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