blob: 3f5099758c5fe4769babe1903fce3136d57a51a9 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
|
/*@targets
* $maxopt $keep_baseline avx512_icl avx512_spr
*/
// policy $keep_baseline is used to avoid skip building avx512_skx
// when its part of baseline features (--cpu-baseline), since
// 'baseline' option isn't specified within targets.
#include "simd_qsort.hpp"
#if defined(NPY_HAVE_AVX512_SPR) && !defined(_MSC_VER)
#include "x86-simd-sort/src/avx512fp16-16bit-qsort.hpp"
#elif defined(NPY_HAVE_AVX512_ICL) && !defined(_MSC_VER)
#include "x86-simd-sort/src/avx512-16bit-qsort.hpp"
#endif
namespace np { namespace qsort_simd {
#if !defined(_MSC_VER)
#if defined(NPY_HAVE_AVX512_ICL) || defined(NPY_HAVE_AVX512_SPR)
template<> void NPY_CPU_DISPATCH_CURFX(QSort)(Half *arr, intptr_t size)
{
#if defined(NPY_HAVE_AVX512_SPR)
avx512_qsort(reinterpret_cast<_Float16*>(arr), size);
#else
avx512_qsort_fp16(reinterpret_cast<uint16_t*>(arr), size);
#endif
}
template<> void NPY_CPU_DISPATCH_CURFX(QSort)(uint16_t *arr, intptr_t size)
{
avx512_qsort(arr, size);
}
template<> void NPY_CPU_DISPATCH_CURFX(QSort)(int16_t *arr, intptr_t size)
{
avx512_qsort(arr, size);
}
#endif // NPY_HAVE_AVX512_ICL || SPR
#endif // _MSC_VER
}} // namespace np::qsort_simd
|
