1 /* { dg-do assemble { target aarch64_asm_sve_ok } } */
2 /* { dg-options "-O -msve-vector-bits=2048 --save-temps" } */
3 
4 #include <stdint.h>
5 
6 typedef int64_t v32di __attribute__((vector_size (256)));
7 typedef int32_t v64si __attribute__((vector_size (256)));
8 typedef int16_t v128hi __attribute__((vector_size (256)));
9 typedef int8_t v256qi __attribute__((vector_size (256)));
10 typedef double v32df __attribute__((vector_size (256)));
11 typedef float v64sf __attribute__((vector_size (256)));
12 typedef _Float16 v128hf __attribute__((vector_size (256)));
13 
14 #define EXTRACT(ELT_TYPE, TYPE, INDEX)		\
15   ELT_TYPE permute_##TYPE##_##INDEX (void)	\
16   {						\
17     TYPE values;				\
18     asm ("" : "=w" (values));			\
19     return values[INDEX];			\
20   }
21 
22 #define TEST_ALL(T)				\
23   T (int64_t, v32di, 0)				\
24   T (int64_t, v32di, 1)				\
25   T (int64_t, v32di, 2)				\
26   T (int64_t, v32di, 7)				\
27   T (int64_t, v32di, 8)				\
28   T (int64_t, v32di, 9)				\
29   T (int64_t, v32di, 15)			\
30   T (int64_t, v32di, 31)			\
31   T (int32_t, v64si, 0)				\
32   T (int32_t, v64si, 1)				\
33   T (int32_t, v64si, 3)				\
34   T (int32_t, v64si, 4)				\
35   T (int32_t, v64si, 15)			\
36   T (int32_t, v64si, 16)			\
37   T (int32_t, v64si, 21)			\
38   T (int32_t, v64si, 31)			\
39   T (int32_t, v64si, 63)			\
40   T (int16_t, v128hi, 0)			\
41   T (int16_t, v128hi, 1)			\
42   T (int16_t, v128hi, 7)			\
43   T (int16_t, v128hi, 8)			\
44   T (int16_t, v128hi, 31)			\
45   T (int16_t, v128hi, 32)			\
46   T (int16_t, v128hi, 47)			\
47   T (int16_t, v128hi, 63)			\
48   T (int16_t, v128hi, 127)			\
49   T (int8_t, v256qi, 0)				\
50   T (int8_t, v256qi, 1)				\
51   T (int8_t, v256qi, 15)			\
52   T (int8_t, v256qi, 16)			\
53   T (int8_t, v256qi, 63)			\
54   T (int8_t, v256qi, 64)			\
55   T (int8_t, v256qi, 100)			\
56   T (int8_t, v256qi, 127)			\
57   T (int8_t, v256qi, 255)			\
58   T (double, v32df, 0)				\
59   T (double, v32df, 1)				\
60   T (double, v32df, 2)				\
61   T (double, v32df, 7)				\
62   T (double, v32df, 8)				\
63   T (double, v32df, 9)				\
64   T (double, v32df, 15)				\
65   T (double, v32df, 31)				\
66   T (float, v64sf, 0)				\
67   T (float, v64sf, 1)				\
68   T (float, v64sf, 3)				\
69   T (float, v64sf, 4)				\
70   T (float, v64sf, 15)				\
71   T (float, v64sf, 16)				\
72   T (float, v64sf, 21)				\
73   T (float, v64sf, 31)				\
74   T (float, v64sf, 63)				\
75   T (_Float16, v128hf, 0)			\
76   T (_Float16, v128hf, 1)			\
77   T (_Float16, v128hf, 7)			\
78   T (_Float16, v128hf, 8)			\
79   T (_Float16, v128hf, 31)			\
80   T (_Float16, v128hf, 32)			\
81   T (_Float16, v128hf, 47)			\
82   T (_Float16, v128hf, 63)			\
83   T (_Float16, v128hf, 127)
84 
85 TEST_ALL (EXTRACT)
86 
87 /* { dg-final { scan-assembler-times {\tumov\tx[0-9]+, v[0-9]+\.d\[0\]\n} 1 } } */
88 /* { dg-final { scan-assembler-times {\tumov\tx[0-9]+, v[0-9]+\.d\[1\]\n} 1 } } */
89 /* { dg-final { scan-assembler-not {\tdup\td[0-9]+, v[0-9]+\.d\[0\]\n} } } */
90 /* { dg-final { scan-assembler-times {\tdup\td[0-9]+, v[0-9]+\.d\[1\]\n} 1 } } */
91 /* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.d, z[0-9]+\.d\[2\]\n} 2 } } */
92 /* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.d, z[0-9]+\.d\[7\]\n} 2 } } */
93 /* { dg-final { scan-assembler-times {\tlastb\tx[0-9]+, p[0-7], z[0-9]+\.d\n} 1 } } */
94 /* { dg-final { scan-assembler-times {\tlastb\td[0-9]+, p[0-7], z[0-9]+\.d\n} 1 } } */
95 
96 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.s\[0\]\n} 1 } } */
97 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.s\[1\]\n} 1 } } */
98 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.s\[3\]\n} 1 } } */
99 /* { dg-final { scan-assembler-not {\tdup\ts[0-9]+, v[0-9]+\.s\[0\]\n} } } */
100 /* { dg-final { scan-assembler-times {\tdup\ts[0-9]+, v[0-9]+\.s\[1\]\n} 1 } } */
101 /* { dg-final { scan-assembler-times {\tdup\ts[0-9]+, v[0-9]+\.s\[3\]\n} 1 } } */
102 /* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.s, z[0-9]+\.s\[4\]\n} 2 } } */
103 /* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.s, z[0-9]+\.s\[15\]\n} 2 } } */
104 /* { dg-final { scan-assembler-times {\tlastb\tw[0-9]+, p[0-7], z[0-9]+\.s\n} 1 } } */
105 /* { dg-final { scan-assembler-times {\tlastb\ts[0-9]+, p[0-7], z[0-9]+\.s\n} 1 } } */
106 
107 /* Also used to move the result of a non-Advanced SIMD extract.  */
108 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.h\[0\]\n} 6 } } */
109 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.h\[1\]\n} 1 } } */
110 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.h\[7\]\n} 1 } } */
111 /* { dg-final { scan-assembler-not {\tdup\th[0-9]+, v[0-9]+\.h\[0\]\n} } } */
112 /* { dg-final { scan-assembler-times {\tdup\th[0-9]+, v[0-9]+\.h\[1\]\n} 1 } } */
113 /* { dg-final { scan-assembler-times {\tdup\th[0-9]+, v[0-9]+\.h\[7\]\n} 1 } } */
114 /* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.h, z[0-9]+\.h\[8\]\n} 2 } } */
115 /* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.h, z[0-9]+\.h\[31\]\n} 2 } } */
116 /* { dg-final { scan-assembler-times {\tlastb\tw[0-9]+, p[0-7], z[0-9]+\.h\n} 1 } } */
117 /* { dg-final { scan-assembler-times {\tlastb\th[0-9]+, p[0-7], z[0-9]+\.h\n} 1 } } */
118 
119 /* Also used to move the result of a non-Advanced SIMD extract.  */
120 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.b\[0\]\n} 6 } } */
121 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.b\[1\]\n} 1 } } */
122 /* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.b\[15\]\n} 1 } } */
123 /* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.b, z[0-9]+\.b\[16\]\n} 1 } } */
124 /* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.b, z[0-9]+\.b\[63\]\n} 1 } } */
125 /* { dg-final { scan-assembler-times {\tlastb\tw[0-9]+, p[0-7], z[0-9]+\.b\n} 1 } } */
126 
127 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #64\n} 7 } } */
128 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #72\n} 2 } } */
129 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #84\n} 2 } } */
130 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #94\n} 2 } } */
131 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #100\n} 1 } } */
132 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #120\n} 2 } } */
133 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #124\n} 2 } } */
134 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #126\n} 2 } } */
135 /* { dg-final { scan-assembler-times {\text\tz[0-9]+\.b, z[0-9]+\.b, z[0-9]+\.b, #127\n} 1 } } */
136