xref: /llvm-project/llvm/test/Transforms/SLPVectorizer/X86/gather-move-out-of-loop.ll (revision 9abb1ffc5cb75465340cb604988d1e386415bd72)
1; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
2; RUN: opt -S -passes=slp-vectorizer -slp-threshold=-100 -mtriple=x86_64-w64-windows-gnu < %s | FileCheck %s
3; RUN: opt -S -passes=slp-vectorizer -slp-threshold=-100 -mtriple=x86_64-w64-windows-gnu\
4; RUN: -slp-skip-early-profitability-check < %s | FileCheck %s --check-prefixes=FORCED
5
6define void @test(i16 %0) {
7; FORCED-LABEL: @test(
8; FORCED-NEXT:  for.body92.preheader:
9; FORCED-NEXT:    [[TMP1:%.*]] = insertelement <2 x i16> <i16 0, i16 poison>, i16 [[TMP0:%.*]], i32 1
10; FORCED-NEXT:    [[TMP2:%.*]] = sext <2 x i16> [[TMP1]] to <2 x i32>
11; FORCED-NEXT:    [[TMP3:%.*]] = zext <2 x i16> [[TMP1]] to <2 x i32>
12; FORCED-NEXT:    [[TMP4:%.*]] = shufflevector <2 x i32> [[TMP2]], <2 x i32> [[TMP3]], <2 x i32> <i32 0, i32 3>
13; FORCED-NEXT:    [[TMP5:%.*]] = shufflevector <2 x i32> [[TMP4]], <2 x i32> poison, <4 x i32> <i32 0, i32 poison, i32 1, i32 poison>
14; FORCED-NEXT:    [[TMP6:%.*]] = shufflevector <4 x i32> <i32 poison, i32 0, i32 poison, i32 0>, <4 x i32> [[TMP5]], <4 x i32> <i32 4, i32 1, i32 6, i32 3>
15; FORCED-NEXT:    br label [[FOR_BODY92:%.*]]
16; FORCED:       for.body92:
17; FORCED-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> zeroinitializer, [[TMP6]]
18; FORCED-NEXT:    store <4 x i32> [[TMP7]], ptr undef, align 8
19; FORCED-NEXT:    br label [[FOR_BODY92]]
20;
21; CHECK-LABEL: @test(
22; CHECK-NEXT:  for.body92.preheader:
23; CHECK-NEXT:    br label [[FOR_BODY92:%.*]]
24; CHECK:       for.body92:
25; CHECK-NEXT:    [[CONV177_I:%.*]] = sext i16 0 to i32
26; CHECK-NEXT:    [[TMP1:%.*]] = zext i16 [[TMP0:%.*]] to i32
27; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i32> <i32 poison, i32 0, i32 poison, i32 0>, i32 [[CONV177_I]], i32 0
28; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i32> [[TMP2]], i32 [[TMP1]], i32 2
29; CHECK-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> zeroinitializer, [[TMP6]]
30; CHECK-NEXT:    store <4 x i32> [[TMP7]], ptr undef, align 8
31; CHECK-NEXT:    br label [[FOR_BODY92]]
32;
33for.body92.preheader:
34  br label %for.body92
35
36for.body92:
37  %conv177.i = sext i16 0 to i32
38  %add178.i = add nsw i32 0, %conv177.i
39  store i32 %add178.i, ptr undef, align 8
40  %1 = zext i16 %0 to i32
41  %sum_mvr_abs.i = getelementptr i32, ptr undef, i32 2
42  %add182.i = add nsw i32 0, %1
43  store i32 %add182.i, ptr %sum_mvr_abs.i, align 8
44  %sum_mvc.i = getelementptr i32, ptr undef, i32 1
45  %add184.i = add nsw i32 0, 0
46  store i32 %add184.i, ptr %sum_mvc.i, align 4
47  %sum_mvc_abs.i = getelementptr i32, ptr undef, i32 3
48  %add188.i = add nsw i32 0, 0
49  store i32 %add188.i, ptr %sum_mvc_abs.i, align 4
50  br label %for.body92
51}
52