| 1 |
/* { dg-do assemble { target aarch64_asm_sve_ok } } */ |
| 2 |
/* { dg-options "-O -msve-vector-bits=512 --save-temps" } */ |
| 3 |
|
| 4 |
#include <stdint.h> |
| 5 |
|
| 6 |
typedef int64_t vnx4di __attribute__((vector_size (64))); |
| 7 |
typedef int32_t vnx8si __attribute__((vector_size (64))); |
| 8 |
typedef int16_t vnx16hi __attribute__((vector_size (64))); |
| 9 |
typedef int8_t vnx32qi __attribute__((vector_size (64))); |
| 10 |
typedef double vnx4df __attribute__((vector_size (64))); |
| 11 |
typedef float vnx8sf __attribute__((vector_size (64))); |
| 12 |
typedef _Float16 vnx16hf __attribute__((vector_size (64))); |
| 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, vnx4di, 0) \ |
| 24 |
T (int64_t, vnx4di, 1) \ |
| 25 |
T (int64_t, vnx4di, 2) \ |
| 26 |
T (int64_t, vnx4di, 7) \ |
| 27 |
T (int32_t, vnx8si, 0) \ |
| 28 |
T (int32_t, vnx8si, 1) \ |
| 29 |
T (int32_t, vnx8si, 3) \ |
| 30 |
T (int32_t, vnx8si, 4) \ |
| 31 |
T (int32_t, vnx8si, 15) \ |
| 32 |
T (int16_t, vnx16hi, 0) \ |
| 33 |
T (int16_t, vnx16hi, 1) \ |
| 34 |
T (int16_t, vnx16hi, 7) \ |
| 35 |
T (int16_t, vnx16hi, 8) \ |
| 36 |
T (int16_t, vnx16hi, 31) \ |
| 37 |
T (int8_t, vnx32qi, 0) \ |
| 38 |
T (int8_t, vnx32qi, 1) \ |
| 39 |
T (int8_t, vnx32qi, 15) \ |
| 40 |
T (int8_t, vnx32qi, 16) \ |
| 41 |
T (int8_t, vnx32qi, 63) \ |
| 42 |
T (double, vnx4df, 0) \ |
| 43 |
T (double, vnx4df, 1) \ |
| 44 |
T (double, vnx4df, 2) \ |
| 45 |
T (double, vnx4df, 7) \ |
| 46 |
T (float, vnx8sf, 0) \ |
| 47 |
T (float, vnx8sf, 1) \ |
| 48 |
T (float, vnx8sf, 3) \ |
| 49 |
T (float, vnx8sf, 4) \ |
| 50 |
T (float, vnx8sf, 15) \ |
| 51 |
T (_Float16, vnx16hf, 0) \ |
| 52 |
T (_Float16, vnx16hf, 1) \ |
| 53 |
T (_Float16, vnx16hf, 7) \ |
| 54 |
T (_Float16, vnx16hf, 8) \ |
| 55 |
T (_Float16, vnx16hf, 31) |
| 56 |
|
| 57 |
TEST_ALL (EXTRACT) |
| 58 |
|
| 59 |
/* { dg-final { scan-assembler-times {\tfmov\tx[0-9]+, d[0-9]+\n} 2 { target aarch64_little_endian } } } */ |
| 60 |
/* { dg-final { scan-assembler-times {\tumov\tx[0-9]+, v[0-9]+\.d\[0\]\n} 1 { target aarch64_big_endian } } } */ |
| 61 |
/* { dg-final { scan-assembler-times {\tumov\tx[0-9]+, v[0-9]+\.d\[1\]\n} 1 } } */ |
| 62 |
/* { dg-final { scan-assembler-not {\tdup\td[0-9]+, v[0-9]+\.d\[0\]\n} } } */ |
| 63 |
/* { dg-final { scan-assembler-times {\tdup\td[0-9]+, v[0-9]+\.d\[1\]\n} 1 } } */ |
| 64 |
/* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.d, z[0-9]+\.d\[2\]\n} 2 } } */ |
| 65 |
/* { dg-final { scan-assembler-times {\tlastb\tx[0-9]+, p[0-7], z[0-9]+\.d\n} 1 } } */ |
| 66 |
/* { dg-final { scan-assembler-times {\tlastb\td[0-9]+, p[0-7], z[0-9]+\.d\n} 1 } } */ |
| 67 |
|
| 68 |
/* { dg-final { scan-assembler-times {\tfmov\tw[0-9]+, s[0-9]+\n} 2 { target aarch64_little_endian } } } */ |
| 69 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.s\[0\]\n} 1 { target aarch64_big_endian } } } */ |
| 70 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.s\[1\]\n} 1 } } */ |
| 71 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.s\[3\]\n} 1 } } */ |
| 72 |
/* { dg-final { scan-assembler-not {\tdup\ts[0-9]+, v[0-9]+\.s\[0\]\n} } } */ |
| 73 |
/* { dg-final { scan-assembler-times {\tdup\ts[0-9]+, v[0-9]+\.s\[1\]\n} 1 } } */ |
| 74 |
/* { dg-final { scan-assembler-times {\tdup\ts[0-9]+, v[0-9]+\.s\[3\]\n} 1 } } */ |
| 75 |
/* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.s, z[0-9]+\.s\[4\]\n} 2 } } */ |
| 76 |
/* { dg-final { scan-assembler-times {\tlastb\tw[0-9]+, p[0-7], z[0-9]+\.s\n} 1 } } */ |
| 77 |
/* { dg-final { scan-assembler-times {\tlastb\ts[0-9]+, p[0-7], z[0-9]+\.s\n} 1 } } */ |
| 78 |
|
| 79 |
/* Also used to move the result of a non-Advanced SIMD extract. */ |
| 80 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.h\[0\]\n} 2 } } */ |
| 81 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.h\[1\]\n} 1 } } */ |
| 82 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.h\[7\]\n} 1 } } */ |
| 83 |
/* { dg-final { scan-assembler-not {\tdup\th[0-9]+, v[0-9]+\.h\[0\]\n} } } */ |
| 84 |
/* { dg-final { scan-assembler-times {\tdup\th[0-9]+, v[0-9]+\.h\[1\]\n} 1 } } */ |
| 85 |
/* { dg-final { scan-assembler-times {\tdup\th[0-9]+, v[0-9]+\.h\[7\]\n} 1 } } */ |
| 86 |
/* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.h, z[0-9]+\.h\[8\]\n} 2 } } */ |
| 87 |
/* { dg-final { scan-assembler-times {\tlastb\tw[0-9]+, p[0-7], z[0-9]+\.h\n} 1 } } */ |
| 88 |
/* { dg-final { scan-assembler-times {\tlastb\th[0-9]+, p[0-7], z[0-9]+\.h\n} 1 } } */ |
| 89 |
|
| 90 |
/* Also used to move the result of a non-Advanced SIMD extract. */ |
| 91 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.b\[0\]\n} 2 } } */ |
| 92 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.b\[1\]\n} 1 } } */ |
| 93 |
/* { dg-final { scan-assembler-times {\tumov\tw[0-9]+, v[0-9]+\.b\[15\]\n} 1 } } */ |
| 94 |
/* { dg-final { scan-assembler-times {\tdup\tz[0-9]+\.b, z[0-9]+\.b\[16\]\n} 1 } } */ |
| 95 |
/* { dg-final { scan-assembler-times {\tlastb\tw[0-9]+, p[0-7], z[0-9]+\.b\n} 1 } } */ |