aboutsummaryrefslogtreecommitdiff
path: root/test/CodeGen/ARM/neon_ld2.ll
blob: 497619ed746a6c83ee0d6e3953adb7311dace020 (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
; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
; RUN: llc < %s -march=arm -mcpu=swift | FileCheck %s --check-prefix=SWIFT

; CHECK: t1
; CHECK: vld1.64
; CHECK: vld1.64
; CHECK: vadd.i64 q
; CHECK: vst1.64
; SWIFT: t1
; SWIFT: vld1.64 {{.d[0-9]+, d[0-9]+}, \[r[0-9]+, :128\]}}
; SWIFT: vld1.64 {{.d[0-9]+, d[0-9]+}, \[r[0-9]+, :128\]}}
; SWIFT: vadd.i64 q
; SWIFT: vst1.64 {{.d[0-9]+, d[0-9]+}, \[r[0-9]+, :128\]}}
define void @t1(<4 x i32>* %r, <2 x i64>* %a, <2 x i64>* %b) nounwind {
entry:
	%0 = load <2 x i64>* %a, align 16		; <<2 x i64>> [#uses=1]
	%1 = load <2 x i64>* %b, align 16		; <<2 x i64>> [#uses=1]
	%2 = add <2 x i64> %0, %1		; <<2 x i64>> [#uses=1]
	%3 = bitcast <2 x i64> %2 to <4 x i32>		; <<4 x i32>> [#uses=1]
	store <4 x i32> %3, <4 x i32>* %r, align 16
	ret void
}

; CHECK: t2
; CHECK: vld1.64
; CHECK: vld1.64
; CHECK: vsub.i64 q
; CHECK: vmov r0, r1, d
; CHECK: vmov r2, r3, d
; SWIFT: t2
; SWIFT: vld1.64 {{.d[0-9]+, d[0-9]+}, \[r[0-9]+, :128\]}}
; SWIFT: vld1.64 {{.d[0-9]+, d[0-9]+}, \[r[0-9]+, :128\]}}
; SWIFT: vsub.i64 q
; SWIFT: vmov r0, r1, d
; SWIFT: vmov r2, r3, d
define <4 x i32> @t2(<2 x i64>* %a, <2 x i64>* %b) nounwind readonly {
entry:
	%0 = load <2 x i64>* %a, align 16		; <<2 x i64>> [#uses=1]
	%1 = load <2 x i64>* %b, align 16		; <<2 x i64>> [#uses=1]
	%2 = sub <2 x i64> %0, %1		; <<2 x i64>> [#uses=1]
	%3 = bitcast <2 x i64> %2 to <4 x i32>		; <<4 x i32>> [#uses=1]
	ret <4 x i32> %3
}

; Limited alignment.
; SWIFT: t3
; SWIFT: vld1.64 {{.d[0-9]+, d[0-9]+}, \[r[0-9]+}}
; SWIFT: vld1.64 {{.d[0-9]+, d[0-9]+}, \[r[0-9]+}}
; SWIFT: vadd.i64 q
; SWIFT: vst1.64 {{.d[0-9]+, d[0-9]+}, \[r[0-9]+}}
define void @t3(<4 x i32>* %r, <2 x i64>* %a, <2 x i64>* %b) nounwind {
entry:
	%0 = load <2 x i64>* %a, align 8
	%1 = load <2 x i64>* %b, align 8
	%2 = add <2 x i64> %0, %1
	%3 = bitcast <2 x i64> %2 to <4 x i32>
	store <4 x i32> %3, <4 x i32>* %r, align 8
	ret void
}