87 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 17define void @test(ptr %C) {18; SSE2-LABEL: 'test'19; SSE2: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 820; SSE2: LV: Found an estimated cost of 2 for VF 2 For instruction: store i64 %valB, ptr %out, align 821; SSE2: LV: Found an estimated cost of 5 for VF 4 For instruction: store i64 %valB, ptr %out, align 822; SSE2: LV: Found an estimated cost of 10 for VF 8 For instruction: store i64 %valB, ptr %out, align 823; SSE2: LV: Found an estimated cost of 20 for VF 16 For instruction: store i64 %valB, ptr %out, align 824;25; SSE42-LABEL: 'test'26; SSE42: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 827; SSE42: LV: Found an estimated cost of 2 for VF 2 For instruction: store i64 %valB, ptr %out, align 828; SSE42: LV: Found an estimated cost of 4 for VF 4 For instruction: store i64 %valB, ptr %out, align 829; SSE42: LV: Found an estimated cost of 8 for VF 8 For instruction: store i64 %valB, ptr %out, align 830; SSE42: LV: Found an estimated cost of 16 for VF 16 For instruction: store i64 %valB, ptr %out, align 831;32; AVX1-LABEL: 'test'33; AVX1: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 834; AVX1: LV: Found an estimated cost of 8 for VF 2 For instruction: store i64 %valB, ptr %out, align 835; AVX1: LV: Found an estimated cost of 8 for VF 4 For instruction: store i64 %valB, ptr %out, align 836; AVX1: LV: Found an estimated cost of 16 for VF 8 For instruction: store i64 %valB, ptr %out, align 837; AVX1: LV: Found an estimated cost of 32 for VF 16 For instruction: store i64 %valB, ptr %out, align 838; AVX1: LV: Found an estimated cost of 64 for VF 32 For instruction: store i64 %valB, ptr %out, align 839;40; AVX2-LABEL: 'test'41; AVX2: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 842; AVX2: LV: Found an estimated cost of 8 for VF 2 For instruction: store i64 %valB, ptr %out, align 843; AVX2: LV: Found an estimated cost of 8 for VF 4 For instruction: store i64 %valB, ptr %out, align 844; AVX2: LV: Found an estimated cost of 16 for VF 8 For instruction: store i64 %valB, ptr %out, align 845; AVX2: LV: Found an estimated cost of 32 for VF 16 For instruction: store i64 %valB, ptr %out, align 846; AVX2: LV: Found an estimated cost of 64 for VF 32 For instruction: store i64 %valB, ptr %out, align 847;48; AVX512-LABEL: 'test'49; AVX512: LV: Found an estimated cost of 1 for VF 1 For instruction: store i64 %valB, ptr %out, align 850; AVX512: LV: Found an estimated cost of 1 for VF 2 For instruction: store i64 %valB, ptr %out, align 851; AVX512: LV: Found an estimated cost of 1 for VF 4 For instruction: store i64 %valB, ptr %out, align 852; AVX512: LV: Found an estimated cost of 1 for VF 8 For instruction: store i64 %valB, ptr %out, align 853; AVX512: LV: Found an estimated cost of 2 for VF 16 For instruction: store i64 %valB, ptr %out, align 854; AVX512: LV: Found an estimated cost of 4 for VF 32 For instruction: store i64 %valB, ptr %out, align 855; AVX512: LV: Found an estimated cost of 8 for VF 64 For instruction: store i64 %valB, ptr %out, align 856;57entry:58 br label %for.body59 60for.body:61 %iv = phi i64 [ 0, %entry ], [ %iv.next, %end ]62 63 %inB = getelementptr inbounds [1024 x i64], ptr @B, i64 0, i64 %iv64 %valB = load i64, ptr %inB65 66 %inA = getelementptr inbounds [1024 x i8], ptr @A, i64 0, i64 %iv67 %valA = load i8, ptr %inA68 %canStore = icmp ne i8 %valA, 069 br i1 %canStore, label %store, label %mask70 71store:72 %out = getelementptr inbounds [1024 x i64], ptr %C, i64 0, i64 %iv73 store i64 %valB, ptr %out74 br label %end75 76mask:77 br label %end78 79end:80 %iv.next = add nuw nsw i64 %iv, 181 %cmp = icmp ult i64 %iv.next, 102482 br i1 %cmp, label %for.body, label %for.cond.cleanup83 84for.cond.cleanup:85 ret void86}87