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
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
|
;; Test float 16 operations.
(module
(memory (data "\40\51\AD\DE"))
(func (export "f32.load_f16") (result f32) (f32.load_f16 (i32.const 0)))
(func (export "f32.store_f16") (f32.store_f16 (i32.const 0) (f32.const 100.5)))
(func (export "i32.load16_u") (result i32) (i32.load16_u (i32.const 2)))
(func (export "f16x8.splat") (param $0 f32) (result v128) (f16x8.splat (local.get $0)))
(func (export "f16x8.extract_lane_first") (param $0 v128) (result f32) (f16x8.extract_lane 0 (local.get $0)))
(func (export "f16x8.extract_lane_last") (param $0 v128) (result f32) (f16x8.extract_lane 7 (local.get $0)))
(func (export "f16x8.replace_lane_first") (param $0 v128) (param $1 f32) (result v128) (f16x8.replace_lane 0 (local.get $0) (local.get $1)))
(func (export "f16x8.replace_lane_last") (param $0 v128) (param $1 f32) (result v128) (f16x8.replace_lane 7 (local.get $0) (local.get $1)))
(func (export "f16x8.eq") (param $0 v128) (param $1 v128) (result v128) (f16x8.eq (local.get $0) (local.get $1)))
(func (export "f16x8.ne") (param $0 v128) (param $1 v128) (result v128) (f16x8.ne (local.get $0) (local.get $1)))
(func (export "f16x8.lt") (param $0 v128) (param $1 v128) (result v128) (f16x8.lt (local.get $0) (local.get $1)))
(func (export "f16x8.gt") (param $0 v128) (param $1 v128) (result v128) (f16x8.gt (local.get $0) (local.get $1)))
(func (export "f16x8.le") (param $0 v128) (param $1 v128) (result v128) (f16x8.le (local.get $0) (local.get $1)))
(func (export "f16x8.ge") (param $0 v128) (param $1 v128) (result v128) (f16x8.ge (local.get $0) (local.get $1)))
(func (export "f16x8.add") (param $0 v128) (param $1 v128) (result v128) (f16x8.add (local.get $0) (local.get $1)))
(func (export "f16x8.sub") (param $0 v128) (param $1 v128) (result v128) (f16x8.sub (local.get $0) (local.get $1)))
(func (export "f16x8.mul") (param $0 v128) (param $1 v128) (result v128) (f16x8.mul (local.get $0) (local.get $1)))
(func (export "f16x8.div") (param $0 v128) (param $1 v128) (result v128) (f16x8.div (local.get $0) (local.get $1)))
(func (export "f16x8.min") (param $0 v128) (param $1 v128) (result v128) (f16x8.min (local.get $0) (local.get $1)))
(func (export "f16x8.max") (param $0 v128) (param $1 v128) (result v128) (f16x8.max (local.get $0) (local.get $1)))
(func (export "f16x8.pmin") (param $0 v128) (param $1 v128) (result v128) (f16x8.pmin (local.get $0) (local.get $1)))
(func (export "f16x8.pmax") (param $0 v128) (param $1 v128) (result v128) (f16x8.pmax (local.get $0) (local.get $1)))
(func (export "f16x8.abs") (param $0 v128) (result v128) (f16x8.abs (local.get $0)))
(func (export "f16x8.neg") (param $0 v128) (result v128) (f16x8.neg (local.get $0)))
(func (export "f16x8.sqrt") (param $0 v128) (result v128) (f16x8.sqrt (local.get $0)))
(func (export "f16x8.ceil") (param $0 v128) (result v128) (f16x8.ceil (local.get $0)))
(func (export "f16x8.floor") (param $0 v128) (result v128) (f16x8.floor (local.get $0)))
(func (export "f16x8.trunc") (param $0 v128) (result v128) (f16x8.trunc (local.get $0)))
(func (export "f16x8.nearest") (param $0 v128) (result v128) (f16x8.nearest (local.get $0)))
(func (export "f16x8.relaxed_madd") (param $0 v128) (param $1 v128) (param $2 v128) (result v128) (f16x8.relaxed_madd (local.get $0) (local.get $1) (local.get $2)))
(func (export "f16x8.relaxed_nmadd") (param $0 v128) (param $1 v128) (param $2 v128) (result v128) (f16x8.relaxed_nmadd (local.get $0) (local.get $1) (local.get $2)))
(func (export "i16x8.trunc_sat_f16x8_s") (param $0 v128) (result v128) (i16x8.trunc_sat_f16x8_s (local.get $0)))
(func (export "i16x8.trunc_sat_f16x8_u") (param $0 v128) (result v128) (i16x8.trunc_sat_f16x8_u (local.get $0)))
(func (export "f16x8.convert_i16x8_s") (param $0 v128) (result v128) (f16x8.convert_i16x8_s (local.get $0)))
(func (export "f16x8.convert_i16x8_u") (param $0 v128) (result v128) (f16x8.convert_i16x8_u (local.get $0)))
;; Multiple operation tests:
(func (export "splat_replace") (result v128) (f16x8.replace_lane 0 (f16x8.splat (f32.const 1)) (f32.const 99))
)
)
(assert_return (invoke "f32.load_f16") (f32.const 42.0))
(invoke "f32.store_f16")
(assert_return (invoke "f32.load_f16") (f32.const 100.5))
;; Ensure that the above operations didn't write to memory they shouldn't have.
(assert_return (invoke "i32.load16_u") (i32.const 0xDEAD))
;; lane accesses
(assert_return (invoke "f16x8.splat" (f32.const 100.5)) (v128.const i16x8 0x5648 0x5648 0x5648 0x5648 0x5648 0x5648 0x5648 0x5648))
(assert_return (invoke "f16x8.extract_lane_first" (v128.const i16x8 0x5648 0 0 0 0 0 0 0)) (f32.const 100.5))
(assert_return (invoke "f16x8.extract_lane_last" (v128.const i16x8 0 0 0 0 0 0 0 0xc500)) (f32.const -5))
(assert_return (invoke "f16x8.replace_lane_first" (v128.const i64x2 0 0) (f32.const 100.5)) (v128.const i16x8 0x5648 0 0 0 0 0 0 0))
(assert_return (invoke "f16x8.replace_lane_last" (v128.const i64x2 0 0) (f32.const 100.5)) (v128.const i16x8 0 0 0 0 0 0 0 0x5648))
;; comparisons
(assert_return (invoke "f16x8.eq"
;; 0 -1 1 nan inf nan inf nan
;; 0 0 -1 nan inf 0 -inf inf
(v128.const i16x8 0 0xbc00 0x3c00 0x7e00 0x7c00 0x7e00 0x7c00 0x7e00)
(v128.const i16x8 0 0 0xbc00 0x7e00 0x7c00 0 0xfc00 0x7c00)
)
(v128.const i16x8 -1 0 0 0 -1 0 0 0)
)
(assert_return (invoke "f16x8.ne"
;; 0 -1 1 nan inf nan inf nan
;; 0 0 -1 nan inf 0 -inf inf
(v128.const i16x8 0 0xbc00 0x3c00 0x7e00 0x7c00 0x7e00 0x7c00 0x7e00)
(v128.const i16x8 0 0 0xbc00 0x7e00 0x7c00 0 0xfc00 0x7c00)
)
(v128.const i16x8 0 -1 -1 -1 0 -1 -1 -1)
)
(assert_return (invoke "f16x8.lt"
;; 0 -1 1 nan inf nan inf nan
;; 0 0 -1 nan inf 0 -inf inf
(v128.const i16x8 0 0xbc00 0x3c00 0x7e00 0x7c00 0x7e00 0x7c00 0x7e00)
(v128.const i16x8 0 0 0xbc00 0x7e00 0x7c00 0 0xfc00 0x7c00)
)
(v128.const i16x8 0 -1 0 0 0 0 0 0)
)
(assert_return (invoke "f16x8.gt"
;; 0 -1 1 nan inf nan inf nan
;; 0 0 -1 nan inf 0 -inf inf
(v128.const i16x8 0 0xbc00 0x3c00 0x7e00 0x7c00 0x7e00 0x7c00 0x7e00)
(v128.const i16x8 0 0 0xbc00 0x7e00 0x7c00 0 0xfc00 0x7c00)
)
(v128.const i16x8 0 0 -1 0 0 0 -1 0)
)
(assert_return (invoke "f16x8.le"
;; 0 -1 1 nan inf nan inf nan
;; 0 0 -1 nan inf 0 -inf inf
(v128.const i16x8 0 0xbc00 0x3c00 0x7e00 0x7c00 0x7e00 0x7c00 0x7e00)
(v128.const i16x8 0 0 0xbc00 0x7e00 0x7c00 0 0xfc00 0x7c00)
)
(v128.const i16x8 -1 -1 0 0 -1 0 0 0)
)
(assert_return (invoke "f16x8.ge"
;; 0 -1 1 nan inf nan inf nan
;; 0 0 -1 nan inf 0 -inf inf
(v128.const i16x8 0 0xbc00 0x3c00 0x7e00 0x7c00 0x7e00 0x7c00 0x7e00)
(v128.const i16x8 0 0 0xbc00 0x7e00 0x7c00 0 0xfc00 0x7c00)
)
(v128.const i16x8 -1 0 -1 0 -1 0 -1 0)
)
;; arithmetic operations
(assert_return (invoke "f16x8.add"
;; nan -nan inf 1.5 -1 -1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0x3c00 0x3c00)
;; 42 -nan inf 1.5 0 1 0 1
(v128.const i16x8 0x5140 0xfe00 0x7c00 0x3e00 0 0x3c00 0 0x3c00))
;; nan nan inf 3 -1 0 1 2
(v128.const i16x8 0x7e00 0x7e00 0x7c00 0x4200 0xbc00 0 0x3c00 0x4000))
(assert_return (invoke "f16x8.sub"
;; nan -nan inf 1.5 -1 -1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0x3c00 0x3c00)
;; 42 -nan inf 1.5 0 1 0 1
(v128.const i16x8 0x5140 0xfe00 0x7c00 0x3e00 0 0x3c00 0 0x3c00))
;; nan nan nan 0 -1 -2 1 0
(v128.const i16x8 0x7e00 0x7e00 0x7e00 0 0xbc00 0xc000 0x3c00 0))
(assert_return (invoke "f16x8.mul"
;; nan -nan inf 1.5 -1 -1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0x3c00 0x3c00)
;; 42 -nan inf 1.5 0 1 0 1
(v128.const i16x8 0x5140 0xfe00 0x7c00 0x3e00 0 0x3c00 0 0x3c00))
;; nan nan inf 2.25 0 -1 0 1
(v128.const i16x8 0x7e00 0x7e00 0x7c00 0x4080 0x8000 0xbc00 0 0x3c00))
(assert_return (invoke "f16x8.div"
;; nan -nan inf 1.5 -1 -1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0x3c00 0x3c00)
;; 42 -nan inf 1.5 0 1 0 1
(v128.const i16x8 0x5140 0xfe00 0x7c00 0x3e00 0 0x3c00 0 0x3c00))
;; nan nan nan 1 -inf -1 inf 1
(v128.const i16x8 0x7e00 0x7e00 0x7e00 0x3c00 0xfc00 0xbc00 0x7c00 0x3c00))
(assert_return (invoke "f16x8.min"
;; nan -nan inf 1.5 -1 -1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0x3c00 0x3c00)
;; 42 -nan inf 1.5 0 1 0 1
(v128.const i16x8 0x5140 0xfe00 0x7c00 0x3e00 0 0x3c00 0 0x3c00))
;; nan nan inf 1.5 -1 -1 0 1
(v128.const i16x8 0x7e00 0x7e00 0x7c00 0x3e00 0xbc00 0xbc00 0 0x3c00))
(assert_return (invoke "f16x8.max"
;; nan -nan inf 1.5 -1 -1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0x3c00 0x3c00)
;; 42 -nan inf 1.5 0 1 0 1
(v128.const i16x8 0x5140 0xfe00 0x7c00 0x3e00 0 0x3c00 0 0x3c00))
;; nan nan inf 1.5 0 1 1 1
(v128.const i16x8 0x7e00 0x7e00 0x7c00 0x3e00 0 0x3c00 0x3c00 0x3c00))
(assert_return (invoke "f16x8.pmin"
;; nan -nan inf 1.5 -1 -1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0x3c00 0x3c00)
;; 42 -nan inf 1.5 0 1 0 1
(v128.const i16x8 0x5140 0xfe00 0x7c00 0x3e00 0 0x3c00 0 0x3c00))
;; nan -nan inf 1.5 -1 -1 0 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0 0x3c00))
(assert_return (invoke "f16x8.pmax"
;; nan -nan inf 1.5 -1 -1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0xbc00 0xbc00 0x3c00 0x3c00)
;; 42 -nan inf 1.5 0 1 0 1
(v128.const i16x8 0x5140 0xfe00 0x7c00 0x3e00 0 0x3c00 0 0x3c00))
;; nan -nan inf 1.5 0 1 1 1
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0x3e00 0 0x3c00 0x3c00 0x3c00))
;; unary arithmetic
(assert_return (invoke "f16x8.abs"
;; nan -nan inf -inf -1 1 1.5 1.2...
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0xfc00 0xbc00 0x3c00 0x3e00 0x3ccd))
;; nan nan inf inf 1 1 1.5 1.2...
(v128.const i16x8 0x7e00 0x7e00 0x7c00 0x7c00 0x3c00 0x3c00 0x3e00 0x3ccd))
(assert_return (invoke "f16x8.neg"
;; nan -nan inf -inf -1 1 1.5 1.2...
(v128.const i16x8 0x7e00 0xfe00 0x7c00 0xfc00 0xbc00 0x3c00 0x3e00 0x3ccd))
;; -nan nan -inf inf 1 -1 -1.5 -1.2...
(v128.const i16x8 0xfe00 0x7e00 0xfc00 0x7c00 0x3c00 0xbc00 0xbe00 0xbccd))
;; XXX Avoid tests that return -nan since it's non-deterministic.
(assert_return (invoke "f16x8.sqrt"
;; nan 0 inf 4 16 1 1.5 1.2...
(v128.const i16x8 0x7e00 0 0x7c00 0x4400 0x4c00 0x3c00 0x3e00 0x3ccd))
;; nan 0 inf 2 4 1 1.22.. 1.09...
(v128.const i16x8 0x7e00 0 0x7c00 0x4000 0x4400 0x3c00 0x3ce6 0x3c62))
(assert_return (invoke "f16x8.ceil"
;; nan 0 inf -inf -1 1 1.5 1.2...
(v128.const i16x8 0x7e00 0 0x7c00 0xfc00 0xbc00 0x3c00 0x3e00 0x3ccd))
;; nan 0 inf -inf -1 1 2 2
(v128.const i16x8 0x7e00 0 0x7c00 0xfc00 0xbc00 0x3c00 0x4000 0x4000))
(assert_return (invoke "f16x8.floor"
;; nan 0 inf -inf -1 1 1.5 1.2...
(v128.const i16x8 0x7e00 0 0x7c00 0xfc00 0xbc00 0x3c00 0x3e00 0x3ccd))
;; nan 0 inf -inf -1 1 1 1
(v128.const i16x8 0x7e00 0 0x7c00 0xfc00 0xbc00 0x3c00 0x3c00 0x3c00))
(assert_return (invoke "f16x8.nearest"
;; nan 0 inf -inf -1 1 1.5 1.2...
(v128.const i16x8 0x7e00 0 0x7c00 0xfc00 0xbc00 0x3c00 0x3e00 0x3ccd))
;; nan 0 inf -inf -1 1 2 1
(v128.const i16x8 0x7e00 0 0x7c00 0xfc00 0xbc00 0x3c00 0x4000 0x3c00))
;; ternary operations
(assert_return (invoke "f16x8.relaxed_madd"
;; Lane 0 illustrates the difference between fused/unfused. e.g.
;; fused: (positive overflow) + -inf = -inf
;; unfused: (inf) + -inf = NaN
;;
;; 1e4 inf -1 0 1 1.5 -2 1
(v128.const i16x8 0x70e2 0x7c00 0xbc00 0 0x3c00 0x3e00 0xc000 0x3c00)
;; 1e4 inf -1 0 1 1.5 4 1
(v128.const i16x8 0x70e2 0x7c00 0xbc00 0 0x3c00 0x3e00 0x4400 0x3c00)
;; -inf inf -1 0 1 2 1 -1
(v128.const i16x8 0xfc00 0x7c00 0xbc00 0 0x3c00 0x4000 0x3c00 0xbc00))
;; -inf inf 0 0 2 4.25 -7 0
(v128.const i16x8 0xfc00 0x7c00 0 0 0x4000 0x4440 0xc700 0))
(assert_return (invoke "f16x8.relaxed_nmadd"
;; Lane 0 illustrates the difference between fused/unfused. e.g.
;; fused: -(positive overflow) + inf = inf
;; unfused: (-inf) + -inf = NaN
;;
;; 1e4 -inf -1 0 1 1.5 -2 1
(v128.const i16x8 0x70e2 0xfc00 0xbc00 0 0x3c00 0x3e00 0xc000 0x3c00)
;; 1e4 inf -1 0 1 1.5 4 1
(v128.const i16x8 0x70e2 0x7c00 0xbc00 0 0x3c00 0x3e00 0x4400 0x3c00)
;; inf inf -1 0 1 2 1 -1
(v128.const i16x8 0x7c00 0x7c00 0xbc00 0 0x3c00 0x4000 0x3c00 0xbc00))
;; inf inf -2 0 0 -0.25 9 -2
(v128.const i16x8 0x7c00 0x7c00 0xc000 0 0 0xb400 0x4880 0xc000))
(assert_return (invoke "splat_replace")
(v128.const i16x8 0x5630 0x3c00 0x3c00 0x3c00 0x3c00 0x3c00 0x3c00 0x3c00)
)
;; conversions
(assert_return (invoke "i16x8.trunc_sat_f16x8_s"
;; 42 nan inf -inf 65504 -65504 0 0
(v128.const i16x8 0x5140 0x7e00 0x7c00 0xfc00 0x7bff 0xfbff 0 0))
(v128.const i16x8 42 0 32767 -32768 32767 -32768 0 0))
(assert_return (invoke "i16x8.trunc_sat_f16x8_u"
;; 42 nan inf -inf 65504 -65504 0 0
(v128.const i16x8 0x5140 0x7e00 0x7c00 0xfc00 0x7bff 0xfbff 0 0))
(v128.const i16x8 42 0 65535 0 65504 0 0 0))
(assert_return (invoke "f16x8.convert_i16x8_s"
;; 32767 is not representable as a whole integer in FP16, so it becomes 32768.
(v128.const i16x8 0 1 -1 32767 -32768 0 0 0))
;; 0 1 -1 32768 -32768 0 0 0
(v128.const i16x8 0 0x3c00 0xbc00 0x7800 0xf800 0 0 0))
(assert_return (invoke "f16x8.convert_i16x8_u"
;; Unlike f32/64, f16 can't represent the full 2^16 integer range so 2^16 becomes infinity.
(v128.const i16x8 0 1 -1 -32 0 0 0 0))
;; 1 inf 65504
(v128.const i16x8 0 0x3c00 0x7c00 0x7bff 0 0 0 0))
|
