81 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*store i64 %valB, ptr %out, align 8"2; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+sse2 --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefixes=SSE23; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+sse4.2 --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefixes=SSE424; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefixes=AVX15; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx2,-fast-gather --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefixes=AVX26; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx2,+fast-gather --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefixes=AVX27; RUN: opt -passes=loop-vectorize -vectorizer-maximize-bandwidth -S -mattr=+avx512bw --debug-only=loop-vectorize --disable-output < %s 2>&1 | FileCheck %s --check-prefixes=AVX5128 9; REQUIRES: asserts10 11target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"12target triple = "x86_64-unknown-linux-gnu"13 14@A = global [1024 x i8] zeroinitializer, align 12815@B = global [1024 x i64] zeroinitializer, align 12816@C = global [1024 x i64] zeroinitializer, align 12817 18define void @test() {19; SSE2-LABEL: 'test'20; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 821; SSE2: LV: Found an estimated cost of 29 for VF 2 For instruction: store i64 %valB, ptr %out, align 822; SSE2: LV: Found an estimated cost of 58 for VF 4 For instruction: store i64 %valB, ptr %out, align 823; SSE2: LV: Found an estimated cost of 116 for VF 8 For instruction: store i64 %valB, ptr %out, align 824; SSE2: LV: Found an estimated cost of 232 for VF 16 For instruction: store i64 %valB, ptr %out, align 825;26; SSE42-LABEL: 'test'27; SSE42: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 828; SSE42: LV: Found an estimated cost of 26 for VF 2 For instruction: store i64 %valB, ptr %out, align 829; SSE42: LV: Found an estimated cost of 52 for VF 4 For instruction: store i64 %valB, ptr %out, align 830; SSE42: LV: Found an estimated cost of 104 for VF 8 For instruction: store i64 %valB, ptr %out, align 831; SSE42: LV: Found an estimated cost of 208 for VF 16 For instruction: store i64 %valB, ptr %out, align 832;33; AVX1-LABEL: 'test'34; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 835; AVX1: LV: Found an estimated cost of 26 for VF 2 For instruction: store i64 %valB, ptr %out, align 836; AVX1: LV: Found an estimated cost of 54 for VF 4 For instruction: store i64 %valB, ptr %out, align 837; AVX1: LV: Found an estimated cost of 108 for VF 8 For instruction: store i64 %valB, ptr %out, align 838; AVX1: LV: Found an estimated cost of 216 for VF 16 For instruction: store i64 %valB, ptr %out, align 839; AVX1: LV: Found an estimated cost of 432 for VF 32 For instruction: store i64 %valB, ptr %out, align 840;41; AVX2-LABEL: 'test'42; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 843; AVX2: LV: Found an estimated cost of 6 for VF 2 For instruction: store i64 %valB, ptr %out, align 844; AVX2: LV: Found an estimated cost of 14 for VF 4 For instruction: store i64 %valB, ptr %out, align 845; AVX2: LV: Found an estimated cost of 28 for VF 8 For instruction: store i64 %valB, ptr %out, align 846; AVX2: LV: Found an estimated cost of 56 for VF 16 For instruction: store i64 %valB, ptr %out, align 847; AVX2: LV: Found an estimated cost of 112 for VF 32 For instruction: store i64 %valB, ptr %out, align 848;49; AVX512-LABEL: 'test'50; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 851; AVX512: LV: Found an estimated cost of 6 for VF 2 For instruction: store i64 %valB, ptr %out, align 852; AVX512: LV: Found an estimated cost of 14 for VF 4 For instruction: store i64 %valB, ptr %out, align 853; AVX512: LV: Found an estimated cost of 10 for VF 8 For instruction: store i64 %valB, ptr %out, align 854; AVX512: LV: Found an estimated cost of 20 for VF 16 For instruction: store i64 %valB, ptr %out, align 855; AVX512: LV: Found an estimated cost of 40 for VF 32 For instruction: store i64 %valB, ptr %out, align 856; AVX512: LV: Found an estimated cost of 80 for VF 64 For instruction: store i64 %valB, ptr %out, align 857;58entry:59 br label %for.body60 61for.body:62 %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]63 64 %inA = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv65 %valA = load i8, ptr %inA66 %valA.ext = sext i8 %valA to i6467 68 %inB = getelementptr inbounds [1024 x i64], ptr @B, i64 0, i64 %iv69 %valB = load i64, ptr %inB70 71 %out = getelementptr inbounds [1024 x i64], ptr @C, i64 0, i64 %valA.ext72 store i64 %valB, ptr %out73 74 %iv.next = add nuw nsw i64 %iv, 175 %cmp = icmp ult i64 %iv.next, 102476 br i1 %cmp, label %for.body, label %for.cond.cleanup77 78for.cond.cleanup:79 ret void80}81