285 lines · plain
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 i8, ptr %in."2; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+sse2 --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefix=SSE23; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefix=AVX14; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx2 --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefix=AVX25; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx512vl --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefix=AVX512DQ6; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx512vl,+avx512bw --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefix=AVX512BW7; REQUIRES: asserts8 9target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"10target triple = "x86_64-unknown-linux-gnu"11 12@A = global [1024 x i8] zeroinitializer, align 12813@B = global [1024 x i8] zeroinitializer, align 12814 15define void @test() {16; SSE2-LABEL: 'test'17; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 118; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 119; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 120; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 121; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 122; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 123; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 124; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 125; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 126; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 127; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 128; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 129; SSE2: LV: Found an estimated cost of 47 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 130; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 131; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 132; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 133; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 134; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 135; SSE2: LV: Found an estimated cost of 90 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 136; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 137; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 138; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 139; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 140; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 141; SSE2: LV: Found an estimated cost of 186 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 142; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 143; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 144; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 145; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 146; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 147; SSE2: LV: Found an estimated cost of 378 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 148; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 149; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 150; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 151; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 152; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 153;54; AVX1-LABEL: 'test'55; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 156; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 157; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 158; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 159; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 160; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 161; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 162; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 163; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 164; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 165; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 166; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 167; AVX1: LV: Found an estimated cost of 27 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 168; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 169; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 170; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 171; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 172; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 173; AVX1: LV: Found an estimated cost of 52 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 174; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 175; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 176; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 177; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 178; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 179; AVX1: LV: Found an estimated cost of 99 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 180; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 181; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 182; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 183; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 184; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 185; AVX1: LV: Found an estimated cost of 198 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 186; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 187; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 188; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 189; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 190; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 191; AVX1: LV: Found an estimated cost of 402 for VF 32 For instruction: %v0 = load i8, ptr %in0, align 192; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i8, ptr %in1, align 193; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i8, ptr %in2, align 194; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i8, ptr %in3, align 195; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i8, ptr %in4, align 196; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v5 = load i8, ptr %in5, align 197;98; AVX2-LABEL: 'test'99; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1100; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1101; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1102; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1103; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1104; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1105; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1106; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1107; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1108; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1109; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1110; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1111; AVX2: LV: Found an estimated cost of 9 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 1112; AVX2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 1113; AVX2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 1114; AVX2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 1115; AVX2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 1116; AVX2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 1117; AVX2: LV: Found an estimated cost of 17 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 1118; AVX2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 1119; AVX2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 1120; AVX2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 1121; AVX2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 1122; AVX2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 1123; AVX2: LV: Found an estimated cost of 20 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 1124; AVX2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 1125; AVX2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 1126; AVX2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 1127; AVX2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 1128; AVX2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 1129; AVX2: LV: Found an estimated cost of 46 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 1130; AVX2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 1131; AVX2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 1132; AVX2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 1133; AVX2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 1134; AVX2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 1135; AVX2: LV: Found an estimated cost of 88 for VF 32 For instruction: %v0 = load i8, ptr %in0, align 1136; AVX2: LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i8, ptr %in1, align 1137; AVX2: LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i8, ptr %in2, align 1138; AVX2: LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i8, ptr %in3, align 1139; AVX2: LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i8, ptr %in4, align 1140; AVX2: LV: Found an estimated cost of 0 for VF 32 For instruction: %v5 = load i8, ptr %in5, align 1141;142; AVX512DQ-LABEL: 'test'143; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1144; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1145; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1146; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1147; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1148; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1149; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1150; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1151; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1152; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1153; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1154; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1155; AVX512DQ: LV: Found an estimated cost of 9 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 1156; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 1157; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 1158; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 1159; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 1160; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 1161; AVX512DQ: LV: Found an estimated cost of 17 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 1162; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 1163; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 1164; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 1165; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 1166; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 1167; AVX512DQ: LV: Found an estimated cost of 21 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 1168; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 1169; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 1170; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 1171; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 1172; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 1173; AVX512DQ: LV: Found an estimated cost of 45 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 1174; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 1175; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 1176; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 1177; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 1178; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 1179; AVX512DQ: LV: Found an estimated cost of 85 for VF 32 For instruction: %v0 = load i8, ptr %in0, align 1180; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i8, ptr %in1, align 1181; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i8, ptr %in2, align 1182; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i8, ptr %in3, align 1183; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i8, ptr %in4, align 1184; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v5 = load i8, ptr %in5, align 1185; AVX512DQ: LV: Found an estimated cost of 810 for VF 64 For instruction: %v0 = load i8, ptr %in0, align 1186; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v1 = load i8, ptr %in1, align 1187; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v2 = load i8, ptr %in2, align 1188; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v3 = load i8, ptr %in3, align 1189; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v4 = load i8, ptr %in4, align 1190; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v5 = load i8, ptr %in5, align 1191;192; AVX512BW-LABEL: 'test'193; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1194; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1195; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1196; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1197; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1198; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1199; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1200; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1201; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1202; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1203; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1204; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1205; AVX512BW: LV: Found an estimated cost of 7 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 1206; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 1207; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 1208; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 1209; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 1210; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 1211; AVX512BW: LV: Found an estimated cost of 25 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 1212; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 1213; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 1214; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 1215; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 1216; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 1217; AVX512BW: LV: Found an estimated cost of 49 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 1218; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 1219; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 1220; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 1221; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 1222; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 1223; AVX512BW: LV: Found an estimated cost of 119 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 1224; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 1225; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 1226; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 1227; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 1228; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 1229; AVX512BW: LV: Found an estimated cost of 237 for VF 32 For instruction: %v0 = load i8, ptr %in0, align 1230; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i8, ptr %in1, align 1231; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i8, ptr %in2, align 1232; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i8, ptr %in3, align 1233; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i8, ptr %in4, align 1234; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v5 = load i8, ptr %in5, align 1235; AVX512BW: LV: Found an estimated cost of 591 for VF 64 For instruction: %v0 = load i8, ptr %in0, align 1236; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v1 = load i8, ptr %in1, align 1237; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v2 = load i8, ptr %in2, align 1238; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v3 = load i8, ptr %in3, align 1239; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v4 = load i8, ptr %in4, align 1240; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v5 = load i8, ptr %in5, align 1241;242entry:243 br label %for.body244 245for.body:246 %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]247 248 %iv.0 = add nuw nsw i64 %iv, 0249 %iv.1 = add nuw nsw i64 %iv, 1250 %iv.2 = add nuw nsw i64 %iv, 2251 %iv.3 = add nuw nsw i64 %iv, 3252 %iv.4 = add nuw nsw i64 %iv, 4253 %iv.5 = add nuw nsw i64 %iv, 5254 255 %in0 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.0256 %in1 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.1257 %in2 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.2258 %in3 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.3259 %in4 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.4260 %in5 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.5261 262 %v0 = load i8, ptr %in0263 %v1 = load i8, ptr %in1264 %v2 = load i8, ptr %in2265 %v3 = load i8, ptr %in3266 %v4 = load i8, ptr %in4267 %v5 = load i8, ptr %in5268 269 %reduce.add.0 = add i8 %v0, %v1270 %reduce.add.1 = add i8 %reduce.add.0, %v2271 %reduce.add.2 = add i8 %reduce.add.1, %v3272 %reduce.add.3 = add i8 %reduce.add.2, %v4273 %reduce.add.4 = add i8 %reduce.add.3, %v5274 275 %out = getelementptr inbounds [1024 x i8], ptr @B, i64 0, i64 %iv.0276 store i8 %reduce.add.4, ptr %out277 278 %iv.next = add nuw nsw i64 %iv.0, 6279 %cmp = icmp ult i64 %iv.next, 1024280 br i1 %cmp, label %for.body, label %for.cond.cleanup281 282for.cond.cleanup:283 ret void284}285