Debian Codegraph @

gcc/testsuite/gcc.target/aarch64/sme2/acle-asm/mls_za32_f32_vg1x4.c

raw
1 /* { dg-final { check-function-bodies "**" "" "-DCHECK_ASM" } } */
2
3 #include "test_sme2_acle.h"
4
5 /*
6 ** mls_0_z0_z0:
7 ** mov (w8|w9|w10|w11), #?0
8 ** fmls za\.s\[\1, 0, vgx4\], {z0\.s - z3\.s}, {z0\.s - z3\.s}
9 ** ret
10 */
11 TEST_ZA_XN (mls_0_z0_z0, svfloat32x4_t,
12 svmls_za32_f32_vg1x4 (0, z0, z0),
13 svmls_za32_vg1x4 (0, z0, z0))
14
15 /*
16 ** mls_w0_z0_z0:
17 ** mov (w8|w9|w10|w11), w0
18 ** fmls za\.s\[\1, 0, vgx4\], {z0\.s - z3\.s}, {z0\.s - z3\.s}
19 ** ret
20 */
21 TEST_ZA_XN (mls_w0_z0_z0, svfloat32x4_t,
22 svmls_za32_f32_vg1x4 (w0, z0, z0),
23 svmls_za32_vg1x4 (w0, z0, z0))
24
25 /*
26 ** mls_w8_z0_z4:
27 ** fmls za\.s\[w8, 0, vgx4\], {z0\.s - z3\.s}, {z4\.s - z7\.s}
28 ** ret
29 */
30 TEST_ZA_XN (mls_w8_z0_z4, svfloat32x4_t,
31 svmls_za32_f32_vg1x4 (w8, z0, z4),
32 svmls_za32_vg1x4 (w8, z0, z4))
33
34 /* Leave the assembler to check for correctness for misaligned registers. */
35
36 /*
37 ** mls_w8_z0_z18:
38 ** ...
39 ** fmls za\.s\[w8, 0, vgx4\], {z0\.s - z3\.s}, [^\n]+
40 ** ret
41 */
42 TEST_ZA_XN (mls_w8_z0_z18, svfloat32x4_t,
43 svmls_za32_f32_vg1x4 (w8, z0, z18),
44 svmls_za32_vg1x4 (w8, z0, z18))
45
46 /*
47 ** mls_w8_z18_z28:
48 ** ...
49 ** fmls za\.s\[w8, 0, vgx4\], [^\n]+, {z28\.s - z31\.s}
50 ** ret
51 */
52 TEST_ZA_XN (mls_w8_z18_z28, svfloat32x4_t,
53 svmls_za32_f32_vg1x4 (w8, z18, z28),
54 svmls_za32_vg1x4 (w8, z18, z28))
55
56 /*
57 ** mls_w8_z28_z23:
58 ** ...
59 ** fmls za\.s\[w8, 0, vgx4\], {z28\.s - z31\.s}, [^\n]+
60 ** ret
61 */
62 TEST_ZA_XN (mls_w8_z28_z23, svfloat32x4_t,
63 svmls_za32_f32_vg1x4 (w8, z28, z23),
64 svmls_za32_vg1x4 (w8, z28, z23))
65
66 /*
67 ** mls_w8p7_z4_z0:
68 ** fmls za\.s\[w8, 7, vgx4\], {z4\.s - z7\.s}, {z0\.s - z3\.s}
69 ** ret
70 */
71 TEST_ZA_XN (mls_w8p7_z4_z0, svfloat32x4_t,
72 svmls_za32_f32_vg1x4 (w8 + 7, z4, z0),
73 svmls_za32_vg1x4 (w8 + 7, z4, z0))
74
75 /*
76 ** mls_w8p8_z4_z4:
77 ** add (w8|w9|w10|w11), w8, #?8
78 ** fmls za\.s\[\1, 0, vgx4\], {z4\.s - z7\.s}, {z4\.s - z7\.s}
79 ** ret
80 */
81 TEST_ZA_XN (mls_w8p8_z4_z4, svfloat32x4_t,
82 svmls_za32_f32_vg1x4 (w8 + 8, z4, z4),
83 svmls_za32_vg1x4 (w8 + 8, z4, z4))
84
85 /*
86 ** mls_w8m1_z4_z0:
87 ** sub (w8|w9|w10|w11), w8, #?1
88 ** fmls za\.s\[\1, 0, vgx4\], {z4\.s - z7\.s}, {z0\.s - z3\.s}
89 ** ret
90 */
91 TEST_ZA_XN (mls_w8m1_z4_z0, svfloat32x4_t,
92 svmls_za32_f32_vg1x4 (w8 - 1, z4, z0),
93 svmls_za32_vg1x4 (w8 - 1, z4, z0))
94
95 /*
96 ** mls_single_0_z1_z0:
97 ** mov (w8|w9|w10|w11), #?0
98 ** fmls za\.s\[\1, 0, vgx4\], {z1\.s - z4\.s}, z0\.s
99 ** ret
100 */
101 TEST_ZA_SINGLE (mls_single_0_z1_z0, svfloat32x4_t, svfloat32_t,
102 svmls_single_za32_f32_vg1x4 (0, z1, z0),
103 svmls_za32_vg1x4 (0, z1, z0))
104
105 /*
106 ** mls_single_w0_z1_z0:
107 ** mov (w8|w9|w10|w11), w0
108 ** fmls za\.s\[\1, 0, vgx4\], {z1\.s - z4\.s}, z0\.s
109 ** ret
110 */
111 TEST_ZA_SINGLE (mls_single_w0_z1_z0, svfloat32x4_t, svfloat32_t,
112 svmls_single_za32_f32_vg1x4 (w0, z1, z0),
113 svmls_za32_vg1x4 (w0, z1, z0))
114
115 /*
116 ** mls_single_w8_z1_z0:
117 ** fmls za\.s\[w8, 0, vgx4\], {z1\.s - z4\.s}, z0\.s
118 ** ret
119 */
120 TEST_ZA_SINGLE (mls_single_w8_z1_z0, svfloat32x4_t, svfloat32_t,
121 svmls_single_za32_f32_vg1x4 (w8, z1, z0),
122 svmls_za32_vg1x4 (w8, z1, z0))
123
124 /*
125 ** mls_single_w8p7_z1_z0:
126 ** fmls za\.s\[w8, 7, vgx4\], {z1\.s - z4\.s}, z0\.s
127 ** ret
128 */
129 TEST_ZA_SINGLE (mls_single_w8p7_z1_z0, svfloat32x4_t, svfloat32_t,
130 svmls_single_za32_f32_vg1x4 (w8 + 7, z1, z0),
131 svmls_za32_vg1x4 (w8 + 7, z1, z0))
132
133 /*
134 ** mls_single_w8p8_z1_z0:
135 ** add (w8|w9|w10|w11), w8, #?8
136 ** fmls za\.s\[\1, 0, vgx4\], {z1\.s - z4\.s}, z0\.s
137 ** ret
138 */
139 TEST_ZA_SINGLE (mls_single_w8p8_z1_z0, svfloat32x4_t, svfloat32_t,
140 svmls_single_za32_f32_vg1x4 (w8 + 8, z1, z0),
141 svmls_za32_vg1x4 (w8 + 8, z1, z0))
142
143 /*
144 ** mls_single_w0m1_z1_z0:
145 ** sub (w8|w9|w10|w11), w0, #?1
146 ** fmls za\.s\[\1, 0, vgx4\], {z1\.s - z4\.s}, z0\.s
147 ** ret
148 */
149 TEST_ZA_SINGLE (mls_single_w0m1_z1_z0, svfloat32x4_t, svfloat32_t,
150 svmls_single_za32_f32_vg1x4 (w0 - 1, z1, z0),
151 svmls_za32_vg1x4 (w0 - 1, z1, z0))
152
153 /*
154 ** mls_single_w8_z0_z15:
155 ** str d15, \[sp, #?-16\]!
156 ** fmls za\.s\[w8, 0, vgx4\], {z0\.s - z3\.s}, z15\.s
157 ** ldr d15, \[sp\], #?16
158 ** ret
159 */
160 TEST_ZA_SINGLE_Z15 (mls_single_w8_z0_z15, svfloat32x4_t, svfloat32_t,
161 svmls_single_za32_f32_vg1x4 (w8, z0, z15),
162 svmls_za32_vg1x4 (w8, z0, z15))
163
164 /*
165 ** mls_single_w8_z20_z16:
166 ** mov (z[0-7]).d, z16.d
167 ** fmls za\.s\[w8, 0, vgx4\], {z20\.s - z23\.s}, \1\.s
168 ** ret
169 */
170 TEST_ZA_SINGLE (mls_single_w8_z20_z16, svfloat32x4_t, svfloat32_t,
171 svmls_single_za32_f32_vg1x4 (w8, z20, z16),
172 svmls_za32_vg1x4 (w8, z20, z16))