brintos

brintos / llvm-project-archived public Read only

0
0
Text · 20.6 KiB · 7534695 Raw
252 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 i32, 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=AVX5126; REQUIRES: asserts7 8target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"9target triple = "x86_64-unknown-linux-gnu"10 11@A = global [1024 x i32] zeroinitializer, align 12812@B = global [1024 x i8] zeroinitializer, align 12813 14define void @test() {15; SSE2-LABEL: 'test'16; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, ptr %in0, align 417; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i32, ptr %in1, align 418; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i32, ptr %in2, align 419; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i32, ptr %in3, align 420; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i32, ptr %in4, align 421; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i32, ptr %in5, align 422; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i32, ptr %in6, align 423; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i32, ptr %in7, align 424; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, ptr %in0, align 425; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i32, ptr %in1, align 426; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i32, ptr %in2, align 427; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i32, ptr %in3, align 428; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i32, ptr %in4, align 429; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i32, ptr %in5, align 430; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i32, ptr %in6, align 431; SSE2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i32, ptr %in7, align 432; SSE2:  LV: Found an estimated cost of 56 for VF 2 For instruction: %v0 = load i32, ptr %in0, align 433; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i32, ptr %in1, align 434; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i32, ptr %in2, align 435; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i32, ptr %in3, align 436; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i32, ptr %in4, align 437; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i32, ptr %in5, align 438; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load i32, ptr %in6, align 439; SSE2:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v7 = load i32, ptr %in7, align 440; SSE2:  LV: Found an estimated cost of 120 for VF 4 For instruction: %v0 = load i32, ptr %in0, align 441; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i32, ptr %in1, align 442; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i32, ptr %in2, align 443; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i32, ptr %in3, align 444; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i32, ptr %in4, align 445; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i32, ptr %in5, align 446; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load i32, ptr %in6, align 447; SSE2:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v7 = load i32, ptr %in7, align 448;49; AVX1-LABEL: 'test'50; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, ptr %in0, align 451; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i32, ptr %in1, align 452; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i32, ptr %in2, align 453; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i32, ptr %in3, align 454; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i32, ptr %in4, align 455; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i32, ptr %in5, align 456; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i32, ptr %in6, align 457; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i32, ptr %in7, align 458; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, ptr %in0, align 459; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i32, ptr %in1, align 460; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i32, ptr %in2, align 461; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i32, ptr %in3, align 462; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i32, ptr %in4, align 463; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i32, ptr %in5, align 464; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i32, ptr %in6, align 465; AVX1:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i32, ptr %in7, align 466; AVX1:  LV: Found an estimated cost of 36 for VF 2 For instruction: %v0 = load i32, ptr %in0, align 467; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i32, ptr %in1, align 468; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i32, ptr %in2, align 469; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i32, ptr %in3, align 470; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i32, ptr %in4, align 471; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i32, ptr %in5, align 472; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load i32, ptr %in6, align 473; AVX1:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v7 = load i32, ptr %in7, align 474; AVX1:  LV: Found an estimated cost of 72 for VF 4 For instruction: %v0 = load i32, ptr %in0, align 475; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i32, ptr %in1, align 476; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i32, ptr %in2, align 477; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i32, ptr %in3, align 478; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i32, ptr %in4, align 479; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i32, ptr %in5, align 480; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load i32, ptr %in6, align 481; AVX1:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v7 = load i32, ptr %in7, align 482; AVX1:  LV: Found an estimated cost of 152 for VF 8 For instruction: %v0 = load i32, ptr %in0, align 483; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i32, ptr %in1, align 484; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i32, ptr %in2, align 485; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i32, ptr %in3, align 486; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i32, ptr %in4, align 487; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i32, ptr %in5, align 488; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load i32, ptr %in6, align 489; AVX1:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v7 = load i32, ptr %in7, align 490;91; AVX2-LABEL: 'test'92; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, ptr %in0, align 493; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i32, ptr %in1, align 494; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i32, ptr %in2, align 495; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i32, ptr %in3, align 496; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i32, ptr %in4, align 497; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i32, ptr %in5, align 498; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i32, ptr %in6, align 499; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i32, ptr %in7, align 4100; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, ptr %in0, align 4101; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i32, ptr %in1, align 4102; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i32, ptr %in2, align 4103; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i32, ptr %in3, align 4104; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i32, ptr %in4, align 4105; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i32, ptr %in5, align 4106; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i32, ptr %in6, align 4107; AVX2:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i32, ptr %in7, align 4108; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v0 = load i32, ptr %in0, align 4109; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v1 = load i32, ptr %in1, align 4110; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v2 = load i32, ptr %in2, align 4111; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v3 = load i32, ptr %in3, align 4112; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v4 = load i32, ptr %in4, align 4113; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v5 = load i32, ptr %in5, align 4114; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v6 = load i32, ptr %in6, align 4115; AVX2:  LV: Found an estimated cost of 4 for VF 2 For instruction: %v7 = load i32, ptr %in7, align 4116; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v0 = load i32, ptr %in0, align 4117; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v1 = load i32, ptr %in1, align 4118; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v2 = load i32, ptr %in2, align 4119; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v3 = load i32, ptr %in3, align 4120; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v4 = load i32, ptr %in4, align 4121; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v5 = load i32, ptr %in5, align 4122; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v6 = load i32, ptr %in6, align 4123; AVX2:  LV: Found an estimated cost of 8 for VF 4 For instruction: %v7 = load i32, ptr %in7, align 4124; AVX2:  LV: Found an estimated cost of 48 for VF 8 For instruction: %v0 = load i32, ptr %in0, align 4125; AVX2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i32, ptr %in1, align 4126; AVX2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i32, ptr %in2, align 4127; AVX2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i32, ptr %in3, align 4128; AVX2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i32, ptr %in4, align 4129; AVX2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i32, ptr %in5, align 4130; AVX2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load i32, ptr %in6, align 4131; AVX2:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v7 = load i32, ptr %in7, align 4132; AVX2:  LV: Found an estimated cost of 34 for VF 16 For instruction: %v0 = load i32, ptr %in0, align 4133; AVX2:  LV: Found an estimated cost of 34 for VF 16 For instruction: %v1 = load i32, ptr %in1, align 4134; AVX2:  LV: Found an estimated cost of 34 for VF 16 For instruction: %v2 = load i32, ptr %in2, align 4135; AVX2:  LV: Found an estimated cost of 34 for VF 16 For instruction: %v3 = load i32, ptr %in3, align 4136; AVX2:  LV: Found an estimated cost of 34 for VF 16 For instruction: %v4 = load i32, ptr %in4, align 4137; AVX2:  LV: Found an estimated cost of 34 for VF 16 For instruction: %v5 = load i32, ptr %in5, align 4138; AVX2:  LV: Found an estimated cost of 34 for VF 16 For instruction: %v6 = load i32, ptr %in6, align 4139; AVX2:  LV: Found an estimated cost of 34 for VF 16 For instruction: %v7 = load i32, ptr %in7, align 4140; AVX2:  LV: Found an estimated cost of 68 for VF 32 For instruction: %v0 = load i32, ptr %in0, align 4141; AVX2:  LV: Found an estimated cost of 68 for VF 32 For instruction: %v1 = load i32, ptr %in1, align 4142; AVX2:  LV: Found an estimated cost of 68 for VF 32 For instruction: %v2 = load i32, ptr %in2, align 4143; AVX2:  LV: Found an estimated cost of 68 for VF 32 For instruction: %v3 = load i32, ptr %in3, align 4144; AVX2:  LV: Found an estimated cost of 68 for VF 32 For instruction: %v4 = load i32, ptr %in4, align 4145; AVX2:  LV: Found an estimated cost of 68 for VF 32 For instruction: %v5 = load i32, ptr %in5, align 4146; AVX2:  LV: Found an estimated cost of 68 for VF 32 For instruction: %v6 = load i32, ptr %in6, align 4147; AVX2:  LV: Found an estimated cost of 68 for VF 32 For instruction: %v7 = load i32, ptr %in7, align 4148;149; AVX512-LABEL: 'test'150; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, ptr %in0, align 4151; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i32, ptr %in1, align 4152; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i32, ptr %in2, align 4153; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i32, ptr %in3, align 4154; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i32, ptr %in4, align 4155; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i32, ptr %in5, align 4156; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i32, ptr %in6, align 4157; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i32, ptr %in7, align 4158; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v0 = load i32, ptr %in0, align 4159; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v1 = load i32, ptr %in1, align 4160; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v2 = load i32, ptr %in2, align 4161; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v3 = load i32, ptr %in3, align 4162; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v4 = load i32, ptr %in4, align 4163; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v5 = load i32, ptr %in5, align 4164; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v6 = load i32, ptr %in6, align 4165; AVX512:  LV: Found an estimated cost of 1 for VF 1 For instruction: %v7 = load i32, ptr %in7, align 4166; AVX512:  LV: Found an estimated cost of 9 for VF 2 For instruction: %v0 = load i32, ptr %in0, align 4167; AVX512:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v1 = load i32, ptr %in1, align 4168; AVX512:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v2 = load i32, ptr %in2, align 4169; AVX512:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v3 = load i32, ptr %in3, align 4170; AVX512:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v4 = load i32, ptr %in4, align 4171; AVX512:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v5 = load i32, ptr %in5, align 4172; AVX512:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v6 = load i32, ptr %in6, align 4173; AVX512:  LV: Found an estimated cost of 0 for VF 2 For instruction: %v7 = load i32, ptr %in7, align 4174; AVX512:  LV: Found an estimated cost of 14 for VF 4 For instruction: %v0 = load i32, ptr %in0, align 4175; AVX512:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v1 = load i32, ptr %in1, align 4176; AVX512:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v2 = load i32, ptr %in2, align 4177; AVX512:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v3 = load i32, ptr %in3, align 4178; AVX512:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v4 = load i32, ptr %in4, align 4179; AVX512:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v5 = load i32, ptr %in5, align 4180; AVX512:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v6 = load i32, ptr %in6, align 4181; AVX512:  LV: Found an estimated cost of 0 for VF 4 For instruction: %v7 = load i32, ptr %in7, align 4182; AVX512:  LV: Found an estimated cost of 40 for VF 8 For instruction: %v0 = load i32, ptr %in0, align 4183; AVX512:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v1 = load i32, ptr %in1, align 4184; AVX512:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v2 = load i32, ptr %in2, align 4185; AVX512:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v3 = load i32, ptr %in3, align 4186; AVX512:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v4 = load i32, ptr %in4, align 4187; AVX512:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v5 = load i32, ptr %in5, align 4188; AVX512:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v6 = load i32, ptr %in6, align 4189; AVX512:  LV: Found an estimated cost of 0 for VF 8 For instruction: %v7 = load i32, ptr %in7, align 4190; AVX512:  LV: Found an estimated cost of 92 for VF 16 For instruction: %v0 = load i32, ptr %in0, align 4191; AVX512:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v1 = load i32, ptr %in1, align 4192; AVX512:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v2 = load i32, ptr %in2, align 4193; AVX512:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v3 = load i32, ptr %in3, align 4194; AVX512:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v4 = load i32, ptr %in4, align 4195; AVX512:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v5 = load i32, ptr %in5, align 4196; AVX512:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v6 = load i32, ptr %in6, align 4197; AVX512:  LV: Found an estimated cost of 0 for VF 16 For instruction: %v7 = load i32, ptr %in7, align 4198;199entry:200  br label %for.body201 202for.body:203  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]204 205  %iv.0 = add nuw nsw i64 %iv, 0206  %iv.1 = add nuw nsw i64 %iv, 1207  %iv.2 = add nuw nsw i64 %iv, 2208  %iv.3 = add nuw nsw i64 %iv, 3209  %iv.4 = add nuw nsw i64 %iv, 4210  %iv.5 = add nuw nsw i64 %iv, 5211  %iv.6 = add nuw nsw i64 %iv, 6212  %iv.7 = add nuw nsw i64 %iv, 7213 214  %in0 = getelementptr inbounds [1024 x i32], ptr @A, i64 0, i64 %iv.0215  %in1 = getelementptr inbounds [1024 x i32], ptr @A, i64 0, i64 %iv.1216  %in2 = getelementptr inbounds [1024 x i32], ptr @A, i64 0, i64 %iv.2217  %in3 = getelementptr inbounds [1024 x i32], ptr @A, i64 0, i64 %iv.3218  %in4 = getelementptr inbounds [1024 x i32], ptr @A, i64 0, i64 %iv.4219  %in5 = getelementptr inbounds [1024 x i32], ptr @A, i64 0, i64 %iv.5220  %in6 = getelementptr inbounds [1024 x i32], ptr @A, i64 0, i64 %iv.6221  %in7 = getelementptr inbounds [1024 x i32], ptr @A, i64 0, i64 %iv.7222 223  %v0 = load i32, ptr %in0224  %v1 = load i32, ptr %in1225  %v2 = load i32, ptr %in2226  %v3 = load i32, ptr %in3227  %v4 = load i32, ptr %in4228  %v5 = load i32, ptr %in5229  %v6 = load i32, ptr %in6230  %v7 = load i32, ptr %in7231 232  %reduce.add.0 = add i32 %v0, %v1233  %reduce.add.1 = add i32 %reduce.add.0, %v2234  %reduce.add.2 = add i32 %reduce.add.1, %v3235  %reduce.add.3 = add i32 %reduce.add.2, %v4236  %reduce.add.4 = add i32 %reduce.add.3, %v5237  %reduce.add.5 = add i32 %reduce.add.4, %v6238  %reduce.add.6 = add i32 %reduce.add.5, %v7239 240  %reduce.add.6.narrow = trunc i32 %reduce.add.6 to i8241 242  %out = getelementptr inbounds [1024 x i8], ptr @B, i64 0, i64 %iv.0243  store i8 %reduce.add.6.narrow, ptr %out244 245  %iv.next = add nuw nsw i64 %iv.0, 8246  %cmp = icmp ult i64 %iv.next, 1024247  br i1 %cmp, label %for.body, label %for.cond.cleanup248 249for.cond.cleanup:250  ret void251}252