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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
; RUN: llc --mtriple=loongarch64 --mattr=+lsx < %s | FileCheck %s
define void @fdiv_v4f32(ptr %res, ptr %a0, ptr %a1) nounwind {
; CHECK-LABEL: fdiv_v4f32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vld $vr0, $a1, 0
; CHECK-NEXT: vld $vr1, $a2, 0
; CHECK-NEXT: vfdiv.s $vr0, $vr0, $vr1
; CHECK-NEXT: vst $vr0, $a0, 0
; CHECK-NEXT: ret
entry:
%v0 = load <4 x float>, ptr %a0
%v1 = load <4 x float>, ptr %a1
%v2 = fdiv <4 x float> %v0, %v1
store <4 x float> %v2, ptr %res
ret void
}
define void @fdiv_v2f64(ptr %res, ptr %a0, ptr %a1) nounwind {
; CHECK-LABEL: fdiv_v2f64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vld $vr0, $a1, 0
; CHECK-NEXT: vld $vr1, $a2, 0
; CHECK-NEXT: vfdiv.d $vr0, $vr0, $vr1
; CHECK-NEXT: vst $vr0, $a0, 0
; CHECK-NEXT: ret
entry:
%v0 = load <2 x double>, ptr %a0
%v1 = load <2 x double>, ptr %a1
%v2 = fdiv <2 x double> %v0, %v1
store <2 x double> %v2, ptr %res
ret void
}
;; 1.0 / vec
define void @one_fdiv_v4f32(ptr %res, ptr %a0) nounwind {
; CHECK-LABEL: one_fdiv_v4f32:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vld $vr0, $a1, 0
; CHECK-NEXT: vfrecip.s $vr0, $vr0
; CHECK-NEXT: vst $vr0, $a0, 0
; CHECK-NEXT: ret
entry:
%v0 = load <4 x float>, ptr %a0
%div = fdiv <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0>, %v0
store <4 x float> %div, ptr %res
ret void
}
define void @one_fdiv_v2f64(ptr %res, ptr %a0) nounwind {
; CHECK-LABEL: one_fdiv_v2f64:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: vld $vr0, $a1, 0
; CHECK-NEXT: vfrecip.d $vr0, $vr0
; CHECK-NEXT: vst $vr0, $a0, 0
; CHECK-NEXT: ret
entry:
%v0 = load <2 x double>, ptr %a0
%div = fdiv <2 x double> <double 1.0, double 1.0>, %v0
store <2 x double> %div, ptr %res
ret void
}
|
