1 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --filter "LV: Found an estimated cost of [0-9]+ for VF [0-9]+ For instruction:\s*%v. = load float, ptr %in."
2 ; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+sse2 --debug-only=loop-vectorize < %s 2>&1 | FileCheck %s --check-prefix=SSE2
3 ; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx --debug-only=loop-vectorize < %s 2>&1 | FileCheck %s --check-prefix=AVX1
4 ; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx2 --debug-only=loop-vectorize < %s 2>&1 | FileCheck %s --check-prefix=AVX2
5 ; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx512vl --debug-only=loop-vectorize < %s 2>&1 | FileCheck %s --check-prefix=AVX512
8 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
9 target triple = "x86_64-unknown-linux-gnu"
11 @A = global [1024 x float] zeroinitializer, align 128
12 @B = global [1024 x i8] zeroinitializer, align 128
16 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, ptr %in0, align 4
17 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load float, ptr %in1, align 4
18 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load float, ptr %in2, align 4
19 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load float, ptr %in3, align 4
20 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load float, ptr %in4, align 4
21 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load float, ptr %in5, align 4
22 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load float, ptr %in6, align 4
23 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, ptr %in0, align 4
24 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load float, ptr %in1, align 4
25 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load float, ptr %in2, align 4
26 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load float, ptr %in3, align 4
27 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load float, ptr %in4, align 4
28 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load float, ptr %in5, align 4
29 ; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load float, ptr %in6, align 4
30 ; SSE2: LV: Found an estimated cost of 21 for VF 2 For instruction: %v0 = load float, ptr %in0, align 4
31 ; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load float, ptr %in1, align 4
32 ; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load float, ptr %in2, align 4
33 ; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load float, ptr %in3, align 4
34 ; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load float, ptr %in4, align 4
35 ; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load float, ptr %in5, align 4
36 ; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load float, ptr %in6, align 4
37 ; SSE2: LV: Found an estimated cost of 49 for VF 4 For instruction: %v0 = load float, ptr %in0, align 4
38 ; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load float, ptr %in1, align 4
39 ; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load float, ptr %in2, align 4
40 ; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load float, ptr %in3, align 4
41 ; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load float, ptr %in4, align 4
42 ; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load float, ptr %in5, align 4
43 ; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load float, ptr %in6, align 4
44 ; SSE2: LV: Found an estimated cost of 98 for VF 8 For instruction: %v0 = load float, ptr %in0, align 4
45 ; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load float, ptr %in1, align 4
46 ; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load float, ptr %in2, align 4
47 ; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load float, ptr %in3, align 4
48 ; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load float, ptr %in4, align 4
49 ; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load float, ptr %in5, align 4
50 ; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load float, ptr %in6, align 4
53 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, ptr %in0, align 4
54 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load float, ptr %in1, align 4
55 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load float, ptr %in2, align 4
56 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load float, ptr %in3, align 4
57 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load float, ptr %in4, align 4
58 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load float, ptr %in5, align 4
59 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load float, ptr %in6, align 4
60 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, ptr %in0, align 4
61 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load float, ptr %in1, align 4
62 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load float, ptr %in2, align 4
63 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load float, ptr %in3, align 4
64 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load float, ptr %in4, align 4
65 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load float, ptr %in5, align 4
66 ; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load float, ptr %in6, align 4
67 ; AVX1: LV: Found an estimated cost of 23 for VF 2 For instruction: %v0 = load float, ptr %in0, align 4
68 ; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load float, ptr %in1, align 4
69 ; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load float, ptr %in2, align 4
70 ; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load float, ptr %in3, align 4
71 ; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load float, ptr %in4, align 4
72 ; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load float, ptr %in5, align 4
73 ; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load float, ptr %in6, align 4
74 ; AVX1: LV: Found an estimated cost of 49 for VF 4 For instruction: %v0 = load float, ptr %in0, align 4
75 ; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load float, ptr %in1, align 4
76 ; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load float, ptr %in2, align 4
77 ; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load float, ptr %in3, align 4
78 ; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load float, ptr %in4, align 4
79 ; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load float, ptr %in5, align 4
80 ; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load float, ptr %in6, align 4
81 ; AVX1: LV: Found an estimated cost of 105 for VF 8 For instruction: %v0 = load float, ptr %in0, align 4
82 ; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load float, ptr %in1, align 4
83 ; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load float, ptr %in2, align 4
84 ; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load float, ptr %in3, align 4
85 ; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load float, ptr %in4, align 4
86 ; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load float, ptr %in5, align 4
87 ; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load float, ptr %in6, align 4
88 ; AVX1: LV: Found an estimated cost of 210 for VF 16 For instruction: %v0 = load float, ptr %in0, align 4
89 ; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load float, ptr %in1, align 4
90 ; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load float, ptr %in2, align 4
91 ; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load float, ptr %in3, align 4
92 ; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load float, ptr %in4, align 4
93 ; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load float, ptr %in5, align 4
94 ; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v6 = load float, ptr %in6, align 4
97 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, ptr %in0, align 4
98 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load float, ptr %in1, align 4
99 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load float, ptr %in2, align 4
100 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load float, ptr %in3, align 4
101 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load float, ptr %in4, align 4
102 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load float, ptr %in5, align 4
103 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load float, ptr %in6, align 4
104 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, ptr %in0, align 4
105 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load float, ptr %in1, align 4
106 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load float, ptr %in2, align 4
107 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load float, ptr %in3, align 4
108 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load float, ptr %in4, align 4
109 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load float, ptr %in5, align 4
110 ; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load float, ptr %in6, align 4
111 ; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v0 = load float, ptr %in0, align 4
112 ; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v1 = load float, ptr %in1, align 4
113 ; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v2 = load float, ptr %in2, align 4
114 ; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v3 = load float, ptr %in3, align 4
115 ; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v4 = load float, ptr %in4, align 4
116 ; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v5 = load float, ptr %in5, align 4
117 ; AVX2: LV: Found an estimated cost of 3 for VF 2 For instruction: %v6 = load float, ptr %in6, align 4
118 ; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v0 = load float, ptr %in0, align 4
119 ; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v1 = load float, ptr %in1, align 4
120 ; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v2 = load float, ptr %in2, align 4
121 ; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v3 = load float, ptr %in3, align 4
122 ; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v4 = load float, ptr %in4, align 4
123 ; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v5 = load float, ptr %in5, align 4
124 ; AVX2: LV: Found an estimated cost of 7 for VF 4 For instruction: %v6 = load float, ptr %in6, align 4
125 ; AVX2: LV: Found an estimated cost of 15 for VF 8 For instruction: %v0 = load float, ptr %in0, align 4
126 ; AVX2: LV: Found an estimated cost of 15 for VF 8 For instruction: %v1 = load float, ptr %in1, align 4
127 ; AVX2: LV: Found an estimated cost of 15 for VF 8 For instruction: %v2 = load float, ptr %in2, align 4
128 ; AVX2: LV: Found an estimated cost of 15 for VF 8 For instruction: %v3 = load float, ptr %in3, align 4
129 ; AVX2: LV: Found an estimated cost of 15 for VF 8 For instruction: %v4 = load float, ptr %in4, align 4
130 ; AVX2: LV: Found an estimated cost of 15 for VF 8 For instruction: %v5 = load float, ptr %in5, align 4
131 ; AVX2: LV: Found an estimated cost of 15 for VF 8 For instruction: %v6 = load float, ptr %in6, align 4
132 ; AVX2: LV: Found an estimated cost of 30 for VF 16 For instruction: %v0 = load float, ptr %in0, align 4
133 ; AVX2: LV: Found an estimated cost of 30 for VF 16 For instruction: %v1 = load float, ptr %in1, align 4
134 ; AVX2: LV: Found an estimated cost of 30 for VF 16 For instruction: %v2 = load float, ptr %in2, align 4
135 ; AVX2: LV: Found an estimated cost of 30 for VF 16 For instruction: %v3 = load float, ptr %in3, align 4
136 ; AVX2: LV: Found an estimated cost of 30 for VF 16 For instruction: %v4 = load float, ptr %in4, align 4
137 ; AVX2: LV: Found an estimated cost of 30 for VF 16 For instruction: %v5 = load float, ptr %in5, align 4
138 ; AVX2: LV: Found an estimated cost of 30 for VF 16 For instruction: %v6 = load float, ptr %in6, align 4
140 ; AVX512-LABEL: 'test'
141 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, ptr %in0, align 4
142 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load float, ptr %in1, align 4
143 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load float, ptr %in2, align 4
144 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load float, ptr %in3, align 4
145 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load float, ptr %in4, align 4
146 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load float, ptr %in5, align 4
147 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load float, ptr %in6, align 4
148 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load float, ptr %in0, align 4
149 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load float, ptr %in1, align 4
150 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load float, ptr %in2, align 4
151 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load float, ptr %in3, align 4
152 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load float, ptr %in4, align 4
153 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load float, ptr %in5, align 4
154 ; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load float, ptr %in6, align 4
155 ; AVX512: LV: Found an estimated cost of 8 for VF 2 For instruction: %v0 = load float, ptr %in0, align 4
156 ; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load float, ptr %in1, align 4
157 ; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load float, ptr %in2, align 4
158 ; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load float, ptr %in3, align 4
159 ; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load float, ptr %in4, align 4
160 ; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load float, ptr %in5, align 4
161 ; AVX512: LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load float, ptr %in6, align 4
162 ; AVX512: LV: Found an estimated cost of 12 for VF 4 For instruction: %v0 = load float, ptr %in0, align 4
163 ; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load float, ptr %in1, align 4
164 ; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load float, ptr %in2, align 4
165 ; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load float, ptr %in3, align 4
166 ; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load float, ptr %in4, align 4
167 ; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load float, ptr %in5, align 4
168 ; AVX512: LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load float, ptr %in6, align 4
169 ; AVX512: LV: Found an estimated cost of 35 for VF 8 For instruction: %v0 = load float, ptr %in0, align 4
170 ; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load float, ptr %in1, align 4
171 ; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load float, ptr %in2, align 4
172 ; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load float, ptr %in3, align 4
173 ; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load float, ptr %in4, align 4
174 ; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load float, ptr %in5, align 4
175 ; AVX512: LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load float, ptr %in6, align 4
176 ; AVX512: LV: Found an estimated cost of 70 for VF 16 For instruction: %v0 = load float, ptr %in0, align 4
177 ; AVX512: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load float, ptr %in1, align 4
178 ; AVX512: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load float, ptr %in2, align 4
179 ; AVX512: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load float, ptr %in3, align 4
180 ; AVX512: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load float, ptr %in4, align 4
181 ; AVX512: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load float, ptr %in5, align 4
182 ; AVX512: LV: Found an estimated cost of 0 for VF 16 For instruction: %v6 = load float, ptr %in6, align 4
183 ; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v0 = load float, ptr %in0, align 4
184 ; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v1 = load float, ptr %in1, align 4
185 ; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v2 = load float, ptr %in2, align 4
186 ; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v3 = load float, ptr %in3, align 4
187 ; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v4 = load float, ptr %in4, align 4
188 ; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v5 = load float, ptr %in5, align 4
189 ; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: %v6 = load float, ptr %in6, align 4
195 %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
197 %iv.0 = add nuw nsw i64 %iv, 0
198 %iv.1 = add nuw nsw i64 %iv, 1
199 %iv.2 = add nuw nsw i64 %iv, 2
200 %iv.3 = add nuw nsw i64 %iv, 3
201 %iv.4 = add nuw nsw i64 %iv, 4
202 %iv.5 = add nuw nsw i64 %iv, 5
203 %iv.6 = add nuw nsw i64 %iv, 6
205 %in0 = getelementptr inbounds [1024 x float], ptr @A, i64 0, i64 %iv.0
206 %in1 = getelementptr inbounds [1024 x float], ptr @A, i64 0, i64 %iv.1
207 %in2 = getelementptr inbounds [1024 x float], ptr @A, i64 0, i64 %iv.2
208 %in3 = getelementptr inbounds [1024 x float], ptr @A, i64 0, i64 %iv.3
209 %in4 = getelementptr inbounds [1024 x float], ptr @A, i64 0, i64 %iv.4
210 %in5 = getelementptr inbounds [1024 x float], ptr @A, i64 0, i64 %iv.5
211 %in6 = getelementptr inbounds [1024 x float], ptr @A, i64 0, i64 %iv.6
213 %v0 = load float, ptr %in0
214 %v1 = load float, ptr %in1
215 %v2 = load float, ptr %in2
216 %v3 = load float, ptr %in3
217 %v4 = load float, ptr %in4
218 %v5 = load float, ptr %in5
219 %v6 = load float, ptr %in6
221 %reduce.add.0 = fadd float %v0, %v1
222 %reduce.add.1 = fadd float %reduce.add.0, %v2
223 %reduce.add.2 = fadd float %reduce.add.1, %v3
224 %reduce.add.3 = fadd float %reduce.add.2, %v4
225 %reduce.add.4 = fadd float %reduce.add.3, %v5
226 %reduce.add.5 = fadd float %reduce.add.4, %v6
228 %reduce.add.5.narrow = fptoui float %reduce.add.5 to i8
230 %out = getelementptr inbounds [1024 x i8], ptr @B, i64 0, i64 %iv.0
231 store i8 %reduce.add.5.narrow, ptr %out
233 %iv.next = add nuw nsw i64 %iv.0, 7
234 %cmp = icmp ult i64 %iv.next, 1024
235 br i1 %cmp, label %for.body, label %for.cond.cleanup