blob: babb49515a03ed7e02b63f39bcfebefb5d4e2655 (
plain)
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
|
; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
; RUN: grep {vpaddl\\.s8} %t | count 2
; RUN: grep {vpaddl\\.s16} %t | count 2
; RUN: grep {vpaddl\\.s32} %t | count 2
; RUN: grep {vpaddl\\.u8} %t | count 2
; RUN: grep {vpaddl\\.u16} %t | count 2
; RUN: grep {vpaddl\\.u32} %t | count 2
define <4 x i16> @vpaddls8(<8 x i8>* %A) nounwind {
%tmp1 = load <8 x i8>* %A
%tmp2 = call <4 x i16> @llvm.arm.neon.vpaddls.v4i16.v8i8(<8 x i8> %tmp1)
ret <4 x i16> %tmp2
}
define <2 x i32> @vpaddls16(<4 x i16>* %A) nounwind {
%tmp1 = load <4 x i16>* %A
%tmp2 = call <2 x i32> @llvm.arm.neon.vpaddls.v2i32.v4i16(<4 x i16> %tmp1)
ret <2 x i32> %tmp2
}
define <1 x i64> @vpaddls32(<2 x i32>* %A) nounwind {
%tmp1 = load <2 x i32>* %A
%tmp2 = call <1 x i64> @llvm.arm.neon.vpaddls.v1i64.v2i32(<2 x i32> %tmp1)
ret <1 x i64> %tmp2
}
define <4 x i16> @vpaddlu8(<8 x i8>* %A) nounwind {
%tmp1 = load <8 x i8>* %A
%tmp2 = call <4 x i16> @llvm.arm.neon.vpaddlu.v4i16.v8i8(<8 x i8> %tmp1)
ret <4 x i16> %tmp2
}
define <2 x i32> @vpaddlu16(<4 x i16>* %A) nounwind {
%tmp1 = load <4 x i16>* %A
%tmp2 = call <2 x i32> @llvm.arm.neon.vpaddlu.v2i32.v4i16(<4 x i16> %tmp1)
ret <2 x i32> %tmp2
}
define <1 x i64> @vpaddlu32(<2 x i32>* %A) nounwind {
%tmp1 = load <2 x i32>* %A
%tmp2 = call <1 x i64> @llvm.arm.neon.vpaddlu.v1i64.v2i32(<2 x i32> %tmp1)
ret <1 x i64> %tmp2
}
define <8 x i16> @vpaddlQs8(<16 x i8>* %A) nounwind {
%tmp1 = load <16 x i8>* %A
%tmp2 = call <8 x i16> @llvm.arm.neon.vpaddls.v8i16.v16i8(<16 x i8> %tmp1)
ret <8 x i16> %tmp2
}
define <4 x i32> @vpaddlQs16(<8 x i16>* %A) nounwind {
%tmp1 = load <8 x i16>* %A
%tmp2 = call <4 x i32> @llvm.arm.neon.vpaddls.v4i32.v8i16(<8 x i16> %tmp1)
ret <4 x i32> %tmp2
}
define <2 x i64> @vpaddlQs32(<4 x i32>* %A) nounwind {
%tmp1 = load <4 x i32>* %A
%tmp2 = call <2 x i64> @llvm.arm.neon.vpaddls.v2i64.v4i32(<4 x i32> %tmp1)
ret <2 x i64> %tmp2
}
define <8 x i16> @vpaddlQu8(<16 x i8>* %A) nounwind {
%tmp1 = load <16 x i8>* %A
%tmp2 = call <8 x i16> @llvm.arm.neon.vpaddlu.v8i16.v16i8(<16 x i8> %tmp1)
ret <8 x i16> %tmp2
}
define <4 x i32> @vpaddlQu16(<8 x i16>* %A) nounwind {
%tmp1 = load <8 x i16>* %A
%tmp2 = call <4 x i32> @llvm.arm.neon.vpaddlu.v4i32.v8i16(<8 x i16> %tmp1)
ret <4 x i32> %tmp2
}
define <2 x i64> @vpaddlQu32(<4 x i32>* %A) nounwind {
%tmp1 = load <4 x i32>* %A
%tmp2 = call <2 x i64> @llvm.arm.neon.vpaddlu.v2i64.v4i32(<4 x i32> %tmp1)
ret <2 x i64> %tmp2
}
declare <4 x i16> @llvm.arm.neon.vpaddls.v4i16.v8i8(<8 x i8>) nounwind readnone
declare <2 x i32> @llvm.arm.neon.vpaddls.v2i32.v4i16(<4 x i16>) nounwind readnone
declare <1 x i64> @llvm.arm.neon.vpaddls.v1i64.v2i32(<2 x i32>) nounwind readnone
declare <4 x i16> @llvm.arm.neon.vpaddlu.v4i16.v8i8(<8 x i8>) nounwind readnone
declare <2 x i32> @llvm.arm.neon.vpaddlu.v2i32.v4i16(<4 x i16>) nounwind readnone
declare <1 x i64> @llvm.arm.neon.vpaddlu.v1i64.v2i32(<2 x i32>) nounwind readnone
declare <8 x i16> @llvm.arm.neon.vpaddls.v8i16.v16i8(<16 x i8>) nounwind readnone
declare <4 x i32> @llvm.arm.neon.vpaddls.v4i32.v8i16(<8 x i16>) nounwind readnone
declare <2 x i64> @llvm.arm.neon.vpaddls.v2i64.v4i32(<4 x i32>) nounwind readnone
declare <8 x i16> @llvm.arm.neon.vpaddlu.v8i16.v16i8(<16 x i8>) nounwind readnone
declare <4 x i32> @llvm.arm.neon.vpaddlu.v4i32.v8i16(<8 x i16>) nounwind readnone
declare <2 x i64> @llvm.arm.neon.vpaddlu.v2i64.v4i32(<4 x i32>) nounwind readnone
|
