xref: /llvm-project/llvm/test/MC/AMDGPU/gfx12_asm_vop2_t16_promote.s (revision c3241a9a4de3ef71a82f9434f84fa7437fe43f9a)
1// NOTE: Assertions have been autogenerated by utils/update_mc_test_checks.py UTC_ARGS: --sort --version 5
2// RUN: llvm-mc -triple=amdgcn -mcpu=gfx1200 -mattr=+real-true16 -show-encoding %s 2>&1 | FileCheck --check-prefix=GFX12 --implicit-check-not=_e32 %s
3
4v_add_f16 v255.h, v1.h, v2.h
5// GFX12: v_add_f16_e64 v255.h, v1.h, v2.h op_sel:[1,1,1] ; encoding: [0xff,0x58,0x32,0xd5,0x01,0x05,0x02,0x00]
6
7v_add_f16 v255.h, v1.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
8// GFX12: v_add_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x58,0x32,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
9
10v_add_f16 v255.h, v1.h, v2.h quad_perm:[3,2,1,0]
11// GFX12: v_add_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x58,0x32,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
12
13v_add_f16 v255.l, v1.l, v2.l
14// GFX12: v_add_f16_e64 v255.l, v1.l, v2.l        ; encoding: [0xff,0x00,0x32,0xd5,0x01,0x05,0x02,0x00]
15
16v_add_f16 v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
17// GFX12: v_add_f16_e64_dpp v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x00,0x32,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
18
19v_add_f16 v255.l, v1.l, v2.l quad_perm:[3,2,1,0]
20// GFX12: v_add_f16_e64_dpp v255.l, v1.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x00,0x32,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
21
22v_add_f16 v5.h, v1.h, v255.h
23// GFX12: v_add_f16_e64 v5.h, v1.h, v255.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x32,0xd5,0x01,0xff,0x03,0x00]
24
25v_add_f16 v5.h, v1.h, v255.h dpp8:[7,6,5,4,3,2,1,0]
26// GFX12: v_add_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x32,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
27
28v_add_f16 v5.h, v1.h, v255.h quad_perm:[3,2,1,0]
29// GFX12: v_add_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x32,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
30
31v_add_f16 v5.h, v255.h, v2.h
32// GFX12: v_add_f16_e64 v5.h, v255.h, v2.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x32,0xd5,0xff,0x05,0x02,0x00]
33
34v_add_f16 v5.h, v255.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
35// GFX12: v_add_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x32,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
36
37v_add_f16 v5.h, v255.h, v2.h quad_perm:[3,2,1,0]
38// GFX12: v_add_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x32,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
39
40v_add_f16 v5.l, v1.l, v255.l
41// GFX12: v_add_f16_e64 v5.l, v1.l, v255.l        ; encoding: [0x05,0x00,0x32,0xd5,0x01,0xff,0x03,0x00]
42
43v_add_f16 v5.l, v1.l, v255.l dpp8:[7,6,5,4,3,2,1,0]
44// GFX12: v_add_f16_e64_dpp v5.l, v1.l, v255.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x32,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
45
46v_add_f16 v5.l, v1.l, v255.l quad_perm:[3,2,1,0]
47// GFX12: v_add_f16_e64_dpp v5.l, v1.l, v255.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x32,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
48
49v_add_f16 v5.l, v255.l, v2.l
50// GFX12: v_add_f16_e64 v5.l, v255.l, v2.l        ; encoding: [0x05,0x00,0x32,0xd5,0xff,0x05,0x02,0x00]
51
52v_add_f16 v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
53// GFX12: v_add_f16_e64_dpp v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x32,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
54
55v_add_f16 v5.l, v255.l, v2.l quad_perm:[3,2,1,0]
56// GFX12: v_add_f16_e64_dpp v5.l, v255.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x32,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
57
58v_fmac_f16 v255, v1, v2
59// GFX12: v_fmac_f16_e64 v255, v1, v2             ; encoding: [0xff,0x00,0x36,0xd5,0x01,0x05,0x02,0x00]
60
61v_fmac_f16 v5, v1, v255
62// GFX12: v_fmac_f16_e64 v5, v1, v255             ; encoding: [0x05,0x00,0x36,0xd5,0x01,0xff,0x03,0x00]
63
64v_fmac_f16 v5, v255, v2
65// GFX12: v_fmac_f16_e64 v5, v255, v2             ; encoding: [0x05,0x00,0x36,0xd5,0xff,0x05,0x02,0x00]
66
67v_ldexp_f16 v255.h, v1.h, v2.h
68// GFX12: v_ldexp_f16_e64 v255.h, v1.h, v2.h op_sel:[1,1,1] ; encoding: [0xff,0x58,0x3b,0xd5,0x01,0x05,0x02,0x00]
69
70v_ldexp_f16 v255.h, v1.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
71// GFX12: v_ldexp_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x58,0x3b,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
72
73v_ldexp_f16 v255.h, v1.h, v2.h quad_perm:[3,2,1,0]
74// GFX12: v_ldexp_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x58,0x3b,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
75
76v_ldexp_f16 v255.l, v1.l, v2.l
77// GFX12: v_ldexp_f16_e64 v255.l, v1.l, v2.l      ; encoding: [0xff,0x00,0x3b,0xd5,0x01,0x05,0x02,0x00]
78
79v_ldexp_f16 v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
80// GFX12: v_ldexp_f16_e64_dpp v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x00,0x3b,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
81
82v_ldexp_f16 v255.l, v1.l, v2.l quad_perm:[3,2,1,0]
83// GFX12: v_ldexp_f16_e64_dpp v255.l, v1.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x00,0x3b,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
84
85v_ldexp_f16 v5.h, v1.h, v255.h
86// GFX12: v_ldexp_f16_e64 v5.h, v1.h, v255.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x3b,0xd5,0x01,0xff,0x03,0x00]
87
88v_ldexp_f16 v5.h, v1.h, v255.h dpp8:[7,6,5,4,3,2,1,0]
89// GFX12: v_ldexp_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x3b,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
90
91v_ldexp_f16 v5.h, v1.h, v255.h quad_perm:[3,2,1,0]
92// GFX12: v_ldexp_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x3b,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
93
94v_ldexp_f16 v5.h, v255.h, v2.h
95// GFX12: v_ldexp_f16_e64 v5.h, v255.h, v2.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x3b,0xd5,0xff,0x05,0x02,0x00]
96
97v_ldexp_f16 v5.h, v255.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
98// GFX12: v_ldexp_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x3b,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
99
100v_ldexp_f16 v5.h, v255.h, v2.h quad_perm:[3,2,1,0]
101// GFX12: v_ldexp_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x3b,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
102
103v_ldexp_f16 v5.l, v255.l, v2.l
104// GFX12: v_ldexp_f16_e64 v5.l, v255.l, v2.l      ; encoding: [0x05,0x00,0x3b,0xd5,0xff,0x05,0x02,0x00]
105
106v_ldexp_f16 v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
107// GFX12: v_ldexp_f16_e64_dpp v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x3b,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
108
109v_ldexp_f16 v5.l, v255.l, v2.l quad_perm:[3,2,1,0]
110// GFX12: v_ldexp_f16_e64_dpp v5.l, v255.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x3b,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
111
112v_max_num_f16 v255, v1, v2
113// GFX12: v_max_num_f16_e64 v255, v1, v2          ; encoding: [0xff,0x00,0x31,0xd5,0x01,0x05,0x02,0x00]
114
115v_max_num_f16 v255, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
116// GFX12: v_max_num_f16_e64_dpp v255, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x00,0x31,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
117
118v_max_num_f16 v255, v1, v2 quad_perm:[3,2,1,0]
119// GFX12: v_max_num_f16_e64_dpp v255, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x00,0x31,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
120
121v_max_num_f16 v5, v1, v255
122// GFX12: v_max_num_f16_e64 v5, v1, v255          ; encoding: [0x05,0x00,0x31,0xd5,0x01,0xff,0x03,0x00]
123
124v_max_num_f16 v5, v1, v255 dpp8:[7,6,5,4,3,2,1,0]
125// GFX12: v_max_num_f16_e64_dpp v5, v1, v255 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x31,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
126
127v_max_num_f16 v5, v1, v255 quad_perm:[3,2,1,0]
128// GFX12: v_max_num_f16_e64_dpp v5, v1, v255 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x31,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
129
130v_max_num_f16 v5, v255, v2
131// GFX12: v_max_num_f16_e64 v5, v255, v2          ; encoding: [0x05,0x00,0x31,0xd5,0xff,0x05,0x02,0x00]
132
133v_max_num_f16 v5, v255, v2 dpp8:[7,6,5,4,3,2,1,0]
134// GFX12: v_max_num_f16_e64_dpp v5, v255, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x31,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
135
136v_max_num_f16 v5, v255, v2 quad_perm:[3,2,1,0]
137// GFX12: v_max_num_f16_e64_dpp v5, v255, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x31,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
138
139v_min_num_f16 v255, v1, v2
140// GFX12: v_min_num_f16_e64 v255, v1, v2          ; encoding: [0xff,0x00,0x30,0xd5,0x01,0x05,0x02,0x00]
141
142v_min_num_f16 v255, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
143// GFX12: v_min_num_f16_e64_dpp v255, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x00,0x30,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
144
145v_min_num_f16 v255, v1, v2 quad_perm:[3,2,1,0]
146// GFX12: v_min_num_f16_e64_dpp v255, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x00,0x30,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
147
148v_min_num_f16 v5, v1, v255
149// GFX12: v_min_num_f16_e64 v5, v1, v255          ; encoding: [0x05,0x00,0x30,0xd5,0x01,0xff,0x03,0x00]
150
151v_min_num_f16 v5, v1, v255 dpp8:[7,6,5,4,3,2,1,0]
152// GFX12: v_min_num_f16_e64_dpp v5, v1, v255 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x30,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
153
154v_min_num_f16 v5, v1, v255 quad_perm:[3,2,1,0]
155// GFX12: v_min_num_f16_e64_dpp v5, v1, v255 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x30,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
156
157v_min_num_f16 v5, v255, v2
158// GFX12: v_min_num_f16_e64 v5, v255, v2          ; encoding: [0x05,0x00,0x30,0xd5,0xff,0x05,0x02,0x00]
159
160v_min_num_f16 v5, v255, v2 dpp8:[7,6,5,4,3,2,1,0]
161// GFX12: v_min_num_f16_e64_dpp v5, v255, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x30,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
162
163v_min_num_f16 v5, v255, v2 quad_perm:[3,2,1,0]
164// GFX12: v_min_num_f16_e64_dpp v5, v255, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x30,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
165
166v_mul_f16 v255.h, v1.h, v2.h
167// GFX12: v_mul_f16_e64 v255.h, v1.h, v2.h op_sel:[1,1,1] ; encoding: [0xff,0x58,0x35,0xd5,0x01,0x05,0x02,0x00]
168
169v_mul_f16 v255.h, v1.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
170// GFX12: v_mul_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x58,0x35,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
171
172v_mul_f16 v255.h, v1.h, v2.h quad_perm:[3,2,1,0]
173// GFX12: v_mul_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x58,0x35,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
174
175v_mul_f16 v255.l, v1.l, v2.l
176// GFX12: v_mul_f16_e64 v255.l, v1.l, v2.l        ; encoding: [0xff,0x00,0x35,0xd5,0x01,0x05,0x02,0x00]
177
178v_mul_f16 v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
179// GFX12: v_mul_f16_e64_dpp v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x00,0x35,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
180
181v_mul_f16 v255.l, v1.l, v2.l quad_perm:[3,2,1,0]
182// GFX12: v_mul_f16_e64_dpp v255.l, v1.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x00,0x35,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
183
184v_mul_f16 v5.h, v1.h, v255.h
185// GFX12: v_mul_f16_e64 v5.h, v1.h, v255.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x35,0xd5,0x01,0xff,0x03,0x00]
186
187v_mul_f16 v5.h, v1.h, v255.h dpp8:[7,6,5,4,3,2,1,0]
188// GFX12: v_mul_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x35,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
189
190v_mul_f16 v5.h, v1.h, v255.h quad_perm:[3,2,1,0]
191// GFX12: v_mul_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x35,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
192
193v_mul_f16 v5.h, v255.h, v2.h
194// GFX12: v_mul_f16_e64 v5.h, v255.h, v2.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x35,0xd5,0xff,0x05,0x02,0x00]
195
196v_mul_f16 v5.h, v255.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
197// GFX12: v_mul_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x35,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
198
199v_mul_f16 v5.h, v255.h, v2.h quad_perm:[3,2,1,0]
200// GFX12: v_mul_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x35,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
201
202v_mul_f16 v5.l, v1.l, v255.l
203// GFX12: v_mul_f16_e64 v5.l, v1.l, v255.l        ; encoding: [0x05,0x00,0x35,0xd5,0x01,0xff,0x03,0x00]
204
205v_mul_f16 v5.l, v1.l, v255.l dpp8:[7,6,5,4,3,2,1,0]
206// GFX12: v_mul_f16_e64_dpp v5.l, v1.l, v255.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x35,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
207
208v_mul_f16 v5.l, v1.l, v255.l quad_perm:[3,2,1,0]
209// GFX12: v_mul_f16_e64_dpp v5.l, v1.l, v255.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x35,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
210
211v_mul_f16 v5.l, v255.l, v2.l
212// GFX12: v_mul_f16_e64 v5.l, v255.l, v2.l        ; encoding: [0x05,0x00,0x35,0xd5,0xff,0x05,0x02,0x00]
213
214v_mul_f16 v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
215// GFX12: v_mul_f16_e64_dpp v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x35,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
216
217v_mul_f16 v5.l, v255.l, v2.l quad_perm:[3,2,1,0]
218// GFX12: v_mul_f16_e64_dpp v5.l, v255.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x35,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
219
220v_sub_f16 v255.h, v1.h, v2.h
221// GFX12: v_sub_f16_e64 v255.h, v1.h, v2.h op_sel:[1,1,1] ; encoding: [0xff,0x58,0x33,0xd5,0x01,0x05,0x02,0x00]
222
223v_sub_f16 v255.h, v1.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
224// GFX12: v_sub_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x58,0x33,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
225
226v_sub_f16 v255.h, v1.h, v2.h quad_perm:[3,2,1,0]
227// GFX12: v_sub_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x58,0x33,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
228
229v_sub_f16 v255.l, v1.l, v2.l
230// GFX12: v_sub_f16_e64 v255.l, v1.l, v2.l        ; encoding: [0xff,0x00,0x33,0xd5,0x01,0x05,0x02,0x00]
231
232v_sub_f16 v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
233// GFX12: v_sub_f16_e64_dpp v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x00,0x33,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
234
235v_sub_f16 v255.l, v1.l, v2.l quad_perm:[3,2,1,0]
236// GFX12: v_sub_f16_e64_dpp v255.l, v1.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x00,0x33,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
237
238v_sub_f16 v5.h, v1.h, v255.h
239// GFX12: v_sub_f16_e64 v5.h, v1.h, v255.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x33,0xd5,0x01,0xff,0x03,0x00]
240
241v_sub_f16 v5.h, v1.h, v255.h dpp8:[7,6,5,4,3,2,1,0]
242// GFX12: v_sub_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x33,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
243
244v_sub_f16 v5.h, v1.h, v255.h quad_perm:[3,2,1,0]
245// GFX12: v_sub_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x33,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
246
247v_sub_f16 v5.h, v255.h, v2.h
248// GFX12: v_sub_f16_e64 v5.h, v255.h, v2.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x33,0xd5,0xff,0x05,0x02,0x00]
249
250v_sub_f16 v5.h, v255.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
251// GFX12: v_sub_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x33,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
252
253v_sub_f16 v5.h, v255.h, v2.h quad_perm:[3,2,1,0]
254// GFX12: v_sub_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x33,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
255
256v_sub_f16 v5.l, v1.l, v255.l
257// GFX12: v_sub_f16_e64 v5.l, v1.l, v255.l        ; encoding: [0x05,0x00,0x33,0xd5,0x01,0xff,0x03,0x00]
258
259v_sub_f16 v5.l, v1.l, v255.l dpp8:[7,6,5,4,3,2,1,0]
260// GFX12: v_sub_f16_e64_dpp v5.l, v1.l, v255.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x33,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
261
262v_sub_f16 v5.l, v1.l, v255.l quad_perm:[3,2,1,0]
263// GFX12: v_sub_f16_e64_dpp v5.l, v1.l, v255.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x33,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
264
265v_sub_f16 v5.l, v255.l, v2.l
266// GFX12: v_sub_f16_e64 v5.l, v255.l, v2.l        ; encoding: [0x05,0x00,0x33,0xd5,0xff,0x05,0x02,0x00]
267
268v_sub_f16 v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
269// GFX12: v_sub_f16_e64_dpp v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x33,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
270
271v_sub_f16 v5.l, v255.l, v2.l quad_perm:[3,2,1,0]
272// GFX12: v_sub_f16_e64_dpp v5.l, v255.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x33,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
273
274v_subrev_f16 v255.h, v1.h, v2.h
275// GFX12: v_subrev_f16_e64 v255.h, v1.h, v2.h op_sel:[1,1,1] ; encoding: [0xff,0x58,0x34,0xd5,0x01,0x05,0x02,0x00]
276
277v_subrev_f16 v255.h, v1.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
278// GFX12: v_subrev_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x58,0x34,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
279
280v_subrev_f16 v255.h, v1.h, v2.h quad_perm:[3,2,1,0]
281// GFX12: v_subrev_f16_e64_dpp v255.h, v1.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x58,0x34,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
282
283v_subrev_f16 v255.l, v1.l, v2.l
284// GFX12: v_subrev_f16_e64 v255.l, v1.l, v2.l     ; encoding: [0xff,0x00,0x34,0xd5,0x01,0x05,0x02,0x00]
285
286v_subrev_f16 v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
287// GFX12: v_subrev_f16_e64_dpp v255.l, v1.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0xff,0x00,0x34,0xd5,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
288
289v_subrev_f16 v255.l, v1.l, v2.l quad_perm:[3,2,1,0]
290// GFX12: v_subrev_f16_e64_dpp v255.l, v1.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0xff,0x00,0x34,0xd5,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
291
292v_subrev_f16 v5.h, v1.h, v255.h
293// GFX12: v_subrev_f16_e64 v5.h, v1.h, v255.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x34,0xd5,0x01,0xff,0x03,0x00]
294
295v_subrev_f16 v5.h, v1.h, v255.h dpp8:[7,6,5,4,3,2,1,0]
296// GFX12: v_subrev_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x34,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
297
298v_subrev_f16 v5.h, v1.h, v255.h quad_perm:[3,2,1,0]
299// GFX12: v_subrev_f16_e64_dpp v5.h, v1.h, v255.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x34,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
300
301v_subrev_f16 v5.h, v255.h, v2.h
302// GFX12: v_subrev_f16_e64 v5.h, v255.h, v2.h op_sel:[1,1,1] ; encoding: [0x05,0x58,0x34,0xd5,0xff,0x05,0x02,0x00]
303
304v_subrev_f16 v5.h, v255.h, v2.h dpp8:[7,6,5,4,3,2,1,0]
305// GFX12: v_subrev_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x58,0x34,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
306
307v_subrev_f16 v5.h, v255.h, v2.h quad_perm:[3,2,1,0]
308// GFX12: v_subrev_f16_e64_dpp v5.h, v255.h, v2.h op_sel:[1,1,1] quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x58,0x34,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
309
310v_subrev_f16 v5.l, v1.l, v255.l
311// GFX12: v_subrev_f16_e64 v5.l, v1.l, v255.l     ; encoding: [0x05,0x00,0x34,0xd5,0x01,0xff,0x03,0x00]
312
313v_subrev_f16 v5.l, v1.l, v255.l dpp8:[7,6,5,4,3,2,1,0]
314// GFX12: v_subrev_f16_e64_dpp v5.l, v1.l, v255.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x34,0xd5,0xe9,0xfe,0x03,0x00,0x01,0x77,0x39,0x05]
315
316v_subrev_f16 v5.l, v1.l, v255.l quad_perm:[3,2,1,0]
317// GFX12: v_subrev_f16_e64_dpp v5.l, v1.l, v255.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x34,0xd5,0xfa,0xfe,0x03,0x00,0x01,0x1b,0x00,0xff]
318
319v_subrev_f16 v5.l, v255.l, v2.l
320// GFX12: v_subrev_f16_e64 v5.l, v255.l, v2.l     ; encoding: [0x05,0x00,0x34,0xd5,0xff,0x05,0x02,0x00]
321
322v_subrev_f16 v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0]
323// GFX12: v_subrev_f16_e64_dpp v5.l, v255.l, v2.l dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x34,0xd5,0xe9,0x04,0x02,0x00,0xff,0x77,0x39,0x05]
324
325v_subrev_f16 v5.l, v255.l, v2.l quad_perm:[3,2,1,0]
326// GFX12: v_subrev_f16_e64_dpp v5.l, v255.l, v2.l quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x34,0xd5,0xfa,0x04,0x02,0x00,0xff,0x1b,0x00,0xff]
327