Debian Codegraph @

gcc/testsuite/gcc.target/aarch64/sme2/acle-asm/mla_za64_u16_vg4x4.c

raw
1 /* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
2
3 #pragma GCC target "+sme-i16i64"
4
5 #include "test_sme2_acle.h"
6
7 /*
8 ** mla_0_z0_z0:
9 ** mov (w8|w9|w10|w11), #?0
10 ** umlall za\.d\[\1, 0:3, vgx4\], {z0\.h - z3\.h}, {z0\.h - z3\.h}
11 ** ret
12 */
13 TEST_ZA_XN (mla_0_z0_z0, svuint16x4_t,
14 svmla_za64_u16_vg4x4 (0, z0, z0),
15 svmla_za64_vg4x4 (0, z0, z0))
16
17 /*
18 ** mla_w0_z0_z0:
19 ** mov (w8|w9|w10|w11), w0
20 ** umlall za\.d\[\1, 0:3, vgx4\], {z0\.h - z3\.h}, {z0\.h - z3\.h}
21 ** ret
22 */
23 TEST_ZA_XN (mla_w0_z0_z0, svuint16x4_t,
24 svmla_za64_u16_vg4x4 (w0, z0, z0),
25 svmla_za64_vg4x4 (w0, z0, z0))
26
27 /*
28 ** mla_w8_z0_z4:
29 ** umlall za\.d\[w8, 0:3, vgx4\], {z0\.h - z3\.h}, {z4\.h - z7\.h}
30 ** ret
31 */
32 TEST_ZA_XN (mla_w8_z0_z4, svuint16x4_t,
33 svmla_za64_u16_vg4x4 (w8, z0, z4),
34 svmla_za64_vg4x4 (w8, z0, z4))
35
36 /* Leave the assembler to check for correctness for misaligned registers. */
37
38 /*
39 ** mla_w8_z0_z18:
40 ** ...
41 ** umlall za\.d\[w8, 0:3, vgx4\], {z0\.h - z3\.h}, [^\n]+
42 ** ret
43 */
44 TEST_ZA_XN (mla_w8_z0_z18, svuint16x4_t,
45 svmla_za64_u16_vg4x4 (w8, z0, z18),
46 svmla_za64_vg4x4 (w8, z0, z18))
47
48 /*
49 ** mla_w8_z18_z0:
50 ** ...
51 ** umlall za\.d\[w8, 0:3, vgx4\], [^\n]+, {z0\.h - z3\.h}
52 ** ret
53 */
54 TEST_ZA_XN (mla_w8_z18_z0, svuint16x4_t,
55 svmla_za64_u16_vg4x4 (w8, z18, z0),
56 svmla_za64_vg4x4 (w8, z18, z0))
57
58 /*
59 ** mla_w8_z0_z23:
60 ** ...
61 ** umlall za\.d\[w8, 0:3, vgx4\], {z0\.h - z3\.h}, [^\n]+
62 ** ret
63 */
64 TEST_ZA_XN (mla_w8_z0_z23, svuint16x4_t,
65 svmla_za64_u16_vg4x4 (w8, z0, z23),
66 svmla_za64_vg4x4 (w8, z0, z23))
67
68 /*
69 ** mla_w8_z23_z0:
70 ** ...
71 ** umlall za\.d\[w8, 0:3, vgx4\], [^\n]+, {z0\.h - z3\.h}
72 ** ret
73 */
74 TEST_ZA_XN (mla_w8_z23_z0, svuint16x4_t,
75 svmla_za64_u16_vg4x4 (w8, z23, z0),
76 svmla_za64_vg4x4 (w8, z23, z0))
77
78 /*
79 ** mla_w8_z4_z28:
80 ** umlall za\.d\[w8, 0:3, vgx4\], {z4\.h - z7\.h}, {z28\.h - z31\.h}
81 ** ret
82 */
83 TEST_ZA_XN (mla_w8_z4_z28, svuint16x4_t,
84 svmla_za64_u16_vg4x4 (w8, z4, z28),
85 svmla_za64_vg4x4 (w8, z4, z28))
86
87 /*
88 ** mla_w8_z28_z0:
89 ** umlall za\.d\[w8, 0:3, vgx4\], {z28\.h - z31\.h}, {z0\.h - z3\.h}
90 ** ret
91 */
92 TEST_ZA_XN (mla_w8_z28_z0, svuint16x4_t,
93 svmla_za64_u16_vg4x4 (w8, z28, z0),
94 svmla_za64_vg4x4 (w8, z28, z0))
95
96 /*
97 ** mla_w8p1_z4_z0:
98 ** add (w8|w9|w10|w11), w8, #?1
99 ** umlall za\.d\[\1, 0:3, vgx4\], {z4\.h - z7\.h}, {z0\.h - z3\.h}
100 ** ret
101 */
102 TEST_ZA_XN (mla_w8p1_z4_z0, svuint16x4_t,
103 svmla_za64_u16_vg4x4 (w8 + 1, z4, z0),
104 svmla_za64_vg4x4 (w8 + 1, z4, z0))
105
106 /*
107 ** mla_w8p2_z4_z0:
108 ** add (w8|w9|w10|w11), w8, #?2
109 ** umlall za\.d\[\1, 0:3, vgx4\], {z4\.h - z7\.h}, {z0\.h - z3\.h}
110 ** ret
111 */
112 TEST_ZA_XN (mla_w8p2_z4_z0, svuint16x4_t,
113 svmla_za64_u16_vg4x4 (w8 + 2, z4, z0),
114 svmla_za64_vg4x4 (w8 + 2, z4, z0))
115
116 /*
117 ** mla_w11p4_z4_z0:
118 ** umlall za\.d\[w11, 4:7, vgx4\], {z4\.h - z7\.h}, {z0\.h - z3\.h}
119 ** ret
120 */
121 TEST_ZA_XN (mla_w11p4_z4_z0, svuint16x4_t,
122 svmla_za64_u16_vg4x4 (w11 + 4, z4, z0),
123 svmla_za64_vg4x4 (w11 + 4, z4, z0))
124
125 /*
126 ** mla_w8p7_z4_z0:
127 ** add (w8|w9|w10|w11), w8, #?7
128 ** umlall za\.d\[\1, 0:3, vgx4\], {z4\.h - z7\.h}, {z0\.h - z3\.h}
129 ** ret
130 */
131 TEST_ZA_XN (mla_w8p7_z4_z0, svuint16x4_t,
132 svmla_za64_u16_vg4x4 (w8 + 7, z4, z0),
133 svmla_za64_vg4x4 (w8 + 7, z4, z0))
134
135 /*
136 ** mla_w8p8_z4_z4:
137 ** add (w8|w9|w10|w11), w8, #?8
138 ** umlall za\.d\[\1, 0:3, vgx4\], {z4\.h - z7\.h}, {z4\.h - z7\.h}
139 ** ret
140 */
141 TEST_ZA_XN (mla_w8p8_z4_z4, svuint16x4_t,
142 svmla_za64_u16_vg4x4 (w8 + 8, z4, z4),
143 svmla_za64_vg4x4 (w8 + 8, z4, z4))
144
145 /*
146 ** mla_w8m1_z4_z0:
147 ** sub (w8|w9|w10|w11), w8, #?1
148 ** umlall za\.d\[\1, 0:3, vgx4\], {z4\.h - z7\.h}, {z0\.h - z3\.h}
149 ** ret
150 */
151 TEST_ZA_XN (mla_w8m1_z4_z0, svuint16x4_t,
152 svmla_za64_u16_vg4x4 (w8 - 1, z4, z0),
153 svmla_za64_vg4x4 (w8 - 1, z4, z0))
154
155 /*
156 ** mla_single_0_z1_z0:
157 ** mov (w8|w9|w10|w11), #?0
158 ** umlall za\.d\[\1, 0:3, vgx4\], {z1\.h - z4\.h}, z0\.h
159 ** ret
160 */
161 TEST_ZA_SINGLE (mla_single_0_z1_z0, svuint16x4_t, svuint16_t,
162 svmla_single_za64_u16_vg4x4 (0, z1, z0),
163 svmla_za64_vg4x4 (0, z1, z0))
164
165 /*
166 ** mla_single_w0_z1_z0:
167 ** mov (w8|w9|w10|w11), w0
168 ** umlall za\.d\[\1, 0:3, vgx4\], {z1\.h - z4\.h}, z0\.h
169 ** ret
170 */
171 TEST_ZA_SINGLE (mla_single_w0_z1_z0, svuint16x4_t, svuint16_t,
172 svmla_single_za64_u16_vg4x4 (w0, z1, z0),
173 svmla_za64_vg4x4 (w0, z1, z0))
174
175 /*
176 ** mla_single_w8_z1_z0:
177 ** umlall za\.d\[w8, 0:3, vgx4\], {z1\.h - z4\.h}, z0\.h
178 ** ret
179 */
180 TEST_ZA_SINGLE (mla_single_w8_z1_z0, svuint16x4_t, svuint16_t,
181 svmla_single_za64_u16_vg4x4 (w8, z1, z0),
182 svmla_za64_vg4x4 (w8, z1, z0))
183
184 /*
185 ** mla_single_w8p1_z1_z0:
186 ** add (w8|w9|w10|w11), w8, #?1
187 ** umlall za\.d\[\1, 0:3, vgx4\], {z1\.h - z4\.h}, z0\.h
188 ** ret
189 */
190 TEST_ZA_SINGLE (mla_single_w8p1_z1_z0, svuint16x4_t, svuint16_t,
191 svmla_single_za64_u16_vg4x4 (w8 + 1, z1, z0),
192 svmla_za64_vg4x4 (w8 + 1, z1, z0))
193
194 /*
195 ** mla_single_w8p4_z20_z0:
196 ** umlall za\.d\[w8, 4:7, vgx4\], {z20\.h - z23\.h}, z0\.h
197 ** ret
198 */
199 TEST_ZA_SINGLE (mla_single_w8p4_z20_z0, svuint16x4_t, svuint16_t,
200 svmla_single_za64_u16_vg4x4 (w8 + 4, z20, z0),
201 svmla_za64_vg4x4 (w8 + 4, z20, z0))
202
203 /*
204 ** mla_single_w8p6_z27_z0:
205 ** add (w8|w9|w10|w11), w8, #?6
206 ** umlall za\.d\[\1, 0:3, vgx4\], {z27\.h - z30\.h}, z0\.h
207 ** ret
208 */
209 TEST_ZA_SINGLE (mla_single_w8p6_z27_z0, svuint16x4_t, svuint16_t,
210 svmla_single_za64_u16_vg4x4 (w8 + 6, z27, z0),
211 svmla_za64_vg4x4 (w8 + 6, z27, z0))
212
213 /*
214 ** mla_single_w8p7_z1_z0:
215 ** add (w8|w9|w10|w11), w8, #?7
216 ** umlall za\.d\[\1, 0:3, vgx4\], {z1\.h - z4\.h}, z0\.h
217 ** ret
218 */
219 TEST_ZA_SINGLE (mla_single_w8p7_z1_z0, svuint16x4_t, svuint16_t,
220 svmla_single_za64_u16_vg4x4 (w8 + 7, z1, z0),
221 svmla_za64_vg4x4 (w8 + 7, z1, z0))
222
223 /*
224 ** mla_single_w8p8_z1_z0:
225 ** add (w8|w9|w10|w11), w8, #?8
226 ** umlall za\.d\[\1, 0:3, vgx4\], {z1\.h - z4\.h}, z0\.h
227 ** ret
228 */
229 TEST_ZA_SINGLE (mla_single_w8p8_z1_z0, svuint16x4_t, svuint16_t,
230 svmla_single_za64_u16_vg4x4 (w8 + 8, z1, z0),
231 svmla_za64_vg4x4 (w8 + 8, z1, z0))
232
233 /*
234 ** mla_single_w0m1_z1_z0:
235 ** sub (w8|w9|w10|w11), w0, #?1
236 ** umlall za\.d\[\1, 0:3, vgx4\], {z1\.h - z4\.h}, z0\.h
237 ** ret
238 */
239 TEST_ZA_SINGLE (mla_single_w0m1_z1_z0, svuint16x4_t, svuint16_t,
240 svmla_single_za64_u16_vg4x4 (w0 - 1, z1, z0),
241 svmla_za64_vg4x4 (w0 - 1, z1, z0))
242
243 /*
244 ** mla_single_w8_z0_z15:
245 ** str d15, \[sp, #?-16\]!
246 ** umlall za\.d\[w8, 0:3, vgx4\], {z0\.h - z3\.h}, z15\.h
247 ** ldr d15, \[sp\], #?16
248 ** ret
249 */
250 TEST_ZA_SINGLE_Z15 (mla_single_w8_z0_z15, svuint16x4_t, svuint16_t,
251 svmla_single_za64_u16_vg4x4 (w8, z0, z15),
252 svmla_za64_vg4x4 (w8, z0, z15))
253
254 /*
255 ** mla_single_w8_z20_z16:
256 ** mov (z[0-7]).d, z16.d
257 ** umlall za\.d\[w8, 0:3, vgx4\], {z20\.h - z23\.h}, \1\.h
258 ** ret
259 */
260 TEST_ZA_SINGLE (mla_single_w8_z20_z16, svuint16x4_t, svuint16_t,
261 svmla_single_za64_u16_vg4x4 (w8, z20, z16),
262 svmla_za64_vg4x4 (w8, z20, z16))