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
145
146
147
148
149
150
151
152
153
154
155
156
157
|
; RUN: llc -amdgpu-scalarize-global-loads=false -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI %s
declare half @llvm.fabs.f16(half %a)
declare i1 @llvm.amdgcn.class.f16(half %a, i32 %b)
; GCN-LABEL: {{^}}class_f16:
; GCN-DAG: buffer_load_ushort v[[A_F16:[0-9]+]]
; GCN-DAG: buffer_load_dword v[[B_I32:[0-9]+]]
; VI: v_cmp_class_f16_e32 vcc, v[[A_F16]], v[[B_I32]]
; GCN: v_cndmask_b32_e64 v[[R_I32:[0-9]+]]
; GCN: buffer_store_dword v[[R_I32]]
; GCN: s_endpgm
define amdgpu_kernel void @class_f16(
ptr addrspace(1) %r,
ptr addrspace(1) %a,
ptr addrspace(1) %b) {
entry:
%a.val = load half, ptr addrspace(1) %a
%b.val = load i32, ptr addrspace(1) %b
%r.val = call i1 @llvm.amdgcn.class.f16(half %a.val, i32 %b.val)
%r.val.sext = sext i1 %r.val to i32
store i32 %r.val.sext, ptr addrspace(1) %r
ret void
}
; GCN-LABEL: {{^}}class_f16_fabs:
; GCN: s_load_dword s[[SB_I32:[0-9]+]]
; GCN: s_load_dword s[[SA_F16:[0-9]+]]
; GCN: v_mov_b32_e32 [[V_B_I32:v[0-9]+]], s[[SB_I32]]
; VI: v_cmp_class_f16_e64 [[CMP:s\[[0-9]+:[0-9]+\]]], |s[[SA_F16]]|, [[V_B_I32]]
; VI: v_cndmask_b32_e64 v[[VR_I32:[0-9]+]], 0, -1, [[CMP]]
; GCN: buffer_store_dword v[[VR_I32]]
; GCN: s_endpgm
define amdgpu_kernel void @class_f16_fabs(
ptr addrspace(1) %r,
[8 x i32],
half %a.val,
[8 x i32],
i32 %b.val) {
entry:
%a.val.fabs = call half @llvm.fabs.f16(half %a.val)
%r.val = call i1 @llvm.amdgcn.class.f16(half %a.val.fabs, i32 %b.val)
%r.val.sext = sext i1 %r.val to i32
store i32 %r.val.sext, ptr addrspace(1) %r
ret void
}
; GCN-LABEL: {{^}}class_f16_fneg:
; GCN: s_load_dword s[[SB_I32:[0-9]+]]
; GCN: s_load_dword s[[SA_F16:[0-9]+]]
; GCN: v_mov_b32_e32 [[V_B_I32:v[0-9]+]], s[[SB_I32]]
; VI: v_cmp_class_f16_e64 [[CMP:s\[[0-9]+:[0-9]+\]]], -s[[SA_F16]], [[V_B_I32]]
; VI: v_cndmask_b32_e64 v[[VR_I32:[0-9]+]], 0, -1, [[CMP]]
; GCN: buffer_store_dword v[[VR_I32]]
; GCN: s_endpgm
define amdgpu_kernel void @class_f16_fneg(
ptr addrspace(1) %r,
[8 x i32],
half %a.val,
[8 x i32],
i32 %b.val) {
entry:
%a.val.fneg = fsub half -0.0, %a.val
%r.val = call i1 @llvm.amdgcn.class.f16(half %a.val.fneg, i32 %b.val)
%r.val.sext = sext i1 %r.val to i32
store i32 %r.val.sext, ptr addrspace(1) %r
ret void
}
; GCN-LABEL: {{^}}class_f16_fabs_fneg:
; GCN: s_load_dword s[[SB_I32:[0-9]+]]
; GCN: s_load_dword s[[SA_F16:[0-9]+]]
; GCN: v_mov_b32_e32 [[V_B_I32:v[0-9]+]], s[[SB_I32]]
; VI: v_cmp_class_f16_e64 [[CMP:s\[[0-9]+:[0-9]+\]]], -|s[[SA_F16]]|, [[V_B_I32]]
; VI: v_cndmask_b32_e64 v[[VR_I32:[0-9]+]], 0, -1, [[CMP]]
; GCN: buffer_store_dword v[[VR_I32]]
; GCN: s_endpgm
define amdgpu_kernel void @class_f16_fabs_fneg(
ptr addrspace(1) %r,
[8 x i32],
half %a.val,
[8 x i32],
i32 %b.val) {
entry:
%a.val.fabs = call half @llvm.fabs.f16(half %a.val)
%a.val.fabs.fneg = fsub half -0.0, %a.val.fabs
%r.val = call i1 @llvm.amdgcn.class.f16(half %a.val.fabs.fneg, i32 %b.val)
%r.val.sext = sext i1 %r.val to i32
store i32 %r.val.sext, ptr addrspace(1) %r
ret void
}
; GCN-LABEL: {{^}}class_f16_1:
; GCN: s_load_dword s[[SA_F16:[0-9]+]]
; VI: v_cmp_class_f16_e64 [[CMP:s\[[0-9]+:[0-9]+\]]], s[[SA_F16]], 1{{$}}
; VI: v_cndmask_b32_e64 v[[VR_I32:[0-9]+]], 0, -1, [[CMP]]
; GCN: buffer_store_dword v[[VR_I32]]
; GCN: s_endpgm
define amdgpu_kernel void @class_f16_1(
ptr addrspace(1) %r,
half %a.val) {
entry:
%r.val = call i1 @llvm.amdgcn.class.f16(half %a.val, i32 1)
%r.val.sext = sext i1 %r.val to i32
store i32 %r.val.sext, ptr addrspace(1) %r
ret void
}
; GCN-LABEL: {{^}}class_f16_64
; GCN: s_load_dword s[[SA_F16:[0-9]+]]
; VI: v_cmp_class_f16_e64 [[CMP:s\[[0-9]+:[0-9]+\]]], s[[SA_F16]], 64{{$}}
; VI: v_cndmask_b32_e64 v[[VR_I32:[0-9]+]], 0, -1, [[CMP]]
; GCN: buffer_store_dword v[[VR_I32]]
; GCN: s_endpgm
define amdgpu_kernel void @class_f16_64(
ptr addrspace(1) %r,
half %a.val) {
entry:
%r.val = call i1 @llvm.amdgcn.class.f16(half %a.val, i32 64)
%r.val.sext = sext i1 %r.val to i32
store i32 %r.val.sext, ptr addrspace(1) %r
ret void
}
; GCN-LABEL: {{^}}class_f16_full_mask:
; GCN: s_load_dword s[[SA_F16:[0-9]+]]
; VI: v_mov_b32_e32 v[[MASK:[0-9]+]], 0x3ff{{$}}
; VI: v_cmp_class_f16_e32 vcc, s[[SA_F16]], v[[MASK]]
; VI: v_cndmask_b32_e64 v[[VR_I32:[0-9]+]], 0, -1, vcc
; GCN: buffer_store_dword v[[VR_I32]]
; GCN: s_endpgm
define amdgpu_kernel void @class_f16_full_mask(
ptr addrspace(1) %r,
half %a.val) {
entry:
%r.val = call i1 @llvm.amdgcn.class.f16(half %a.val, i32 1023)
%r.val.sext = sext i1 %r.val to i32
store i32 %r.val.sext, ptr addrspace(1) %r
ret void
}
; GCN-LABEL: {{^}}class_f16_nine_bit_mask:
; GCN: s_load_dword s[[SA_F16:[0-9]+]]
; VI: v_mov_b32_e32 v[[MASK:[0-9]+]], 0x1ff{{$}}
; VI: v_cmp_class_f16_e32 vcc, s[[SA_F16]], v[[MASK]]
; VI: v_cndmask_b32_e64 v[[VR_I32:[0-9]+]], 0, -1, vcc
; GCN: buffer_store_dword v[[VR_I32]]
; GCN: s_endpgm
define amdgpu_kernel void @class_f16_nine_bit_mask(
ptr addrspace(1) %r,
half %a.val) {
entry:
%r.val = call i1 @llvm.amdgcn.class.f16(half %a.val, i32 511)
%r.val.sext = sext i1 %r.val to i32
store i32 %r.val.sext, ptr addrspace(1) %r
ret void
}
|
