1; RUN: opt -S -mtriple=x86_64-apple-darwin -mcpu=core-avx2 -cost-model -analyze < %s | FileCheck %s -check-prefix=AVX2
2
3
4; AVX2-LABEL: test1
5; AVX2: Found an estimated cost of 4 {{.*}}.masked
6define <2 x double> @test1(<2 x i64> %trigger, <2 x double>* %addr, <2 x double> %dst) {
7  %mask = icmp eq <2 x i64> %trigger, zeroinitializer
8  %res = call <2 x double> @llvm.masked.load.v2f64(<2 x double>* %addr, i32 4, <2 x i1>%mask, <2 x double>%dst)
9  ret <2 x double> %res
10}
11
12; AVX2-LABEL: test2
13; AVX2: Found an estimated cost of 4 {{.*}}.masked
14define <4 x i32> @test2(<4 x i32> %trigger, <4 x i32>* %addr, <4 x i32> %dst) {
15  %mask = icmp eq <4 x i32> %trigger, zeroinitializer
16  %res = call <4 x i32> @llvm.masked.load.v4i32(<4 x i32>* %addr, i32 4, <4 x i1>%mask, <4 x i32>%dst)
17  ret <4 x i32> %res
18}
19
20; AVX2-LABEL: test3
21; AVX2: Found an estimated cost of 4 {{.*}}.masked
22define void @test3(<4 x i32> %trigger, <4 x i32>* %addr, <4 x i32> %val) {
23  %mask = icmp eq <4 x i32> %trigger, zeroinitializer
24  call void @llvm.masked.store.v4i32(<4 x i32>%val, <4 x i32>* %addr, i32 4, <4 x i1>%mask)
25  ret void
26}
27
28; AVX2-LABEL: test4
29; AVX2: Found an estimated cost of 4 {{.*}}.masked
30define <8 x float> @test4(<8 x i32> %trigger, <8 x float>* %addr, <8 x float> %dst) {
31  %mask = icmp eq <8 x i32> %trigger, zeroinitializer
32  %res = call <8 x float> @llvm.masked.load.v8f32(<8 x float>* %addr, i32 4, <8 x i1>%mask, <8 x float>%dst)
33  ret <8 x float> %res
34}
35
36; AVX2-LABEL: test5
37; AVX2: Found an estimated cost of 5 {{.*}}.masked
38define void @test5(<2 x i32> %trigger, <2 x float>* %addr, <2 x float> %val) {
39  %mask = icmp eq <2 x i32> %trigger, zeroinitializer
40  call void @llvm.masked.store.v2f32(<2 x float>%val, <2 x float>* %addr, i32 4, <2 x i1>%mask)
41  ret void
42}
43
44; AVX2-LABEL: test6
45; AVX2: Found an estimated cost of 6 {{.*}}.masked
46define void @test6(<2 x i32> %trigger, <2 x i32>* %addr, <2 x i32> %val) {
47  %mask = icmp eq <2 x i32> %trigger, zeroinitializer
48  call void @llvm.masked.store.v2i32(<2 x i32>%val, <2 x i32>* %addr, i32 4, <2 x i1>%mask)
49  ret void
50}
51
52; AVX2-LABEL: test7
53; AVX2: Found an estimated cost of 5 {{.*}}.masked
54define <2 x float> @test7(<2 x i32> %trigger, <2 x float>* %addr, <2 x float> %dst) {
55  %mask = icmp eq <2 x i32> %trigger, zeroinitializer
56  %res = call <2 x float> @llvm.masked.load.v2f32(<2 x float>* %addr, i32 4, <2 x i1>%mask, <2 x float>%dst)
57  ret <2 x float> %res
58}
59
60; AVX2-LABEL: test8
61; AVX2: Found an estimated cost of 6 {{.*}}.masked
62define <2 x i32> @test8(<2 x i32> %trigger, <2 x i32>* %addr, <2 x i32> %dst) {
63  %mask = icmp eq <2 x i32> %trigger, zeroinitializer
64  %res = call <2 x i32> @llvm.masked.load.v2i32(<2 x i32>* %addr, i32 4, <2 x i1>%mask, <2 x i32>%dst)
65  ret <2 x i32> %res
66}
67
68
69declare <16 x i32> @llvm.masked.load.v16i32(<16 x i32>*, i32, <16 x i1>, <16 x i32>)
70declare <4 x i32> @llvm.masked.load.v4i32(<4 x i32>*, i32, <4 x i1>, <4 x i32>)
71declare <2 x i32> @llvm.masked.load.v2i32(<2 x i32>*, i32, <2 x i1>, <2 x i32>)
72declare void @llvm.masked.store.v16i32(<16 x i32>, <16 x i32>*, i32, <16 x i1>)
73declare void @llvm.masked.store.v8i32(<8 x i32>, <8 x i32>*, i32, <8 x i1>)
74declare void @llvm.masked.store.v4i32(<4 x i32>, <4 x i32>*, i32, <4 x i1>)
75declare void @llvm.masked.store.v2f32(<2 x float>, <2 x float>*, i32, <2 x i1>)
76declare void @llvm.masked.store.v2i32(<2 x i32>, <2 x i32>*, i32, <2 x i1>)
77declare void @llvm.masked.store.v16f32(<16 x float>, <16 x float>*, i32, <16 x i1>)
78declare void @llvm.masked.store.v16f32p(<16 x float>*, <16 x float>**, i32, <16 x i1>)
79declare <16 x float> @llvm.masked.load.v16f32(<16 x float>*, i32, <16 x i1>, <16 x float>)
80declare <8 x float> @llvm.masked.load.v8f32(<8 x float>*, i32, <8 x i1>, <8 x float>)
81declare <4 x float> @llvm.masked.load.v4f32(<4 x float>*, i32, <4 x i1>, <4 x float>)
82declare <2 x float> @llvm.masked.load.v2f32(<2 x float>*, i32, <2 x i1>, <2 x float>)
83declare <8 x double> @llvm.masked.load.v8f64(<8 x double>*, i32, <8 x i1>, <8 x double>)
84declare <4 x double> @llvm.masked.load.v4f64(<4 x double>*, i32, <4 x i1>, <4 x double>)
85declare <2 x double> @llvm.masked.load.v2f64(<2 x double>*, i32, <2 x i1>, <2 x double>)
86declare void @llvm.masked.store.v8f64(<8 x double>, <8 x double>*, i32, <8 x i1>)
87declare void @llvm.masked.store.v2f64(<2 x double>, <2 x double>*, i32, <2 x i1>)
88declare void @llvm.masked.store.v2i64(<2 x i64>, <2 x i64>*, i32, <2 x i1>)
89
90