0
0
Fork 0
mirror of https://github.com/matrix-construct/construct synced 2024-11-25 16:22:35 +01:00
construct/include/ircd/simd/shift.h

268 lines
5.5 KiB
C++

// The Construct
//
// Copyright (C) The Construct Developers, Authors & Contributors
// Copyright (C) 2016-2020 Jason Volk <jason@zemos.net>
//
// Permission to use, copy, modify, and/or distribute this software for any
// purpose with or without fee is hereby granted, provided that the above
// copyright notice and this permission notice is present in all copies. The
// full license for this software is available in the LICENSE file.
#pragma once
#define HAVE_IRCD_SIMD_SHIFT_H
// xmmx shifter interface
namespace ircd::simd
{
template<int b,
class T>
typename std::enable_if<sizeof(T) == 16, T>::type
shl(const T a) noexcept;
template<int b,
class T>
typename std::enable_if<sizeof(T) == 32, T>::type
shl(const T a) noexcept;
template<int b,
class T>
typename std::enable_if<sizeof(T) == 64, T>::type
shl(const T a) noexcept;
template<int b,
class T>
typename std::enable_if<sizeof(T) == 16, T>::type
shr(const T a) noexcept;
template<int b,
class T>
typename std::enable_if<sizeof(T) == 32, T>::type
shr(const T a) noexcept;
template<int b,
class T>
typename std::enable_if<sizeof(T) == 64, T>::type
shr(const T a) noexcept;
}
// xmmx shifter fallback template (internal)
namespace ircd::simd
{
template<int b,
class V,
class T>
T _shl(const T) noexcept;
template<int b,
class V,
class T>
T _shr(const T) noexcept;
}
#ifdef HAVE_X86INTRIN_H
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 16, T>::type
ircd::simd::shr(const T a)
noexcept
{
static_assert
(
b % 8 == 0, "xmmx register only shifts right at bytewise resolution."
);
return T(_mm_bsrli_si128(u128x1(a), b / 8));
}
#endif
#ifdef HAVE_X86INTRIN_H
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 16, T>::type
ircd::simd::shl(const T a)
noexcept
{
static_assert
(
b % 8 == 0, "xmmx register only shifts left at bytewise resolution."
);
return T(_mm_bslli_si128(u128x1(a), b / 8));
}
#endif
#if defined(HAVE_X86INTRIN_H) && defined(__AVX2__)
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 32, T>::type
ircd::simd::shr(const T a)
noexcept
{
static_assert
(
b % 8 == 0, "ymmx register only shifts right at bytewise resolution."
);
return T(_mm256_srli_si256(u256x1(a), b / 8));
}
#else
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 32, T>::type
ircd::simd::shr(const T a)
noexcept
{
return _shr<b, u8x32>(a);
}
#endif
#if defined(HAVE_X86INTRIN_H) && defined(__AVX2__)
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 32, T>::type
ircd::simd::shl(const T a)
noexcept
{
static_assert
(
b % 8 == 0, "ymmx register only shifts left at bytewise resolution."
);
return T(_mm256_slli_si256(u256x1(a), b / 8));
}
#else
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 32, T>::type
ircd::simd::shl(const T a)
noexcept
{
return _shl<b, u8x32>(a);
}
#endif
#if defined(HAVE_X86INTRIN_H) && defined(__AVX512F__)
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 64, T>::type
ircd::simd::shr(const T a)
noexcept
{
static_assert
(
b % 8 == 0, "ymmx register only shifts right at bytewise resolution."
);
return T(_mm512_srli_si512(u512x1(a), b / 8));
}
#else
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 64, T>::type
ircd::simd::shr(const T a)
noexcept
{
return _shr<b, u8x64>(a);
}
#endif
#if defined(HAVE_X86INTRIN_H) && defined(__AVX512F__)
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 64, T>::type
ircd::simd::shl(const T a)
noexcept
{
static_assert
(
b % 8 == 0, "ymmx register only shifts left at bytewise resolution."
);
return T(_mm512_slli_si512(u512x1(a), b / 8));
}
#else
template<int b,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline typename std::enable_if<sizeof(T) == 64, T>::type
ircd::simd::shl(const T a)
noexcept
{
return _shl<b, u8x64>(a);
}
#endif
template<int b,
class V,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline T
ircd::simd::_shr(const T a)
noexcept
{
static_assert
(
sizeof(T) == sizeof(V)
);
static_assert
(
b % 8 == 0, "[emulated] xmmx register only shifts right at bytewise resolution."
);
constexpr int B
{
b / 8
};
V ret;
for(size_t i(0); i < sizeof(V) - B; ++i)
ret[i] = a[i + B];
for(size_t i(sizeof(V) - B); i < sizeof(V); ++i)
ret[i] = 0;
return T(ret);
}
template<int b,
class V,
class T>
[[using gnu: always_inline, gnu_inline, artificial]]
extern inline T
ircd::simd::_shl(const T a)
noexcept
{
static_assert
(
sizeof(T) == sizeof(V)
);
static_assert
(
b % 8 == 0, "[emulated] xmmx register only shifts left at bytewise resolution."
);
constexpr int B
{
b / 8
};
V ret;
for(size_t i(0); i < B; ++i)
ret[i] = 0;
for(size_t i(B); i < sizeof(V); ++i)
ret[i] = a[i - B];
return T(ret);
}