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
|
; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
; RUN: grep {vsubl\\.s8} %t | count 1
; RUN: grep {vsubl\\.s16} %t | count 1
; RUN: grep {vsubl\\.s32} %t | count 1
; RUN: grep {vsubl\\.u8} %t | count 1
; RUN: grep {vsubl\\.u16} %t | count 1
; RUN: grep {vsubl\\.u32} %t | count 1
define <8 x i16> @vsubls8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
%tmp1 = load <8 x i8>* %A
%tmp2 = load <8 x i8>* %B
%tmp3 = call <8 x i16> @llvm.arm.neon.vsubls.v8i16(<8 x i8> %tmp1, <8 x i8> %tmp2)
ret <8 x i16> %tmp3
}
define <4 x i32> @vsubls16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
%tmp1 = load <4 x i16>* %A
%tmp2 = load <4 x i16>* %B
%tmp3 = call <4 x i32> @llvm.arm.neon.vsubls.v4i32(<4 x i16> %tmp1, <4 x i16> %tmp2)
ret <4 x i32> %tmp3
}
define <2 x i64> @vsubls32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
%tmp1 = load <2 x i32>* %A
%tmp2 = load <2 x i32>* %B
%tmp3 = call <2 x i64> @llvm.arm.neon.vsubls.v2i64(<2 x i32> %tmp1, <2 x i32> %tmp2)
ret <2 x i64> %tmp3
}
define <8 x i16> @vsublu8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
%tmp1 = load <8 x i8>* %A
%tmp2 = load <8 x i8>* %B
%tmp3 = call <8 x i16> @llvm.arm.neon.vsublu.v8i16(<8 x i8> %tmp1, <8 x i8> %tmp2)
ret <8 x i16> %tmp3
}
define <4 x i32> @vsublu16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
%tmp1 = load <4 x i16>* %A
%tmp2 = load <4 x i16>* %B
%tmp3 = call <4 x i32> @llvm.arm.neon.vsublu.v4i32(<4 x i16> %tmp1, <4 x i16> %tmp2)
ret <4 x i32> %tmp3
}
define <2 x i64> @vsublu32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
%tmp1 = load <2 x i32>* %A
%tmp2 = load <2 x i32>* %B
%tmp3 = call <2 x i64> @llvm.arm.neon.vsublu.v2i64(<2 x i32> %tmp1, <2 x i32> %tmp2)
ret <2 x i64> %tmp3
}
declare <8 x i16> @llvm.arm.neon.vsubls.v8i16(<8 x i8>, <8 x i8>) nounwind readnone
declare <4 x i32> @llvm.arm.neon.vsubls.v4i32(<4 x i16>, <4 x i16>) nounwind readnone
declare <2 x i64> @llvm.arm.neon.vsubls.v2i64(<2 x i32>, <2 x i32>) nounwind readnone
declare <8 x i16> @llvm.arm.neon.vsublu.v8i16(<8 x i8>, <8 x i8>) nounwind readnone
declare <4 x i32> @llvm.arm.neon.vsublu.v4i32(<4 x i16>, <4 x i16>) nounwind readnone
declare <2 x i64> @llvm.arm.neon.vsublu.v2i64(<2 x i32>, <2 x i32>) nounwind readnone
|
