1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
|
;RUN: llc < %s -march=r600 -mcpu=verde | FileCheck %s
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+_VGPR[0-9]+_VGPR[0-9]+_VGPR[0-9]+}}, 15
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+_VGPR[0-9]+}}, 3
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+}}, 2
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+}}, 1
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+}}, 4
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+}}, 8
;CHECK: IMAGE_SAMPLE_C {{VGPR[0-9]+_VGPR[0-9]+}}, 5
;CHECK: IMAGE_SAMPLE_C {{VGPR[0-9]+_VGPR[0-9]+}}, 9
;CHECK: IMAGE_SAMPLE_C {{VGPR[0-9]+_VGPR[0-9]+}}, 6
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+_VGPR[0-9]+}}, 10
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+_VGPR[0-9]+}}, 12
;CHECK: IMAGE_SAMPLE_C {{VGPR[0-9]+_VGPR[0-9]+_VGPR[0-9]+}}, 7
;CHECK: IMAGE_SAMPLE_C {{VGPR[0-9]+_VGPR[0-9]+_VGPR[0-9]+}}, 11
;CHECK: IMAGE_SAMPLE_C {{VGPR[0-9]+_VGPR[0-9]+_VGPR[0-9]+}}, 13
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+_VGPR[0-9]+_VGPR[0-9]+}}, 14
;CHECK: IMAGE_SAMPLE {{VGPR[0-9]+}}, 8
define void @test(i32 %a1, i32 %a2, i32 %a3, i32 %a4) {
%v1 = insertelement <4 x i32> undef, i32 %a1, i32 0
%v2 = insertelement <4 x i32> undef, i32 %a1, i32 1
%v3 = insertelement <4 x i32> undef, i32 %a1, i32 2
%v4 = insertelement <4 x i32> undef, i32 %a1, i32 3
%v5 = insertelement <4 x i32> undef, i32 %a2, i32 0
%v6 = insertelement <4 x i32> undef, i32 %a2, i32 1
%v7 = insertelement <4 x i32> undef, i32 %a2, i32 2
%v8 = insertelement <4 x i32> undef, i32 %a2, i32 3
%v9 = insertelement <4 x i32> undef, i32 %a3, i32 0
%v10 = insertelement <4 x i32> undef, i32 %a3, i32 1
%v11 = insertelement <4 x i32> undef, i32 %a3, i32 2
%v12 = insertelement <4 x i32> undef, i32 %a3, i32 3
%v13 = insertelement <4 x i32> undef, i32 %a4, i32 0
%v14 = insertelement <4 x i32> undef, i32 %a4, i32 1
%v15 = insertelement <4 x i32> undef, i32 %a4, i32 2
%v16 = insertelement <4 x i32> undef, i32 %a4, i32 3
%res1 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v1,
<8 x i32> undef, <4 x i32> undef, i32 1)
%res2 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v2,
<8 x i32> undef, <4 x i32> undef, i32 2)
%res3 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v3,
<8 x i32> undef, <4 x i32> undef, i32 3)
%res4 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v4,
<8 x i32> undef, <4 x i32> undef, i32 4)
%res5 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v5,
<8 x i32> undef, <4 x i32> undef, i32 5)
%res6 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v6,
<8 x i32> undef, <4 x i32> undef, i32 6)
%res7 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v7,
<8 x i32> undef, <4 x i32> undef, i32 7)
%res8 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v8,
<8 x i32> undef, <4 x i32> undef, i32 8)
%res9 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v9,
<8 x i32> undef, <4 x i32> undef, i32 9)
%res10 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v10,
<8 x i32> undef, <4 x i32> undef, i32 10)
%res11 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v11,
<8 x i32> undef, <4 x i32> undef, i32 11)
%res12 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v12,
<8 x i32> undef, <4 x i32> undef, i32 12)
%res13 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v13,
<8 x i32> undef, <4 x i32> undef, i32 13)
%res14 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v14,
<8 x i32> undef, <4 x i32> undef, i32 14)
%res15 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v15,
<8 x i32> undef, <4 x i32> undef, i32 15)
%res16 = call <4 x float> @llvm.SI.sample.(<4 x i32> %v16,
<8 x i32> undef, <4 x i32> undef, i32 16)
%e1 = extractelement <4 x float> %res1, i32 0
%e2 = extractelement <4 x float> %res2, i32 1
%e3 = extractelement <4 x float> %res3, i32 2
%e4 = extractelement <4 x float> %res4, i32 3
%t0 = extractelement <4 x float> %res5, i32 0
%t1 = extractelement <4 x float> %res5, i32 1
%e5 = fadd float %t0, %t1
%t2 = extractelement <4 x float> %res6, i32 0
%t3 = extractelement <4 x float> %res6, i32 2
%e6 = fadd float %t2, %t3
%t4 = extractelement <4 x float> %res7, i32 0
%t5 = extractelement <4 x float> %res7, i32 3
%e7 = fadd float %t4, %t5
%t6 = extractelement <4 x float> %res8, i32 1
%t7 = extractelement <4 x float> %res8, i32 2
%e8 = fadd float %t6, %t7
%t8 = extractelement <4 x float> %res9, i32 1
%t9 = extractelement <4 x float> %res9, i32 3
%e9 = fadd float %t8, %t9
%t10 = extractelement <4 x float> %res10, i32 2
%t11 = extractelement <4 x float> %res10, i32 3
%e10 = fadd float %t10, %t11
%t12 = extractelement <4 x float> %res11, i32 0
%t13 = extractelement <4 x float> %res11, i32 1
%t14 = extractelement <4 x float> %res11, i32 2
%t15 = fadd float %t12, %t13
%e11 = fadd float %t14, %t15
%t16 = extractelement <4 x float> %res12, i32 0
%t17 = extractelement <4 x float> %res12, i32 1
%t18 = extractelement <4 x float> %res12, i32 3
%t19 = fadd float %t16, %t17
%e12 = fadd float %t18, %t19
%t20 = extractelement <4 x float> %res13, i32 0
%t21 = extractelement <4 x float> %res13, i32 2
%t22 = extractelement <4 x float> %res13, i32 3
%t23 = fadd float %t20, %t21
%e13 = fadd float %t22, %t23
%t24 = extractelement <4 x float> %res14, i32 1
%t25 = extractelement <4 x float> %res14
|
