brintos

brintos / llvm-project-archived public Read only

0
0
Text · 21.0 KiB · 60db7a5 Raw
247 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 i16, 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 i16] 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 i16, ptr %in0, align 218; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 219; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 220; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 221; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 222; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 223; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 224; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 225; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 226; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 227; SSE2:  LV: Found an estimated cost of 22 for VF 2 For instruction: %v0 = load i16, ptr %in0, align 228; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i16, ptr %in1, align 229; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i16, ptr %in2, align 230; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i16, ptr %in3, align 231; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i16, ptr %in4, align 232; SSE2:  LV: Found an estimated cost of 43 for VF 4 For instruction: %v0 = load i16, ptr %in0, align 233; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i16, ptr %in1, align 234; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i16, ptr %in2, align 235; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i16, ptr %in3, align 236; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i16, ptr %in4, align 237; SSE2:  LV: Found an estimated cost of 85 for VF 8 For instruction: %v0 = load i16, ptr %in0, align 238; SSE2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i16, ptr %in1, align 239; SSE2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i16, ptr %in2, align 240; SSE2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i16, ptr %in3, align 241; SSE2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i16, ptr %in4, align 242; SSE2:  LV: Found an estimated cost of 170 for VF 16 For instruction: %v0 = load i16, ptr %in0, align 243; SSE2:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i16, ptr %in1, align 244; SSE2:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i16, ptr %in2, align 245; SSE2:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i16, ptr %in3, align 246; SSE2:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i16, ptr %in4, align 247;48; AVX1-LABEL: 'test'49; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 250; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 251; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 252; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 253; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 254; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 255; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 256; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 257; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 258; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 259; AVX1:  LV: Found an estimated cost of 25 for VF 2 For instruction: %v0 = load i16, ptr %in0, align 260; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i16, ptr %in1, align 261; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i16, ptr %in2, align 262; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i16, ptr %in3, align 263; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i16, ptr %in4, align 264; AVX1:  LV: Found an estimated cost of 43 for VF 4 For instruction: %v0 = load i16, ptr %in0, align 265; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i16, ptr %in1, align 266; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i16, ptr %in2, align 267; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i16, ptr %in3, align 268; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i16, ptr %in4, align 269; AVX1:  LV: Found an estimated cost of 85 for VF 8 For instruction: %v0 = load i16, ptr %in0, align 270; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i16, ptr %in1, align 271; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i16, ptr %in2, align 272; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i16, ptr %in3, align 273; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i16, ptr %in4, align 274; AVX1:  LV: Found an estimated cost of 175 for VF 16 For instruction: %v0 = load i16, ptr %in0, align 275; AVX1:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i16, ptr %in1, align 276; AVX1:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i16, ptr %in2, align 277; AVX1:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i16, ptr %in3, align 278; AVX1:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i16, ptr %in4, align 279; AVX1:  LV: Found an estimated cost of 350 for VF 32 For instruction: %v0 = load i16, ptr %in0, align 280; AVX1:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i16, ptr %in1, align 281; AVX1:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i16, ptr %in2, align 282; AVX1:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i16, ptr %in3, align 283; AVX1:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i16, ptr %in4, align 284;85; AVX2-LABEL: 'test'86; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 287; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 288; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 289; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 290; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 291; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 292; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 293; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 294; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 295; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 296; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v0 = load i16, ptr %in0, align 297; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v1 = load i16, ptr %in1, align 298; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v2 = load i16, ptr %in2, align 299; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v3 = load i16, ptr %in3, align 2100; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v4 = load i16, ptr %in4, align 2101; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v0 = load i16, ptr %in0, align 2102; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v1 = load i16, ptr %in1, align 2103; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v2 = load i16, ptr %in2, align 2104; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v3 = load i16, ptr %in3, align 2105; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v4 = load i16, ptr %in4, align 2106; AVX2:  LV: Found an estimated cost of 16 for VF 8 For instruction: %v0 = load i16, ptr %in0, align 2107; AVX2:  LV: Found an estimated cost of 16 for VF 8 For instruction: %v1 = load i16, ptr %in1, align 2108; AVX2:  LV: Found an estimated cost of 16 for VF 8 For instruction: %v2 = load i16, ptr %in2, align 2109; AVX2:  LV: Found an estimated cost of 16 for VF 8 For instruction: %v3 = load i16, ptr %in3, align 2110; AVX2:  LV: Found an estimated cost of 16 for VF 8 For instruction: %v4 = load i16, ptr %in4, align 2111; AVX2:  LV: Found an estimated cost of 33 for VF 16 For instruction: %v0 = load i16, ptr %in0, align 2112; AVX2:  LV: Found an estimated cost of 33 for VF 16 For instruction: %v1 = load i16, ptr %in1, align 2113; AVX2:  LV: Found an estimated cost of 33 for VF 16 For instruction: %v2 = load i16, ptr %in2, align 2114; AVX2:  LV: Found an estimated cost of 33 for VF 16 For instruction: %v3 = load i16, ptr %in3, align 2115; AVX2:  LV: Found an estimated cost of 33 for VF 16 For instruction: %v4 = load i16, ptr %in4, align 2116; AVX2:  LV: Found an estimated cost of 66 for VF 32 For instruction: %v0 = load i16, ptr %in0, align 2117; AVX2:  LV: Found an estimated cost of 66 for VF 32 For instruction: %v1 = load i16, ptr %in1, align 2118; AVX2:  LV: Found an estimated cost of 66 for VF 32 For instruction: %v2 = load i16, ptr %in2, align 2119; AVX2:  LV: Found an estimated cost of 66 for VF 32 For instruction: %v3 = load i16, ptr %in3, align 2120; AVX2:  LV: Found an estimated cost of 66 for VF 32 For instruction: %v4 = load i16, ptr %in4, align 2121;122; AVX512DQ-LABEL: 'test'123; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 2124; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 2125; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 2126; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 2127; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 2128; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 2129; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 2130; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 2131; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 2132; AVX512DQ:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 2133; AVX512DQ:  LV: Found an estimated cost of 25 for VF 2 For instruction: %v0 = load i16, ptr %in0, align 2134; AVX512DQ:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i16, ptr %in1, align 2135; AVX512DQ:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i16, ptr %in2, align 2136; AVX512DQ:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i16, ptr %in3, align 2137; AVX512DQ:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i16, ptr %in4, align 2138; AVX512DQ:  LV: Found an estimated cost of 45 for VF 4 For instruction: %v0 = load i16, ptr %in0, align 2139; AVX512DQ:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i16, ptr %in1, align 2140; AVX512DQ:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i16, ptr %in2, align 2141; AVX512DQ:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i16, ptr %in3, align 2142; AVX512DQ:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i16, ptr %in4, align 2143; AVX512DQ:  LV: Found an estimated cost of 85 for VF 8 For instruction: %v0 = load i16, ptr %in0, align 2144; AVX512DQ:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i16, ptr %in1, align 2145; AVX512DQ:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i16, ptr %in2, align 2146; AVX512DQ:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i16, ptr %in3, align 2147; AVX512DQ:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i16, ptr %in4, align 2148; AVX512DQ:  LV: Found an estimated cost of 175 for VF 16 For instruction: %v0 = load i16, ptr %in0, align 2149; AVX512DQ:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i16, ptr %in1, align 2150; AVX512DQ:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i16, ptr %in2, align 2151; AVX512DQ:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i16, ptr %in3, align 2152; AVX512DQ:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i16, ptr %in4, align 2153; AVX512DQ:  LV: Found an estimated cost of 355 for VF 32 For instruction: %v0 = load i16, ptr %in0, align 2154; AVX512DQ:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i16, ptr %in1, align 2155; AVX512DQ:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i16, ptr %in2, align 2156; AVX512DQ:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i16, ptr %in3, align 2157; AVX512DQ:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i16, ptr %in4, align 2158; AVX512DQ:  LV: Found an estimated cost of 710 for VF 64 For instruction: %v0 = load i16, ptr %in0, align 2159; AVX512DQ:  LV: Found an estimated cost of 0 for VF 64 For instruction: %v1 = load i16, ptr %in1, align 2160; AVX512DQ:  LV: Found an estimated cost of 0 for VF 64 For instruction: %v2 = load i16, ptr %in2, align 2161; AVX512DQ:  LV: Found an estimated cost of 0 for VF 64 For instruction: %v3 = load i16, ptr %in3, align 2162; AVX512DQ:  LV: Found an estimated cost of 0 for VF 64 For instruction: %v4 = load i16, ptr %in4, align 2163;164; AVX512BW-LABEL: 'test'165; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 2166; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 2167; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 2168; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 2169; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 2170; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i16, ptr %in0, align 2171; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i16, ptr %in1, align 2172; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i16, ptr %in2, align 2173; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i16, ptr %in3, align 2174; AVX512BW:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i16, ptr %in4, align 2175; AVX512BW:  LV: Found an estimated cost of 11 for VF 2 For instruction: %v0 = load i16, ptr %in0, align 2176; AVX512BW:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i16, ptr %in1, align 2177; AVX512BW:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i16, ptr %in2, align 2178; AVX512BW:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i16, ptr %in3, align 2179; AVX512BW:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i16, ptr %in4, align 2180; AVX512BW:  LV: Found an estimated cost of 11 for VF 4 For instruction: %v0 = load i16, ptr %in0, align 2181; AVX512BW:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i16, ptr %in1, align 2182; AVX512BW:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i16, ptr %in2, align 2183; AVX512BW:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i16, ptr %in3, align 2184; AVX512BW:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i16, ptr %in4, align 2185; AVX512BW:  LV: Found an estimated cost of 14 for VF 8 For instruction: %v0 = load i16, ptr %in0, align 2186; AVX512BW:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i16, ptr %in1, align 2187; AVX512BW:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i16, ptr %in2, align 2188; AVX512BW:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i16, ptr %in3, align 2189; AVX512BW:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i16, ptr %in4, align 2190; AVX512BW:  LV: Found an estimated cost of 28 for VF 16 For instruction: %v0 = load i16, ptr %in0, align 2191; AVX512BW:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i16, ptr %in1, align 2192; AVX512BW:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i16, ptr %in2, align 2193; AVX512BW:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i16, ptr %in3, align 2194; AVX512BW:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i16, ptr %in4, align 2195; AVX512BW:  LV: Found an estimated cost of 55 for VF 32 For instruction: %v0 = load i16, ptr %in0, align 2196; AVX512BW:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v1 = load i16, ptr %in1, align 2197; AVX512BW:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v2 = load i16, ptr %in2, align 2198; AVX512BW:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v3 = load i16, ptr %in3, align 2199; AVX512BW:  LV: Found an estimated cost of 0 for VF 32 For instruction: %v4 = load i16, ptr %in4, align 2200; AVX512BW:  LV: Found an estimated cost of 235 for VF 64 For instruction: %v0 = load i16, ptr %in0, align 2201; AVX512BW:  LV: Found an estimated cost of 0 for VF 64 For instruction: %v1 = load i16, ptr %in1, align 2202; AVX512BW:  LV: Found an estimated cost of 0 for VF 64 For instruction: %v2 = load i16, ptr %in2, align 2203; AVX512BW:  LV: Found an estimated cost of 0 for VF 64 For instruction: %v3 = load i16, ptr %in3, align 2204; AVX512BW:  LV: Found an estimated cost of 0 for VF 64 For instruction: %v4 = load i16, ptr %in4, align 2205;206entry:207  br label %for.body208 209for.body:210  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]211 212  %iv.0 = add nuw nsw i64 %iv, 0213  %iv.1 = add nuw nsw i64 %iv, 1214  %iv.2 = add nuw nsw i64 %iv, 2215  %iv.3 = add nuw nsw i64 %iv, 3216  %iv.4 = add nuw nsw i64 %iv, 4217 218  %in0 = getelementptr inbounds [1024 x i16], ptr @A, i64 0, i64 %iv.0219  %in1 = getelementptr inbounds [1024 x i16], ptr @A, i64 0, i64 %iv.1220  %in2 = getelementptr inbounds [1024 x i16], ptr @A, i64 0, i64 %iv.2221  %in3 = getelementptr inbounds [1024 x i16], ptr @A, i64 0, i64 %iv.3222  %in4 = getelementptr inbounds [1024 x i16], ptr @A, i64 0, i64 %iv.4223 224  %v0 = load i16, ptr %in0225  %v1 = load i16, ptr %in1226  %v2 = load i16, ptr %in2227  %v3 = load i16, ptr %in3228  %v4 = load i16, ptr %in4229 230  %reduce.add.0 = add i16 %v0, %v1231  %reduce.add.1 = add i16 %reduce.add.0, %v2232  %reduce.add.2 = add i16 %reduce.add.1, %v3233  %reduce.add.3 = add i16 %reduce.add.2, %v4234 235  %reduce.add.3.narrow = trunc i16 %reduce.add.3 to i8236 237  %out = getelementptr inbounds [1024 x i8], ptr @B, i64 0, i64 %iv.0238  store i8 %reduce.add.3.narrow, ptr %out239 240  %iv.next = add nuw nsw i64 %iv.0, 5241  %cmp = icmp ult i64 %iv.next, 1024242  br i1 %cmp, label %for.body, label %for.cond.cleanup243 244for.cond.cleanup:245  ret void246}247