NEXT:  entry:
14 // CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]])
15 // CHECK-NEXT:    br label [[OMP_PARALLEL:%.*]]
16 // CHECK:       omp_parallel:
17 // CHECK-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* @_Z14parallel_for_0v..omp_par to void (i32*, i32*, ...)*))
18 // CHECK-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT:%.*]]
19 // CHECK:       omp.par.outlined.exit:
20 // CHECK-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
21 // CHECK:       omp.par.exit.split:
22 // CHECK-NEXT:    ret void
23 //
24 // CHECK-DEBUG-LABEL: @_Z14parallel_for_0v(
25 // CHECK-DEBUG-NEXT:  entry:
26 // CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1:[0-9]+]]), !dbg [[DBG13:![0-9]+]]
27 // CHECK-DEBUG-NEXT:    br label [[OMP_PARALLEL:%.*]]
28 // CHECK-DEBUG:       omp_parallel:
29 // CHECK-DEBUG-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* @_Z14parallel_for_0v..omp_par to void (i32*, i32*, ...)*)), !dbg [[DBG14:![0-9]+]]
30 // CHECK-DEBUG-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT:%.*]]
31 // CHECK-DEBUG:       omp.par.outlined.exit:
32 // CHECK-DEBUG-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
33 // CHECK-DEBUG:       omp.par.exit.split:
34 // CHECK-DEBUG-NEXT:    ret void, !dbg [[DBG18:![0-9]+]]
35 //
36 void parallel_for_0(void) {
37 #pragma omp parallel
38   {
39 #pragma omp for
40     for (int i = 0; i < 100; ++i) {
41     }
42   }
43 }
44 
45 // CHECK-LABEL: @_Z14parallel_for_1Pfid(
46 // CHECK-NEXT:  entry:
47 // CHECK-NEXT:    [[STRUCTARG17:%.*]] = alloca { { i32*, double*, float** }*, i32*, double*, float** }, align 8
48 // CHECK-NEXT:    [[STRUCTARG:%.*]] = alloca { i32*, double*, float** }, align 8
49 // CHECK-NEXT:    [[R_ADDR:%.*]] = alloca float*, align 8
50 // CHECK-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
51 // CHECK-NEXT:    [[B_ADDR:%.*]] = alloca double, align 8
52 // CHECK-NEXT:    store float* [[R:%.*]], float** [[R_ADDR]], align 8
53 // CHECK-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
54 // CHECK-NEXT:    store double [[B:%.*]], double* [[B_ADDR]], align 8
55 // CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1]])
56 // CHECK-NEXT:    br label [[OMP_PARALLEL:%.*]]
57 // CHECK:       omp_parallel:
58 // CHECK-NEXT:    [[GEP_STRUCTARG:%.*]] = getelementptr { { i32*, double*, float** }*, i32*, double*, float** }, { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]], i32 0, i32 0
59 // CHECK-NEXT:    store { i32*, double*, float** }* [[STRUCTARG]], { i32*, double*, float** }** [[GEP_STRUCTARG]], align 8
60 // CHECK-NEXT:    [[GEP_A_ADDR18:%.*]] = getelementptr { { i32*, double*, float** }*, i32*, double*, float** }, { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]], i32 0, i32 1
61 // CHECK-NEXT:    store i32* [[A_ADDR]], i32** [[GEP_A_ADDR18]], align 8
62 // CHECK-NEXT:    [[GEP_B_ADDR19:%.*]] = getelementptr { { i32*, double*, float** }*, i32*, double*, float** }, { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]], i32 0, i32 2
63 // CHECK-NEXT:    store double* [[B_ADDR]], double** [[GEP_B_ADDR19]], align 8
64 // CHECK-NEXT:    [[GEP_R_ADDR20:%.*]] = getelementptr { { i32*, double*, float** }*, i32*, double*, float** }, { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]], i32 0, i32 3
65 // CHECK-NEXT:    store float** [[R_ADDR]], float*** [[GEP_R_ADDR20]], align 8
66 // CHECK-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, { { i32*, double*, float** }*, i32*, double*, float** }*)* @_Z14parallel_for_1Pfid..omp_par.4 to void (i32*, i32*, ...)*), { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]])
67 // CHECK-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT16:%.*]]
68 // CHECK:       omp.par.outlined.exit16:
69 // CHECK-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
70 // CHECK:       omp.par.exit.split:
71 // CHECK-NEXT:    ret void
72 //
73 // CHECK-DEBUG-LABEL: @_Z14parallel_for_1Pfid(
74 // CHECK-DEBUG-NEXT:  entry:
75 // CHECK-DEBUG-NEXT:    [[STRUCTARG17:%.*]] = alloca { { i32*, double*, float** }*, i32*, double*, float** }, align 8
76 // CHECK-DEBUG-NEXT:    [[STRUCTARG:%.*]] = alloca { i32*, double*, float** }, align 8
77 // CHECK-DEBUG-NEXT:    [[R_ADDR:%.*]] = alloca float*, align 8
78 // CHECK-DEBUG-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
79 // CHECK-DEBUG-NEXT:    [[B_ADDR:%.*]] = alloca double, align 8
80 // CHECK-DEBUG-NEXT:    store float* [[R:%.*]], float** [[R_ADDR]], align 8
81 // CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata float** [[R_ADDR]], metadata [[META72:![0-9]+]], metadata !DIExpression()), !dbg [[DBG73:![0-9]+]]
82 // CHECK-DEBUG-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
83 // CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[A_ADDR]], metadata [[META74:![0-9]+]], metadata !DIExpression()), !dbg [[DBG75:![0-9]+]]
84 // CHECK-DEBUG-NEXT:    store double [[B:%.*]], double* [[B_ADDR]], align 8
85 // CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata double* [[B_ADDR]], metadata [[META76:![0-9]+]], metadata !DIExpression()), !dbg [[DBG77:![0-9]+]]
86 // CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB6:[0-9]+]]), !dbg [[DBG78:![0-9]+]]
87 // CHECK-DEBUG-NEXT:    br label [[OMP_PARALLEL:%.*]]
88 // CHECK-DEBUG:       omp_parallel:
89 // CHECK-DEBUG-NEXT:    [[GEP_STRUCTARG:%.*]] = getelementptr { { i32*, double*, float** }*, i32*, double*, float** }, { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]], i32 0, i32 0
90 // CHECK-DEBUG-NEXT:    store { i32*, double*, float** }* [[STRUCTARG]], { i32*, double*, float** }** [[GEP_STRUCTARG]], align 8
91 // CHECK-DEBUG-NEXT:    [[GEP_A_ADDR18:%.*]] = getelementptr { { i32*, double*, float** }*, i32*, double*, float** }, { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]], i32 0, i32 1
92 // CHECK-DEBUG-NEXT:    store i32* [[A_ADDR]], i32** [[GEP_A_ADDR18]], align 8
93 // CHECK-DEBUG-NEXT:    [[GEP_B_ADDR19:%.*]] = getelementptr { { i32*, double*, float** }*, i32*, double*, float** }, { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]], i32 0, i32 2
94 // CHECK-DEBUG-NEXT:    store double* [[B_ADDR]], double** [[GEP_B_ADDR19]], align 8
95 // CHECK-DEBUG-NEXT:    [[GEP_R_ADDR20:%.*]] = getelementptr { { i32*, double*, float** }*, i32*, double*, float** }, { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]], i32 0, i32 3
96 // CHECK-DEBUG-NEXT:    store float** [[R_ADDR]], float*** [[GEP_R_ADDR20]], align 8
97 // CHECK-DEBUG-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB6]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, { { i32*, double*, float** }*, i32*, double*, float** }*)* @_Z14parallel_for_1Pfid..omp_par.4 to void (i32*, i32*, ...)*), { { i32*, double*, float** }*, i32*, double*, float** }* [[STRUCTARG17]]), !dbg [[DBG79:![0-9]+]]
98 // CHECK-DEBUG-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT16:%.*]]
99 // CHECK-DEBUG:       omp.par.outlined.exit16:
100 // CHECK-DEBUG-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
101 // CHECK-DEBUG:       omp.par.exit.split:
102 // CHECK-DEBUG-NEXT:    ret void, !dbg [[DBG81:![0-9]+]]
103 //
104 void parallel_for_1(float *r, int a, double b) {
105 #pragma omp parallel
106   {
107 #pragma omp parallel
108     {
109 #pragma omp for
110       for (int i = 0; i < 100; ++i) {
111         *r = a + b;
112       }
113     }
114   }
115 }
116 
117 // CHECK-LABEL: @_Z14parallel_for_2Pfid(
118 // CHECK-NEXT:  entry:
119 // CHECK-NEXT:    [[STRUCTARG218:%.*]] = alloca { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, align 8
120 // CHECK-NEXT:    [[STRUCTARG214:%.*]] = alloca { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, align 8
121 // CHECK-NEXT:    [[STRUCTARG209:%.*]] = alloca { i32*, double*, float** }, align 8
122 // CHECK-NEXT:    [[STRUCTARG:%.*]] = alloca { i32*, double*, float** }, align 8
123 // CHECK-NEXT:    [[R_ADDR:%.*]] = alloca float*, align 8
124 // CHECK-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
125 // CHECK-NEXT:    [[B_ADDR:%.*]] = alloca double, align 8
126 // CHECK-NEXT:    [[I185:%.*]] = alloca i32, align 4
127 // CHECK-NEXT:    [[AGG_CAPTURED186:%.*]] = alloca [[STRUCT_ANON_17:%.*]], align 8
128 // CHECK-NEXT:    [[AGG_CAPTURED187:%.*]] = alloca [[STRUCT_ANON_18:%.*]], align 4
129 // CHECK-NEXT:    [[DOTCOUNT_ADDR188:%.*]] = alloca i32, align 4
130 // CHECK-NEXT:    [[P_LASTITER203:%.*]] = alloca i32, align 4
131 // CHECK-NEXT:    [[P_LOWERBOUND204:%.*]] = alloca i32, align 4
132 // CHECK-NEXT:    [[P_UPPERBOUND205:%.*]] = alloca i32, align 4
133 // CHECK-NEXT:    [[P_STRIDE206:%.*]] = alloca i32, align 4
134 // CHECK-NEXT:    store float* [[R:%.*]], float** [[R_ADDR]], align 8
135 // CHECK-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
136 // CHECK-NEXT:    store double [[B:%.*]], double* [[B_ADDR]], align 8
137 // CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1]])
138 // CHECK-NEXT:    br label [[OMP_PARALLEL:%.*]]
139 // CHECK:       omp_parallel:
140 // CHECK-NEXT:    [[GEP_STRUCTARG214:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 0
141 // CHECK-NEXT:    store { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG214]], { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }** [[GEP_STRUCTARG214]], align 8
142 // CHECK-NEXT:    [[GEP_STRUCTARG219:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 1
143 // CHECK-NEXT:    store { i32*, double*, float** }* [[STRUCTARG]], { i32*, double*, float** }** [[GEP_STRUCTARG219]], align 8
144 // CHECK-NEXT:    [[GEP_A_ADDR:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 2
145 // CHECK-NEXT:    store i32* [[A_ADDR]], i32** [[GEP_A_ADDR]], align 8
146 // CHECK-NEXT:    [[GEP_B_ADDR:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 3
147 // CHECK-NEXT:    store double* [[B_ADDR]], double** [[GEP_B_ADDR]], align 8
148 // CHECK-NEXT:    [[GEP_R_ADDR:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 4
149 // CHECK-NEXT:    store float** [[R_ADDR]], float*** [[GEP_R_ADDR]], align 8
150 // CHECK-NEXT:    [[GEP_STRUCTARG209220:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 5
151 // CHECK-NEXT:    store { i32*, double*, float** }* [[STRUCTARG209]], { i32*, double*, float** }** [[GEP_STRUCTARG209220]], align 8
152 // CHECK-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB1]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*)* @_Z14parallel_for_2Pfid..omp_par.23 to void (i32*, i32*, ...)*), { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]])
153 // CHECK-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT184:%.*]]
154 // CHECK:       omp.par.outlined.exit184:
155 // CHECK-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
156 // CHECK:       omp.par.exit.split:
157 // CHECK-NEXT:    store i32 0, i32* [[I185]], align 4
158 // CHECK-NEXT:    [[TMP0:%.*]] = getelementptr inbounds [[STRUCT_ANON_17]], %struct.anon.17* [[AGG_CAPTURED186]], i32 0, i32 0
159 // CHECK-NEXT:    store i32* [[I185]], i32** [[TMP0]], align 8
160 // CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [[STRUCT_ANON_18]], %struct.anon.18* [[AGG_CAPTURED187]], i32 0, i32 0
161 // CHECK-NEXT:    [[TMP2:%.*]] = load i32, i32* [[I185]], align 4
162 // CHECK-NEXT:    store i32 [[TMP2]], i32* [[TMP1]], align 4
163 // CHECK-NEXT:    call void @__captured_stmt.19(i32* [[DOTCOUNT_ADDR188]], %struct.anon.17* [[AGG_CAPTURED186]])
164 // CHECK-NEXT:    [[DOTCOUNT189:%.*]] = load i32, i32* [[DOTCOUNT_ADDR188]], align 4
165 // CHECK-NEXT:    br label [[OMP_LOOP_PREHEADER190:%.*]]
166 // CHECK:       omp_loop.preheader190:
167 // CHECK-NEXT:    store i32 0, i32* [[P_LOWERBOUND204]], align 4
168 // CHECK-NEXT:    [[TMP3:%.*]] = sub i32 [[DOTCOUNT189]], 1
169 // CHECK-NEXT:    store i32 [[TMP3]], i32* [[P_UPPERBOUND205]], align 4
170 // CHECK-NEXT:    store i32 1, i32* [[P_STRIDE206]], align 4
171 // CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM207:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1]])
172 // CHECK-NEXT:    call void @__kmpc_for_static_init_4u(%struct.ident_t* @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM207]], i32 34, i32* [[P_LASTITER203]], i32* [[P_LOWERBOUND204]], i32* [[P_UPPERBOUND205]], i32* [[P_STRIDE206]], i32 1, i32 1)
173 // CHECK-NEXT:    [[TMP4:%.*]] = load i32, i32* [[P_LOWERBOUND204]], align 4
174 // CHECK-NEXT:    [[TMP5:%.*]] = load i32, i32* [[P_UPPERBOUND205]], align 4
175 // CHECK-NEXT:    [[TMP6:%.*]] = sub i32 [[TMP5]], [[TMP4]]
176 // CHECK-NEXT:    [[TMP7:%.*]] = add i32 [[TMP6]], 1
177 // CHECK-NEXT:    br label [[OMP_LOOP_HEADER191:%.*]]
178 // CHECK:       omp_loop.header191:
179 // CHECK-NEXT:    [[OMP_LOOP_IV197:%.*]] = phi i32 [ 0, [[OMP_LOOP_PREHEADER190]] ], [ [[OMP_LOOP_NEXT199:%.*]], [[OMP_LOOP_INC194:%.*]] ]
180 // CHECK-NEXT:    br label [[OMP_LOOP_COND192:%.*]]
181 // CHECK:       omp_loop.cond192:
182 // CHECK-NEXT:    [[OMP_LOOP_CMP198:%.*]] = icmp ult i32 [[OMP_LOOP_IV197]], [[TMP7]]
183 // CHECK-NEXT:    br i1 [[OMP_LOOP_CMP198]], label [[OMP_LOOP_BODY193:%.*]], label [[OMP_LOOP_EXIT195:%.*]]
184 // CHECK:       omp_loop.body193:
185 // CHECK-NEXT:    [[TMP8:%.*]] = add i32 [[OMP_LOOP_IV197]], [[TMP4]]
186 // CHECK-NEXT:    call void @__captured_stmt.20(i32* [[I185]], i32 [[TMP8]], %struct.anon.18* [[AGG_CAPTURED187]])
187 // CHECK-NEXT:    [[TMP9:%.*]] = load i32, i32* [[A_ADDR]], align 4
188 // CHECK-NEXT:    [[CONV200:%.*]] = sitofp i32 [[TMP9]] to double
189 // CHECK-NEXT:    [[TMP10:%.*]] = load double, double* [[B_ADDR]], align 8
190 // CHECK-NEXT:    [[ADD201:%.*]] = fadd double [[CONV200]], [[TMP10]]
191 // CHECK-NEXT:    [[CONV202:%.*]] = fptrunc double [[ADD201]] to float
192 // CHECK-NEXT:    [[TMP11:%.*]] = load float*, float** [[R_ADDR]], align 8
193 // CHECK-NEXT:    store float [[CONV202]], float* [[TMP11]], align 4
194 // CHECK-NEXT:    br label [[OMP_LOOP_INC194]]
195 // CHECK:       omp_loop.inc194:
196 // CHECK-NEXT:    [[OMP_LOOP_NEXT199]] = add nuw i32 [[OMP_LOOP_IV197]], 1
197 // CHECK-NEXT:    br label [[OMP_LOOP_HEADER191]]
198 // CHECK:       omp_loop.exit195:
199 // CHECK-NEXT:    call void @__kmpc_for_static_fini(%struct.ident_t* @[[GLOB1]], i32 [[OMP_GLOBAL_THREAD_NUM207]])
200 // CHECK-NEXT:    [[OMP_GLOBAL_THREAD_NUM208:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB1]])
201 // CHECK-NEXT:    call void @__kmpc_barrier(%struct.ident_t* @[[GLOB2:[0-9]+]], i32 [[OMP_GLOBAL_THREAD_NUM208]])
202 // CHECK-NEXT:    br label [[OMP_LOOP_AFTER196:%.*]]
203 // CHECK:       omp_loop.after196:
204 // CHECK-NEXT:    ret void
205 //
206 // CHECK-DEBUG-LABEL: @_Z14parallel_for_2Pfid(
207 // CHECK-DEBUG-NEXT:  entry:
208 // CHECK-DEBUG-NEXT:    [[STRUCTARG218:%.*]] = alloca { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, align 8
209 // CHECK-DEBUG-NEXT:    [[STRUCTARG214:%.*]] = alloca { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, align 8
210 // CHECK-DEBUG-NEXT:    [[STRUCTARG209:%.*]] = alloca { i32*, double*, float** }, align 8
211 // CHECK-DEBUG-NEXT:    [[STRUCTARG:%.*]] = alloca { i32*, double*, float** }, align 8
212 // CHECK-DEBUG-NEXT:    [[R_ADDR:%.*]] = alloca float*, align 8
213 // CHECK-DEBUG-NEXT:    [[A_ADDR:%.*]] = alloca i32, align 4
214 // CHECK-DEBUG-NEXT:    [[B_ADDR:%.*]] = alloca double, align 8
215 // CHECK-DEBUG-NEXT:    [[I185:%.*]] = alloca i32, align 4
216 // CHECK-DEBUG-NEXT:    [[AGG_CAPTURED186:%.*]] = alloca [[STRUCT_ANON_17:%.*]], align 8
217 // CHECK-DEBUG-NEXT:    [[AGG_CAPTURED187:%.*]] = alloca [[STRUCT_ANON_18:%.*]], align 4
218 // CHECK-DEBUG-NEXT:    [[DOTCOUNT_ADDR188:%.*]] = alloca i32, align 4
219 // CHECK-DEBUG-NEXT:    [[P_LASTITER203:%.*]] = alloca i32, align 4
220 // CHECK-DEBUG-NEXT:    [[P_LOWERBOUND204:%.*]] = alloca i32, align 4
221 // CHECK-DEBUG-NEXT:    [[P_UPPERBOUND205:%.*]] = alloca i32, align 4
222 // CHECK-DEBUG-NEXT:    [[P_STRIDE206:%.*]] = alloca i32, align 4
223 // CHECK-DEBUG-NEXT:    store float* [[R:%.*]], float** [[R_ADDR]], align 8
224 // CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata float** [[R_ADDR]], metadata [[META133:![0-9]+]], metadata !DIExpression()), !dbg [[DBG134:![0-9]+]]
225 // CHECK-DEBUG-NEXT:    store i32 [[A:%.*]], i32* [[A_ADDR]], align 4
226 // CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[A_ADDR]], metadata [[META135:![0-9]+]], metadata !DIExpression()), !dbg [[DBG136:![0-9]+]]
227 // CHECK-DEBUG-NEXT:    store double [[B:%.*]], double* [[B_ADDR]], align 8
228 // CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata double* [[B_ADDR]], metadata [[META137:![0-9]+]], metadata !DIExpression()), !dbg [[DBG138:![0-9]+]]
229 // CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB13:[0-9]+]]), !dbg [[DBG139:![0-9]+]]
230 // CHECK-DEBUG-NEXT:    br label [[OMP_PARALLEL:%.*]]
231 // CHECK-DEBUG:       omp_parallel:
232 // CHECK-DEBUG-NEXT:    [[GEP_STRUCTARG214:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 0
233 // CHECK-DEBUG-NEXT:    store { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG214]], { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }** [[GEP_STRUCTARG214]], align 8
234 // CHECK-DEBUG-NEXT:    [[GEP_STRUCTARG219:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 1
235 // CHECK-DEBUG-NEXT:    store { i32*, double*, float** }* [[STRUCTARG]], { i32*, double*, float** }** [[GEP_STRUCTARG219]], align 8
236 // CHECK-DEBUG-NEXT:    [[GEP_A_ADDR:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 2
237 // CHECK-DEBUG-NEXT:    store i32* [[A_ADDR]], i32** [[GEP_A_ADDR]], align 8
238 // CHECK-DEBUG-NEXT:    [[GEP_B_ADDR:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 3
239 // CHECK-DEBUG-NEXT:    store double* [[B_ADDR]], double** [[GEP_B_ADDR]], align 8
240 // CHECK-DEBUG-NEXT:    [[GEP_R_ADDR:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 4
241 // CHECK-DEBUG-NEXT:    store float** [[R_ADDR]], float*** [[GEP_R_ADDR]], align 8
242 // CHECK-DEBUG-NEXT:    [[GEP_STRUCTARG209220:%.*]] = getelementptr { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]], i32 0, i32 5
243 // CHECK-DEBUG-NEXT:    store { i32*, double*, float** }* [[STRUCTARG209]], { i32*, double*, float** }** [[GEP_STRUCTARG209220]], align 8
244 // CHECK-DEBUG-NEXT:    call void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%struct.ident_t* @[[GLOB13]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*)* @_Z14parallel_for_2Pfid..omp_par.23 to void (i32*, i32*, ...)*), { { { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }*, { i32*, double*, float** }*, i32*, double*, float**, { i32*, double*, float** }* }* [[STRUCTARG218]]), !dbg [[DBG140:![0-9]+]]
245 // CHECK-DEBUG-NEXT:    br label [[OMP_PAR_OUTLINED_EXIT184:%.*]]
246 // CHECK-DEBUG:       omp.par.outlined.exit184:
247 // CHECK-DEBUG-NEXT:    br label [[OMP_PAR_EXIT_SPLIT:%.*]]
248 // CHECK-DEBUG:       omp.par.exit.split:
249 // CHECK-DEBUG-NEXT:    call void @llvm.dbg.declare(metadata i32* [[I185]], metadata [[META144:![0-9]+]], metadata !DIExpression()), !dbg [[DBG147:![0-9]+]]
250 // CHECK-DEBUG-NEXT:    store i32 0, i32* [[I185]], align 4, !dbg [[DBG147]]
251 // CHECK-DEBUG-NEXT:    [[TMP0:%.*]] = getelementptr inbounds [[STRUCT_ANON_17]], %struct.anon.17* [[AGG_CAPTURED186]], i32 0, i32 0, !dbg [[DBG148:![0-9]+]]
252 // CHECK-DEBUG-NEXT:    store i32* [[I185]], i32** [[TMP0]], align 8, !dbg [[DBG148]]
253 // CHECK-DEBUG-NEXT:    [[TMP1:%.*]] = getelementptr inbounds [[STRUCT_ANON_18]], %struct.anon.18* [[AGG_CAPTURED187]], i32 0, i32 0, !dbg [[DBG148]]
254 // CHECK-DEBUG-NEXT:    [[TMP2:%.*]] = load i32, i32* [[I185]], align 4, !dbg [[DBG149:![0-9]+]]
255 // CHECK-DEBUG-NEXT:    store i32 [[TMP2]], i32* [[TMP1]], align 4, !dbg [[DBG148]]
256 // CHECK-DEBUG-NEXT:    call void @__captured_stmt.19(i32* [[DOTCOUNT_ADDR188]], %struct.anon.17* [[AGG_CAPTURED186]]), !dbg [[DBG148]]
257 // CHECK-DEBUG-NEXT:    [[DOTCOUNT189:%.*]] = load i32, i32* [[DOTCOUNT_ADDR188]], align 4, !dbg [[DBG148]]
258 // CHECK-DEBUG-NEXT:    br label [[OMP_LOOP_PREHEADER190:%.*]], !dbg [[DBG148]]
259 // CHECK-DEBUG:       omp_loop.preheader190:
260 // CHECK-DEBUG-NEXT:    store i32 0, i32* [[P_LOWERBOUND204]], align 4, !dbg [[DBG148]]
261 // CHECK-DEBUG-NEXT:    [[TMP3:%.*]] = sub i32 [[DOTCOUNT189]], 1, !dbg [[DBG148]]
262 // CHECK-DEBUG-NEXT:    store i32 [[TMP3]], i32* [[P_UPPERBOUND205]], align 4, !dbg [[DBG148]]
263 // CHECK-DEBUG-NEXT:    store i32 1, i32* [[P_STRIDE206]], align 4, !dbg [[DBG148]]
264 // CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM207:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB42:[0-9]+]]), !dbg [[DBG148]]
265 // CHECK-DEBUG-NEXT:    call void @__kmpc_for_static_init_4u(%struct.ident_t* @[[GLOB42]], i32 [[OMP_GLOBAL_THREAD_NUM207]], i32 34, i32* [[P_LASTITER203]], i32* [[P_LOWERBOUND204]], i32* [[P_UPPERBOUND205]], i32* [[P_STRIDE206]], i32 1, i32 1), !dbg [[DBG148]]
266 // CHECK-DEBUG-NEXT:    [[TMP4:%.*]] = load i32, i32* [[P_LOWERBOUND204]], align 4, !dbg [[DBG148]]
267 // CHECK-DEBUG-NEXT:    [[TMP5:%.*]] = load i32, i32* [[P_UPPERBOUND205]], align 4, !dbg [[DBG148]]
268 // CHECK-DEBUG-NEXT:    [[TMP6:%.*]] = sub i32 [[TMP5]], [[TMP4]], !dbg [[DBG148]]
269 // CHECK-DEBUG-NEXT:    [[TMP7:%.*]] = add i32 [[TMP6]], 1, !dbg [[DBG148]]
270 // CHECK-DEBUG-NEXT:    br label [[OMP_LOOP_HEADER191:%.*]], !dbg [[DBG148]]
271 // CHECK-DEBUG:       omp_loop.header191:
272 // CHECK-DEBUG-NEXT:    [[OMP_LOOP_IV197:%.*]] = phi i32 [ 0, [[OMP_LOOP_PREHEADER190]] ], [ [[OMP_LOOP_NEXT199:%.*]], [[OMP_LOOP_INC194:%.*]] ], !dbg [[DBG148]]
273 // CHECK-DEBUG-NEXT:    br label [[OMP_LOOP_COND192:%.*]], !dbg [[DBG148]]
274 // CHECK-DEBUG:       omp_loop.cond192:
275 // CHECK-DEBUG-NEXT:    [[OMP_LOOP_CMP198:%.*]] = icmp ult i32 [[OMP_LOOP_IV197]], [[TMP7]], !dbg [[DBG148]]
276 // CHECK-DEBUG-NEXT:    br i1 [[OMP_LOOP_CMP198]], label [[OMP_LOOP_BODY193:%.*]], label [[OMP_LOOP_EXIT195:%.*]], !dbg [[DBG148]]
277 // CHECK-DEBUG:       omp_loop.body193:
278 // CHECK-DEBUG-NEXT:    [[TMP8:%.*]] = add i32 [[OMP_LOOP_IV197]], [[TMP4]], !dbg [[DBG148]]
279 // CHECK-DEBUG-NEXT:    call void @__captured_stmt.20(i32* [[I185]], i32 [[TMP8]], %struct.anon.18* [[AGG_CAPTURED187]]), !dbg [[DBG148]]
280 // CHECK-DEBUG-NEXT:    [[TMP9:%.*]] = load i32, i32* [[A_ADDR]], align 4, !dbg [[DBG150:![0-9]+]]
281 // CHECK-DEBUG-NEXT:    [[CONV200:%.*]] = sitofp i32 [[TMP9]] to double, !dbg [[DBG150]]
282 // CHECK-DEBUG-NEXT:    [[TMP10:%.*]] = load double, double* [[B_ADDR]], align 8, !dbg [[DBG151:![0-9]+]]
283 // CHECK-DEBUG-NEXT:    [[ADD201:%.*]] = fadd double [[CONV200]], [[TMP10]], !dbg [[DBG152:![0-9]+]]
284 // CHECK-DEBUG-NEXT:    [[CONV202:%.*]] = fptrunc double [[ADD201]] to float, !dbg [[DBG150]]
285 // CHECK-DEBUG-NEXT:    [[TMP11:%.*]] = load float*, float** [[R_ADDR]], align 8, !dbg [[DBG153:![0-9]+]]
286 // CHECK-DEBUG-NEXT:    store float [[CONV202]], float* [[TMP11]], align 4, !dbg [[DBG154:![0-9]+]]
287 // CHECK-DEBUG-NEXT:    br label [[OMP_LOOP_INC194]], !dbg [[DBG148]]
288 // CHECK-DEBUG:       omp_loop.inc194:
289 // CHECK-DEBUG-NEXT:    [[OMP_LOOP_NEXT199]] = add nuw i32 [[OMP_LOOP_IV197]], 1, !dbg [[DBG148]]
290 // CHECK-DEBUG-NEXT:    br label [[OMP_LOOP_HEADER191]], !dbg [[DBG148]]
291 // CHECK-DEBUG:       omp_loop.exit195:
292 // CHECK-DEBUG-NEXT:    call void @__kmpc_for_static_fini(%struct.ident_t* @[[GLOB42]], i32 [[OMP_GLOBAL_THREAD_NUM207]]), !dbg [[DBG148]]
293 // CHECK-DEBUG-NEXT:    [[OMP_GLOBAL_THREAD_NUM208:%.*]] = call i32 @__kmpc_global_thread_num(%struct.ident_t* @[[GLOB42]]), !dbg [[DBG151]]
294 // CHECK-DEBUG-NEXT:    call void @__kmpc_barrier(%struct.ident_t* @[[GLOB43:[0-9]+]], i32 [[OMP_GLOBAL_THREAD_NUM208]]), !dbg [[DBG151]]
295 // CHECK-DEBUG-NEXT:    br label [[OMP_LOOP_AFTER196:%.*]], !dbg [[DBG148]]
296 // CHECK-DEBUG:       omp_loop.after196:
297 // CHECK-DEBUG-NEXT:    ret void, !dbg [[DBG155:![0-9]+]]
298 //
299 void parallel_for_2(float *r, int a, double b) {
300 #pragma omp parallel
301   {
302 #pragma omp for
303     for (int i = 0; i < 100; ++i)
304       *r = a + b;
305 #pragma omp parallel
306     {
307 #pragma omp for
308       for (int i = 0; i < 100; ++i)
309         *r = a + b;
310 #pragma omp parallel
311       {
312 #pragma omp for
313         for (int i = 0; i < 100; ++i)
314           *r = a + b;
315       }
316 #pragma omp for
317       for (int i = 0; i < 100; ++i)
318         *r = a + b;
319 #pragma omp parallel
320       {
321 #pragma omp for
322         for (int i = 0; i < 100; ++i)
323           *r = a + b;
324       }
325 #pragma omp for
326       for (int i = 0; i < 100; ++i)
327         *r = a + b;
328     }
329 #pragma omp for
330     for (int i = 0; i < 100; ++i)
331       *r = a + b;
332   }
333 #pragma omp for
334   for (int i = 0; i < 100; ++i)
335     *r = a + b;
336 }
337 
338 #endif
339