365 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: %v6 = load i8, ptr %in6, align 124; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 125; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 126; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 127; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 128; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 129; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 130; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 131; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 132; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 133; SSE2: LV: Found an estimated cost of 56 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 134; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 135; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 136; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 137; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 138; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 139; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load i8, ptr %in6, align 140; SSE2: LV: Found an estimated cost of 0 for VF 2 For instruction: %v7 = load i8, ptr %in7, align 141; SSE2: LV: Found an estimated cost of 120 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 142; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 143; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 144; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 145; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 146; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 147; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load i8, ptr %in6, align 148; SSE2: LV: Found an estimated cost of 0 for VF 4 For instruction: %v7 = load i8, ptr %in7, align 149; SSE2: LV: Found an estimated cost of 248 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 150; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 151; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 152; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 153; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 154; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 155; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load i8, ptr %in6, align 156; SSE2: LV: Found an estimated cost of 0 for VF 8 For instruction: %v7 = load i8, ptr %in7, align 157; SSE2: LV: Found an estimated cost of 504 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 158; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 159; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 160; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 161; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 162; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 163; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v6 = load i8, ptr %in6, align 164; SSE2: LV: Found an estimated cost of 0 for VF 16 For instruction: %v7 = load i8, ptr %in7, align 165;66; AVX1-LABEL: 'test'67; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 168; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 169; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 170; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 171; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 172; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 173; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 174; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 175; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 176; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 177; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 178; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 179; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 180; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 181; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 182; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 183; AVX1: LV: Found an estimated cost of 33 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 184; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 185; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 186; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 187; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 188; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 189; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load i8, ptr %in6, align 190; AVX1: LV: Found an estimated cost of 0 for VF 2 For instruction: %v7 = load i8, ptr %in7, align 191; AVX1: LV: Found an estimated cost of 66 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 192; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 193; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 194; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 195; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 196; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 197; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load i8, ptr %in6, align 198; AVX1: LV: Found an estimated cost of 0 for VF 4 For instruction: %v7 = load i8, ptr %in7, align 199; AVX1: LV: Found an estimated cost of 132 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 1100; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 1101; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 1102; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 1103; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 1104; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 1105; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load i8, ptr %in6, align 1106; AVX1: LV: Found an estimated cost of 0 for VF 8 For instruction: %v7 = load i8, ptr %in7, align 1107; AVX1: LV: Found an estimated cost of 264 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 1108; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 1109; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 1110; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 1111; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 1112; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 1113; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v6 = load i8, ptr %in6, align 1114; AVX1: LV: Found an estimated cost of 0 for VF 16 For instruction: %v7 = load i8, ptr %in7, align 1115; AVX1: LV: Found an estimated cost of 536 for VF 32 For instruction: %v0 = load i8, ptr %in0, align 1116; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i8, ptr %in1, align 1117; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i8, ptr %in2, align 1118; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i8, ptr %in3, align 1119; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i8, ptr %in4, align 1120; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v5 = load i8, ptr %in5, align 1121; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v6 = load i8, ptr %in6, align 1122; AVX1: LV: Found an estimated cost of 0 for VF 32 For instruction: %v7 = load i8, ptr %in7, align 1123;124; AVX2-LABEL: 'test'125; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1126; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1127; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1128; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1129; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1130; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1131; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 1132; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 1133; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1134; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1135; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1136; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1137; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1138; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1139; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 1140; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 1141; AVX2: LV: Found an estimated cost of 4 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 1142; AVX2: LV: Found an estimated cost of 4 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 1143; AVX2: LV: Found an estimated cost of 4 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 1144; AVX2: LV: Found an estimated cost of 4 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 1145; AVX2: LV: Found an estimated cost of 4 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 1146; AVX2: LV: Found an estimated cost of 4 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 1147; AVX2: LV: Found an estimated cost of 4 for VF 2 For instruction: %v6 = load i8, ptr %in6, align 1148; AVX2: LV: Found an estimated cost of 4 for VF 2 For instruction: %v7 = load i8, ptr %in7, align 1149; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 1150; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 1151; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 1152; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 1153; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 1154; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 1155; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: %v6 = load i8, ptr %in6, align 1156; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: %v7 = load i8, ptr %in7, align 1157; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 1158; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 1159; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 1160; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 1161; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 1162; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 1163; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: %v6 = load i8, ptr %in6, align 1164; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: %v7 = load i8, ptr %in7, align 1165; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 1166; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 1167; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 1168; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 1169; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 1170; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 1171; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: %v6 = load i8, ptr %in6, align 1172; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: %v7 = load i8, ptr %in7, align 1173; AVX2: LV: Found an estimated cost of 65 for VF 32 For instruction: %v0 = load i8, ptr %in0, align 1174; AVX2: LV: Found an estimated cost of 65 for VF 32 For instruction: %v1 = load i8, ptr %in1, align 1175; AVX2: LV: Found an estimated cost of 65 for VF 32 For instruction: %v2 = load i8, ptr %in2, align 1176; AVX2: LV: Found an estimated cost of 65 for VF 32 For instruction: %v3 = load i8, ptr %in3, align 1177; AVX2: LV: Found an estimated cost of 65 for VF 32 For instruction: %v4 = load i8, ptr %in4, align 1178; AVX2: LV: Found an estimated cost of 65 for VF 32 For instruction: %v5 = load i8, ptr %in5, align 1179; AVX2: LV: Found an estimated cost of 65 for VF 32 For instruction: %v6 = load i8, ptr %in6, align 1180; AVX2: LV: Found an estimated cost of 65 for VF 32 For instruction: %v7 = load i8, ptr %in7, align 1181;182; AVX512DQ-LABEL: 'test'183; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1184; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1185; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1186; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1187; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1188; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1189; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 1190; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 1191; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1192; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1193; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1194; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1195; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1196; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1197; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 1198; AVX512DQ: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 1199; AVX512DQ: LV: Found an estimated cost of 33 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 1200; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 1201; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 1202; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 1203; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 1204; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 1205; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load i8, ptr %in6, align 1206; AVX512DQ: LV: Found an estimated cost of 0 for VF 2 For instruction: %v7 = load i8, ptr %in7, align 1207; AVX512DQ: LV: Found an estimated cost of 66 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 1208; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 1209; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 1210; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 1211; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 1212; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 1213; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load i8, ptr %in6, align 1214; AVX512DQ: LV: Found an estimated cost of 0 for VF 4 For instruction: %v7 = load i8, ptr %in7, align 1215; AVX512DQ: LV: Found an estimated cost of 132 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 1216; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 1217; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 1218; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 1219; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 1220; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 1221; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load i8, ptr %in6, align 1222; AVX512DQ: LV: Found an estimated cost of 0 for VF 8 For instruction: %v7 = load i8, ptr %in7, align 1223; AVX512DQ: LV: Found an estimated cost of 264 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 1224; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 1225; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 1226; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 1227; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 1228; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 1229; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v6 = load i8, ptr %in6, align 1230; AVX512DQ: LV: Found an estimated cost of 0 for VF 16 For instruction: %v7 = load i8, ptr %in7, align 1231; AVX512DQ: LV: Found an estimated cost of 536 for VF 32 For instruction: %v0 = load i8, ptr %in0, align 1232; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i8, ptr %in1, align 1233; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i8, ptr %in2, align 1234; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i8, ptr %in3, align 1235; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i8, ptr %in4, align 1236; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v5 = load i8, ptr %in5, align 1237; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v6 = load i8, ptr %in6, align 1238; AVX512DQ: LV: Found an estimated cost of 0 for VF 32 For instruction: %v7 = load i8, ptr %in7, align 1239; AVX512DQ: LV: Found an estimated cost of 1080 for VF 64 For instruction: %v0 = load i8, ptr %in0, align 1240; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v1 = load i8, ptr %in1, align 1241; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v2 = load i8, ptr %in2, align 1242; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v3 = load i8, ptr %in3, align 1243; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v4 = load i8, ptr %in4, align 1244; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v5 = load i8, ptr %in5, align 1245; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v6 = load i8, ptr %in6, align 1246; AVX512DQ: LV: Found an estimated cost of 0 for VF 64 For instruction: %v7 = load i8, ptr %in7, align 1247;248; AVX512BW-LABEL: 'test'249; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1250; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1251; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1252; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1253; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1254; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1255; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 1256; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 1257; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i8, ptr %in0, align 1258; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i8, ptr %in1, align 1259; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i8, ptr %in2, align 1260; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i8, ptr %in3, align 1261; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i8, ptr %in4, align 1262; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i8, ptr %in5, align 1263; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i8, ptr %in6, align 1264; AVX512BW: LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i8, ptr %in7, align 1265; AVX512BW: LV: Found an estimated cost of 9 for VF 2 For instruction: %v0 = load i8, ptr %in0, align 1266; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i8, ptr %in1, align 1267; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i8, ptr %in2, align 1268; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i8, ptr %in3, align 1269; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i8, ptr %in4, align 1270; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i8, ptr %in5, align 1271; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load i8, ptr %in6, align 1272; AVX512BW: LV: Found an estimated cost of 0 for VF 2 For instruction: %v7 = load i8, ptr %in7, align 1273; AVX512BW: LV: Found an estimated cost of 33 for VF 4 For instruction: %v0 = load i8, ptr %in0, align 1274; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i8, ptr %in1, align 1275; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i8, ptr %in2, align 1276; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i8, ptr %in3, align 1277; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i8, ptr %in4, align 1278; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i8, ptr %in5, align 1279; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load i8, ptr %in6, align 1280; AVX512BW: LV: Found an estimated cost of 0 for VF 4 For instruction: %v7 = load i8, ptr %in7, align 1281; AVX512BW: LV: Found an estimated cost of 65 for VF 8 For instruction: %v0 = load i8, ptr %in0, align 1282; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i8, ptr %in1, align 1283; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i8, ptr %in2, align 1284; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i8, ptr %in3, align 1285; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i8, ptr %in4, align 1286; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i8, ptr %in5, align 1287; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load i8, ptr %in6, align 1288; AVX512BW: LV: Found an estimated cost of 0 for VF 8 For instruction: %v7 = load i8, ptr %in7, align 1289; AVX512BW: LV: Found an estimated cost of 158 for VF 16 For instruction: %v0 = load i8, ptr %in0, align 1290; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i8, ptr %in1, align 1291; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i8, ptr %in2, align 1292; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i8, ptr %in3, align 1293; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i8, ptr %in4, align 1294; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i8, ptr %in5, align 1295; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v6 = load i8, ptr %in6, align 1296; AVX512BW: LV: Found an estimated cost of 0 for VF 16 For instruction: %v7 = load i8, ptr %in7, align 1297; AVX512BW: LV: Found an estimated cost of 472 for VF 32 For instruction: %v0 = load i8, ptr %in0, align 1298; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i8, ptr %in1, align 1299; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i8, ptr %in2, align 1300; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i8, ptr %in3, align 1301; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i8, ptr %in4, align 1302; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v5 = load i8, ptr %in5, align 1303; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v6 = load i8, ptr %in6, align 1304; AVX512BW: LV: Found an estimated cost of 0 for VF 32 For instruction: %v7 = load i8, ptr %in7, align 1305; AVX512BW: LV: Found an estimated cost of 1100 for VF 64 For instruction: %v0 = load i8, ptr %in0, align 1306; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v1 = load i8, ptr %in1, align 1307; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v2 = load i8, ptr %in2, align 1308; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v3 = load i8, ptr %in3, align 1309; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v4 = load i8, ptr %in4, align 1310; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v5 = load i8, ptr %in5, align 1311; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v6 = load i8, ptr %in6, align 1312; AVX512BW: LV: Found an estimated cost of 0 for VF 64 For instruction: %v7 = load i8, ptr %in7, align 1313;314entry:315 br label %for.body316 317for.body:318 %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]319 320 %iv.0 = add nuw nsw i64 %iv, 0321 %iv.1 = add nuw nsw i64 %iv, 1322 %iv.2 = add nuw nsw i64 %iv, 2323 %iv.3 = add nuw nsw i64 %iv, 3324 %iv.4 = add nuw nsw i64 %iv, 4325 %iv.5 = add nuw nsw i64 %iv, 5326 %iv.6 = add nuw nsw i64 %iv, 6327 %iv.7 = add nuw nsw i64 %iv, 7328 329 %in0 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.0330 %in1 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.1331 %in2 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.2332 %in3 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.3333 %in4 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.4334 %in5 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.5335 %in6 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.6336 %in7 = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv.7337 338 %v0 = load i8, ptr %in0339 %v1 = load i8, ptr %in1340 %v2 = load i8, ptr %in2341 %v3 = load i8, ptr %in3342 %v4 = load i8, ptr %in4343 %v5 = load i8, ptr %in5344 %v6 = load i8, ptr %in6345 %v7 = load i8, ptr %in7346 347 %reduce.add.0 = add i8 %v0, %v1348 %reduce.add.1 = add i8 %reduce.add.0, %v2349 %reduce.add.2 = add i8 %reduce.add.1, %v3350 %reduce.add.3 = add i8 %reduce.add.2, %v4351 %reduce.add.4 = add i8 %reduce.add.3, %v5352 %reduce.add.5 = add i8 %reduce.add.4, %v6353 %reduce.add.6 = add i8 %reduce.add.5, %v7354 355 %out = getelementptr inbounds [1024 x i8], ptr @B, i64 0, i64 %iv.0356 store i8 %reduce.add.6, ptr %out357 358 %iv.next = add nuw nsw i64 %iv.0, 8359 %cmp = icmp ult i64 %iv.next, 1024360 br i1 %cmp, label %for.body, label %for.cond.cleanup361 362for.cond.cleanup:363 ret void364}365