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
141
142
143
144
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
; RUN: llc -mtriple=aarch64--linux-gnu -mattr=+sve2p1 < %s | FileCheck %s
; LD1W
define <vscale x 4 x i32> @test_svld1uwq_i32_ss(<vscale x 1 x i1> %pred, ptr %base, i64 %offset) {
; CHECK-LABEL: test_svld1uwq_i32_ss:
; CHECK: // %bb.0:
; CHECK-NEXT: ld1w { z0.q }, p0/z, [x0, x1, lsl #2]
; CHECK-NEXT: ret
%gep = getelementptr i32, ptr %base, i64 %offset
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.ld1uwq.nxv4i32(<vscale x 1 x i1> %pred, ptr %gep)
ret <vscale x 4 x i32> %res
}
define <vscale x 4 x i32> @test_svld1uwq_i32_si(<vscale x 1 x i1> %pred, ptr %base) {
; CHECK-LABEL: test_svld1uwq_i32_si:
; CHECK: // %bb.0:
; CHECK-NEXT: ld1w { z0.q }, p0/z, [x0, #-8, mul vl]
; CHECK-NEXT: ld1w { z1.q }, p0/z, [x0, #7, mul vl]
; CHECK-NEXT: add z0.s, z0.s, z1.s
; CHECK-NEXT: ret
%gep1 = getelementptr inbounds <vscale x 1 x i32>, ptr %base, i64 -8
%res1 = call <vscale x 4 x i32> @llvm.aarch64.sve.ld1uwq.nxv4i32(<vscale x 1 x i1> %pred, ptr %gep1)
%gep2 = getelementptr inbounds <vscale x 1 x i32>, ptr %base, i64 7
%res2 = call <vscale x 4 x i32> @llvm.aarch64.sve.ld1uwq.nxv4i32(<vscale x 1 x i1> %pred, ptr %gep2)
%res = add <vscale x 4 x i32> %res1, %res2
ret <vscale x 4 x i32> %res
}
define <vscale x 4 x i32> @test_svld1uwq_i32_out_of_bound(<vscale x 1 x i1> %pred, ptr %base) {
; CHECK-LABEL: test_svld1uwq_i32_out_of_bound:
; CHECK: // %bb.0:
; CHECK-NEXT: addvl x8, x0, #2
; CHECK-NEXT: ld1w { z0.q }, p0/z, [x8]
; CHECK-NEXT: ret
%gep = getelementptr inbounds <vscale x 1 x i32>, ptr %base, i64 8
%res = call <vscale x 4 x i32> @llvm.aarch64.sve.ld1uwq.nxv4i32(<vscale x 1 x i1> %pred, ptr %gep)
ret <vscale x 4 x i32> %res
}
define <vscale x 4 x float> @test_svld1uwq_f32_ss(<vscale x 1 x i1> %pred, ptr %base, i64 %offset) {
; CHECK-LABEL: test_svld1uwq_f32_ss:
; CHECK: // %bb.0:
; CHECK-NEXT: ld1w { z0.q }, p0/z, [x0, x1, lsl #2]
; CHECK-NEXT: ret
%gep = getelementptr float, ptr %base, i64 %offset
%res = call <vscale x 4 x float> @llvm.aarch64.sve.ld1uwq.nxv4f32(<vscale x 1 x i1> %pred, ptr %gep)
ret <vscale x 4 x float> %res
}
define <vscale x 4 x float> @test_svld1uwq_f32_si(<vscale x 1 x i1> %pred, ptr %base) {
; CHECK-LABEL: test_svld1uwq_f32_si:
; CHECK: // %bb.0:
; CHECK-NEXT: ld1w { z0.q }, p0/z, [x0, #-8, mul vl]
; CHECK-NEXT: ld1w { z1.q }, p0/z, [x0, #7, mul vl]
; CHECK-NEXT: fadd z0.s, z0.s, z1.s
; CHECK-NEXT: ret
%gep1 = getelementptr inbounds <vscale x 1 x float>, ptr %base, i64 -8
%res1 = call <vscale x 4 x float> @llvm.aarch64.sve.ld1uwq.nxv4f32(<vscale x 1 x i1> %pred, ptr %gep1)
%gep2 = getelementptr inbounds <vscale x 1 x float>, ptr %base, i64 7
%res2 = call <vscale x 4 x float> @llvm.aarch64.sve.ld1uwq.nxv4f32(<vscale x 1 x i1> %pred, ptr %gep2)
%res = fadd <vscale x 4 x float> %res1, %res2
ret <vscale x 4 x float> %res
}
; LD1D
define <vscale x 2 x i64> @test_svld1udq_i64_ss(<vscale x 1 x i1> %pred, ptr %base, i64 %offset) {
; CHECK-LABEL: test_svld1udq_i64_ss:
; CHECK: // %bb.0:
; CHECK-NEXT: ld1d { z0.q }, p0/z, [x0, x1, lsl #3]
; CHECK-NEXT: ret
%gep = getelementptr i64, ptr %base, i64 %offset
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.ld1udq.nxv2i64(<vscale x 1 x i1> %pred, ptr %gep)
ret <vscale x 2 x i64> %res
}
define <vscale x 2 x i64> @test_svld1udq_i64_si(<vscale x 1 x i1> %pred, ptr %base) {
; CHECK-LABEL: test_svld1udq_i64_si:
; CHECK: // %bb.0:
; CHECK-NEXT: ld1d { z0.q }, p0/z, [x0, #-8, mul vl]
; CHECK-NEXT: ld1d { z1.q }, p0/z, [x0, #7, mul vl]
; CHECK-NEXT: add z0.d, z0.d, z1.d
; CHECK-NEXT: ret
%gep1 = getelementptr inbounds <vscale x 1 x i64>, ptr %base, i64 -8
%res1 = call <vscale x 2 x i64> @llvm.aarch64.sve.ld1udq.nxv2i64(<vscale x 1 x i1> %pred, ptr %gep1)
%gep2 = getelementptr inbounds <vscale x 1 x i64>, ptr %base, i64 7
%res2 = call <vscale x 2 x i64> @llvm.aarch64.sve.ld1udq.nxv2i64(<vscale x 1 x i1> %pred, ptr %gep2)
%res = add <vscale x 2 x i64> %res1, %res2
ret <vscale x 2 x i64> %res
}
define <vscale x 2 x i64> @test_svld1udq_i64_out_of_bound(<vscale x 1 x i1> %pred, ptr %base) {
; CHECK-LABEL: test_svld1udq_i64_out_of_bound:
; CHECK: // %bb.0:
; CHECK-NEXT: addvl x8, x0, #-5
; CHECK-NEXT: ld1d { z0.q }, p0/z, [x8]
; CHECK-NEXT: ret
%gep = getelementptr inbounds <vscale x 1 x i64>, ptr %base, i64 -10
%res = call <vscale x 2 x i64> @llvm.aarch64.sve.ld1udq.nxv2i64(<vscale x 1 x i1> %pred, ptr %gep)
ret <vscale x 2 x i64> %res
}
define <vscale x 2 x double> @test_svld1udq_f64_ss(<vscale x 1 x i1> %pred, ptr %base, i64 %offset) {
; CHECK-LABEL: test_svld1udq_f64_ss:
; CHECK: // %bb.0:
; CHECK-NEXT: ld1d { z0.q }, p0/z, [x0, x1, lsl #3]
; CHECK-NEXT: ret
%gep = getelementptr double, ptr %base, i64 %offset
%res = call <vscale x 2 x double> @llvm.aarch64.sve.ld1udq.nxv2f64(<vscale x 1 x i1> %pred, ptr %gep)
ret <vscale x 2 x double> %res
}
define <vscale x 2 x double> @test_svld1udq_f64_si(<vscale x 1 x i1> %pred, ptr %base) {
; CHECK-LABEL: test_svld1udq_f64_si:
; CHECK: // %bb.0:
; CHECK-NEXT: ld1d { z0.q }, p0/z, [x0, #-8, mul vl]
; CHECK-NEXT: ld1d { z1.q }, p0/z, [x0, #7, mul vl]
; CHECK-NEXT: fadd z0.d, z0.d, z1.d
; CHECK-NEXT: ret
%gep1 = getelementptr inbounds <vscale x 1 x double>, ptr %base, i64 -8
%res1 = call <vscale x 2 x double> @llvm.aarch64.sve.ld1udq.nxv2f64(<vscale x 1 x i1> %pred, ptr %gep1)
%gep2 = getelementptr inbounds <vscale x 1 x double>, ptr %base, i64 7
%res2 = call <vscale x 2 x double> @llvm.aarch64.sve.ld1udq.nxv2f64(<vscale x 1 x i1> %pred, ptr %gep2)
%res = fadd <vscale x 2 x double> %res1, %res2
ret <vscale x 2 x double> %res
}
declare <vscale x 4 x i32> @llvm.aarch64.sve.ld1uwq.nxv4i32(<vscale x 1 x i1>, ptr)
declare <vscale x 4 x float> @llvm.aarch64.sve.ld1uwq.nxv4f32(<vscale x 1 x i1>, ptr)
declare <vscale x 2 x i64> @llvm.aarch64.sve.ld1udq.nxv2i64(<vscale x 1 x i1>, ptr)
declare <vscale x 2 x double> @llvm.aarch64.sve.ld1udq.nxv2f64(<vscale x 1 x i1>, ptr)
|
