use crate::{
    core_arch::{simd::*, x86::*},
    intrinsics::simd::*,
    ptr,
};
use core::hint::unreachable_unchecked;
#[cfg(test)]
use stdarch_test::assert_instr;
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsw))]
pub unsafe fn _mm512_abs_epi16(a: __m512i) -> __m512i {
    let a = a.as_i16x32();
    let cmp: i16x32 = simd_gt(a, i16x32::ZERO);
    transmute(simd_select(cmp, a, simd_neg(a)))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsw))]
pub unsafe fn _mm512_mask_abs_epi16(src: __m512i, k: __mmask32, a: __m512i) -> __m512i {
    let abs = _mm512_abs_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, abs, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsw))]
pub unsafe fn _mm512_maskz_abs_epi16(k: __mmask32, a: __m512i) -> __m512i {
    let abs = _mm512_abs_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, abs, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsw))]
pub unsafe fn _mm256_mask_abs_epi16(src: __m256i, k: __mmask16, a: __m256i) -> __m256i {
    let abs = _mm256_abs_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, abs, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsw))]
pub unsafe fn _mm256_maskz_abs_epi16(k: __mmask16, a: __m256i) -> __m256i {
    let abs = _mm256_abs_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, abs, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsw))]
pub unsafe fn _mm_mask_abs_epi16(src: __m128i, k: __mmask8, a: __m128i) -> __m128i {
    let abs = _mm_abs_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, abs, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsw))]
pub unsafe fn _mm_maskz_abs_epi16(k: __mmask8, a: __m128i) -> __m128i {
    let abs = _mm_abs_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, abs, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsb))]
pub unsafe fn _mm512_abs_epi8(a: __m512i) -> __m512i {
    let a = a.as_i8x64();
    let cmp: i8x64 = simd_gt(a, i8x64::ZERO);
    transmute(simd_select(cmp, a, simd_neg(a)))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsb))]
pub unsafe fn _mm512_mask_abs_epi8(src: __m512i, k: __mmask64, a: __m512i) -> __m512i {
    let abs = _mm512_abs_epi8(a).as_i8x64();
    transmute(simd_select_bitmask(k, abs, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsb))]
pub unsafe fn _mm512_maskz_abs_epi8(k: __mmask64, a: __m512i) -> __m512i {
    let abs = _mm512_abs_epi8(a).as_i8x64();
    transmute(simd_select_bitmask(k, abs, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsb))]
pub unsafe fn _mm256_mask_abs_epi8(src: __m256i, k: __mmask32, a: __m256i) -> __m256i {
    let abs = _mm256_abs_epi8(a).as_i8x32();
    transmute(simd_select_bitmask(k, abs, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsb))]
pub unsafe fn _mm256_maskz_abs_epi8(k: __mmask32, a: __m256i) -> __m256i {
    let abs = _mm256_abs_epi8(a).as_i8x32();
    transmute(simd_select_bitmask(k, abs, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsb))]
pub unsafe fn _mm_mask_abs_epi8(src: __m128i, k: __mmask16, a: __m128i) -> __m128i {
    let abs = _mm_abs_epi8(a).as_i8x16();
    transmute(simd_select_bitmask(k, abs, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpabsb))]
pub unsafe fn _mm_maskz_abs_epi8(k: __mmask16, a: __m128i) -> __m128i {
    let abs = _mm_abs_epi8(a).as_i8x16();
    transmute(simd_select_bitmask(k, abs, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddw))]
pub unsafe fn _mm512_add_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_add(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddw))]
pub unsafe fn _mm512_mask_add_epi16(src: __m512i, k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_add_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, add, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddw))]
pub unsafe fn _mm512_maskz_add_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_add_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, add, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddw))]
pub unsafe fn _mm256_mask_add_epi16(src: __m256i, k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_add_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, add, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddw))]
pub unsafe fn _mm256_maskz_add_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_add_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, add, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddw))]
pub unsafe fn _mm_mask_add_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_add_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, add, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddw))]
pub unsafe fn _mm_maskz_add_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_add_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, add, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddb))]
pub unsafe fn _mm512_add_epi8(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_add(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddb))]
pub unsafe fn _mm512_mask_add_epi8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_add_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, add, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddb))]
pub unsafe fn _mm512_maskz_add_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_add_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, add, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddb))]
pub unsafe fn _mm256_mask_add_epi8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_add_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, add, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddb))]
pub unsafe fn _mm256_maskz_add_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_add_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, add, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddb))]
pub unsafe fn _mm_mask_add_epi8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_add_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, add, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddb))]
pub unsafe fn _mm_maskz_add_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_add_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, add, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusw))]
pub unsafe fn _mm512_adds_epu16(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_saturating_add(a.as_u16x32(), b.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusw))]
pub unsafe fn _mm512_mask_adds_epu16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let add = _mm512_adds_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, add, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusw))]
pub unsafe fn _mm512_maskz_adds_epu16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_adds_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, add, u16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusw))]
pub unsafe fn _mm256_mask_adds_epu16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let add = _mm256_adds_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, add, src.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusw))]
pub unsafe fn _mm256_maskz_adds_epu16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_adds_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, add, u16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusw))]
pub unsafe fn _mm_mask_adds_epu16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_adds_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, add, src.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusw))]
pub unsafe fn _mm_maskz_adds_epu16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_adds_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, add, u16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusb))]
pub unsafe fn _mm512_adds_epu8(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_saturating_add(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusb))]
pub unsafe fn _mm512_mask_adds_epu8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_adds_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, add, src.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusb))]
pub unsafe fn _mm512_maskz_adds_epu8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_adds_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, add, u8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusb))]
pub unsafe fn _mm256_mask_adds_epu8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_adds_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, add, src.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusb))]
pub unsafe fn _mm256_maskz_adds_epu8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_adds_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, add, u8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusb))]
pub unsafe fn _mm_mask_adds_epu8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_adds_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, add, src.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddusb))]
pub unsafe fn _mm_maskz_adds_epu8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_adds_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, add, u8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsw))]
pub unsafe fn _mm512_adds_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_saturating_add(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsw))]
pub unsafe fn _mm512_mask_adds_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let add = _mm512_adds_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, add, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsw))]
pub unsafe fn _mm512_maskz_adds_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_adds_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, add, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsw))]
pub unsafe fn _mm256_mask_adds_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let add = _mm256_adds_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, add, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsw))]
pub unsafe fn _mm256_maskz_adds_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_adds_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, add, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsw))]
pub unsafe fn _mm_mask_adds_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_adds_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, add, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsw))]
pub unsafe fn _mm_maskz_adds_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_adds_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, add, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsb))]
pub unsafe fn _mm512_adds_epi8(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_saturating_add(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsb))]
pub unsafe fn _mm512_mask_adds_epi8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_adds_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, add, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsb))]
pub unsafe fn _mm512_maskz_adds_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let add = _mm512_adds_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, add, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsb))]
pub unsafe fn _mm256_mask_adds_epi8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_adds_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, add, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsb))]
pub unsafe fn _mm256_maskz_adds_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let add = _mm256_adds_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, add, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsb))]
pub unsafe fn _mm_mask_adds_epi8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_adds_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, add, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpaddsb))]
pub unsafe fn _mm_maskz_adds_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let add = _mm_adds_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, add, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubw))]
pub unsafe fn _mm512_sub_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_sub(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubw))]
pub unsafe fn _mm512_mask_sub_epi16(src: __m512i, k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_sub_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, sub, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubw))]
pub unsafe fn _mm512_maskz_sub_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_sub_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, sub, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubw))]
pub unsafe fn _mm256_mask_sub_epi16(src: __m256i, k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_sub_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, sub, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubw))]
pub unsafe fn _mm256_maskz_sub_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_sub_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, sub, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubw))]
pub unsafe fn _mm_mask_sub_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_sub_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, sub, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubw))]
pub unsafe fn _mm_maskz_sub_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_sub_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, sub, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubb))]
pub unsafe fn _mm512_sub_epi8(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_sub(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubb))]
pub unsafe fn _mm512_mask_sub_epi8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_sub_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, sub, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubb))]
pub unsafe fn _mm512_maskz_sub_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_sub_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, sub, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubb))]
pub unsafe fn _mm256_mask_sub_epi8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_sub_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, sub, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubb))]
pub unsafe fn _mm256_maskz_sub_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_sub_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, sub, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubb))]
pub unsafe fn _mm_mask_sub_epi8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_sub_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, sub, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubb))]
pub unsafe fn _mm_maskz_sub_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_sub_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, sub, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusw))]
pub unsafe fn _mm512_subs_epu16(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_saturating_sub(a.as_u16x32(), b.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusw))]
pub unsafe fn _mm512_mask_subs_epu16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let sub = _mm512_subs_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, sub, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusw))]
pub unsafe fn _mm512_maskz_subs_epu16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_subs_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, sub, u16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusw))]
pub unsafe fn _mm256_mask_subs_epu16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let sub = _mm256_subs_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, sub, src.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusw))]
pub unsafe fn _mm256_maskz_subs_epu16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_subs_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, sub, u16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusw))]
pub unsafe fn _mm_mask_subs_epu16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_subs_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, sub, src.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusw))]
pub unsafe fn _mm_maskz_subs_epu16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_subs_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, sub, u16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusb))]
pub unsafe fn _mm512_subs_epu8(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_saturating_sub(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusb))]
pub unsafe fn _mm512_mask_subs_epu8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_subs_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, sub, src.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusb))]
pub unsafe fn _mm512_maskz_subs_epu8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_subs_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, sub, u8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusb))]
pub unsafe fn _mm256_mask_subs_epu8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_subs_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, sub, src.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusb))]
pub unsafe fn _mm256_maskz_subs_epu8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_subs_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, sub, u8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusb))]
pub unsafe fn _mm_mask_subs_epu8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_subs_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, sub, src.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubusb))]
pub unsafe fn _mm_maskz_subs_epu8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_subs_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, sub, u8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsw))]
pub unsafe fn _mm512_subs_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_saturating_sub(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsw))]
pub unsafe fn _mm512_mask_subs_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let sub = _mm512_subs_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, sub, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsw))]
pub unsafe fn _mm512_maskz_subs_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_subs_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, sub, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsw))]
pub unsafe fn _mm256_mask_subs_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let sub = _mm256_subs_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, sub, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsw))]
pub unsafe fn _mm256_maskz_subs_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_subs_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, sub, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsw))]
pub unsafe fn _mm_mask_subs_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_subs_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, sub, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsw))]
pub unsafe fn _mm_maskz_subs_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_subs_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, sub, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsb))]
pub unsafe fn _mm512_subs_epi8(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_saturating_sub(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsb))]
pub unsafe fn _mm512_mask_subs_epi8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_subs_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, sub, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsb))]
pub unsafe fn _mm512_maskz_subs_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let sub = _mm512_subs_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, sub, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsb))]
pub unsafe fn _mm256_mask_subs_epi8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_subs_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, sub, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsb))]
pub unsafe fn _mm256_maskz_subs_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let sub = _mm256_subs_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, sub, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsb))]
pub unsafe fn _mm_mask_subs_epi8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_subs_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, sub, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsubsb))]
pub unsafe fn _mm_maskz_subs_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let sub = _mm_subs_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, sub, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhuw))]
pub unsafe fn _mm512_mulhi_epu16(a: __m512i, b: __m512i) -> __m512i {
    let a = simd_cast::<_, u32x32>(a.as_u16x32());
    let b = simd_cast::<_, u32x32>(b.as_u16x32());
    let r = simd_shr(simd_mul(a, b), u32x32::splat(16));
    transmute(simd_cast::<u32x32, u16x32>(r))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhuw))]
pub unsafe fn _mm512_mask_mulhi_epu16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let mul = _mm512_mulhi_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, mul, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhuw))]
pub unsafe fn _mm512_maskz_mulhi_epu16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let mul = _mm512_mulhi_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, mul, u16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhuw))]
pub unsafe fn _mm256_mask_mulhi_epu16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let mul = _mm256_mulhi_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, mul, src.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhuw))]
pub unsafe fn _mm256_maskz_mulhi_epu16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let mul = _mm256_mulhi_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, mul, u16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhuw))]
pub unsafe fn _mm_mask_mulhi_epu16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let mul = _mm_mulhi_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, mul, src.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhuw))]
pub unsafe fn _mm_maskz_mulhi_epu16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let mul = _mm_mulhi_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, mul, u16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhw))]
pub unsafe fn _mm512_mulhi_epi16(a: __m512i, b: __m512i) -> __m512i {
    let a = simd_cast::<_, i32x32>(a.as_i16x32());
    let b = simd_cast::<_, i32x32>(b.as_i16x32());
    let r = simd_shr(simd_mul(a, b), i32x32::splat(16));
    transmute(simd_cast::<i32x32, i16x32>(r))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhw))]
pub unsafe fn _mm512_mask_mulhi_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let mul = _mm512_mulhi_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, mul, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhw))]
pub unsafe fn _mm512_maskz_mulhi_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let mul = _mm512_mulhi_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, mul, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhw))]
pub unsafe fn _mm256_mask_mulhi_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let mul = _mm256_mulhi_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, mul, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhw))]
pub unsafe fn _mm256_maskz_mulhi_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let mul = _mm256_mulhi_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, mul, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhw))]
pub unsafe fn _mm_mask_mulhi_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let mul = _mm_mulhi_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, mul, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhw))]
pub unsafe fn _mm_maskz_mulhi_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let mul = _mm_mulhi_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, mul, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhrsw))]
pub unsafe fn _mm512_mulhrs_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpmulhrsw(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhrsw))]
pub unsafe fn _mm512_mask_mulhrs_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let mul = _mm512_mulhrs_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, mul, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhrsw))]
pub unsafe fn _mm512_maskz_mulhrs_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let mul = _mm512_mulhrs_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, mul, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhrsw))]
pub unsafe fn _mm256_mask_mulhrs_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let mul = _mm256_mulhrs_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, mul, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhrsw))]
pub unsafe fn _mm256_maskz_mulhrs_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let mul = _mm256_mulhrs_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, mul, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhrsw))]
pub unsafe fn _mm_mask_mulhrs_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let mul = _mm_mulhrs_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, mul, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmulhrsw))]
pub unsafe fn _mm_maskz_mulhrs_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let mul = _mm_mulhrs_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, mul, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmullw))]
pub unsafe fn _mm512_mullo_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_mul(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmullw))]
pub unsafe fn _mm512_mask_mullo_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let mul = _mm512_mullo_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, mul, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmullw))]
pub unsafe fn _mm512_maskz_mullo_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let mul = _mm512_mullo_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, mul, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmullw))]
pub unsafe fn _mm256_mask_mullo_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let mul = _mm256_mullo_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, mul, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmullw))]
pub unsafe fn _mm256_maskz_mullo_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let mul = _mm256_mullo_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, mul, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmullw))]
pub unsafe fn _mm_mask_mullo_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let mul = _mm_mullo_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, mul, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmullw))]
pub unsafe fn _mm_maskz_mullo_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let mul = _mm_mullo_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, mul, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxuw))]
pub unsafe fn _mm512_max_epu16(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_u16x32();
    let b = b.as_u16x32();
    transmute(simd_select::<i16x32, _>(simd_gt(a, b), a, b))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxuw))]
pub unsafe fn _mm512_mask_max_epu16(src: __m512i, k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let max = _mm512_max_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, max, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxuw))]
pub unsafe fn _mm512_maskz_max_epu16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let max = _mm512_max_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, max, u16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxuw))]
pub unsafe fn _mm256_mask_max_epu16(src: __m256i, k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let max = _mm256_max_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, max, src.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxuw))]
pub unsafe fn _mm256_maskz_max_epu16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let max = _mm256_max_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, max, u16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxuw))]
pub unsafe fn _mm_mask_max_epu16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let max = _mm_max_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, max, src.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxuw))]
pub unsafe fn _mm_maskz_max_epu16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let max = _mm_max_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, max, u16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxub))]
pub unsafe fn _mm512_max_epu8(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_u8x64();
    let b = b.as_u8x64();
    transmute(simd_select::<i8x64, _>(simd_gt(a, b), a, b))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxub))]
pub unsafe fn _mm512_mask_max_epu8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let max = _mm512_max_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, max, src.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxub))]
pub unsafe fn _mm512_maskz_max_epu8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let max = _mm512_max_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, max, u8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxub))]
pub unsafe fn _mm256_mask_max_epu8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let max = _mm256_max_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, max, src.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxub))]
pub unsafe fn _mm256_maskz_max_epu8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let max = _mm256_max_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, max, u8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxub))]
pub unsafe fn _mm_mask_max_epu8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let max = _mm_max_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, max, src.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxub))]
pub unsafe fn _mm_maskz_max_epu8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let max = _mm_max_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, max, u8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsw))]
pub unsafe fn _mm512_max_epi16(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_i16x32();
    let b = b.as_i16x32();
    transmute(simd_select::<i16x32, _>(simd_gt(a, b), a, b))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsw))]
pub unsafe fn _mm512_mask_max_epi16(src: __m512i, k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let max = _mm512_max_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, max, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsw))]
pub unsafe fn _mm512_maskz_max_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let max = _mm512_max_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, max, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsw))]
pub unsafe fn _mm256_mask_max_epi16(src: __m256i, k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let max = _mm256_max_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, max, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsw))]
pub unsafe fn _mm256_maskz_max_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let max = _mm256_max_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, max, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsw))]
pub unsafe fn _mm_mask_max_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let max = _mm_max_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, max, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsw))]
pub unsafe fn _mm_maskz_max_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let max = _mm_max_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, max, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsb))]
pub unsafe fn _mm512_max_epi8(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_i8x64();
    let b = b.as_i8x64();
    transmute(simd_select::<i8x64, _>(simd_gt(a, b), a, b))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsb))]
pub unsafe fn _mm512_mask_max_epi8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let max = _mm512_max_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, max, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsb))]
pub unsafe fn _mm512_maskz_max_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let max = _mm512_max_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, max, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsb))]
pub unsafe fn _mm256_mask_max_epi8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let max = _mm256_max_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, max, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsb))]
pub unsafe fn _mm256_maskz_max_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let max = _mm256_max_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, max, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsb))]
pub unsafe fn _mm_mask_max_epi8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let max = _mm_max_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, max, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaxsb))]
pub unsafe fn _mm_maskz_max_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let max = _mm_max_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, max, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminuw))]
pub unsafe fn _mm512_min_epu16(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_u16x32();
    let b = b.as_u16x32();
    transmute(simd_select::<i16x32, _>(simd_lt(a, b), a, b))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminuw))]
pub unsafe fn _mm512_mask_min_epu16(src: __m512i, k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let min = _mm512_min_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, min, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminuw))]
pub unsafe fn _mm512_maskz_min_epu16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let min = _mm512_min_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, min, u16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminuw))]
pub unsafe fn _mm256_mask_min_epu16(src: __m256i, k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let min = _mm256_min_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, min, src.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminuw))]
pub unsafe fn _mm256_maskz_min_epu16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let min = _mm256_min_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, min, u16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminuw))]
pub unsafe fn _mm_mask_min_epu16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let min = _mm_min_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, min, src.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminuw))]
pub unsafe fn _mm_maskz_min_epu16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let min = _mm_min_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, min, u16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminub))]
pub unsafe fn _mm512_min_epu8(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_u8x64();
    let b = b.as_u8x64();
    transmute(simd_select::<i8x64, _>(simd_lt(a, b), a, b))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminub))]
pub unsafe fn _mm512_mask_min_epu8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let min = _mm512_min_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, min, src.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminub))]
pub unsafe fn _mm512_maskz_min_epu8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let min = _mm512_min_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, min, u8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminub))]
pub unsafe fn _mm256_mask_min_epu8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let min = _mm256_min_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, min, src.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminub))]
pub unsafe fn _mm256_maskz_min_epu8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let min = _mm256_min_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, min, u8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminub))]
pub unsafe fn _mm_mask_min_epu8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let min = _mm_min_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, min, src.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminub))]
pub unsafe fn _mm_maskz_min_epu8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let min = _mm_min_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, min, u8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsw))]
pub unsafe fn _mm512_min_epi16(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_i16x32();
    let b = b.as_i16x32();
    transmute(simd_select::<i16x32, _>(simd_lt(a, b), a, b))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsw))]
pub unsafe fn _mm512_mask_min_epi16(src: __m512i, k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let min = _mm512_min_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, min, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsw))]
pub unsafe fn _mm512_maskz_min_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let min = _mm512_min_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, min, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsw))]
pub unsafe fn _mm256_mask_min_epi16(src: __m256i, k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let min = _mm256_min_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, min, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsw))]
pub unsafe fn _mm256_maskz_min_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let min = _mm256_min_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, min, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsw))]
pub unsafe fn _mm_mask_min_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let min = _mm_min_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, min, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsw))]
pub unsafe fn _mm_maskz_min_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let min = _mm_min_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, min, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsb))]
pub unsafe fn _mm512_min_epi8(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_i8x64();
    let b = b.as_i8x64();
    transmute(simd_select::<i8x64, _>(simd_lt(a, b), a, b))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsb))]
pub unsafe fn _mm512_mask_min_epi8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let min = _mm512_min_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, min, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsb))]
pub unsafe fn _mm512_maskz_min_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let min = _mm512_min_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, min, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsb))]
pub unsafe fn _mm256_mask_min_epi8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let min = _mm256_min_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, min, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsb))]
pub unsafe fn _mm256_maskz_min_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let min = _mm256_min_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, min, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsb))]
pub unsafe fn _mm_mask_min_epi8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let min = _mm_min_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, min, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpminsb))]
pub unsafe fn _mm_maskz_min_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let min = _mm_min_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, min, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmplt_epu16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<u16x32, _>(simd_lt(a.as_u16x32(), b.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmplt_epu16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epu16_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmplt_epu16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<u16x16, _>(simd_lt(a.as_u16x16(), b.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmplt_epu16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epu16_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmplt_epu16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<u16x8, _>(simd_lt(a.as_u16x8(), b.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmplt_epu16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epu16_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmplt_epu8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<u8x64, _>(simd_lt(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmplt_epu8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epu8_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmplt_epu8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<u8x32, _>(simd_lt(a.as_u8x32(), b.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmplt_epu8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epu8_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmplt_epu8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<u8x16, _>(simd_lt(a.as_u8x16(), b.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmplt_epu8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epu8_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmplt_epi16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<i16x32, _>(simd_lt(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmplt_epi16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epi16_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmplt_epi16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<i16x16, _>(simd_lt(a.as_i16x16(), b.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmplt_epi16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epi16_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmplt_epi16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<i16x8, _>(simd_lt(a.as_i16x8(), b.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmplt_epi16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epi16_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmplt_epi8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<i8x64, _>(simd_lt(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmplt_epi8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epi8_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmplt_epi8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<i8x32, _>(simd_lt(a.as_i8x32(), b.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmplt_epi8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epi8_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmplt_epi8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<i8x16, _>(simd_lt(a.as_i8x16(), b.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmplt_epi8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epi8_mask::<_MM_CMPINT_LT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpgt_epu16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<u16x32, _>(simd_gt(a.as_u16x32(), b.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpgt_epu16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epu16_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpgt_epu16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<u16x16, _>(simd_gt(a.as_u16x16(), b.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpgt_epu16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epu16_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpgt_epu16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<u16x8, _>(simd_gt(a.as_u16x8(), b.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpgt_epu16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epu16_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpgt_epu8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<u8x64, _>(simd_gt(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpgt_epu8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epu8_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpgt_epu8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<u8x32, _>(simd_gt(a.as_u8x32(), b.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpgt_epu8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epu8_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpgt_epu8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<u8x16, _>(simd_gt(a.as_u8x16(), b.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpgt_epu8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epu8_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpgt_epi16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<i16x32, _>(simd_gt(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpgt_epi16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epi16_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpgt_epi16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<i16x16, _>(simd_gt(a.as_i16x16(), b.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpgt_epi16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epi16_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpgt_epi16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<i16x8, _>(simd_gt(a.as_i16x8(), b.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpgt_epi16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epi16_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpgt_epi8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<i8x64, _>(simd_gt(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpgt_epi8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epi8_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpgt_epi8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<i8x32, _>(simd_gt(a.as_i8x32(), b.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpgt_epi8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epi8_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpgt_epi8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<i8x16, _>(simd_gt(a.as_i8x16(), b.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpgt_epi8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epi8_mask::<_MM_CMPINT_NLE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmple_epu16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<u16x32, _>(simd_le(a.as_u16x32(), b.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmple_epu16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epu16_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmple_epu16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<u16x16, _>(simd_le(a.as_u16x16(), b.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmple_epu16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epu16_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmple_epu16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<u16x8, _>(simd_le(a.as_u16x8(), b.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmple_epu16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epu16_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmple_epu8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<u8x64, _>(simd_le(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmple_epu8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epu8_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmple_epu8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<u8x32, _>(simd_le(a.as_u8x32(), b.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmple_epu8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epu8_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmple_epu8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<u8x16, _>(simd_le(a.as_u8x16(), b.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmple_epu8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epu8_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmple_epi16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<i16x32, _>(simd_le(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmple_epi16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epi16_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmple_epi16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<i16x16, _>(simd_le(a.as_i16x16(), b.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmple_epi16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epi16_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmple_epi16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<i16x8, _>(simd_le(a.as_i16x8(), b.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmple_epi16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epi16_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmple_epi8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<i8x64, _>(simd_le(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmple_epi8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epi8_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmple_epi8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<i8x32, _>(simd_le(a.as_i8x32(), b.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmple_epi8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epi8_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmple_epi8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<i8x16, _>(simd_le(a.as_i8x16(), b.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmple_epi8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epi8_mask::<_MM_CMPINT_LE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpge_epu16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<u16x32, _>(simd_ge(a.as_u16x32(), b.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpge_epu16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epu16_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpge_epu16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<u16x16, _>(simd_ge(a.as_u16x16(), b.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpge_epu16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epu16_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpge_epu16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<u16x8, _>(simd_ge(a.as_u16x8(), b.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpge_epu16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epu16_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpge_epu8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<u8x64, _>(simd_ge(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpge_epu8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epu8_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpge_epu8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<u8x32, _>(simd_ge(a.as_u8x32(), b.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpge_epu8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epu8_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpge_epu8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<u8x16, _>(simd_ge(a.as_u8x16(), b.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpge_epu8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epu8_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpge_epi16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<i16x32, _>(simd_ge(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpge_epi16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epi16_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpge_epi16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<i16x16, _>(simd_ge(a.as_i16x16(), b.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpge_epi16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epi16_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpge_epi16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<i16x8, _>(simd_ge(a.as_i16x8(), b.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpge_epi16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epi16_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpge_epi8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<i8x64, _>(simd_ge(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpge_epi8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epi8_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpge_epi8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<i8x32, _>(simd_ge(a.as_i8x32(), b.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpge_epi8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epi8_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpge_epi8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<i8x16, _>(simd_ge(a.as_i8x16(), b.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpge_epi8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epi8_mask::<_MM_CMPINT_NLT>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpeq_epu16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<u16x32, _>(simd_eq(a.as_u16x32(), b.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpeq_epu16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epu16_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpeq_epu16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<u16x16, _>(simd_eq(a.as_u16x16(), b.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpeq_epu16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epu16_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpeq_epu16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<u16x8, _>(simd_eq(a.as_u16x8(), b.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpeq_epu16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epu16_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpeq_epu8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<u8x64, _>(simd_eq(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpeq_epu8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epu8_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpeq_epu8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<u8x32, _>(simd_eq(a.as_u8x32(), b.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpeq_epu8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epu8_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpeq_epu8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<u8x16, _>(simd_eq(a.as_u8x16(), b.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpeq_epu8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epu8_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpeq_epi16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<i16x32, _>(simd_eq(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpeq_epi16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epi16_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpeq_epi16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<i16x16, _>(simd_eq(a.as_i16x16(), b.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpeq_epi16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epi16_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpeq_epi16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<i16x8, _>(simd_eq(a.as_i16x8(), b.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpeq_epi16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epi16_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpeq_epi8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<i8x64, _>(simd_eq(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpeq_epi8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epi8_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpeq_epi8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<i8x32, _>(simd_eq(a.as_i8x32(), b.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpeq_epi8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epi8_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpeq_epi8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<i8x16, _>(simd_eq(a.as_i8x16(), b.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpeq_epi8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epi8_mask::<_MM_CMPINT_EQ>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpneq_epu16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<u16x32, _>(simd_ne(a.as_u16x32(), b.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpneq_epu16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epu16_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpneq_epu16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<u16x16, _>(simd_ne(a.as_u16x16(), b.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpneq_epu16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epu16_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpneq_epu16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<u16x8, _>(simd_ne(a.as_u16x8(), b.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpneq_epu16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epu16_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpneq_epu8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<u8x64, _>(simd_ne(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpneq_epu8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epu8_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpneq_epu8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<u8x32, _>(simd_ne(a.as_u8x32(), b.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpneq_epu8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epu8_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpneq_epu8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<u8x16, _>(simd_ne(a.as_u8x16(), b.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpneq_epu8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epu8_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpneq_epi16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    simd_bitmask::<i16x32, _>(simd_ne(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpneq_epi16_mask(k1: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    _mm512_mask_cmp_epi16_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpneq_epi16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    simd_bitmask::<i16x16, _>(simd_ne(a.as_i16x16(), b.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpneq_epi16_mask(k1: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    _mm256_mask_cmp_epi16_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpneq_epi16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    simd_bitmask::<i16x8, _>(simd_ne(a.as_i16x8(), b.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpneq_epi16_mask(k1: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    _mm_mask_cmp_epi16_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_cmpneq_epi8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    simd_bitmask::<i8x64, _>(simd_ne(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm512_mask_cmpneq_epi8_mask(k1: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    _mm512_mask_cmp_epi8_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_cmpneq_epi8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    simd_bitmask::<i8x32, _>(simd_ne(a.as_i8x32(), b.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm256_mask_cmpneq_epi8_mask(k1: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    _mm256_mask_cmp_epi8_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_cmpneq_epi8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    simd_bitmask::<i8x16, _>(simd_ne(a.as_i8x16(), b.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpcmp))]
pub unsafe fn _mm_mask_cmpneq_epi8_mask(k1: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    _mm_mask_cmp_epi8_mask::<_MM_CMPINT_NE>(k1, a, b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm512_cmp_epu16_mask<const IMM8: i32>(a: __m512i, b: __m512i) -> __mmask32 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u16x32();
    let b = b.as_u16x32();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i16x32::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i16x32::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm512_mask_cmp_epu16_mask<const IMM8: i32>(
    k1: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __mmask32 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u16x32();
    let b = b.as_u16x32();
    let k1 = simd_select_bitmask(k1, i16x32::splat(-1), i16x32::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i16x32::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i16x32::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm256_cmp_epu16_mask<const IMM8: i32>(a: __m256i, b: __m256i) -> __mmask16 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u16x16();
    let b = b.as_u16x16();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i16x16::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i16x16::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm256_mask_cmp_epu16_mask<const IMM8: i32>(
    k1: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __mmask16 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u16x16();
    let b = b.as_u16x16();
    let k1 = simd_select_bitmask(k1, i16x16::splat(-1), i16x16::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i16x16::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i16x16::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm_cmp_epu16_mask<const IMM8: i32>(a: __m128i, b: __m128i) -> __mmask8 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u16x8();
    let b = b.as_u16x8();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i16x8::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i16x8::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm_mask_cmp_epu16_mask<const IMM8: i32>(
    k1: __mmask8,
    a: __m128i,
    b: __m128i,
) -> __mmask8 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u16x8();
    let b = b.as_u16x8();
    let k1 = simd_select_bitmask(k1, i16x8::splat(-1), i16x8::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i16x8::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i16x8::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm512_cmp_epu8_mask<const IMM8: i32>(a: __m512i, b: __m512i) -> __mmask64 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u8x64();
    let b = b.as_u8x64();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i8x64::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i8x64::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm512_mask_cmp_epu8_mask<const IMM8: i32>(
    k1: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __mmask64 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u8x64();
    let b = b.as_u8x64();
    let k1 = simd_select_bitmask(k1, i8x64::splat(-1), i8x64::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i8x64::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i8x64::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm256_cmp_epu8_mask<const IMM8: i32>(a: __m256i, b: __m256i) -> __mmask32 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u8x32();
    let b = b.as_u8x32();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i8x32::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i8x32::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm256_mask_cmp_epu8_mask<const IMM8: i32>(
    k1: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __mmask32 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u8x32();
    let b = b.as_u8x32();
    let k1 = simd_select_bitmask(k1, i8x32::splat(-1), i8x32::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i8x32::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i8x32::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm_cmp_epu8_mask<const IMM8: i32>(a: __m128i, b: __m128i) -> __mmask16 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u8x16();
    let b = b.as_u8x16();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i8x16::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i8x16::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm_mask_cmp_epu8_mask<const IMM8: i32>(
    k1: __mmask16,
    a: __m128i,
    b: __m128i,
) -> __mmask16 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_u8x16();
    let b = b.as_u8x16();
    let k1 = simd_select_bitmask(k1, i8x16::splat(-1), i8x16::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i8x16::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i8x16::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm512_cmp_epi16_mask<const IMM8: i32>(a: __m512i, b: __m512i) -> __mmask32 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i16x32();
    let b = b.as_i16x32();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i16x32::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i16x32::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm512_mask_cmp_epi16_mask<const IMM8: i32>(
    k1: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __mmask32 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i16x32();
    let b = b.as_i16x32();
    let k1 = simd_select_bitmask(k1, i16x32::splat(-1), i16x32::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i16x32::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i16x32::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm256_cmp_epi16_mask<const IMM8: i32>(a: __m256i, b: __m256i) -> __mmask16 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i16x16();
    let b = b.as_i16x16();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i16x16::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i16x16::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm256_mask_cmp_epi16_mask<const IMM8: i32>(
    k1: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __mmask16 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i16x16();
    let b = b.as_i16x16();
    let k1 = simd_select_bitmask(k1, i16x16::splat(-1), i16x16::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i16x16::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i16x16::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm_cmp_epi16_mask<const IMM8: i32>(a: __m128i, b: __m128i) -> __mmask8 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i16x8();
    let b = b.as_i16x8();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i16x8::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i16x8::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm_mask_cmp_epi16_mask<const IMM8: i32>(
    k1: __mmask8,
    a: __m128i,
    b: __m128i,
) -> __mmask8 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i16x8();
    let b = b.as_i16x8();
    let k1 = simd_select_bitmask(k1, i16x8::splat(-1), i16x8::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i16x8::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i16x8::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm512_cmp_epi8_mask<const IMM8: i32>(a: __m512i, b: __m512i) -> __mmask64 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i8x64();
    let b = b.as_i8x64();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i8x64::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i8x64::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm512_mask_cmp_epi8_mask<const IMM8: i32>(
    k1: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __mmask64 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i8x64();
    let b = b.as_i8x64();
    let k1 = simd_select_bitmask(k1, i8x64::splat(-1), i8x64::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i8x64::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i8x64::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm256_cmp_epi8_mask<const IMM8: i32>(a: __m256i, b: __m256i) -> __mmask32 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i8x32();
    let b = b.as_i8x32();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i8x32::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i8x32::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm256_mask_cmp_epi8_mask<const IMM8: i32>(
    k1: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __mmask32 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i8x32();
    let b = b.as_i8x32();
    let k1 = simd_select_bitmask(k1, i8x32::splat(-1), i8x32::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i8x32::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i8x32::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm_cmp_epi8_mask<const IMM8: i32>(a: __m128i, b: __m128i) -> __mmask16 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i8x16();
    let b = b.as_i8x16();
    let r = match IMM8 {
        0 => simd_eq(a, b),
        1 => simd_lt(a, b),
        2 => simd_le(a, b),
        3 => i8x16::ZERO,
        4 => simd_ne(a, b),
        5 => simd_ge(a, b),
        6 => simd_gt(a, b),
        _ => i8x16::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpcmp, IMM8 = 0))]
pub unsafe fn _mm_mask_cmp_epi8_mask<const IMM8: i32>(
    k1: __mmask16,
    a: __m128i,
    b: __m128i,
) -> __mmask16 {
    static_assert_uimm_bits!(IMM8, 3);
    let a = a.as_i8x16();
    let b = b.as_i8x16();
    let k1 = simd_select_bitmask(k1, i8x16::splat(-1), i8x16::ZERO);
    let r = match IMM8 {
        0 => simd_and(k1, simd_eq(a, b)),
        1 => simd_and(k1, simd_lt(a, b)),
        2 => simd_and(k1, simd_le(a, b)),
        3 => i8x16::ZERO,
        4 => simd_and(k1, simd_ne(a, b)),
        5 => simd_and(k1, simd_ge(a, b)),
        6 => simd_and(k1, simd_gt(a, b)),
        _ => i8x16::splat(-1),
    };
    simd_bitmask(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_add_epi16(a: __m256i) -> i16 {
    simd_reduce_add_unordered(a.as_i16x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_add_epi16(k: __mmask16, a: __m256i) -> i16 {
    simd_reduce_add_unordered(simd_select_bitmask(k, a.as_i16x16(), i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_add_epi16(a: __m128i) -> i16 {
    simd_reduce_add_unordered(a.as_i16x8())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_add_epi16(k: __mmask8, a: __m128i) -> i16 {
    simd_reduce_add_unordered(simd_select_bitmask(k, a.as_i16x8(), i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_add_epi8(a: __m256i) -> i8 {
    simd_reduce_add_unordered(a.as_i8x32())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_add_epi8(k: __mmask32, a: __m256i) -> i8 {
    simd_reduce_add_unordered(simd_select_bitmask(k, a.as_i8x32(), i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_add_epi8(a: __m128i) -> i8 {
    simd_reduce_add_unordered(a.as_i8x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_add_epi8(k: __mmask16, a: __m128i) -> i8 {
    simd_reduce_add_unordered(simd_select_bitmask(k, a.as_i8x16(), i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_and_epi16(a: __m256i) -> i16 {
    simd_reduce_and(a.as_i16x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_and_epi16(k: __mmask16, a: __m256i) -> i16 {
    simd_reduce_and(simd_select_bitmask(
        k,
        a.as_i16x16(),
        _mm256_set1_epi64x(-1).as_i16x16(),
    ))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_and_epi16(a: __m128i) -> i16 {
    simd_reduce_and(a.as_i16x8())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_and_epi16(k: __mmask8, a: __m128i) -> i16 {
    simd_reduce_and(simd_select_bitmask(
        k,
        a.as_i16x8(),
        _mm_set1_epi64x(-1).as_i16x8(),
    ))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_and_epi8(a: __m256i) -> i8 {
    simd_reduce_and(a.as_i8x32())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_and_epi8(k: __mmask32, a: __m256i) -> i8 {
    simd_reduce_and(simd_select_bitmask(
        k,
        a.as_i8x32(),
        _mm256_set1_epi64x(-1).as_i8x32(),
    ))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_and_epi8(a: __m128i) -> i8 {
    simd_reduce_and(a.as_i8x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_and_epi8(k: __mmask16, a: __m128i) -> i8 {
    simd_reduce_and(simd_select_bitmask(
        k,
        a.as_i8x16(),
        _mm_set1_epi64x(-1).as_i8x16(),
    ))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_max_epi16(a: __m256i) -> i16 {
    simd_reduce_max(a.as_i16x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_max_epi16(k: __mmask16, a: __m256i) -> i16 {
    simd_reduce_max(simd_select_bitmask(k, a.as_i16x16(), i16x16::splat(-32768)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_max_epi16(a: __m128i) -> i16 {
    simd_reduce_max(a.as_i16x8())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_max_epi16(k: __mmask8, a: __m128i) -> i16 {
    simd_reduce_max(simd_select_bitmask(k, a.as_i16x8(), i16x8::splat(-32768)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_max_epi8(a: __m256i) -> i8 {
    simd_reduce_max(a.as_i8x32())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_max_epi8(k: __mmask32, a: __m256i) -> i8 {
    simd_reduce_max(simd_select_bitmask(k, a.as_i8x32(), i8x32::splat(-128)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_max_epi8(a: __m128i) -> i8 {
    simd_reduce_max(a.as_i8x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_max_epi8(k: __mmask16, a: __m128i) -> i8 {
    simd_reduce_max(simd_select_bitmask(k, a.as_i8x16(), i8x16::splat(-128)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_max_epu16(a: __m256i) -> u16 {
    simd_reduce_max(a.as_u16x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_max_epu16(k: __mmask16, a: __m256i) -> u16 {
    simd_reduce_max(simd_select_bitmask(k, a.as_u16x16(), u16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_max_epu16(a: __m128i) -> u16 {
    simd_reduce_max(a.as_u16x8())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_max_epu16(k: __mmask8, a: __m128i) -> u16 {
    simd_reduce_max(simd_select_bitmask(k, a.as_u16x8(), u16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_max_epu8(a: __m256i) -> u8 {
    simd_reduce_max(a.as_u8x32())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_max_epu8(k: __mmask32, a: __m256i) -> u8 {
    simd_reduce_max(simd_select_bitmask(k, a.as_u8x32(), u8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_max_epu8(a: __m128i) -> u8 {
    simd_reduce_max(a.as_u8x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_max_epu8(k: __mmask16, a: __m128i) -> u8 {
    simd_reduce_max(simd_select_bitmask(k, a.as_u8x16(), u8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_min_epi16(a: __m256i) -> i16 {
    simd_reduce_min(a.as_i16x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_min_epi16(k: __mmask16, a: __m256i) -> i16 {
    simd_reduce_min(simd_select_bitmask(k, a.as_i16x16(), i16x16::splat(0x7fff)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_min_epi16(a: __m128i) -> i16 {
    simd_reduce_min(a.as_i16x8())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_min_epi16(k: __mmask8, a: __m128i) -> i16 {
    simd_reduce_min(simd_select_bitmask(k, a.as_i16x8(), i16x8::splat(0x7fff)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_min_epi8(a: __m256i) -> i8 {
    simd_reduce_min(a.as_i8x32())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_min_epi8(k: __mmask32, a: __m256i) -> i8 {
    simd_reduce_min(simd_select_bitmask(k, a.as_i8x32(), i8x32::splat(0x7f)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_min_epi8(a: __m128i) -> i8 {
    simd_reduce_min(a.as_i8x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_min_epi8(k: __mmask16, a: __m128i) -> i8 {
    simd_reduce_min(simd_select_bitmask(k, a.as_i8x16(), i8x16::splat(0x7f)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_min_epu16(a: __m256i) -> u16 {
    simd_reduce_min(a.as_u16x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_min_epu16(k: __mmask16, a: __m256i) -> u16 {
    simd_reduce_min(simd_select_bitmask(k, a.as_u16x16(), u16x16::splat(0xffff)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_min_epu16(a: __m128i) -> u16 {
    simd_reduce_min(a.as_u16x8())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_min_epu16(k: __mmask8, a: __m128i) -> u16 {
    simd_reduce_min(simd_select_bitmask(k, a.as_u16x8(), u16x8::splat(0xffff)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_min_epu8(a: __m256i) -> u8 {
    simd_reduce_min(a.as_u8x32())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_min_epu8(k: __mmask32, a: __m256i) -> u8 {
    simd_reduce_min(simd_select_bitmask(k, a.as_u8x32(), u8x32::splat(0xff)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_min_epu8(a: __m128i) -> u8 {
    simd_reduce_min(a.as_u8x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_min_epu8(k: __mmask16, a: __m128i) -> u8 {
    simd_reduce_min(simd_select_bitmask(k, a.as_u8x16(), u8x16::splat(0xff)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_mul_epi16(a: __m256i) -> i16 {
    simd_reduce_mul_unordered(a.as_i16x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_mul_epi16(k: __mmask16, a: __m256i) -> i16 {
    simd_reduce_mul_unordered(simd_select_bitmask(k, a.as_i16x16(), i16x16::splat(1)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_mul_epi16(a: __m128i) -> i16 {
    simd_reduce_mul_unordered(a.as_i16x8())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_mul_epi16(k: __mmask8, a: __m128i) -> i16 {
    simd_reduce_mul_unordered(simd_select_bitmask(k, a.as_i16x8(), i16x8::splat(1)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_mul_epi8(a: __m256i) -> i8 {
    simd_reduce_mul_unordered(a.as_i8x32())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_mul_epi8(k: __mmask32, a: __m256i) -> i8 {
    simd_reduce_mul_unordered(simd_select_bitmask(k, a.as_i8x32(), i8x32::splat(1)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_mul_epi8(a: __m128i) -> i8 {
    simd_reduce_mul_unordered(a.as_i8x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_mul_epi8(k: __mmask16, a: __m128i) -> i8 {
    simd_reduce_mul_unordered(simd_select_bitmask(k, a.as_i8x16(), i8x16::splat(1)))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_or_epi16(a: __m256i) -> i16 {
    simd_reduce_or(a.as_i16x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_or_epi16(k: __mmask16, a: __m256i) -> i16 {
    simd_reduce_or(simd_select_bitmask(k, a.as_i16x16(), i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_or_epi16(a: __m128i) -> i16 {
    simd_reduce_or(a.as_i16x8())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_or_epi16(k: __mmask8, a: __m128i) -> i16 {
    simd_reduce_or(simd_select_bitmask(k, a.as_i16x8(), i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_reduce_or_epi8(a: __m256i) -> i8 {
    simd_reduce_or(a.as_i8x32())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_reduce_or_epi8(k: __mmask32, a: __m256i) -> i8 {
    simd_reduce_or(simd_select_bitmask(k, a.as_i8x32(), i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_reduce_or_epi8(a: __m128i) -> i8 {
    simd_reduce_or(a.as_i8x16())
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_reduce_or_epi8(k: __mmask16, a: __m128i) -> i8 {
    simd_reduce_or(simd_select_bitmask(k, a.as_i8x16(), i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm512_loadu_epi16(mem_addr: *const i16) -> __m512i {
    ptr::read_unaligned(mem_addr as *const __m512i)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm256_loadu_epi16(mem_addr: *const i16) -> __m256i {
    ptr::read_unaligned(mem_addr as *const __m256i)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm_loadu_epi16(mem_addr: *const i16) -> __m128i {
    ptr::read_unaligned(mem_addr as *const __m128i)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm512_loadu_epi8(mem_addr: *const i8) -> __m512i {
    ptr::read_unaligned(mem_addr as *const __m512i)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm256_loadu_epi8(mem_addr: *const i8) -> __m256i {
    ptr::read_unaligned(mem_addr as *const __m256i)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm_loadu_epi8(mem_addr: *const i8) -> __m128i {
    ptr::read_unaligned(mem_addr as *const __m128i)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm512_storeu_epi16(mem_addr: *mut i16, a: __m512i) {
    ptr::write_unaligned(mem_addr as *mut __m512i, a);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm256_storeu_epi16(mem_addr: *mut i16, a: __m256i) {
    ptr::write_unaligned(mem_addr as *mut __m256i, a);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm_storeu_epi16(mem_addr: *mut i16, a: __m128i) {
    ptr::write_unaligned(mem_addr as *mut __m128i, a);
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm512_storeu_epi8(mem_addr: *mut i8, a: __m512i) {
    ptr::write_unaligned(mem_addr as *mut __m512i, a);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm256_storeu_epi8(mem_addr: *mut i8, a: __m256i) {
    ptr::write_unaligned(mem_addr as *mut __m256i, a);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovups))] pub unsafe fn _mm_storeu_epi8(mem_addr: *mut i8, a: __m128i) {
    ptr::write_unaligned(mem_addr as *mut __m128i, a);
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm512_mask_loadu_epi16(src: __m512i, k: __mmask32, mem_addr: *const i16) -> __m512i {
    transmute(loaddqu16_512(mem_addr, src.as_i16x32(), k))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm512_maskz_loadu_epi16(k: __mmask32, mem_addr: *const i16) -> __m512i {
    _mm512_mask_loadu_epi16(_mm512_setzero_si512(), k, mem_addr)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm512_mask_loadu_epi8(src: __m512i, k: __mmask64, mem_addr: *const i8) -> __m512i {
    transmute(loaddqu8_512(mem_addr, src.as_i8x64(), k))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm512_maskz_loadu_epi8(k: __mmask64, mem_addr: *const i8) -> __m512i {
    _mm512_mask_loadu_epi8(_mm512_setzero_si512(), k, mem_addr)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_loadu_epi16(src: __m256i, k: __mmask16, mem_addr: *const i16) -> __m256i {
    transmute(loaddqu16_256(mem_addr, src.as_i16x16(), k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_maskz_loadu_epi16(k: __mmask16, mem_addr: *const i16) -> __m256i {
    _mm256_mask_loadu_epi16(_mm256_setzero_si256(), k, mem_addr)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_loadu_epi8(src: __m256i, k: __mmask32, mem_addr: *const i8) -> __m256i {
    transmute(loaddqu8_256(mem_addr, src.as_i8x32(), k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_maskz_loadu_epi8(k: __mmask32, mem_addr: *const i8) -> __m256i {
    _mm256_mask_loadu_epi8(_mm256_setzero_si256(), k, mem_addr)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_loadu_epi16(src: __m128i, k: __mmask8, mem_addr: *const i16) -> __m128i {
    transmute(loaddqu16_128(mem_addr, src.as_i16x8(), k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_maskz_loadu_epi16(k: __mmask8, mem_addr: *const i16) -> __m128i {
    _mm_mask_loadu_epi16(_mm_setzero_si128(), k, mem_addr)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_loadu_epi8(src: __m128i, k: __mmask16, mem_addr: *const i8) -> __m128i {
    transmute(loaddqu8_128(mem_addr, src.as_i8x16(), k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_maskz_loadu_epi8(k: __mmask16, mem_addr: *const i8) -> __m128i {
    _mm_mask_loadu_epi8(_mm_setzero_si128(), k, mem_addr)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm512_mask_storeu_epi16(mem_addr: *mut i16, mask: __mmask32, a: __m512i) {
    storedqu16_512(mem_addr, a.as_i16x32(), mask)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm512_mask_storeu_epi8(mem_addr: *mut i8, mask: __mmask64, a: __m512i) {
    storedqu8_512(mem_addr, a.as_i8x64(), mask)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_storeu_epi16(mem_addr: *mut i16, mask: __mmask16, a: __m256i) {
    storedqu16_256(mem_addr, a.as_i16x16(), mask)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm256_mask_storeu_epi8(mem_addr: *mut i8, mask: __mmask32, a: __m256i) {
    storedqu8_256(mem_addr, a.as_i8x32(), mask)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_storeu_epi16(mem_addr: *mut i16, mask: __mmask8, a: __m128i) {
    storedqu16_128(mem_addr, a.as_i16x8(), mask)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _mm_mask_storeu_epi8(mem_addr: *mut i8, mask: __mmask16, a: __m128i) {
    storedqu8_128(mem_addr, a.as_i8x16(), mask)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddwd))]
pub unsafe fn _mm512_madd_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpmaddwd(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddwd))]
pub unsafe fn _mm512_mask_madd_epi16(
    src: __m512i,
    k: __mmask16,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let madd = _mm512_madd_epi16(a, b).as_i32x16();
    transmute(simd_select_bitmask(k, madd, src.as_i32x16()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddwd))]
pub unsafe fn _mm512_maskz_madd_epi16(k: __mmask16, a: __m512i, b: __m512i) -> __m512i {
    let madd = _mm512_madd_epi16(a, b).as_i32x16();
    transmute(simd_select_bitmask(k, madd, i32x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddwd))]
pub unsafe fn _mm256_mask_madd_epi16(src: __m256i, k: __mmask8, a: __m256i, b: __m256i) -> __m256i {
    let madd = _mm256_madd_epi16(a, b).as_i32x8();
    transmute(simd_select_bitmask(k, madd, src.as_i32x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddwd))]
pub unsafe fn _mm256_maskz_madd_epi16(k: __mmask8, a: __m256i, b: __m256i) -> __m256i {
    let madd = _mm256_madd_epi16(a, b).as_i32x8();
    transmute(simd_select_bitmask(k, madd, i32x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddwd))]
pub unsafe fn _mm_mask_madd_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let madd = _mm_madd_epi16(a, b).as_i32x4();
    transmute(simd_select_bitmask(k, madd, src.as_i32x4()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddwd))]
pub unsafe fn _mm_maskz_madd_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let madd = _mm_madd_epi16(a, b).as_i32x4();
    transmute(simd_select_bitmask(k, madd, i32x4::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddubsw))]
pub unsafe fn _mm512_maddubs_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpmaddubsw(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddubsw))]
pub unsafe fn _mm512_mask_maddubs_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let madd = _mm512_maddubs_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, madd, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddubsw))]
pub unsafe fn _mm512_maskz_maddubs_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let madd = _mm512_maddubs_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, madd, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddubsw))]
pub unsafe fn _mm256_mask_maddubs_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let madd = _mm256_maddubs_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, madd, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddubsw))]
pub unsafe fn _mm256_maskz_maddubs_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let madd = _mm256_maddubs_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, madd, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddubsw))]
pub unsafe fn _mm_mask_maddubs_epi16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let madd = _mm_maddubs_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, madd, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmaddubsw))]
pub unsafe fn _mm_maskz_maddubs_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let madd = _mm_maddubs_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, madd, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackssdw))]
pub unsafe fn _mm512_packs_epi32(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpackssdw(a.as_i32x16(), b.as_i32x16()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackssdw))]
pub unsafe fn _mm512_mask_packs_epi32(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let pack = _mm512_packs_epi32(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, pack, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackssdw))]
pub unsafe fn _mm512_maskz_packs_epi32(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let pack = _mm512_packs_epi32(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, pack, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackssdw))]
pub unsafe fn _mm256_mask_packs_epi32(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let pack = _mm256_packs_epi32(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, pack, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackssdw))]
pub unsafe fn _mm256_maskz_packs_epi32(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let pack = _mm256_packs_epi32(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, pack, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackssdw))]
pub unsafe fn _mm_mask_packs_epi32(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let pack = _mm_packs_epi32(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, pack, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackssdw))]
pub unsafe fn _mm_maskz_packs_epi32(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let pack = _mm_packs_epi32(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, pack, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpacksswb))]
pub unsafe fn _mm512_packs_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpacksswb(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpacksswb))]
pub unsafe fn _mm512_mask_packs_epi16(
    src: __m512i,
    k: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let pack = _mm512_packs_epi16(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, pack, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpacksswb))]
pub unsafe fn _mm512_maskz_packs_epi16(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let pack = _mm512_packs_epi16(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, pack, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpacksswb))]
pub unsafe fn _mm256_mask_packs_epi16(
    src: __m256i,
    k: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let pack = _mm256_packs_epi16(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, pack, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpacksswb))]
pub unsafe fn _mm256_maskz_packs_epi16(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let pack = _mm256_packs_epi16(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, pack, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpacksswb))]
pub unsafe fn _mm_mask_packs_epi16(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let pack = _mm_packs_epi16(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, pack, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpacksswb))]
pub unsafe fn _mm_maskz_packs_epi16(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let pack = _mm_packs_epi16(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, pack, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackusdw))]
pub unsafe fn _mm512_packus_epi32(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpackusdw(a.as_i32x16(), b.as_i32x16()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackusdw))]
pub unsafe fn _mm512_mask_packus_epi32(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let pack = _mm512_packus_epi32(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, pack, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackusdw))]
pub unsafe fn _mm512_maskz_packus_epi32(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let pack = _mm512_packus_epi32(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, pack, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackusdw))]
pub unsafe fn _mm256_mask_packus_epi32(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let pack = _mm256_packus_epi32(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, pack, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackusdw))]
pub unsafe fn _mm256_maskz_packus_epi32(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let pack = _mm256_packus_epi32(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, pack, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackusdw))]
pub unsafe fn _mm_mask_packus_epi32(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let pack = _mm_packus_epi32(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, pack, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackusdw))]
pub unsafe fn _mm_maskz_packus_epi32(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let pack = _mm_packus_epi32(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, pack, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackuswb))]
pub unsafe fn _mm512_packus_epi16(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpackuswb(a.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackuswb))]
pub unsafe fn _mm512_mask_packus_epi16(
    src: __m512i,
    k: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let pack = _mm512_packus_epi16(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, pack, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackuswb))]
pub unsafe fn _mm512_maskz_packus_epi16(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let pack = _mm512_packus_epi16(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, pack, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackuswb))]
pub unsafe fn _mm256_mask_packus_epi16(
    src: __m256i,
    k: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let pack = _mm256_packus_epi16(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, pack, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackuswb))]
pub unsafe fn _mm256_maskz_packus_epi16(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let pack = _mm256_packus_epi16(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, pack, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackuswb))]
pub unsafe fn _mm_mask_packus_epi16(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let pack = _mm_packus_epi16(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, pack, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpackuswb))]
pub unsafe fn _mm_maskz_packus_epi16(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let pack = _mm_packus_epi16(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, pack, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgw))]
pub unsafe fn _mm512_avg_epu16(a: __m512i, b: __m512i) -> __m512i {
    let a = simd_cast::<_, u32x32>(a.as_u16x32());
    let b = simd_cast::<_, u32x32>(b.as_u16x32());
    let r = simd_shr(simd_add(simd_add(a, b), u32x32::splat(1)), u32x32::splat(1));
    transmute(simd_cast::<_, u16x32>(r))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgw))]
pub unsafe fn _mm512_mask_avg_epu16(src: __m512i, k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let avg = _mm512_avg_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, avg, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgw))]
pub unsafe fn _mm512_maskz_avg_epu16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let avg = _mm512_avg_epu16(a, b).as_u16x32();
    transmute(simd_select_bitmask(k, avg, u16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgw))]
pub unsafe fn _mm256_mask_avg_epu16(src: __m256i, k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let avg = _mm256_avg_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, avg, src.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgw))]
pub unsafe fn _mm256_maskz_avg_epu16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let avg = _mm256_avg_epu16(a, b).as_u16x16();
    transmute(simd_select_bitmask(k, avg, u16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgw))]
pub unsafe fn _mm_mask_avg_epu16(src: __m128i, k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let avg = _mm_avg_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, avg, src.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgw))]
pub unsafe fn _mm_maskz_avg_epu16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let avg = _mm_avg_epu16(a, b).as_u16x8();
    transmute(simd_select_bitmask(k, avg, u16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgb))]
pub unsafe fn _mm512_avg_epu8(a: __m512i, b: __m512i) -> __m512i {
    let a = simd_cast::<_, u16x64>(a.as_u8x64());
    let b = simd_cast::<_, u16x64>(b.as_u8x64());
    let r = simd_shr(simd_add(simd_add(a, b), u16x64::splat(1)), u16x64::splat(1));
    transmute(simd_cast::<_, u8x64>(r))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgb))]
pub unsafe fn _mm512_mask_avg_epu8(src: __m512i, k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let avg = _mm512_avg_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, avg, src.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgb))]
pub unsafe fn _mm512_maskz_avg_epu8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let avg = _mm512_avg_epu8(a, b).as_u8x64();
    transmute(simd_select_bitmask(k, avg, u8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgb))]
pub unsafe fn _mm256_mask_avg_epu8(src: __m256i, k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let avg = _mm256_avg_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, avg, src.as_u8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgb))]
pub unsafe fn _mm256_maskz_avg_epu8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let avg = _mm256_avg_epu8(a, b).as_u8x32();
    transmute(simd_select_bitmask(k, avg, u8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgb))]
pub unsafe fn _mm_mask_avg_epu8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let avg = _mm_avg_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, avg, src.as_u8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpavgb))]
pub unsafe fn _mm_maskz_avg_epu8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let avg = _mm_avg_epu8(a, b).as_u8x16();
    transmute(simd_select_bitmask(k, avg, u8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw))]
pub unsafe fn _mm512_sll_epi16(a: __m512i, count: __m128i) -> __m512i {
    transmute(vpsllw(a.as_i16x32(), count.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw))]
pub unsafe fn _mm512_mask_sll_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    count: __m128i,
) -> __m512i {
    let shf = _mm512_sll_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw))]
pub unsafe fn _mm512_maskz_sll_epi16(k: __mmask32, a: __m512i, count: __m128i) -> __m512i {
    let shf = _mm512_sll_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw))]
pub unsafe fn _mm256_mask_sll_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    count: __m128i,
) -> __m256i {
    let shf = _mm256_sll_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw))]
pub unsafe fn _mm256_maskz_sll_epi16(k: __mmask16, a: __m256i, count: __m128i) -> __m256i {
    let shf = _mm256_sll_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw))]
pub unsafe fn _mm_mask_sll_epi16(src: __m128i, k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_sll_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw))]
pub unsafe fn _mm_maskz_sll_epi16(k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_sll_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw, IMM8 = 5))]
#[rustc_legacy_const_generics(1)]
pub unsafe fn _mm512_slli_epi16<const IMM8: u32>(a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    if IMM8 >= 16 {
        _mm512_setzero_si512()
    } else {
        transmute(simd_shl(a.as_u16x32(), u16x32::splat(IMM8 as u16)))
    }
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm512_mask_slli_epi16<const IMM8: u32>(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = if IMM8 >= 16 {
        u16x32::ZERO
    } else {
        simd_shl(a.as_u16x32(), u16x32::splat(IMM8 as u16))
    };
    transmute(simd_select_bitmask(k, shf, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm512_maskz_slli_epi16<const IMM8: u32>(k: __mmask32, a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    if IMM8 >= 16 {
        _mm512_setzero_si512()
    } else {
        let shf = simd_shl(a.as_u16x32(), u16x32::splat(IMM8 as u16));
        transmute(simd_select_bitmask(k, shf, u16x32::ZERO))
    }
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm256_mask_slli_epi16<const IMM8: u32>(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = if IMM8 >= 16 {
        u16x16::ZERO
    } else {
        simd_shl(a.as_u16x16(), u16x16::splat(IMM8 as u16))
    };
    transmute(simd_select_bitmask(k, shf, src.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm256_maskz_slli_epi16<const IMM8: u32>(k: __mmask16, a: __m256i) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    if IMM8 >= 16 {
        _mm256_setzero_si256()
    } else {
        let shf = simd_shl(a.as_u16x16(), u16x16::splat(IMM8 as u16));
        transmute(simd_select_bitmask(k, shf, u16x16::ZERO))
    }
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm_mask_slli_epi16<const IMM8: u32>(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = if IMM8 >= 16 {
        u16x8::ZERO
    } else {
        simd_shl(a.as_u16x8(), u16x8::splat(IMM8 as u16))
    };
    transmute(simd_select_bitmask(k, shf, src.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm_maskz_slli_epi16<const IMM8: u32>(k: __mmask8, a: __m128i) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    if IMM8 >= 16 {
        _mm_setzero_si128()
    } else {
        let shf = simd_shl(a.as_u16x8(), u16x8::splat(IMM8 as u16));
        transmute(simd_select_bitmask(k, shf, u16x8::ZERO))
    }
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm512_sllv_epi16(a: __m512i, count: __m512i) -> __m512i {
    transmute(vpsllvw(a.as_i16x32(), count.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm512_mask_sllv_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    count: __m512i,
) -> __m512i {
    let shf = _mm512_sllv_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm512_maskz_sllv_epi16(k: __mmask32, a: __m512i, count: __m512i) -> __m512i {
    let shf = _mm512_sllv_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm256_sllv_epi16(a: __m256i, count: __m256i) -> __m256i {
    transmute(vpsllvw256(a.as_i16x16(), count.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm256_mask_sllv_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    count: __m256i,
) -> __m256i {
    let shf = _mm256_sllv_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm256_maskz_sllv_epi16(k: __mmask16, a: __m256i, count: __m256i) -> __m256i {
    let shf = _mm256_sllv_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm_sllv_epi16(a: __m128i, count: __m128i) -> __m128i {
    transmute(vpsllvw128(a.as_i16x8(), count.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm_mask_sllv_epi16(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
    count: __m128i,
) -> __m128i {
    let shf = _mm_sllv_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsllvw))]
pub unsafe fn _mm_maskz_sllv_epi16(k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_sllv_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw))]
pub unsafe fn _mm512_srl_epi16(a: __m512i, count: __m128i) -> __m512i {
    transmute(vpsrlw(a.as_i16x32(), count.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw))]
pub unsafe fn _mm512_mask_srl_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    count: __m128i,
) -> __m512i {
    let shf = _mm512_srl_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw))]
pub unsafe fn _mm512_maskz_srl_epi16(k: __mmask32, a: __m512i, count: __m128i) -> __m512i {
    let shf = _mm512_srl_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw))]
pub unsafe fn _mm256_mask_srl_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    count: __m128i,
) -> __m256i {
    let shf = _mm256_srl_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw))]
pub unsafe fn _mm256_maskz_srl_epi16(k: __mmask16, a: __m256i, count: __m128i) -> __m256i {
    let shf = _mm256_srl_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw))]
pub unsafe fn _mm_mask_srl_epi16(src: __m128i, k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_srl_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw))]
pub unsafe fn _mm_maskz_srl_epi16(k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_srl_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw, IMM8 = 5))]
#[rustc_legacy_const_generics(1)]
pub unsafe fn _mm512_srli_epi16<const IMM8: u32>(a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    if IMM8 >= 16 {
        _mm512_setzero_si512()
    } else {
        transmute(simd_shr(a.as_u16x32(), u16x32::splat(IMM8 as u16)))
    }
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm512_mask_srli_epi16<const IMM8: u32>(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = if IMM8 >= 16 {
        u16x32::ZERO
    } else {
        simd_shr(a.as_u16x32(), u16x32::splat(IMM8 as u16))
    };
    transmute(simd_select_bitmask(k, shf, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm512_maskz_srli_epi16<const IMM8: i32>(k: __mmask32, a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    if IMM8 >= 16 {
        _mm512_setzero_si512()
    } else {
        let shf = simd_shr(a.as_u16x32(), u16x32::splat(IMM8 as u16));
        transmute(simd_select_bitmask(k, shf, u16x32::ZERO))
    }
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm256_mask_srli_epi16<const IMM8: i32>(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = _mm256_srli_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shf.as_i16x16(), src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm256_maskz_srli_epi16<const IMM8: i32>(k: __mmask16, a: __m256i) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = _mm256_srli_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shf.as_i16x16(), i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm_mask_srli_epi16<const IMM8: i32>(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = _mm_srli_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shf.as_i16x8(), src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm_maskz_srli_epi16<const IMM8: i32>(k: __mmask8, a: __m128i) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = _mm_srli_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shf.as_i16x8(), i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm512_srlv_epi16(a: __m512i, count: __m512i) -> __m512i {
    transmute(vpsrlvw(a.as_i16x32(), count.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm512_mask_srlv_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    count: __m512i,
) -> __m512i {
    let shf = _mm512_srlv_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm512_maskz_srlv_epi16(k: __mmask32, a: __m512i, count: __m512i) -> __m512i {
    let shf = _mm512_srlv_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm256_srlv_epi16(a: __m256i, count: __m256i) -> __m256i {
    transmute(vpsrlvw256(a.as_i16x16(), count.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm256_mask_srlv_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    count: __m256i,
) -> __m256i {
    let shf = _mm256_srlv_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm256_maskz_srlv_epi16(k: __mmask16, a: __m256i, count: __m256i) -> __m256i {
    let shf = _mm256_srlv_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm_srlv_epi16(a: __m128i, count: __m128i) -> __m128i {
    transmute(vpsrlvw128(a.as_i16x8(), count.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm_mask_srlv_epi16(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
    count: __m128i,
) -> __m128i {
    let shf = _mm_srlv_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrlvw))]
pub unsafe fn _mm_maskz_srlv_epi16(k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_srlv_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw))]
pub unsafe fn _mm512_sra_epi16(a: __m512i, count: __m128i) -> __m512i {
    transmute(vpsraw(a.as_i16x32(), count.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw))]
pub unsafe fn _mm512_mask_sra_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    count: __m128i,
) -> __m512i {
    let shf = _mm512_sra_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw))]
pub unsafe fn _mm512_maskz_sra_epi16(k: __mmask32, a: __m512i, count: __m128i) -> __m512i {
    let shf = _mm512_sra_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw))]
pub unsafe fn _mm256_mask_sra_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    count: __m128i,
) -> __m256i {
    let shf = _mm256_sra_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw))]
pub unsafe fn _mm256_maskz_sra_epi16(k: __mmask16, a: __m256i, count: __m128i) -> __m256i {
    let shf = _mm256_sra_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw))]
pub unsafe fn _mm_mask_sra_epi16(src: __m128i, k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_sra_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw))]
pub unsafe fn _mm_maskz_sra_epi16(k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_sra_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw, IMM8 = 1))]
#[rustc_legacy_const_generics(1)]
pub unsafe fn _mm512_srai_epi16<const IMM8: u32>(a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    transmute(simd_shr(a.as_i16x32(), i16x32::splat(IMM8.min(15) as i16)))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw, IMM8 = 1))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm512_mask_srai_epi16<const IMM8: u32>(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = simd_shr(a.as_i16x32(), i16x32::splat(IMM8.min(15) as i16));
    transmute(simd_select_bitmask(k, shf, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw, IMM8 = 1))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm512_maskz_srai_epi16<const IMM8: u32>(k: __mmask32, a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let shf = simd_shr(a.as_i16x32(), i16x32::splat(IMM8.min(15) as i16));
    transmute(simd_select_bitmask(k, shf, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw, IMM8 = 1))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm256_mask_srai_epi16<const IMM8: u32>(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = simd_shr(a.as_i16x16(), i16x16::splat(IMM8.min(15) as i16));
    transmute(simd_select_bitmask(k, r, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw, IMM8 = 1))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm256_maskz_srai_epi16<const IMM8: u32>(k: __mmask16, a: __m256i) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = simd_shr(a.as_i16x16(), i16x16::splat(IMM8.min(15) as i16));
    transmute(simd_select_bitmask(k, r, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw, IMM8 = 1))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm_mask_srai_epi16<const IMM8: u32>(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = simd_shr(a.as_i16x8(), i16x8::splat(IMM8.min(15) as i16));
    transmute(simd_select_bitmask(k, r, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsraw, IMM8 = 1))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm_maskz_srai_epi16<const IMM8: u32>(k: __mmask8, a: __m128i) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = simd_shr(a.as_i16x8(), i16x8::splat(IMM8.min(15) as i16));
    transmute(simd_select_bitmask(k, r, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm512_srav_epi16(a: __m512i, count: __m512i) -> __m512i {
    transmute(vpsravw(a.as_i16x32(), count.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm512_mask_srav_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    count: __m512i,
) -> __m512i {
    let shf = _mm512_srav_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm512_maskz_srav_epi16(k: __mmask32, a: __m512i, count: __m512i) -> __m512i {
    let shf = _mm512_srav_epi16(a, count).as_i16x32();
    transmute(simd_select_bitmask(k, shf, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm256_srav_epi16(a: __m256i, count: __m256i) -> __m256i {
    transmute(vpsravw256(a.as_i16x16(), count.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm256_mask_srav_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    count: __m256i,
) -> __m256i {
    let shf = _mm256_srav_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm256_maskz_srav_epi16(k: __mmask16, a: __m256i, count: __m256i) -> __m256i {
    let shf = _mm256_srav_epi16(a, count).as_i16x16();
    transmute(simd_select_bitmask(k, shf, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm_srav_epi16(a: __m128i, count: __m128i) -> __m128i {
    transmute(vpsravw128(a.as_i16x8(), count.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm_mask_srav_epi16(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
    count: __m128i,
) -> __m128i {
    let shf = _mm_srav_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsravw))]
pub unsafe fn _mm_maskz_srav_epi16(k: __mmask8, a: __m128i, count: __m128i) -> __m128i {
    let shf = _mm_srav_epi16(a, count).as_i16x8();
    transmute(simd_select_bitmask(k, shf, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vperm))] pub unsafe fn _mm512_permutex2var_epi16(a: __m512i, idx: __m512i, b: __m512i) -> __m512i {
    transmute(vpermi2w(a.as_i16x32(), idx.as_i16x32(), b.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermt2w))]
pub unsafe fn _mm512_mask_permutex2var_epi16(
    a: __m512i,
    k: __mmask32,
    idx: __m512i,
    b: __m512i,
) -> __m512i {
    let permute = _mm512_permutex2var_epi16(a, idx, b).as_i16x32();
    transmute(simd_select_bitmask(k, permute, a.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vperm))] pub unsafe fn _mm512_maskz_permutex2var_epi16(
    k: __mmask32,
    a: __m512i,
    idx: __m512i,
    b: __m512i,
) -> __m512i {
    let permute = _mm512_permutex2var_epi16(a, idx, b).as_i16x32();
    transmute(simd_select_bitmask(k, permute, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermi2w))]
pub unsafe fn _mm512_mask2_permutex2var_epi16(
    a: __m512i,
    idx: __m512i,
    k: __mmask32,
    b: __m512i,
) -> __m512i {
    let permute = _mm512_permutex2var_epi16(a, idx, b).as_i16x32();
    transmute(simd_select_bitmask(k, permute, idx.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vperm))] pub unsafe fn _mm256_permutex2var_epi16(a: __m256i, idx: __m256i, b: __m256i) -> __m256i {
    transmute(vpermi2w256(a.as_i16x16(), idx.as_i16x16(), b.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermt2w))]
pub unsafe fn _mm256_mask_permutex2var_epi16(
    a: __m256i,
    k: __mmask16,
    idx: __m256i,
    b: __m256i,
) -> __m256i {
    let permute = _mm256_permutex2var_epi16(a, idx, b).as_i16x16();
    transmute(simd_select_bitmask(k, permute, a.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vperm))] pub unsafe fn _mm256_maskz_permutex2var_epi16(
    k: __mmask16,
    a: __m256i,
    idx: __m256i,
    b: __m256i,
) -> __m256i {
    let permute = _mm256_permutex2var_epi16(a, idx, b).as_i16x16();
    transmute(simd_select_bitmask(k, permute, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermi2w))]
pub unsafe fn _mm256_mask2_permutex2var_epi16(
    a: __m256i,
    idx: __m256i,
    k: __mmask16,
    b: __m256i,
) -> __m256i {
    let permute = _mm256_permutex2var_epi16(a, idx, b).as_i16x16();
    transmute(simd_select_bitmask(k, permute, idx.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vperm))] pub unsafe fn _mm_permutex2var_epi16(a: __m128i, idx: __m128i, b: __m128i) -> __m128i {
    transmute(vpermi2w128(a.as_i16x8(), idx.as_i16x8(), b.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermt2w))]
pub unsafe fn _mm_mask_permutex2var_epi16(
    a: __m128i,
    k: __mmask8,
    idx: __m128i,
    b: __m128i,
) -> __m128i {
    let permute = _mm_permutex2var_epi16(a, idx, b).as_i16x8();
    transmute(simd_select_bitmask(k, permute, a.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vperm))] pub unsafe fn _mm_maskz_permutex2var_epi16(
    k: __mmask8,
    a: __m128i,
    idx: __m128i,
    b: __m128i,
) -> __m128i {
    let permute = _mm_permutex2var_epi16(a, idx, b).as_i16x8();
    transmute(simd_select_bitmask(k, permute, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermi2w))]
pub unsafe fn _mm_mask2_permutex2var_epi16(
    a: __m128i,
    idx: __m128i,
    k: __mmask8,
    b: __m128i,
) -> __m128i {
    let permute = _mm_permutex2var_epi16(a, idx, b).as_i16x8();
    transmute(simd_select_bitmask(k, permute, idx.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm512_permutexvar_epi16(idx: __m512i, a: __m512i) -> __m512i {
    transmute(vpermw(a.as_i16x32(), idx.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm512_mask_permutexvar_epi16(
    src: __m512i,
    k: __mmask32,
    idx: __m512i,
    a: __m512i,
) -> __m512i {
    let permute = _mm512_permutexvar_epi16(idx, a).as_i16x32();
    transmute(simd_select_bitmask(k, permute, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm512_maskz_permutexvar_epi16(k: __mmask32, idx: __m512i, a: __m512i) -> __m512i {
    let permute = _mm512_permutexvar_epi16(idx, a).as_i16x32();
    transmute(simd_select_bitmask(k, permute, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm256_permutexvar_epi16(idx: __m256i, a: __m256i) -> __m256i {
    transmute(vpermw256(a.as_i16x16(), idx.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm256_mask_permutexvar_epi16(
    src: __m256i,
    k: __mmask16,
    idx: __m256i,
    a: __m256i,
) -> __m256i {
    let permute = _mm256_permutexvar_epi16(idx, a).as_i16x16();
    transmute(simd_select_bitmask(k, permute, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm256_maskz_permutexvar_epi16(k: __mmask16, idx: __m256i, a: __m256i) -> __m256i {
    let permute = _mm256_permutexvar_epi16(idx, a).as_i16x16();
    transmute(simd_select_bitmask(k, permute, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm_permutexvar_epi16(idx: __m128i, a: __m128i) -> __m128i {
    transmute(vpermw128(a.as_i16x8(), idx.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm_mask_permutexvar_epi16(
    src: __m128i,
    k: __mmask8,
    idx: __m128i,
    a: __m128i,
) -> __m128i {
    let permute = _mm_permutexvar_epi16(idx, a).as_i16x8();
    transmute(simd_select_bitmask(k, permute, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpermw))]
pub unsafe fn _mm_maskz_permutexvar_epi16(k: __mmask8, idx: __m128i, a: __m128i) -> __m128i {
    let permute = _mm_permutexvar_epi16(idx, a).as_i16x8();
    transmute(simd_select_bitmask(k, permute, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))] pub unsafe fn _mm512_mask_blend_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_select_bitmask(k, b.as_i16x32(), a.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))] pub unsafe fn _mm256_mask_blend_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    transmute(simd_select_bitmask(k, b.as_i16x16(), a.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))] pub unsafe fn _mm_mask_blend_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    transmute(simd_select_bitmask(k, b.as_i16x8(), a.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))] pub unsafe fn _mm512_mask_blend_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    transmute(simd_select_bitmask(k, b.as_i8x64(), a.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))] pub unsafe fn _mm256_mask_blend_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    transmute(simd_select_bitmask(k, b.as_i8x32(), a.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))] pub unsafe fn _mm_mask_blend_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    transmute(simd_select_bitmask(k, b.as_i8x16(), a.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm512_broadcastw_epi16(a: __m128i) -> __m512i {
    let a = _mm512_castsi128_si512(a).as_i16x32();
    let ret: i16x32 = simd_shuffle!(
        a,
        a,
        [
            0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
            0, 0, 0,
        ],
    );
    transmute(ret)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm512_mask_broadcastw_epi16(src: __m512i, k: __mmask32, a: __m128i) -> __m512i {
    let broadcast = _mm512_broadcastw_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, broadcast, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm512_maskz_broadcastw_epi16(k: __mmask32, a: __m128i) -> __m512i {
    let broadcast = _mm512_broadcastw_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, broadcast, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm256_mask_broadcastw_epi16(src: __m256i, k: __mmask16, a: __m128i) -> __m256i {
    let broadcast = _mm256_broadcastw_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, broadcast, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm256_maskz_broadcastw_epi16(k: __mmask16, a: __m128i) -> __m256i {
    let broadcast = _mm256_broadcastw_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, broadcast, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm_mask_broadcastw_epi16(src: __m128i, k: __mmask8, a: __m128i) -> __m128i {
    let broadcast = _mm_broadcastw_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, broadcast, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm_maskz_broadcastw_epi16(k: __mmask8, a: __m128i) -> __m128i {
    let broadcast = _mm_broadcastw_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, broadcast, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastb))]
pub unsafe fn _mm512_broadcastb_epi8(a: __m128i) -> __m512i {
    let a = _mm512_castsi128_si512(a).as_i8x64();
    let ret: i8x64 = simd_shuffle!(
        a,
        a,
        [
            0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
            0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
            0, 0, 0, 0, 0, 0,
        ],
    );
    transmute(ret)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastb))]
pub unsafe fn _mm512_mask_broadcastb_epi8(src: __m512i, k: __mmask64, a: __m128i) -> __m512i {
    let broadcast = _mm512_broadcastb_epi8(a).as_i8x64();
    transmute(simd_select_bitmask(k, broadcast, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastb))]
pub unsafe fn _mm512_maskz_broadcastb_epi8(k: __mmask64, a: __m128i) -> __m512i {
    let broadcast = _mm512_broadcastb_epi8(a).as_i8x64();
    transmute(simd_select_bitmask(k, broadcast, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastb))]
pub unsafe fn _mm256_mask_broadcastb_epi8(src: __m256i, k: __mmask32, a: __m128i) -> __m256i {
    let broadcast = _mm256_broadcastb_epi8(a).as_i8x32();
    transmute(simd_select_bitmask(k, broadcast, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastb))]
pub unsafe fn _mm256_maskz_broadcastb_epi8(k: __mmask32, a: __m128i) -> __m256i {
    let broadcast = _mm256_broadcastb_epi8(a).as_i8x32();
    transmute(simd_select_bitmask(k, broadcast, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastb))]
pub unsafe fn _mm_mask_broadcastb_epi8(src: __m128i, k: __mmask16, a: __m128i) -> __m128i {
    let broadcast = _mm_broadcastb_epi8(a).as_i8x16();
    transmute(simd_select_bitmask(k, broadcast, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastb))]
pub unsafe fn _mm_maskz_broadcastb_epi8(k: __mmask16, a: __m128i) -> __m128i {
    let broadcast = _mm_broadcastb_epi8(a).as_i8x16();
    transmute(simd_select_bitmask(k, broadcast, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhwd))]
pub unsafe fn _mm512_unpackhi_epi16(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_i16x32();
    let b = b.as_i16x32();
    #[rustfmt::skip]
    let r: i16x32 = simd_shuffle!(
        a,
        b,
        [
            4, 32 + 4, 5, 32 + 5,
            6, 32 + 6, 7, 32 + 7,
            12, 32 + 12, 13, 32 + 13,
            14, 32 + 14, 15, 32 + 15,
            20, 32 + 20, 21, 32 + 21,
            22, 32 + 22, 23, 32 + 23,
            28, 32 + 28, 29, 32 + 29,
            30, 32 + 30, 31, 32 + 31,
        ],
    );
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhwd))]
pub unsafe fn _mm512_mask_unpackhi_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let unpackhi = _mm512_unpackhi_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, unpackhi, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhwd))]
pub unsafe fn _mm512_maskz_unpackhi_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let unpackhi = _mm512_unpackhi_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, unpackhi, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhwd))]
pub unsafe fn _mm256_mask_unpackhi_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let unpackhi = _mm256_unpackhi_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, unpackhi, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhwd))]
pub unsafe fn _mm256_maskz_unpackhi_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let unpackhi = _mm256_unpackhi_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, unpackhi, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhwd))]
pub unsafe fn _mm_mask_unpackhi_epi16(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
    b: __m128i,
) -> __m128i {
    let unpackhi = _mm_unpackhi_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, unpackhi, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhwd))]
pub unsafe fn _mm_maskz_unpackhi_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let unpackhi = _mm_unpackhi_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, unpackhi, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhbw))]
pub unsafe fn _mm512_unpackhi_epi8(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_i8x64();
    let b = b.as_i8x64();
    #[rustfmt::skip]
    let r: i8x64 = simd_shuffle!(
        a,
        b,
        [
            8,  64+8,   9, 64+9,
            10, 64+10, 11, 64+11,
            12, 64+12, 13, 64+13,
            14, 64+14, 15, 64+15,
            24, 64+24, 25, 64+25,
            26, 64+26, 27, 64+27,
            28, 64+28, 29, 64+29,
            30, 64+30, 31, 64+31,
            40, 64+40, 41, 64+41,
            42, 64+42, 43, 64+43,
            44, 64+44, 45, 64+45,
            46, 64+46, 47, 64+47,
            56, 64+56, 57, 64+57,
            58, 64+58, 59, 64+59,
            60, 64+60, 61, 64+61,
            62, 64+62, 63, 64+63,
        ],
    );
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhbw))]
pub unsafe fn _mm512_mask_unpackhi_epi8(
    src: __m512i,
    k: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let unpackhi = _mm512_unpackhi_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, unpackhi, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhbw))]
pub unsafe fn _mm512_maskz_unpackhi_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let unpackhi = _mm512_unpackhi_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, unpackhi, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhbw))]
pub unsafe fn _mm256_mask_unpackhi_epi8(
    src: __m256i,
    k: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let unpackhi = _mm256_unpackhi_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, unpackhi, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhbw))]
pub unsafe fn _mm256_maskz_unpackhi_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let unpackhi = _mm256_unpackhi_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, unpackhi, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhbw))]
pub unsafe fn _mm_mask_unpackhi_epi8(
    src: __m128i,
    k: __mmask16,
    a: __m128i,
    b: __m128i,
) -> __m128i {
    let unpackhi = _mm_unpackhi_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, unpackhi, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpckhbw))]
pub unsafe fn _mm_maskz_unpackhi_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let unpackhi = _mm_unpackhi_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, unpackhi, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklwd))]
pub unsafe fn _mm512_unpacklo_epi16(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_i16x32();
    let b = b.as_i16x32();
    #[rustfmt::skip]
    let r: i16x32 = simd_shuffle!(
        a,
        b,
        [
            0,  32+0,   1, 32+1,
            2,  32+2,   3, 32+3,
            8,  32+8,   9, 32+9,
            10, 32+10, 11, 32+11,
            16, 32+16, 17, 32+17,
            18, 32+18, 19, 32+19,
            24, 32+24, 25, 32+25,
            26, 32+26, 27, 32+27
        ],
    );
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklwd))]
pub unsafe fn _mm512_mask_unpacklo_epi16(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let unpacklo = _mm512_unpacklo_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, unpacklo, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklwd))]
pub unsafe fn _mm512_maskz_unpacklo_epi16(k: __mmask32, a: __m512i, b: __m512i) -> __m512i {
    let unpacklo = _mm512_unpacklo_epi16(a, b).as_i16x32();
    transmute(simd_select_bitmask(k, unpacklo, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklwd))]
pub unsafe fn _mm256_mask_unpacklo_epi16(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let unpacklo = _mm256_unpacklo_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, unpacklo, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklwd))]
pub unsafe fn _mm256_maskz_unpacklo_epi16(k: __mmask16, a: __m256i, b: __m256i) -> __m256i {
    let unpacklo = _mm256_unpacklo_epi16(a, b).as_i16x16();
    transmute(simd_select_bitmask(k, unpacklo, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklwd))]
pub unsafe fn _mm_mask_unpacklo_epi16(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
    b: __m128i,
) -> __m128i {
    let unpacklo = _mm_unpacklo_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, unpacklo, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklwd))]
pub unsafe fn _mm_maskz_unpacklo_epi16(k: __mmask8, a: __m128i, b: __m128i) -> __m128i {
    let unpacklo = _mm_unpacklo_epi16(a, b).as_i16x8();
    transmute(simd_select_bitmask(k, unpacklo, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklbw))]
pub unsafe fn _mm512_unpacklo_epi8(a: __m512i, b: __m512i) -> __m512i {
    let a = a.as_i8x64();
    let b = b.as_i8x64();
    #[rustfmt::skip]
    let r: i8x64 = simd_shuffle!(
        a,
        b,
        [
            0,  64+0,   1, 64+1,
            2,  64+2,   3, 64+3,
            4,  64+4,   5, 64+5,
            6,  64+6,   7, 64+7,
            16, 64+16, 17, 64+17,
            18, 64+18, 19, 64+19,
            20, 64+20, 21, 64+21,
            22, 64+22, 23, 64+23,
            32, 64+32, 33, 64+33,
            34, 64+34, 35, 64+35,
            36, 64+36, 37, 64+37,
            38, 64+38, 39, 64+39,
            48, 64+48, 49, 64+49,
            50, 64+50, 51, 64+51,
            52, 64+52, 53, 64+53,
            54, 64+54, 55, 64+55,
        ],
    );
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklbw))]
pub unsafe fn _mm512_mask_unpacklo_epi8(
    src: __m512i,
    k: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let unpacklo = _mm512_unpacklo_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, unpacklo, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklbw))]
pub unsafe fn _mm512_maskz_unpacklo_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let unpacklo = _mm512_unpacklo_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, unpacklo, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklbw))]
pub unsafe fn _mm256_mask_unpacklo_epi8(
    src: __m256i,
    k: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let unpacklo = _mm256_unpacklo_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, unpacklo, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklbw))]
pub unsafe fn _mm256_maskz_unpacklo_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let unpacklo = _mm256_unpacklo_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, unpacklo, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklbw))]
pub unsafe fn _mm_mask_unpacklo_epi8(
    src: __m128i,
    k: __mmask16,
    a: __m128i,
    b: __m128i,
) -> __m128i {
    let unpacklo = _mm_unpacklo_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, unpacklo, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpunpcklbw))]
pub unsafe fn _mm_maskz_unpacklo_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let unpacklo = _mm_unpacklo_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, unpacklo, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
pub unsafe fn _mm512_mask_mov_epi16(src: __m512i, k: __mmask32, a: __m512i) -> __m512i {
    let mov = a.as_i16x32();
    transmute(simd_select_bitmask(k, mov, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
pub unsafe fn _mm512_maskz_mov_epi16(k: __mmask32, a: __m512i) -> __m512i {
    let mov = a.as_i16x32();
    transmute(simd_select_bitmask(k, mov, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
pub unsafe fn _mm256_mask_mov_epi16(src: __m256i, k: __mmask16, a: __m256i) -> __m256i {
    let mov = a.as_i16x16();
    transmute(simd_select_bitmask(k, mov, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
pub unsafe fn _mm256_maskz_mov_epi16(k: __mmask16, a: __m256i) -> __m256i {
    let mov = a.as_i16x16();
    transmute(simd_select_bitmask(k, mov, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
pub unsafe fn _mm_mask_mov_epi16(src: __m128i, k: __mmask8, a: __m128i) -> __m128i {
    let mov = a.as_i16x8();
    transmute(simd_select_bitmask(k, mov, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu16))]
pub unsafe fn _mm_maskz_mov_epi16(k: __mmask8, a: __m128i) -> __m128i {
    let mov = a.as_i16x8();
    transmute(simd_select_bitmask(k, mov, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
pub unsafe fn _mm512_mask_mov_epi8(src: __m512i, k: __mmask64, a: __m512i) -> __m512i {
    let mov = a.as_i8x64();
    transmute(simd_select_bitmask(k, mov, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
pub unsafe fn _mm512_maskz_mov_epi8(k: __mmask64, a: __m512i) -> __m512i {
    let mov = a.as_i8x64();
    transmute(simd_select_bitmask(k, mov, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
pub unsafe fn _mm256_mask_mov_epi8(src: __m256i, k: __mmask32, a: __m256i) -> __m256i {
    let mov = a.as_i8x32();
    transmute(simd_select_bitmask(k, mov, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
pub unsafe fn _mm256_maskz_mov_epi8(k: __mmask32, a: __m256i) -> __m256i {
    let mov = a.as_i8x32();
    transmute(simd_select_bitmask(k, mov, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
pub unsafe fn _mm_mask_mov_epi8(src: __m128i, k: __mmask16, a: __m128i) -> __m128i {
    let mov = a.as_i8x16();
    transmute(simd_select_bitmask(k, mov, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vmovdqu8))]
pub unsafe fn _mm_maskz_mov_epi8(k: __mmask16, a: __m128i) -> __m128i {
    let mov = a.as_i8x16();
    transmute(simd_select_bitmask(k, mov, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm512_mask_set1_epi16(src: __m512i, k: __mmask32, a: i16) -> __m512i {
    let r = _mm512_set1_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, r, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm512_maskz_set1_epi16(k: __mmask32, a: i16) -> __m512i {
    let r = _mm512_set1_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, r, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm256_mask_set1_epi16(src: __m256i, k: __mmask16, a: i16) -> __m256i {
    let r = _mm256_set1_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, r, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm256_maskz_set1_epi16(k: __mmask16, a: i16) -> __m256i {
    let r = _mm256_set1_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, r, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm_mask_set1_epi16(src: __m128i, k: __mmask8, a: i16) -> __m128i {
    let r = _mm_set1_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, r, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcastw))]
pub unsafe fn _mm_maskz_set1_epi16(k: __mmask8, a: i16) -> __m128i {
    let r = _mm_set1_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, r, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcast))]
pub unsafe fn _mm512_mask_set1_epi8(src: __m512i, k: __mmask64, a: i8) -> __m512i {
    let r = _mm512_set1_epi8(a).as_i8x64();
    transmute(simd_select_bitmask(k, r, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcast))]
pub unsafe fn _mm512_maskz_set1_epi8(k: __mmask64, a: i8) -> __m512i {
    let r = _mm512_set1_epi8(a).as_i8x64();
    transmute(simd_select_bitmask(k, r, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcast))]
pub unsafe fn _mm256_mask_set1_epi8(src: __m256i, k: __mmask32, a: i8) -> __m256i {
    let r = _mm256_set1_epi8(a).as_i8x32();
    transmute(simd_select_bitmask(k, r, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcast))]
pub unsafe fn _mm256_maskz_set1_epi8(k: __mmask32, a: i8) -> __m256i {
    let r = _mm256_set1_epi8(a).as_i8x32();
    transmute(simd_select_bitmask(k, r, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcast))]
pub unsafe fn _mm_mask_set1_epi8(src: __m128i, k: __mmask16, a: i8) -> __m128i {
    let r = _mm_set1_epi8(a).as_i8x16();
    transmute(simd_select_bitmask(k, r, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpbroadcast))]
pub unsafe fn _mm_maskz_set1_epi8(k: __mmask16, a: i8) -> __m128i {
    let r = _mm_set1_epi8(a).as_i8x16();
    transmute(simd_select_bitmask(k, r, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshuflw, IMM8 = 0))]
#[rustc_legacy_const_generics(1)]
pub unsafe fn _mm512_shufflelo_epi16<const IMM8: i32>(a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_i16x32();
    let r: i16x32 = simd_shuffle!(
        a,
        a,
        [
            IMM8 as u32 & 0b11,
            (IMM8 as u32 >> 2) & 0b11,
            (IMM8 as u32 >> 4) & 0b11,
            (IMM8 as u32 >> 6) & 0b11,
            4,
            5,
            6,
            7,
            (IMM8 as u32 & 0b11) + 8,
            ((IMM8 as u32 >> 2) & 0b11) + 8,
            ((IMM8 as u32 >> 4) & 0b11) + 8,
            ((IMM8 as u32 >> 6) & 0b11) + 8,
            12,
            13,
            14,
            15,
            (IMM8 as u32 & 0b11) + 16,
            ((IMM8 as u32 >> 2) & 0b11) + 16,
            ((IMM8 as u32 >> 4) & 0b11) + 16,
            ((IMM8 as u32 >> 6) & 0b11) + 16,
            20,
            21,
            22,
            23,
            (IMM8 as u32 & 0b11) + 24,
            ((IMM8 as u32 >> 2) & 0b11) + 24,
            ((IMM8 as u32 >> 4) & 0b11) + 24,
            ((IMM8 as u32 >> 6) & 0b11) + 24,
            28,
            29,
            30,
            31,
        ],
    );
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshuflw, IMM8 = 0))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm512_mask_shufflelo_epi16<const IMM8: i32>(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm512_shufflelo_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, r.as_i16x32(), src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshuflw, IMM8 = 0))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm512_maskz_shufflelo_epi16<const IMM8: i32>(k: __mmask32, a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm512_shufflelo_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, r.as_i16x32(), i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshuflw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm256_mask_shufflelo_epi16<const IMM8: i32>(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let shuffle = _mm256_shufflelo_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shuffle.as_i16x16(), src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshuflw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm256_maskz_shufflelo_epi16<const IMM8: i32>(k: __mmask16, a: __m256i) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let shuffle = _mm256_shufflelo_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shuffle.as_i16x16(), i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshuflw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm_mask_shufflelo_epi16<const IMM8: i32>(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let shuffle = _mm_shufflelo_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shuffle.as_i16x8(), src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshuflw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm_maskz_shufflelo_epi16<const IMM8: i32>(k: __mmask8, a: __m128i) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let shuffle = _mm_shufflelo_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shuffle.as_i16x8(), i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufhw, IMM8 = 0))]
#[rustc_legacy_const_generics(1)]
pub unsafe fn _mm512_shufflehi_epi16<const IMM8: i32>(a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_i16x32();
    let r: i16x32 = simd_shuffle!(
        a,
        a,
        [
            0,
            1,
            2,
            3,
            (IMM8 as u32 & 0b11) + 4,
            ((IMM8 as u32 >> 2) & 0b11) + 4,
            ((IMM8 as u32 >> 4) & 0b11) + 4,
            ((IMM8 as u32 >> 6) & 0b11) + 4,
            8,
            9,
            10,
            11,
            (IMM8 as u32 & 0b11) + 12,
            ((IMM8 as u32 >> 2) & 0b11) + 12,
            ((IMM8 as u32 >> 4) & 0b11) + 12,
            ((IMM8 as u32 >> 6) & 0b11) + 12,
            16,
            17,
            18,
            19,
            (IMM8 as u32 & 0b11) + 20,
            ((IMM8 as u32 >> 2) & 0b11) + 20,
            ((IMM8 as u32 >> 4) & 0b11) + 20,
            ((IMM8 as u32 >> 6) & 0b11) + 20,
            24,
            25,
            26,
            27,
            (IMM8 as u32 & 0b11) + 28,
            ((IMM8 as u32 >> 2) & 0b11) + 28,
            ((IMM8 as u32 >> 4) & 0b11) + 28,
            ((IMM8 as u32 >> 6) & 0b11) + 28,
        ],
    );
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufhw, IMM8 = 0))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm512_mask_shufflehi_epi16<const IMM8: i32>(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm512_shufflehi_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, r.as_i16x32(), src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufhw, IMM8 = 0))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm512_maskz_shufflehi_epi16<const IMM8: i32>(k: __mmask32, a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm512_shufflehi_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, r.as_i16x32(), i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufhw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm256_mask_shufflehi_epi16<const IMM8: i32>(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let shuffle = _mm256_shufflehi_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shuffle.as_i16x16(), src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufhw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm256_maskz_shufflehi_epi16<const IMM8: i32>(k: __mmask16, a: __m256i) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let shuffle = _mm256_shufflehi_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shuffle.as_i16x16(), i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufhw, IMM8 = 5))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm_mask_shufflehi_epi16<const IMM8: i32>(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let shuffle = _mm_shufflehi_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shuffle.as_i16x8(), src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufhw, IMM8 = 5))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm_maskz_shufflehi_epi16<const IMM8: i32>(k: __mmask8, a: __m128i) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let shuffle = _mm_shufflehi_epi16::<IMM8>(a);
    transmute(simd_select_bitmask(k, shuffle.as_i16x8(), i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufb))]
pub unsafe fn _mm512_shuffle_epi8(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpshufb(a.as_i8x64(), b.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufb))]
pub unsafe fn _mm512_mask_shuffle_epi8(
    src: __m512i,
    k: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    let shuffle = _mm512_shuffle_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, shuffle, src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufb))]
pub unsafe fn _mm512_maskz_shuffle_epi8(k: __mmask64, a: __m512i, b: __m512i) -> __m512i {
    let shuffle = _mm512_shuffle_epi8(a, b).as_i8x64();
    transmute(simd_select_bitmask(k, shuffle, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufb))]
pub unsafe fn _mm256_mask_shuffle_epi8(
    src: __m256i,
    k: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    let shuffle = _mm256_shuffle_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, shuffle, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufb))]
pub unsafe fn _mm256_maskz_shuffle_epi8(k: __mmask32, a: __m256i, b: __m256i) -> __m256i {
    let shuffle = _mm256_shuffle_epi8(a, b).as_i8x32();
    transmute(simd_select_bitmask(k, shuffle, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufb))]
pub unsafe fn _mm_mask_shuffle_epi8(src: __m128i, k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let shuffle = _mm_shuffle_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, shuffle, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpshufb))]
pub unsafe fn _mm_maskz_shuffle_epi8(k: __mmask16, a: __m128i, b: __m128i) -> __m128i {
    let shuffle = _mm_shuffle_epi8(a, b).as_i8x16();
    transmute(simd_select_bitmask(k, shuffle, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmw))]
pub unsafe fn _mm512_test_epi16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    let and = _mm512_and_si512(a, b);
    let zero = _mm512_setzero_si512();
    _mm512_cmpneq_epi16_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmw))]
pub unsafe fn _mm512_mask_test_epi16_mask(k: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    let and = _mm512_and_si512(a, b);
    let zero = _mm512_setzero_si512();
    _mm512_mask_cmpneq_epi16_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmw))]
pub unsafe fn _mm256_test_epi16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    let and = _mm256_and_si256(a, b);
    let zero = _mm256_setzero_si256();
    _mm256_cmpneq_epi16_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmw))]
pub unsafe fn _mm256_mask_test_epi16_mask(k: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    let and = _mm256_and_si256(a, b);
    let zero = _mm256_setzero_si256();
    _mm256_mask_cmpneq_epi16_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmw))]
pub unsafe fn _mm_test_epi16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    let and = _mm_and_si128(a, b);
    let zero = _mm_setzero_si128();
    _mm_cmpneq_epi16_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmw))]
pub unsafe fn _mm_mask_test_epi16_mask(k: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    let and = _mm_and_si128(a, b);
    let zero = _mm_setzero_si128();
    _mm_mask_cmpneq_epi16_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmb))]
pub unsafe fn _mm512_test_epi8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    let and = _mm512_and_si512(a, b);
    let zero = _mm512_setzero_si512();
    _mm512_cmpneq_epi8_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmb))]
pub unsafe fn _mm512_mask_test_epi8_mask(k: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    let and = _mm512_and_si512(a, b);
    let zero = _mm512_setzero_si512();
    _mm512_mask_cmpneq_epi8_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmb))]
pub unsafe fn _mm256_test_epi8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    let and = _mm256_and_si256(a, b);
    let zero = _mm256_setzero_si256();
    _mm256_cmpneq_epi8_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmb))]
pub unsafe fn _mm256_mask_test_epi8_mask(k: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    let and = _mm256_and_si256(a, b);
    let zero = _mm256_setzero_si256();
    _mm256_mask_cmpneq_epi8_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmb))]
pub unsafe fn _mm_test_epi8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    let and = _mm_and_si128(a, b);
    let zero = _mm_setzero_si128();
    _mm_cmpneq_epi8_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestmb))]
pub unsafe fn _mm_mask_test_epi8_mask(k: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    let and = _mm_and_si128(a, b);
    let zero = _mm_setzero_si128();
    _mm_mask_cmpneq_epi8_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmw))]
pub unsafe fn _mm512_testn_epi16_mask(a: __m512i, b: __m512i) -> __mmask32 {
    let and = _mm512_and_si512(a, b);
    let zero = _mm512_setzero_si512();
    _mm512_cmpeq_epi16_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmw))]
pub unsafe fn _mm512_mask_testn_epi16_mask(k: __mmask32, a: __m512i, b: __m512i) -> __mmask32 {
    let and = _mm512_and_si512(a, b);
    let zero = _mm512_setzero_si512();
    _mm512_mask_cmpeq_epi16_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmw))]
pub unsafe fn _mm256_testn_epi16_mask(a: __m256i, b: __m256i) -> __mmask16 {
    let and = _mm256_and_si256(a, b);
    let zero = _mm256_setzero_si256();
    _mm256_cmpeq_epi16_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmw))]
pub unsafe fn _mm256_mask_testn_epi16_mask(k: __mmask16, a: __m256i, b: __m256i) -> __mmask16 {
    let and = _mm256_and_si256(a, b);
    let zero = _mm256_setzero_si256();
    _mm256_mask_cmpeq_epi16_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmw))]
pub unsafe fn _mm_testn_epi16_mask(a: __m128i, b: __m128i) -> __mmask8 {
    let and = _mm_and_si128(a, b);
    let zero = _mm_setzero_si128();
    _mm_cmpeq_epi16_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmw))]
pub unsafe fn _mm_mask_testn_epi16_mask(k: __mmask8, a: __m128i, b: __m128i) -> __mmask8 {
    let and = _mm_and_si128(a, b);
    let zero = _mm_setzero_si128();
    _mm_mask_cmpeq_epi16_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmb))]
pub unsafe fn _mm512_testn_epi8_mask(a: __m512i, b: __m512i) -> __mmask64 {
    let and = _mm512_and_si512(a, b);
    let zero = _mm512_setzero_si512();
    _mm512_cmpeq_epi8_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmb))]
pub unsafe fn _mm512_mask_testn_epi8_mask(k: __mmask64, a: __m512i, b: __m512i) -> __mmask64 {
    let and = _mm512_and_si512(a, b);
    let zero = _mm512_setzero_si512();
    _mm512_mask_cmpeq_epi8_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmb))]
pub unsafe fn _mm256_testn_epi8_mask(a: __m256i, b: __m256i) -> __mmask32 {
    let and = _mm256_and_si256(a, b);
    let zero = _mm256_setzero_si256();
    _mm256_cmpeq_epi8_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmb))]
pub unsafe fn _mm256_mask_testn_epi8_mask(k: __mmask32, a: __m256i, b: __m256i) -> __mmask32 {
    let and = _mm256_and_si256(a, b);
    let zero = _mm256_setzero_si256();
    _mm256_mask_cmpeq_epi8_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmb))]
pub unsafe fn _mm_testn_epi8_mask(a: __m128i, b: __m128i) -> __mmask16 {
    let and = _mm_and_si128(a, b);
    let zero = _mm_setzero_si128();
    _mm_cmpeq_epi8_mask(and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vptestnmb))]
pub unsafe fn _mm_mask_testn_epi8_mask(k: __mmask16, a: __m128i, b: __m128i) -> __mmask16 {
    let and = _mm_and_si128(a, b);
    let zero = _mm_setzero_si128();
    _mm_mask_cmpeq_epi8_mask(k, and, zero)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(mov))] pub unsafe fn _store_mask64(mem_addr: *mut __mmask64, a: __mmask64) {
    ptr::write(mem_addr as *mut __mmask64, a);
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(mov))] pub unsafe fn _store_mask32(mem_addr: *mut __mmask32, a: __mmask32) {
    ptr::write(mem_addr as *mut __mmask32, a);
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(mov))] pub unsafe fn _load_mask64(mem_addr: *const __mmask64) -> __mmask64 {
    ptr::read(mem_addr as *const __mmask64)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(mov))] pub unsafe fn _load_mask32(mem_addr: *const __mmask32) -> __mmask32 {
    ptr::read(mem_addr as *const __mmask32)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsadbw))]
pub unsafe fn _mm512_sad_epu8(a: __m512i, b: __m512i) -> __m512i {
    transmute(vpsadbw(a.as_u8x64(), b.as_u8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm512_dbsad_epu8<const IMM8: i32>(a: __m512i, b: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x64();
    let b = b.as_u8x64();
    let r = vdbpsadbw(a, b, IMM8);
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(4)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm512_mask_dbsad_epu8<const IMM8: i32>(
    src: __m512i,
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x64();
    let b = b.as_u8x64();
    let r = vdbpsadbw(a, b, IMM8);
    transmute(simd_select_bitmask(k, r, src.as_u16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm512_maskz_dbsad_epu8<const IMM8: i32>(
    k: __mmask32,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x64();
    let b = b.as_u8x64();
    let r = vdbpsadbw(a, b, IMM8);
    transmute(simd_select_bitmask(k, r, u16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm256_dbsad_epu8<const IMM8: i32>(a: __m256i, b: __m256i) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x32();
    let b = b.as_u8x32();
    let r = vdbpsadbw256(a, b, IMM8);
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(4)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm256_mask_dbsad_epu8<const IMM8: i32>(
    src: __m256i,
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x32();
    let b = b.as_u8x32();
    let r = vdbpsadbw256(a, b, IMM8);
    transmute(simd_select_bitmask(k, r, src.as_u16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm256_maskz_dbsad_epu8<const IMM8: i32>(
    k: __mmask16,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x32();
    let b = b.as_u8x32();
    let r = vdbpsadbw256(a, b, IMM8);
    transmute(simd_select_bitmask(k, r, u16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(2)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm_dbsad_epu8<const IMM8: i32>(a: __m128i, b: __m128i) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x16();
    let b = b.as_u8x16();
    let r = vdbpsadbw128(a, b, IMM8);
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(4)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm_mask_dbsad_epu8<const IMM8: i32>(
    src: __m128i,
    k: __mmask8,
    a: __m128i,
    b: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x16();
    let b = b.as_u8x16();
    let r = vdbpsadbw128(a, b, IMM8);
    transmute(simd_select_bitmask(k, r, src.as_u16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vdbpsadbw, IMM8 = 0))]
pub unsafe fn _mm_maskz_dbsad_epu8<const IMM8: i32>(
    k: __mmask8,
    a: __m128i,
    b: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_u8x16();
    let b = b.as_u8x16();
    let r = vdbpsadbw128(a, b, IMM8);
    transmute(simd_select_bitmask(k, r, u16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovw2m))]
pub unsafe fn _mm512_movepi16_mask(a: __m512i) -> __mmask32 {
    let filter = _mm512_set1_epi16(1 << 15);
    let a = _mm512_and_si512(a, filter);
    _mm512_cmpeq_epi16_mask(a, filter)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovw2m))]
pub unsafe fn _mm256_movepi16_mask(a: __m256i) -> __mmask16 {
    let filter = _mm256_set1_epi16(1 << 15);
    let a = _mm256_and_si256(a, filter);
    _mm256_cmpeq_epi16_mask(a, filter)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovw2m))]
pub unsafe fn _mm_movepi16_mask(a: __m128i) -> __mmask8 {
    let filter = _mm_set1_epi16(1 << 15);
    let a = _mm_and_si128(a, filter);
    _mm_cmpeq_epi16_mask(a, filter)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovb2m))]
pub unsafe fn _mm512_movepi8_mask(a: __m512i) -> __mmask64 {
    let filter = _mm512_set1_epi8(1 << 7);
    let a = _mm512_and_si512(a, filter);
    _mm512_cmpeq_epi8_mask(a, filter)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovmskb))] pub unsafe fn _mm256_movepi8_mask(a: __m256i) -> __mmask32 {
    let filter = _mm256_set1_epi8(1 << 7);
    let a = _mm256_and_si256(a, filter);
    _mm256_cmpeq_epi8_mask(a, filter)
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovmskb))] pub unsafe fn _mm_movepi8_mask(a: __m128i) -> __mmask16 {
    let filter = _mm_set1_epi8(1 << 7);
    let a = _mm_and_si128(a, filter);
    _mm_cmpeq_epi8_mask(a, filter)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovm2w))]
pub unsafe fn _mm512_movm_epi16(k: __mmask32) -> __m512i {
    let one = _mm512_set1_epi16(
        1 << 15
            | 1 << 14
            | 1 << 13
            | 1 << 12
            | 1 << 11
            | 1 << 10
            | 1 << 9
            | 1 << 8
            | 1 << 7
            | 1 << 6
            | 1 << 5
            | 1 << 4
            | 1 << 3
            | 1 << 2
            | 1 << 1
            | 1 << 0,
    )
    .as_i16x32();
    transmute(simd_select_bitmask(k, one, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovm2w))]
pub unsafe fn _mm256_movm_epi16(k: __mmask16) -> __m256i {
    let one = _mm256_set1_epi16(
        1 << 15
            | 1 << 14
            | 1 << 13
            | 1 << 12
            | 1 << 11
            | 1 << 10
            | 1 << 9
            | 1 << 8
            | 1 << 7
            | 1 << 6
            | 1 << 5
            | 1 << 4
            | 1 << 3
            | 1 << 2
            | 1 << 1
            | 1 << 0,
    )
    .as_i16x16();
    transmute(simd_select_bitmask(k, one, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovm2w))]
pub unsafe fn _mm_movm_epi16(k: __mmask8) -> __m128i {
    let one = _mm_set1_epi16(
        1 << 15
            | 1 << 14
            | 1 << 13
            | 1 << 12
            | 1 << 11
            | 1 << 10
            | 1 << 9
            | 1 << 8
            | 1 << 7
            | 1 << 6
            | 1 << 5
            | 1 << 4
            | 1 << 3
            | 1 << 2
            | 1 << 1
            | 1 << 0,
    )
    .as_i16x8();
    transmute(simd_select_bitmask(k, one, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovm2b))]
pub unsafe fn _mm512_movm_epi8(k: __mmask64) -> __m512i {
    let one =
        _mm512_set1_epi8(1 << 7 | 1 << 6 | 1 << 5 | 1 << 4 | 1 << 3 | 1 << 2 | 1 << 1 | 1 << 0)
            .as_i8x64();
    transmute(simd_select_bitmask(k, one, i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovm2b))]
pub unsafe fn _mm256_movm_epi8(k: __mmask32) -> __m256i {
    let one =
        _mm256_set1_epi8(1 << 7 | 1 << 6 | 1 << 5 | 1 << 4 | 1 << 3 | 1 << 2 | 1 << 1 | 1 << 0)
            .as_i8x32();
    transmute(simd_select_bitmask(k, one, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovm2b))]
pub unsafe fn _mm_movm_epi8(k: __mmask16) -> __m128i {
    let one = _mm_set1_epi8(1 << 7 | 1 << 6 | 1 << 5 | 1 << 4 | 1 << 3 | 1 << 2 | 1 << 1 | 1 << 0)
        .as_i8x16();
    transmute(simd_select_bitmask(k, one, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _cvtmask32_u32(a: __mmask32) -> u32 {
    a
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _cvtu32_mask32(a: u32) -> __mmask32 {
    a
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kadd_mask32(a: __mmask32, b: __mmask32) -> __mmask32 {
    a + b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kadd_mask64(a: __mmask64, b: __mmask64) -> __mmask64 {
    a + b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kand_mask32(a: __mmask32, b: __mmask32) -> __mmask32 {
    a & b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kand_mask64(a: __mmask64, b: __mmask64) -> __mmask64 {
    a & b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _knot_mask32(a: __mmask32) -> __mmask32 {
    !a
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _knot_mask64(a: __mmask64) -> __mmask64 {
    !a
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kandn_mask32(a: __mmask32, b: __mmask32) -> __mmask32 {
    _knot_mask32(a) & b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kandn_mask64(a: __mmask64, b: __mmask64) -> __mmask64 {
    _knot_mask64(a) & b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kor_mask32(a: __mmask32, b: __mmask32) -> __mmask32 {
    a | b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kor_mask64(a: __mmask64, b: __mmask64) -> __mmask64 {
    a | b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kxor_mask32(a: __mmask32, b: __mmask32) -> __mmask32 {
    a ^ b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kxor_mask64(a: __mmask64, b: __mmask64) -> __mmask64 {
    a ^ b
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kxnor_mask32(a: __mmask32, b: __mmask32) -> __mmask32 {
    _knot_mask32(a ^ b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kxnor_mask64(a: __mmask64, b: __mmask64) -> __mmask64 {
    _knot_mask64(a ^ b)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kortest_mask32_u8(a: __mmask32, b: __mmask32, all_ones: *mut u8) -> u8 {
    let tmp = _kor_mask32(a, b);
    *all_ones = (tmp == 0xffffffff) as u8;
    (tmp == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kortest_mask64_u8(a: __mmask64, b: __mmask64, all_ones: *mut u8) -> u8 {
    let tmp = _kor_mask64(a, b);
    *all_ones = (tmp == 0xffffffff_ffffffff) as u8;
    (tmp == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kortestc_mask32_u8(a: __mmask32, b: __mmask32) -> u8 {
    (_kor_mask32(a, b) == 0xffffffff) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kortestc_mask64_u8(a: __mmask64, b: __mmask64) -> u8 {
    (_kor_mask64(a, b) == 0xffffffff_ffffffff) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kortestz_mask32_u8(a: __mmask32, b: __mmask32) -> u8 {
    (_kor_mask32(a, b) == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kortestz_mask64_u8(a: __mmask64, b: __mmask64) -> u8 {
    (_kor_mask64(a, b) == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[rustc_legacy_const_generics(1)]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kshiftli_mask32<const COUNT: u32>(a: __mmask32) -> __mmask32 {
    a << COUNT
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[rustc_legacy_const_generics(1)]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kshiftli_mask64<const COUNT: u32>(a: __mmask64) -> __mmask64 {
    a << COUNT
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[rustc_legacy_const_generics(1)]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kshiftri_mask32<const COUNT: u32>(a: __mmask32) -> __mmask32 {
    a >> COUNT
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[rustc_legacy_const_generics(1)]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _kshiftri_mask64<const COUNT: u32>(a: __mmask64) -> __mmask64 {
    a >> COUNT
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _ktest_mask32_u8(a: __mmask32, b: __mmask32, and_not: *mut u8) -> u8 {
    *and_not = (_kandn_mask32(a, b) == 0) as u8;
    (_kand_mask32(a, b) == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _ktest_mask64_u8(a: __mmask64, b: __mmask64, and_not: *mut u8) -> u8 {
    *and_not = (_kandn_mask64(a, b) == 0) as u8;
    (_kand_mask64(a, b) == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _ktestc_mask32_u8(a: __mmask32, b: __mmask32) -> u8 {
    (_kandn_mask32(a, b) == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _ktestc_mask64_u8(a: __mmask64, b: __mmask64) -> u8 {
    (_kandn_mask64(a, b) == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _ktestz_mask32_u8(a: __mmask32, b: __mmask32) -> u8 {
    (_kand_mask32(a, b) == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
pub unsafe fn _ktestz_mask64_u8(a: __mmask64, b: __mmask64) -> u8 {
    (_kand_mask64(a, b) == 0) as u8
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(mov))] pub unsafe fn _mm512_kunpackw(a: __mmask32, b: __mmask32) -> __mmask32 {
    ((a & 0xffff) << 16) | (b & 0xffff)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(mov))] pub unsafe fn _mm512_kunpackd(a: __mmask64, b: __mmask64) -> __mmask64 {
    ((a & 0xffffffff) << 32) | (b & 0xffffffff)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm512_cvtepi16_epi8(a: __m512i) -> __m256i {
    let a = a.as_i16x32();
    transmute::<i8x32, _>(simd_cast(a))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm512_mask_cvtepi16_epi8(src: __m256i, k: __mmask32, a: __m512i) -> __m256i {
    let convert = _mm512_cvtepi16_epi8(a).as_i8x32();
    transmute(simd_select_bitmask(k, convert, src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm512_maskz_cvtepi16_epi8(k: __mmask32, a: __m512i) -> __m256i {
    let convert = _mm512_cvtepi16_epi8(a).as_i8x32();
    transmute(simd_select_bitmask(k, convert, i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm256_cvtepi16_epi8(a: __m256i) -> __m128i {
    let a = a.as_i16x16();
    transmute::<i8x16, _>(simd_cast(a))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm256_mask_cvtepi16_epi8(src: __m128i, k: __mmask16, a: __m256i) -> __m128i {
    let convert = _mm256_cvtepi16_epi8(a).as_i8x16();
    transmute(simd_select_bitmask(k, convert, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm256_maskz_cvtepi16_epi8(k: __mmask16, a: __m256i) -> __m128i {
    let convert = _mm256_cvtepi16_epi8(a).as_i8x16();
    transmute(simd_select_bitmask(k, convert, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm_cvtepi16_epi8(a: __m128i) -> __m128i {
    let a = a.as_i16x8();
    let v256: i16x16 = simd_shuffle!(
        a,
        i16x8::ZERO,
        [0, 1, 2, 3, 4, 5, 6, 7, 8, 8, 8, 8, 8, 8, 8, 8]
    );
    transmute::<i8x16, _>(simd_cast(v256))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm_mask_cvtepi16_epi8(src: __m128i, k: __mmask8, a: __m128i) -> __m128i {
    let convert = _mm_cvtepi16_epi8(a).as_i8x16();
    let k: __mmask16 = 0b11111111_11111111 & k as __mmask16;
    transmute(simd_select_bitmask(k, convert, src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm_maskz_cvtepi16_epi8(k: __mmask8, a: __m128i) -> __m128i {
    let convert = _mm_cvtepi16_epi8(a).as_i8x16();
    let k: __mmask16 = 0b11111111_11111111 & k as __mmask16;
    transmute(simd_select_bitmask(k, convert, i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm512_cvtsepi16_epi8(a: __m512i) -> __m256i {
    transmute(vpmovswb(
        a.as_i16x32(),
        i8x32::ZERO,
        0b11111111_11111111_11111111_11111111,
    ))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm512_mask_cvtsepi16_epi8(src: __m256i, k: __mmask32, a: __m512i) -> __m256i {
    transmute(vpmovswb(a.as_i16x32(), src.as_i8x32(), k))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm512_maskz_cvtsepi16_epi8(k: __mmask32, a: __m512i) -> __m256i {
    transmute(vpmovswb(a.as_i16x32(), i8x32::ZERO, k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm256_cvtsepi16_epi8(a: __m256i) -> __m128i {
    transmute(vpmovswb256(a.as_i16x16(), i8x16::ZERO, 0b11111111_11111111))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm256_mask_cvtsepi16_epi8(src: __m128i, k: __mmask16, a: __m256i) -> __m128i {
    transmute(vpmovswb256(a.as_i16x16(), src.as_i8x16(), k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm256_maskz_cvtsepi16_epi8(k: __mmask16, a: __m256i) -> __m128i {
    transmute(vpmovswb256(a.as_i16x16(), i8x16::ZERO, k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm_cvtsepi16_epi8(a: __m128i) -> __m128i {
    transmute(vpmovswb128(a.as_i16x8(), i8x16::ZERO, 0b11111111))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm_mask_cvtsepi16_epi8(src: __m128i, k: __mmask8, a: __m128i) -> __m128i {
    transmute(vpmovswb128(a.as_i16x8(), src.as_i8x16(), k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm_maskz_cvtsepi16_epi8(k: __mmask8, a: __m128i) -> __m128i {
    transmute(vpmovswb128(a.as_i16x8(), i8x16::ZERO, k))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm512_cvtusepi16_epi8(a: __m512i) -> __m256i {
    transmute(vpmovuswb(
        a.as_u16x32(),
        u8x32::ZERO,
        0b11111111_11111111_11111111_11111111,
    ))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm512_mask_cvtusepi16_epi8(src: __m256i, k: __mmask32, a: __m512i) -> __m256i {
    transmute(vpmovuswb(a.as_u16x32(), src.as_u8x32(), k))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm512_maskz_cvtusepi16_epi8(k: __mmask32, a: __m512i) -> __m256i {
    transmute(vpmovuswb(a.as_u16x32(), u8x32::ZERO, k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm256_cvtusepi16_epi8(a: __m256i) -> __m128i {
    transmute(vpmovuswb256(
        a.as_u16x16(),
        u8x16::ZERO,
        0b11111111_11111111,
    ))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm256_mask_cvtusepi16_epi8(src: __m128i, k: __mmask16, a: __m256i) -> __m128i {
    transmute(vpmovuswb256(a.as_u16x16(), src.as_u8x16(), k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm256_maskz_cvtusepi16_epi8(k: __mmask16, a: __m256i) -> __m128i {
    transmute(vpmovuswb256(a.as_u16x16(), u8x16::ZERO, k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm_cvtusepi16_epi8(a: __m128i) -> __m128i {
    transmute(vpmovuswb128(a.as_u16x8(), u8x16::ZERO, 0b11111111))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm_mask_cvtusepi16_epi8(src: __m128i, k: __mmask8, a: __m128i) -> __m128i {
    transmute(vpmovuswb128(a.as_u16x8(), src.as_u8x16(), k))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm_maskz_cvtusepi16_epi8(k: __mmask8, a: __m128i) -> __m128i {
    transmute(vpmovuswb128(a.as_u16x8(), u8x16::ZERO, k))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovsxbw))]
pub unsafe fn _mm512_cvtepi8_epi16(a: __m256i) -> __m512i {
    let a = a.as_i8x32();
    transmute::<i16x32, _>(simd_cast(a))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovsxbw))]
pub unsafe fn _mm512_mask_cvtepi8_epi16(src: __m512i, k: __mmask32, a: __m256i) -> __m512i {
    let convert = _mm512_cvtepi8_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, convert, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovsxbw))]
pub unsafe fn _mm512_maskz_cvtepi8_epi16(k: __mmask32, a: __m256i) -> __m512i {
    let convert = _mm512_cvtepi8_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, convert, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovsxbw))]
pub unsafe fn _mm256_mask_cvtepi8_epi16(src: __m256i, k: __mmask16, a: __m128i) -> __m256i {
    let convert = _mm256_cvtepi8_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, convert, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovsxbw))]
pub unsafe fn _mm256_maskz_cvtepi8_epi16(k: __mmask16, a: __m128i) -> __m256i {
    let convert = _mm256_cvtepi8_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, convert, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovsxbw))]
pub unsafe fn _mm_mask_cvtepi8_epi16(src: __m128i, k: __mmask8, a: __m128i) -> __m128i {
    let convert = _mm_cvtepi8_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, convert, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovsxbw))]
pub unsafe fn _mm_maskz_cvtepi8_epi16(k: __mmask8, a: __m128i) -> __m128i {
    let convert = _mm_cvtepi8_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, convert, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovzxbw))]
pub unsafe fn _mm512_cvtepu8_epi16(a: __m256i) -> __m512i {
    let a = a.as_u8x32();
    transmute::<i16x32, _>(simd_cast(a))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovzxbw))]
pub unsafe fn _mm512_mask_cvtepu8_epi16(src: __m512i, k: __mmask32, a: __m256i) -> __m512i {
    let convert = _mm512_cvtepu8_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, convert, src.as_i16x32()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovzxbw))]
pub unsafe fn _mm512_maskz_cvtepu8_epi16(k: __mmask32, a: __m256i) -> __m512i {
    let convert = _mm512_cvtepu8_epi16(a).as_i16x32();
    transmute(simd_select_bitmask(k, convert, i16x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovzxbw))]
pub unsafe fn _mm256_mask_cvtepu8_epi16(src: __m256i, k: __mmask16, a: __m128i) -> __m256i {
    let convert = _mm256_cvtepu8_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, convert, src.as_i16x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovzxbw))]
pub unsafe fn _mm256_maskz_cvtepu8_epi16(k: __mmask16, a: __m128i) -> __m256i {
    let convert = _mm256_cvtepu8_epi16(a).as_i16x16();
    transmute(simd_select_bitmask(k, convert, i16x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovzxbw))]
pub unsafe fn _mm_mask_cvtepu8_epi16(src: __m128i, k: __mmask8, a: __m128i) -> __m128i {
    let convert = _mm_cvtepu8_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, convert, src.as_i16x8()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovzxbw))]
pub unsafe fn _mm_maskz_cvtepu8_epi16(k: __mmask8, a: __m128i) -> __m128i {
    let convert = _mm_cvtepu8_epi16(a).as_i16x8();
    transmute(simd_select_bitmask(k, convert, i16x8::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpslldq, IMM8 = 3))]
#[rustc_legacy_const_generics(1)]
pub unsafe fn _mm512_bslli_epi128<const IMM8: i32>(a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    const fn mask(shift: i32, i: u32) -> u32 {
        let shift = shift as u32 & 0xff;
        if shift > 15 || i % 16 < shift {
            0
        } else {
            64 + (i - shift)
        }
    }
    let a = a.as_i8x64();
    let zero = i8x64::ZERO;
    let r: i8x64 = simd_shuffle!(
        zero,
        a,
        [
            mask(IMM8, 0),
            mask(IMM8, 1),
            mask(IMM8, 2),
            mask(IMM8, 3),
            mask(IMM8, 4),
            mask(IMM8, 5),
            mask(IMM8, 6),
            mask(IMM8, 7),
            mask(IMM8, 8),
            mask(IMM8, 9),
            mask(IMM8, 10),
            mask(IMM8, 11),
            mask(IMM8, 12),
            mask(IMM8, 13),
            mask(IMM8, 14),
            mask(IMM8, 15),
            mask(IMM8, 16),
            mask(IMM8, 17),
            mask(IMM8, 18),
            mask(IMM8, 19),
            mask(IMM8, 20),
            mask(IMM8, 21),
            mask(IMM8, 22),
            mask(IMM8, 23),
            mask(IMM8, 24),
            mask(IMM8, 25),
            mask(IMM8, 26),
            mask(IMM8, 27),
            mask(IMM8, 28),
            mask(IMM8, 29),
            mask(IMM8, 30),
            mask(IMM8, 31),
            mask(IMM8, 32),
            mask(IMM8, 33),
            mask(IMM8, 34),
            mask(IMM8, 35),
            mask(IMM8, 36),
            mask(IMM8, 37),
            mask(IMM8, 38),
            mask(IMM8, 39),
            mask(IMM8, 40),
            mask(IMM8, 41),
            mask(IMM8, 42),
            mask(IMM8, 43),
            mask(IMM8, 44),
            mask(IMM8, 45),
            mask(IMM8, 46),
            mask(IMM8, 47),
            mask(IMM8, 48),
            mask(IMM8, 49),
            mask(IMM8, 50),
            mask(IMM8, 51),
            mask(IMM8, 52),
            mask(IMM8, 53),
            mask(IMM8, 54),
            mask(IMM8, 55),
            mask(IMM8, 56),
            mask(IMM8, 57),
            mask(IMM8, 58),
            mask(IMM8, 59),
            mask(IMM8, 60),
            mask(IMM8, 61),
            mask(IMM8, 62),
            mask(IMM8, 63),
        ],
    );
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpsrldq, IMM8 = 3))]
#[rustc_legacy_const_generics(1)]
pub unsafe fn _mm512_bsrli_epi128<const IMM8: i32>(a: __m512i) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let a = a.as_i8x64();
    let zero = i8x64::ZERO;
    let r: i8x64 = match IMM8 % 16 {
        0 => simd_shuffle!(
            a,
            zero,
            [
                0, 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, 40, 41, 42, 43,
                44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63,
            ],
        ),
        1 => simd_shuffle!(
            a,
            zero,
            [
                1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 17, 18, 19, 20, 21, 22, 23,
                24, 25, 26, 27, 28, 29, 30, 31, 80, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44,
                45, 46, 47, 96, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112,
            ],
        ),
        2 => simd_shuffle!(
            a,
            zero,
            [
                2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 18, 19, 20, 21, 22, 23, 24,
                25, 26, 27, 28, 29, 30, 31, 80, 81, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45,
                46, 47, 96, 97, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113,
            ],
        ),
        3 => simd_shuffle!(
            a,
            zero,
            [
                3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 19, 20, 21, 22, 23, 24,
                25, 26, 27, 28, 29, 30, 31, 80, 81, 82, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45,
                46, 47, 96, 97, 98, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113,
                114,
            ],
        ),
        4 => simd_shuffle!(
            a,
            zero,
            [
                4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 20, 21, 22, 23, 24, 25,
                26, 27, 28, 29, 30, 31, 80, 81, 82, 83, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46,
                47, 96, 97, 98, 99, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114,
                115,
            ],
        ),
        5 => simd_shuffle!(
            a,
            zero,
            [
                5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 21, 22, 23, 24, 25, 26,
                27, 28, 29, 30, 31, 80, 81, 82, 83, 84, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47,
                96, 97, 98, 99, 100, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114,
                115, 116,
            ],
        ),
        6 => simd_shuffle!(
            a,
            zero,
            [
                6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 22, 23, 24, 25, 26, 27,
                28, 29, 30, 31, 80, 81, 82, 83, 84, 85, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 96,
                97, 98, 99, 100, 101, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115,
                116, 117,
            ],
        ),
        7 => simd_shuffle!(
            a,
            zero,
            [
                7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 23, 24, 25, 26, 27,
                28, 29, 30, 31, 80, 81, 82, 83, 84, 85, 86, 39, 40, 41, 42, 43, 44, 45, 46, 47, 96,
                97, 98, 99, 100, 101, 102, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115,
                116, 117, 118,
            ],
        ),
        8 => simd_shuffle!(
            a,
            zero,
            [
                8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 24, 25, 26, 27, 28,
                29, 30, 31, 80, 81, 82, 83, 84, 85, 86, 87, 40, 41, 42, 43, 44, 45, 46, 47, 96, 97,
                98, 99, 100, 101, 102, 103, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115,
                116, 117, 118, 119,
            ],
        ),
        9 => simd_shuffle!(
            a,
            zero,
            [
                9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 25, 26, 27, 28, 29,
                30, 31, 80, 81, 82, 83, 84, 85, 86, 87, 88, 41, 42, 43, 44, 45, 46, 47, 96, 97, 98,
                99, 100, 101, 102, 103, 104, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115, 116,
                117, 118, 119, 120,
            ],
        ),
        10 => simd_shuffle!(
            a,
            zero,
            [
                10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 26, 27, 28, 29, 30,
                31, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 42, 43, 44, 45, 46, 47, 96, 97, 98, 99,
                100, 101, 102, 103, 104, 105, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115, 116, 117,
                118, 119, 120, 121,
            ],
        ),
        11 => simd_shuffle!(
            a,
            zero,
            [
                11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 27, 28, 29, 30, 31,
                80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 43, 44, 45, 46, 47, 96, 97, 98, 99,
                100, 101, 102, 103, 104, 105, 106, 59, 60, 61, 62, 63, 112, 113, 114, 115, 116,
                117, 118, 119, 120, 121, 122,
            ],
        ),
        12 => simd_shuffle!(
            a,
            zero,
            [
                12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 28, 29, 30, 31, 80,
                81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 44, 45, 46, 47, 96, 97, 98, 99, 100,
                101, 102, 103, 104, 105, 106, 107, 60, 61, 62, 63, 112, 113, 114, 115, 116, 117,
                118, 119, 120, 121, 122, 123,
            ],
        ),
        13 => simd_shuffle!(
            a,
            zero,
            [
                13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 29, 30, 31, 80, 81,
                82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 45, 46, 47, 96, 97, 98, 99, 100, 101,
                102, 103, 104, 105, 106, 107, 108, 61, 62, 63, 112, 113, 114, 115, 116, 117, 118,
                119, 120, 121, 122, 123, 124,
            ],
        ),
        14 => simd_shuffle!(
            a,
            zero,
            [
                14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 30, 31, 80, 81, 82,
                83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 46, 47, 96, 97, 98, 99, 100, 101, 102,
                103, 104, 105, 106, 107, 108, 109, 62, 63, 112, 113, 114, 115, 116, 117, 118, 119,
                120, 121, 122, 123, 124, 125,
            ],
        ),
        15 => simd_shuffle!(
            a,
            zero,
            [
                15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 31, 80, 81, 82, 83,
                84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 47, 96, 97, 98, 99, 100, 101, 102, 103,
                104, 105, 106, 107, 108, 109, 110, 63, 112, 113, 114, 115, 116, 117, 118, 119, 120,
                121, 122, 123, 124, 125, 126,
            ],
        ),
        _ => zero,
    };
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpalignr, IMM8 = 1))]
#[rustc_legacy_const_generics(2)]
pub unsafe fn _mm512_alignr_epi8<const IMM8: i32>(a: __m512i, b: __m512i) -> __m512i {
    if IMM8 >= 32 {
        return _mm512_setzero_si512();
    }
    let (a, b) = if IMM8 > 16 {
        (_mm512_setzero_si512(), a)
    } else {
        (a, b)
    };
    let a = a.as_i8x64();
    let b = b.as_i8x64();
    if IMM8 == 16 {
        return transmute(a);
    }
    let r: i8x64 = match IMM8 % 16 {
        0 => simd_shuffle!(
            b,
            a,
            [
                0, 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, 40, 41, 42, 43,
                44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63,
            ],
        ),
        1 => simd_shuffle!(
            b,
            a,
            [
                1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 17, 18, 19, 20, 21, 22, 23,
                24, 25, 26, 27, 28, 29, 30, 31, 80, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44,
                45, 46, 47, 96, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112,
            ],
        ),
        2 => simd_shuffle!(
            b,
            a,
            [
                2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 18, 19, 20, 21, 22, 23, 24,
                25, 26, 27, 28, 29, 30, 31, 80, 81, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45,
                46, 47, 96, 97, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113,
            ],
        ),
        3 => simd_shuffle!(
            b,
            a,
            [
                3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 19, 20, 21, 22, 23, 24,
                25, 26, 27, 28, 29, 30, 31, 80, 81, 82, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45,
                46, 47, 96, 97, 98, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113,
                114,
            ],
        ),
        4 => simd_shuffle!(
            b,
            a,
            [
                4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 20, 21, 22, 23, 24, 25,
                26, 27, 28, 29, 30, 31, 80, 81, 82, 83, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46,
                47, 96, 97, 98, 99, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114,
                115,
            ],
        ),
        5 => simd_shuffle!(
            b,
            a,
            [
                5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 21, 22, 23, 24, 25, 26,
                27, 28, 29, 30, 31, 80, 81, 82, 83, 84, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47,
                96, 97, 98, 99, 100, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114,
                115, 116,
            ],
        ),
        6 => simd_shuffle!(
            b,
            a,
            [
                6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 22, 23, 24, 25, 26, 27,
                28, 29, 30, 31, 80, 81, 82, 83, 84, 85, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 96,
                97, 98, 99, 100, 101, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115,
                116, 117,
            ],
        ),
        7 => simd_shuffle!(
            b,
            a,
            [
                7, 8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 23, 24, 25, 26, 27,
                28, 29, 30, 31, 80, 81, 82, 83, 84, 85, 86, 39, 40, 41, 42, 43, 44, 45, 46, 47, 96,
                97, 98, 99, 100, 101, 102, 55, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115,
                116, 117, 118,
            ],
        ),
        8 => simd_shuffle!(
            b,
            a,
            [
                8, 9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 24, 25, 26, 27, 28,
                29, 30, 31, 80, 81, 82, 83, 84, 85, 86, 87, 40, 41, 42, 43, 44, 45, 46, 47, 96, 97,
                98, 99, 100, 101, 102, 103, 56, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115,
                116, 117, 118, 119,
            ],
        ),
        9 => simd_shuffle!(
            b,
            a,
            [
                9, 10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 25, 26, 27, 28, 29,
                30, 31, 80, 81, 82, 83, 84, 85, 86, 87, 88, 41, 42, 43, 44, 45, 46, 47, 96, 97, 98,
                99, 100, 101, 102, 103, 104, 57, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115, 116,
                117, 118, 119, 120,
            ],
        ),
        10 => simd_shuffle!(
            b,
            a,
            [
                10, 11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 26, 27, 28, 29, 30,
                31, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 42, 43, 44, 45, 46, 47, 96, 97, 98, 99,
                100, 101, 102, 103, 104, 105, 58, 59, 60, 61, 62, 63, 112, 113, 114, 115, 116, 117,
                118, 119, 120, 121,
            ],
        ),
        11 => simd_shuffle!(
            b,
            a,
            [
                11, 12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 27, 28, 29, 30, 31,
                80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 43, 44, 45, 46, 47, 96, 97, 98, 99,
                100, 101, 102, 103, 104, 105, 106, 59, 60, 61, 62, 63, 112, 113, 114, 115, 116,
                117, 118, 119, 120, 121, 122,
            ],
        ),
        12 => simd_shuffle!(
            b,
            a,
            [
                12, 13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 28, 29, 30, 31, 80,
                81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 44, 45, 46, 47, 96, 97, 98, 99, 100,
                101, 102, 103, 104, 105, 106, 107, 60, 61, 62, 63, 112, 113, 114, 115, 116, 117,
                118, 119, 120, 121, 122, 123,
            ],
        ),
        13 => simd_shuffle!(
            b,
            a,
            [
                13, 14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 29, 30, 31, 80, 81,
                82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 45, 46, 47, 96, 97, 98, 99, 100, 101,
                102, 103, 104, 105, 106, 107, 108, 61, 62, 63, 112, 113, 114, 115, 116, 117, 118,
                119, 120, 121, 122, 123, 124,
            ],
        ),
        14 => simd_shuffle!(
            b,
            a,
            [
                14, 15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 30, 31, 80, 81, 82,
                83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 46, 47, 96, 97, 98, 99, 100, 101, 102,
                103, 104, 105, 106, 107, 108, 109, 62, 63, 112, 113, 114, 115, 116, 117, 118, 119,
                120, 121, 122, 123, 124, 125,
            ],
        ),
        15 => simd_shuffle!(
            b,
            a,
            [
                15, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 31, 80, 81, 82, 83,
                84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 47, 96, 97, 98, 99, 100, 101, 102, 103,
                104, 105, 106, 107, 108, 109, 110, 63, 112, 113, 114, 115, 116, 117, 118, 119, 120,
                121, 122, 123, 124, 125, 126,
            ],
        ),
        _ => unreachable_unchecked(),
    };
    transmute(r)
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpalignr, IMM8 = 1))]
#[rustc_legacy_const_generics(4)]
pub unsafe fn _mm512_mask_alignr_epi8<const IMM8: i32>(
    src: __m512i,
    k: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm512_alignr_epi8::<IMM8>(a, b);
    transmute(simd_select_bitmask(k, r.as_i8x64(), src.as_i8x64()))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpalignr, IMM8 = 1))]
#[rustc_legacy_const_generics(3)]
pub unsafe fn _mm512_maskz_alignr_epi8<const IMM8: i32>(
    k: __mmask64,
    a: __m512i,
    b: __m512i,
) -> __m512i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm512_alignr_epi8::<IMM8>(a, b);
    transmute(simd_select_bitmask(k, r.as_i8x64(), i8x64::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(4)]
#[cfg_attr(test, assert_instr(vpalignr, IMM8 = 5))]
pub unsafe fn _mm256_mask_alignr_epi8<const IMM8: i32>(
    src: __m256i,
    k: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm256_alignr_epi8::<IMM8>(a, b);
    transmute(simd_select_bitmask(k, r.as_i8x32(), src.as_i8x32()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpalignr, IMM8 = 5))]
pub unsafe fn _mm256_maskz_alignr_epi8<const IMM8: i32>(
    k: __mmask32,
    a: __m256i,
    b: __m256i,
) -> __m256i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm256_alignr_epi8::<IMM8>(a, b);
    transmute(simd_select_bitmask(k, r.as_i8x32(), i8x32::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(4)]
#[cfg_attr(test, assert_instr(vpalignr, IMM8 = 5))]
pub unsafe fn _mm_mask_alignr_epi8<const IMM8: i32>(
    src: __m128i,
    k: __mmask16,
    a: __m128i,
    b: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm_alignr_epi8::<IMM8>(a, b);
    transmute(simd_select_bitmask(k, r.as_i8x16(), src.as_i8x16()))
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[rustc_legacy_const_generics(3)]
#[cfg_attr(test, assert_instr(vpalignr, IMM8 = 5))]
pub unsafe fn _mm_maskz_alignr_epi8<const IMM8: i32>(
    k: __mmask16,
    a: __m128i,
    b: __m128i,
) -> __m128i {
    static_assert_uimm_bits!(IMM8, 8);
    let r = _mm_alignr_epi8::<IMM8>(a, b);
    transmute(simd_select_bitmask(k, r.as_i8x16(), i8x16::ZERO))
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm512_mask_cvtsepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask32, a: __m512i) {
    vpmovswbmem(mem_addr, a.as_i16x32(), k);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm256_mask_cvtsepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask16, a: __m256i) {
    vpmovswbmem256(mem_addr, a.as_i16x16(), k);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovswb))]
pub unsafe fn _mm_mask_cvtsepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask8, a: __m128i) {
    vpmovswbmem128(mem_addr, a.as_i16x8(), k);
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm512_mask_cvtepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask32, a: __m512i) {
    vpmovwbmem(mem_addr, a.as_i16x32(), k);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm256_mask_cvtepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask16, a: __m256i) {
    vpmovwbmem256(mem_addr, a.as_i16x16(), k);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovwb))]
pub unsafe fn _mm_mask_cvtepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask8, a: __m128i) {
    vpmovwbmem128(mem_addr, a.as_i16x8(), k);
}
#[inline]
#[target_feature(enable = "avx512bw")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm512_mask_cvtusepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask32, a: __m512i) {
    vpmovuswbmem(mem_addr, a.as_i16x32(), k);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm256_mask_cvtusepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask16, a: __m256i) {
    vpmovuswbmem256(mem_addr, a.as_i16x16(), k);
}
#[inline]
#[target_feature(enable = "avx512bw,avx512vl")]
#[unstable(feature = "stdarch_x86_avx512", issue = "111137")]
#[cfg_attr(test, assert_instr(vpmovuswb))]
pub unsafe fn _mm_mask_cvtusepi16_storeu_epi8(mem_addr: *mut i8, k: __mmask8, a: __m128i) {
    vpmovuswbmem128(mem_addr, a.as_i16x8(), k);
}
#[allow(improper_ctypes)]
extern "C" {
    #[link_name = "llvm.x86.avx512.pmul.hr.sw.512"]
    fn vpmulhrsw(a: i16x32, b: i16x32) -> i16x32;
    #[link_name = "llvm.x86.avx512.pmaddw.d.512"]
    fn vpmaddwd(a: i16x32, b: i16x32) -> i32x16;
    #[link_name = "llvm.x86.avx512.pmaddubs.w.512"]
    fn vpmaddubsw(a: i8x64, b: i8x64) -> i16x32;
    #[link_name = "llvm.x86.avx512.packssdw.512"]
    fn vpackssdw(a: i32x16, b: i32x16) -> i16x32;
    #[link_name = "llvm.x86.avx512.packsswb.512"]
    fn vpacksswb(a: i16x32, b: i16x32) -> i8x64;
    #[link_name = "llvm.x86.avx512.packusdw.512"]
    fn vpackusdw(a: i32x16, b: i32x16) -> u16x32;
    #[link_name = "llvm.x86.avx512.packuswb.512"]
    fn vpackuswb(a: i16x32, b: i16x32) -> u8x64;
    #[link_name = "llvm.x86.avx512.psll.w.512"]
    fn vpsllw(a: i16x32, count: i16x8) -> i16x32;
    #[link_name = "llvm.x86.avx512.psllv.w.512"]
    fn vpsllvw(a: i16x32, b: i16x32) -> i16x32;
    #[link_name = "llvm.x86.avx512.psllv.w.256"]
    fn vpsllvw256(a: i16x16, b: i16x16) -> i16x16;
    #[link_name = "llvm.x86.avx512.psllv.w.128"]
    fn vpsllvw128(a: i16x8, b: i16x8) -> i16x8;
    #[link_name = "llvm.x86.avx512.psrl.w.512"]
    fn vpsrlw(a: i16x32, count: i16x8) -> i16x32;
    #[link_name = "llvm.x86.avx512.psrlv.w.512"]
    fn vpsrlvw(a: i16x32, b: i16x32) -> i16x32;
    #[link_name = "llvm.x86.avx512.psrlv.w.256"]
    fn vpsrlvw256(a: i16x16, b: i16x16) -> i16x16;
    #[link_name = "llvm.x86.avx512.psrlv.w.128"]
    fn vpsrlvw128(a: i16x8, b: i16x8) -> i16x8;
    #[link_name = "llvm.x86.avx512.psra.w.512"]
    fn vpsraw(a: i16x32, count: i16x8) -> i16x32;
    #[link_name = "llvm.x86.avx512.psrav.w.512"]
    fn vpsravw(a: i16x32, count: i16x32) -> i16x32;
    #[link_name = "llvm.x86.avx512.psrav.w.256"]
    fn vpsravw256(a: i16x16, count: i16x16) -> i16x16;
    #[link_name = "llvm.x86.avx512.psrav.w.128"]
    fn vpsravw128(a: i16x8, count: i16x8) -> i16x8;
    #[link_name = "llvm.x86.avx512.vpermi2var.hi.512"]
    fn vpermi2w(a: i16x32, idx: i16x32, b: i16x32) -> i16x32;
    #[link_name = "llvm.x86.avx512.vpermi2var.hi.256"]
    fn vpermi2w256(a: i16x16, idx: i16x16, b: i16x16) -> i16x16;
    #[link_name = "llvm.x86.avx512.vpermi2var.hi.128"]
    fn vpermi2w128(a: i16x8, idx: i16x8, b: i16x8) -> i16x8;
    #[link_name = "llvm.x86.avx512.permvar.hi.512"]
    fn vpermw(a: i16x32, idx: i16x32) -> i16x32;
    #[link_name = "llvm.x86.avx512.permvar.hi.256"]
    fn vpermw256(a: i16x16, idx: i16x16) -> i16x16;
    #[link_name = "llvm.x86.avx512.permvar.hi.128"]
    fn vpermw128(a: i16x8, idx: i16x8) -> i16x8;
    #[link_name = "llvm.x86.avx512.pshuf.b.512"]
    fn vpshufb(a: i8x64, b: i8x64) -> i8x64;
    #[link_name = "llvm.x86.avx512.psad.bw.512"]
    fn vpsadbw(a: u8x64, b: u8x64) -> u64x8;
    #[link_name = "llvm.x86.avx512.dbpsadbw.512"]
    fn vdbpsadbw(a: u8x64, b: u8x64, imm8: i32) -> u16x32;
    #[link_name = "llvm.x86.avx512.dbpsadbw.256"]
    fn vdbpsadbw256(a: u8x32, b: u8x32, imm8: i32) -> u16x16;
    #[link_name = "llvm.x86.avx512.dbpsadbw.128"]
    fn vdbpsadbw128(a: u8x16, b: u8x16, imm8: i32) -> u16x8;
    #[link_name = "llvm.x86.avx512.mask.pmovs.wb.512"]
    fn vpmovswb(a: i16x32, src: i8x32, mask: u32) -> i8x32;
    #[link_name = "llvm.x86.avx512.mask.pmovs.wb.256"]
    fn vpmovswb256(a: i16x16, src: i8x16, mask: u16) -> i8x16;
    #[link_name = "llvm.x86.avx512.mask.pmovs.wb.128"]
    fn vpmovswb128(a: i16x8, src: i8x16, mask: u8) -> i8x16;
    #[link_name = "llvm.x86.avx512.mask.pmovus.wb.512"]
    fn vpmovuswb(a: u16x32, src: u8x32, mask: u32) -> u8x32;
    #[link_name = "llvm.x86.avx512.mask.pmovus.wb.256"]
    fn vpmovuswb256(a: u16x16, src: u8x16, mask: u16) -> u8x16;
    #[link_name = "llvm.x86.avx512.mask.pmovus.wb.128"]
    fn vpmovuswb128(a: u16x8, src: u8x16, mask: u8) -> u8x16;
    #[link_name = "llvm.x86.avx512.mask.pmovs.wb.mem.512"]
    fn vpmovswbmem(mem_addr: *mut i8, a: i16x32, mask: u32);
    #[link_name = "llvm.x86.avx512.mask.pmovs.wb.mem.256"]
    fn vpmovswbmem256(mem_addr: *mut i8, a: i16x16, mask: u16);
    #[link_name = "llvm.x86.avx512.mask.pmovs.wb.mem.128"]
    fn vpmovswbmem128(mem_addr: *mut i8, a: i16x8, mask: u8);
    #[link_name = "llvm.x86.avx512.mask.pmov.wb.mem.512"]
    fn vpmovwbmem(mem_addr: *mut i8, a: i16x32, mask: u32);
    #[link_name = "llvm.x86.avx512.mask.pmov.wb.mem.256"]
    fn vpmovwbmem256(mem_addr: *mut i8, a: i16x16, mask: u16);
    #[link_name = "llvm.x86.avx512.mask.pmov.wb.mem.128"]
    fn vpmovwbmem128(mem_addr: *mut i8, a: i16x8, mask: u8);
    #[link_name = "llvm.x86.avx512.mask.pmovus.wb.mem.512"]
    fn vpmovuswbmem(mem_addr: *mut i8, a: i16x32, mask: u32);
    #[link_name = "llvm.x86.avx512.mask.pmovus.wb.mem.256"]
    fn vpmovuswbmem256(mem_addr: *mut i8, a: i16x16, mask: u16);
    #[link_name = "llvm.x86.avx512.mask.pmovus.wb.mem.128"]
    fn vpmovuswbmem128(mem_addr: *mut i8, a: i16x8, mask: u8);
    #[link_name = "llvm.x86.avx512.mask.loadu.b.128"]
    fn loaddqu8_128(mem_addr: *const i8, a: i8x16, mask: u16) -> i8x16;
    #[link_name = "llvm.x86.avx512.mask.loadu.w.128"]
    fn loaddqu16_128(mem_addr: *const i16, a: i16x8, mask: u8) -> i16x8;
    #[link_name = "llvm.x86.avx512.mask.loadu.b.256"]
    fn loaddqu8_256(mem_addr: *const i8, a: i8x32, mask: u32) -> i8x32;
    #[link_name = "llvm.x86.avx512.mask.loadu.w.256"]
    fn loaddqu16_256(mem_addr: *const i16, a: i16x16, mask: u16) -> i16x16;
    #[link_name = "llvm.x86.avx512.mask.loadu.b.512"]
    fn loaddqu8_512(mem_addr: *const i8, a: i8x64, mask: u64) -> i8x64;
    #[link_name = "llvm.x86.avx512.mask.loadu.w.512"]
    fn loaddqu16_512(mem_addr: *const i16, a: i16x32, mask: u32) -> i16x32;
    #[link_name = "llvm.x86.avx512.mask.storeu.b.128"]
    fn storedqu8_128(mem_addr: *mut i8, a: i8x16, mask: u16);
    #[link_name = "llvm.x86.avx512.mask.storeu.w.128"]
    fn storedqu16_128(mem_addr: *mut i16, a: i16x8, mask: u8);
    #[link_name = "llvm.x86.avx512.mask.storeu.b.256"]
    fn storedqu8_256(mem_addr: *mut i8, a: i8x32, mask: u32);
    #[link_name = "llvm.x86.avx512.mask.storeu.w.256"]
    fn storedqu16_256(mem_addr: *mut i16, a: i16x16, mask: u16);
    #[link_name = "llvm.x86.avx512.mask.storeu.b.512"]
    fn storedqu8_512(mem_addr: *mut i8, a: i8x64, mask: u64);
    #[link_name = "llvm.x86.avx512.mask.storeu.w.512"]
    fn storedqu16_512(mem_addr: *mut i16, a: i16x32, mask: u32);
}
#[cfg(test)]
mod tests {
    use stdarch_test::simd_test;
    use crate::core_arch::x86::*;
    use crate::hint::black_box;
    use crate::mem::{self};
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_abs_epi16() {
        let a = _mm512_set1_epi16(-1);
        let r = _mm512_abs_epi16(a);
        let e = _mm512_set1_epi16(1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_abs_epi16() {
        let a = _mm512_set1_epi16(-1);
        let r = _mm512_mask_abs_epi16(a, 0, a);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_abs_epi16(a, 0b00000000_11111111_00000000_11111111, a);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(-1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 -1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_abs_epi16() {
        let a = _mm512_set1_epi16(-1);
        let r = _mm512_maskz_abs_epi16(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_abs_epi16(0b00000000_11111111_00000000_11111111, a);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1,
                                  0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_abs_epi16() {
        let a = _mm256_set1_epi16(-1);
        let r = _mm256_mask_abs_epi16(a, 0, a);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_abs_epi16(a, 0b00000000_11111111, a);
        let e = _mm256_set_epi16(-1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_abs_epi16() {
        let a = _mm256_set1_epi16(-1);
        let r = _mm256_maskz_abs_epi16(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_abs_epi16(0b00000000_11111111, a);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_abs_epi16() {
        let a = _mm_set1_epi16(-1);
        let r = _mm_mask_abs_epi16(a, 0, a);
        assert_eq_m128i(r, a);
        let r = _mm_mask_abs_epi16(a, 0b00001111, a);
        let e = _mm_set_epi16(-1, -1, -1, -1, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_abs_epi16() {
        let a = _mm_set1_epi16(-1);
        let r = _mm_maskz_abs_epi16(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_abs_epi16(0b00001111, a);
        let e = _mm_set_epi16(0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_abs_epi8() {
        let a = _mm512_set1_epi8(-1);
        let r = _mm512_abs_epi8(a);
        let e = _mm512_set1_epi8(1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_abs_epi8() {
        let a = _mm512_set1_epi8(-1);
        let r = _mm512_mask_abs_epi8(a, 0, a);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_abs_epi8(
            a,
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(-1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1,
                                -1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1,
                                -1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1,
                                -1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_abs_epi8() {
        let a = _mm512_set1_epi8(-1);
        let r = _mm512_maskz_abs_epi8(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_abs_epi8(
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1,
                                0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1,
                                0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1,
                                0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_abs_epi8() {
        let a = _mm256_set1_epi8(-1);
        let r = _mm256_mask_abs_epi8(a, 0, a);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_abs_epi8(a, 0b00000000_11111111_00000000_11111111, a);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(-1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1,
                                -1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_abs_epi8() {
        let a = _mm256_set1_epi8(-1);
        let r = _mm256_maskz_abs_epi8(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_abs_epi8(0b00000000_11111111_00000000_11111111, a);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1,
                                0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_abs_epi8() {
        let a = _mm_set1_epi8(-1);
        let r = _mm_mask_abs_epi8(a, 0, a);
        assert_eq_m128i(r, a);
        let r = _mm_mask_abs_epi8(a, 0b00000000_11111111, a);
        let e = _mm_set_epi8(-1, -1, -1, -1, -1, -1, -1, -1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_abs_epi8() {
        let a = _mm_set1_epi8(-1);
        let r = _mm_maskz_abs_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_abs_epi8(0b00000000_11111111, a);
        #[rustfmt::skip]
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_add_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(2);
        let r = _mm512_add_epi16(a, b);
        let e = _mm512_set1_epi16(3);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_add_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(2);
        let r = _mm512_mask_add_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_add_epi16(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3,
                                 1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_add_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(2);
        let r = _mm512_maskz_add_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_add_epi16(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3,
                                 0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_add_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(2);
        let r = _mm256_mask_add_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_add_epi16(a, 0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_add_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(2);
        let r = _mm256_maskz_add_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_add_epi16(0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_add_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(2);
        let r = _mm_mask_add_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_add_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 3, 3, 3, 3);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_add_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(2);
        let r = _mm_maskz_add_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_add_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 3, 3, 3, 3);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_add_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(2);
        let r = _mm512_add_epi8(a, b);
        let e = _mm512_set1_epi8(3);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_add_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(2);
        let r = _mm512_mask_add_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_add_epi8(
            a,
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3,
                                1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3,
                                1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3,
                                1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_add_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(2);
        let r = _mm512_maskz_add_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_add_epi8(
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3,
                                0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3,
                                0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3,
                                0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_add_epi8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(2);
        let r = _mm256_mask_add_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_add_epi8(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3,
                                1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_add_epi8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(2);
        let r = _mm256_maskz_add_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_add_epi8(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3,
                                0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_add_epi8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(2);
        let r = _mm_mask_add_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_add_epi8(a, 0b00000000_11111111, a, b);
        let e = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_add_epi8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(2);
        let r = _mm_maskz_add_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_add_epi8(0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 3, 3, 3, 3);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_adds_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(u16::MAX as i16);
        let r = _mm512_adds_epu16(a, b);
        let e = _mm512_set1_epi16(u16::MAX as i16);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_adds_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(u16::MAX as i16);
        let r = _mm512_mask_adds_epu16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_adds_epu16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_adds_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(u16::MAX as i16);
        let r = _mm512_maskz_adds_epu16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_adds_epu16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_adds_epu16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(u16::MAX as i16);
        let r = _mm256_mask_adds_epu16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_adds_epu16(a, 0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_adds_epu16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(u16::MAX as i16);
        let r = _mm256_maskz_adds_epu16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_adds_epu16(0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_adds_epu16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(u16::MAX as i16);
        let r = _mm_mask_adds_epu16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_adds_epu16(a, 0b00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi16(1, 1, 1, 1, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_adds_epu16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(u16::MAX as i16);
        let r = _mm_maskz_adds_epu16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_adds_epu16(0b00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi16(0, 0, 0, 0, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16, u16::MAX as i16);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_adds_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(u8::MAX as i8);
        let r = _mm512_adds_epu8(a, b);
        let e = _mm512_set1_epi8(u8::MAX as i8);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_adds_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(u8::MAX as i8);
        let r = _mm512_mask_adds_epu8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_adds_epu8(
            a,
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_adds_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(u8::MAX as i8);
        let r = _mm512_maskz_adds_epu8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_adds_epu8(
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_adds_epu8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(u8::MAX as i8);
        let r = _mm256_mask_adds_epu8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_adds_epu8(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_adds_epu8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(u8::MAX as i8);
        let r = _mm256_maskz_adds_epu8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_adds_epu8(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_adds_epu8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(u8::MAX as i8);
        let r = _mm_mask_adds_epu8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_adds_epu8(a, 0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_adds_epu8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(u8::MAX as i8);
        let r = _mm_maskz_adds_epu8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_adds_epu8(0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_adds_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_adds_epi16(a, b);
        let e = _mm512_set1_epi16(i16::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_adds_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_mask_adds_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_adds_epi16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_adds_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_maskz_adds_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_adds_epi16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_adds_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(i16::MAX);
        let r = _mm256_mask_adds_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_adds_epi16(a, 0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_adds_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(i16::MAX);
        let r = _mm256_maskz_adds_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_adds_epi16(0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_adds_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(i16::MAX);
        let r = _mm_mask_adds_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_adds_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_adds_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(i16::MAX);
        let r = _mm_maskz_adds_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_adds_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_adds_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(i8::MAX);
        let r = _mm512_adds_epi8(a, b);
        let e = _mm512_set1_epi8(i8::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_adds_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(i8::MAX);
        let r = _mm512_mask_adds_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_adds_epi8(
            a,
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_adds_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(i8::MAX);
        let r = _mm512_maskz_adds_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_adds_epi8(
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_adds_epi8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(i8::MAX);
        let r = _mm256_mask_adds_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_adds_epi8(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_adds_epi8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(i8::MAX);
        let r = _mm256_maskz_adds_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_adds_epi8(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_adds_epi8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(i8::MAX);
        let r = _mm_mask_adds_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_adds_epi8(a, 0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_adds_epi8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(i8::MAX);
        let r = _mm_maskz_adds_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_adds_epi8(0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_sub_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(2);
        let r = _mm512_sub_epi16(a, b);
        let e = _mm512_set1_epi16(-1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_sub_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(2);
        let r = _mm512_mask_sub_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_sub_epi16(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1,
                                 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_sub_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(2);
        let r = _mm512_maskz_sub_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_sub_epi16(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1,
                                 0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_sub_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(2);
        let r = _mm256_mask_sub_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_sub_epi16(a, 0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_sub_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(2);
        let r = _mm256_maskz_sub_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_sub_epi16(0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_sub_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(2);
        let r = _mm_mask_sub_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_sub_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, -1, -1, -1, -1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_sub_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(2);
        let r = _mm_maskz_sub_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_sub_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, -1, -1, -1, -1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_sub_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(2);
        let r = _mm512_sub_epi8(a, b);
        let e = _mm512_set1_epi8(-1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_sub_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(2);
        let r = _mm512_mask_sub_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_sub_epi8(
            a,
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1,
                                1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1,
                                1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1,
                                1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_sub_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(2);
        let r = _mm512_maskz_sub_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_sub_epi8(
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1,
                                0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1,
                                0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1,
                                0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_sub_epi8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(2);
        let r = _mm256_mask_sub_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_sub_epi8(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1,
                                1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_sub_epi8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(2);
        let r = _mm256_maskz_sub_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_sub_epi8(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1,
                                0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_sub_epi8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(2);
        let r = _mm_mask_sub_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_sub_epi8(a, 0b00000000_11111111, a, b);
        let e = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_sub_epi8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(2);
        let r = _mm_maskz_sub_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_sub_epi8(0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_subs_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(u16::MAX as i16);
        let r = _mm512_subs_epu16(a, b);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_subs_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(u16::MAX as i16);
        let r = _mm512_mask_subs_epu16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_subs_epu16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_subs_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(u16::MAX as i16);
        let r = _mm512_maskz_subs_epu16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_subs_epu16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_subs_epu16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(u16::MAX as i16);
        let r = _mm256_mask_subs_epu16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_subs_epu16(a, 0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_subs_epu16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(u16::MAX as i16);
        let r = _mm256_maskz_subs_epu16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_subs_epu16(0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_subs_epu16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(u16::MAX as i16);
        let r = _mm_mask_subs_epu16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_subs_epu16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_subs_epu16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(u16::MAX as i16);
        let r = _mm_maskz_subs_epu16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_subs_epu16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_subs_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(u8::MAX as i8);
        let r = _mm512_subs_epu8(a, b);
        let e = _mm512_set1_epi8(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_subs_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(u8::MAX as i8);
        let r = _mm512_mask_subs_epu8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_subs_epu8(
            a,
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_subs_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(u8::MAX as i8);
        let r = _mm512_maskz_subs_epu8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_subs_epu8(
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_subs_epu8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(u8::MAX as i8);
        let r = _mm256_mask_subs_epu8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_subs_epu8(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_subs_epu8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(u8::MAX as i8);
        let r = _mm256_maskz_subs_epu8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_subs_epu8(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_subs_epu8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(u8::MAX as i8);
        let r = _mm_mask_subs_epu8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_subs_epu8(a, 0b00000000_00001111, a, b);
        let e = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_subs_epu8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(u8::MAX as i8);
        let r = _mm_maskz_subs_epu8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_subs_epu8(0b00000000_00001111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_subs_epi16() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_subs_epi16(a, b);
        let e = _mm512_set1_epi16(i16::MIN);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_subs_epi16() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_mask_subs_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_subs_epi16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
                                 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, i16::MIN, i16::MIN, i16::MIN, i16::MIN);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_subs_epi16() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_maskz_subs_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_subs_epi16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i16::MIN, i16::MIN, i16::MIN, i16::MIN);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_subs_epi16() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(i16::MAX);
        let r = _mm256_mask_subs_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_subs_epi16(a, 0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, i16::MIN, i16::MIN, i16::MIN, i16::MIN);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_subs_epi16() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(i16::MAX);
        let r = _mm256_maskz_subs_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_subs_epi16(0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i16::MIN, i16::MIN, i16::MIN, i16::MIN);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_subs_epi16() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(i16::MAX);
        let r = _mm_mask_subs_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_subs_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(-1, -1, -1, -1, i16::MIN, i16::MIN, i16::MIN, i16::MIN);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_subs_epi16() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(i16::MAX);
        let r = _mm_maskz_subs_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_subs_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, i16::MIN, i16::MIN, i16::MIN, i16::MIN);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_subs_epi8() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(i8::MAX);
        let r = _mm512_subs_epi8(a, b);
        let e = _mm512_set1_epi8(i8::MIN);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_subs_epi8() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(i8::MAX);
        let r = _mm512_mask_subs_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_subs_epi8(
            a,
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
                                -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
                                -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
                                -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, i8::MIN, i8::MIN, i8::MIN, i8::MIN);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_subs_epi8() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(i8::MAX);
        let r = _mm512_maskz_subs_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_subs_epi8(
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MIN, i8::MIN, i8::MIN, i8::MIN);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_subs_epi8() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(i8::MAX);
        let r = _mm256_mask_subs_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_subs_epi8(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
                                -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, i8::MIN, i8::MIN, i8::MIN, i8::MIN);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_subs_epi8() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(i8::MAX);
        let r = _mm256_maskz_subs_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_subs_epi8(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MIN, i8::MIN, i8::MIN, i8::MIN);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_subs_epi8() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(i8::MAX);
        let r = _mm_mask_subs_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_subs_epi8(a, 0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi8(-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, i8::MIN, i8::MIN, i8::MIN, i8::MIN);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_subs_epi8() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(i8::MAX);
        let r = _mm_maskz_subs_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_subs_epi8(0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MIN, i8::MIN, i8::MIN, i8::MIN);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mulhi_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mulhi_epu16(a, b);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_mulhi_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mask_mulhi_epu16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_mulhi_epu16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_mulhi_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_maskz_mulhi_epu16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_mulhi_epu16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_mulhi_epu16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_mask_mulhi_epu16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_mulhi_epu16(a, 0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_mulhi_epu16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_maskz_mulhi_epu16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_mulhi_epu16(0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_mulhi_epu16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_mask_mulhi_epu16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_mulhi_epu16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_mulhi_epu16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_maskz_mulhi_epu16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_mulhi_epu16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mulhi_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mulhi_epi16(a, b);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_mulhi_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mask_mulhi_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_mulhi_epi16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_mulhi_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_maskz_mulhi_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_mulhi_epi16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_mulhi_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_mask_mulhi_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_mulhi_epi16(a, 0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_mulhi_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_maskz_mulhi_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_mulhi_epi16(0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_mulhi_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_mask_mulhi_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_mulhi_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_mulhi_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_maskz_mulhi_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_mulhi_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mulhrs_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mulhrs_epi16(a, b);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_mulhrs_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mask_mulhrs_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_mulhrs_epi16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_mulhrs_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_maskz_mulhrs_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_mulhrs_epi16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_mulhrs_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_mask_mulhrs_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_mulhrs_epi16(a, 0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_mulhrs_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_maskz_mulhrs_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_mulhrs_epi16(0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_mulhrs_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_mask_mulhrs_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_mulhrs_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_mulhrs_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_maskz_mulhrs_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_mulhrs_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mullo_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mullo_epi16(a, b);
        let e = _mm512_set1_epi16(1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_mullo_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mask_mullo_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_mullo_epi16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_mullo_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_maskz_mullo_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_mullo_epi16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_mullo_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_mask_mullo_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_mullo_epi16(a, 0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_mullo_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_maskz_mullo_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_mullo_epi16(0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_mullo_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_mask_mullo_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_mullo_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_mullo_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_maskz_mullo_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_mullo_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_max_epu16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_max_epu16(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15,
                                 15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_mask_max_epu16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_mask_max_epu16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_max_epu16(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_maskz_max_epu16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_maskz_max_epu16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_max_epu16(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_mask_max_epu16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm256_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_mask_max_epu16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_max_epu16(a, 0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_maskz_max_epu16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm256_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_maskz_max_epu16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_max_epu16(0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_mask_max_epu16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let b = _mm_set_epi16(7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_mask_max_epu16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_max_epu16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_maskz_max_epu16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let b = _mm_set_epi16(7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_maskz_max_epu16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_max_epu16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 4, 5, 6, 7);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_max_epu8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_max_epu8(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15,
                                15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15,
                                15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15,
                                15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_mask_max_epu8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_mask_max_epu8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_max_epu8(
            a,
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_maskz_max_epu8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_maskz_max_epu8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_max_epu8(
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_mask_max_epu8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_mask_max_epu8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_max_epu8(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_maskz_max_epu8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_maskz_max_epu8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_max_epu8(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_mask_max_epu8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_mask_max_epu8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_max_epu8(a, 0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_maskz_max_epu8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_maskz_max_epu8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_max_epu8(0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_max_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_max_epi16(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15,
                                 15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_mask_max_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_mask_max_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_max_epi16(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_maskz_max_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_maskz_max_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_max_epi16(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_mask_max_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm256_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_mask_max_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_max_epi16(a, 0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_maskz_max_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm256_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_maskz_max_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_max_epi16(0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_mask_max_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let b = _mm_set_epi16(7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_mask_max_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_max_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_maskz_max_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let b = _mm_set_epi16(7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_maskz_max_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_max_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 4, 5, 6, 7);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_max_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_max_epi8(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15,
                                15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15,
                                15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15,
                                15, 14, 13, 12, 11, 10, 9, 8, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_mask_max_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_mask_max_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_max_epi8(
            a,
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_maskz_max_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_maskz_max_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_max_epi8(
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_mask_max_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_mask_max_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_max_epi8(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_maskz_max_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_maskz_max_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_max_epi8(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_mask_max_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_mask_max_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_max_epi8(a, 0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_maskz_max_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_maskz_max_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_max_epi8(0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 8, 9, 10, 11, 12, 13, 14, 15);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_min_epu16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_min_epu16(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                 0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_mask_min_epu16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_mask_min_epu16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_min_epu16(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                 0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_maskz_min_epu16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_maskz_min_epu16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_min_epu16(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_mask_min_epu16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm256_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_mask_min_epu16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_min_epu16(a, 0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_maskz_min_epu16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm256_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_maskz_min_epu16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_min_epu16(0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_mask_min_epu16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let b = _mm_set_epi16(7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_mask_min_epu16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_min_epu16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(0, 1, 2, 3, 3, 2, 1, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_maskz_min_epu16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let b = _mm_set_epi16(7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_maskz_min_epu16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_min_epu16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 3, 2, 1, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_min_epu8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_min_epu8(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_mask_min_epu8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_mask_min_epu8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_min_epu8(
            a,
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_maskz_min_epu8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_maskz_min_epu8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_min_epu8(
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_mask_min_epu8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_mask_min_epu8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_min_epu8(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_maskz_min_epu8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_maskz_min_epu8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_min_epu8(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_mask_min_epu8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_mask_min_epu8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_min_epu8(a, 0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_maskz_min_epu8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_maskz_min_epu8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_min_epu8(0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_min_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_min_epi16(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                 0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_mask_min_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_mask_min_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_min_epi16(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                 0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_maskz_min_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_maskz_min_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_min_epi16(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_mask_min_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm256_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_mask_min_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_min_epi16(a, 0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_maskz_min_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm256_set_epi16(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_maskz_min_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_min_epi16(0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_mask_min_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let b = _mm_set_epi16(7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_mask_min_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_min_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(0, 1, 2, 3, 3, 2, 1, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_maskz_min_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let b = _mm_set_epi16(7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_maskz_min_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_min_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 3, 2, 1, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_min_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_min_epi8(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_mask_min_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_mask_min_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_min_epi8(
            a,
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f")]
    unsafe fn test_mm512_maskz_min_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm512_maskz_min_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_min_epi8(
            0b00000000_11111111_00000000_11111111_00000000_11111111_00000000_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_mask_min_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_mask_min_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_min_epi8(a, 0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm256_maskz_min_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
                                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0,
                                15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm256_maskz_min_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_min_epi8(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_mask_min_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_mask_min_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_min_epi8(a, 0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512vl")]
    unsafe fn test_mm_maskz_min_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set_epi8(15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
        let r = _mm_maskz_min_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_min_epi8(0b00000000_11111111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 7, 6, 5, 4, 3, 2, 1, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmplt_epu16_mask() {
        let a = _mm512_set1_epi16(-2);
        let b = _mm512_set1_epi16(-1);
        let m = _mm512_cmplt_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmplt_epu16_mask() {
        let a = _mm512_set1_epi16(-2);
        let b = _mm512_set1_epi16(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmplt_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmplt_epu16_mask() {
        let a = _mm256_set1_epi16(-2);
        let b = _mm256_set1_epi16(-1);
        let m = _mm256_cmplt_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmplt_epu16_mask() {
        let a = _mm256_set1_epi16(-2);
        let b = _mm256_set1_epi16(-1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmplt_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmplt_epu16_mask() {
        let a = _mm_set1_epi16(-2);
        let b = _mm_set1_epi16(-1);
        let m = _mm_cmplt_epu16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmplt_epu16_mask() {
        let a = _mm_set1_epi16(-2);
        let b = _mm_set1_epi16(-1);
        let mask = 0b01010101;
        let r = _mm_mask_cmplt_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmplt_epu8_mask() {
        let a = _mm512_set1_epi8(-2);
        let b = _mm512_set1_epi8(-1);
        let m = _mm512_cmplt_epu8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmplt_epu8_mask() {
        let a = _mm512_set1_epi8(-2);
        let b = _mm512_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmplt_epu8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmplt_epu8_mask() {
        let a = _mm256_set1_epi8(-2);
        let b = _mm256_set1_epi8(-1);
        let m = _mm256_cmplt_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmplt_epu8_mask() {
        let a = _mm256_set1_epi8(-2);
        let b = _mm256_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmplt_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmplt_epu8_mask() {
        let a = _mm_set1_epi8(-2);
        let b = _mm_set1_epi8(-1);
        let m = _mm_cmplt_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmplt_epu8_mask() {
        let a = _mm_set1_epi8(-2);
        let b = _mm_set1_epi8(-1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmplt_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmplt_epi16_mask() {
        let a = _mm512_set1_epi16(-2);
        let b = _mm512_set1_epi16(-1);
        let m = _mm512_cmplt_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmplt_epi16_mask() {
        let a = _mm512_set1_epi16(-2);
        let b = _mm512_set1_epi16(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmplt_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmplt_epi16_mask() {
        let a = _mm256_set1_epi16(-2);
        let b = _mm256_set1_epi16(-1);
        let m = _mm256_cmplt_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmplt_epi16_mask() {
        let a = _mm256_set1_epi16(-2);
        let b = _mm256_set1_epi16(-1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmplt_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmplt_epi16_mask() {
        let a = _mm_set1_epi16(-2);
        let b = _mm_set1_epi16(-1);
        let m = _mm_cmplt_epi16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmplt_epi16_mask() {
        let a = _mm_set1_epi16(-2);
        let b = _mm_set1_epi16(-1);
        let mask = 0b01010101;
        let r = _mm_mask_cmplt_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmplt_epi8_mask() {
        let a = _mm512_set1_epi8(-2);
        let b = _mm512_set1_epi8(-1);
        let m = _mm512_cmplt_epi8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmplt_epi8_mask() {
        let a = _mm512_set1_epi8(-2);
        let b = _mm512_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmplt_epi8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmplt_epi8_mask() {
        let a = _mm256_set1_epi8(-2);
        let b = _mm256_set1_epi8(-1);
        let m = _mm256_cmplt_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmplt_epi8_mask() {
        let a = _mm256_set1_epi8(-2);
        let b = _mm256_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmplt_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmplt_epi8_mask() {
        let a = _mm_set1_epi8(-2);
        let b = _mm_set1_epi8(-1);
        let m = _mm_cmplt_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmplt_epi8_mask() {
        let a = _mm_set1_epi8(-2);
        let b = _mm_set1_epi8(-1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmplt_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpgt_epu16_mask() {
        let a = _mm512_set1_epi16(2);
        let b = _mm512_set1_epi16(1);
        let m = _mm512_cmpgt_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpgt_epu16_mask() {
        let a = _mm512_set1_epi16(2);
        let b = _mm512_set1_epi16(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpgt_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpgt_epu16_mask() {
        let a = _mm256_set1_epi16(2);
        let b = _mm256_set1_epi16(1);
        let m = _mm256_cmpgt_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpgt_epu16_mask() {
        let a = _mm256_set1_epi16(2);
        let b = _mm256_set1_epi16(1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmpgt_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpgt_epu16_mask() {
        let a = _mm_set1_epi16(2);
        let b = _mm_set1_epi16(1);
        let m = _mm_cmpgt_epu16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpgt_epu16_mask() {
        let a = _mm_set1_epi16(2);
        let b = _mm_set1_epi16(1);
        let mask = 0b01010101;
        let r = _mm_mask_cmpgt_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpgt_epu8_mask() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(1);
        let m = _mm512_cmpgt_epu8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpgt_epu8_mask() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpgt_epu8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpgt_epu8_mask() {
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(1);
        let m = _mm256_cmpgt_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpgt_epu8_mask() {
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmpgt_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpgt_epu8_mask() {
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(1);
        let m = _mm_cmpgt_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpgt_epu8_mask() {
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmpgt_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpgt_epi16_mask() {
        let a = _mm512_set1_epi16(2);
        let b = _mm512_set1_epi16(-1);
        let m = _mm512_cmpgt_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpgt_epi16_mask() {
        let a = _mm512_set1_epi16(2);
        let b = _mm512_set1_epi16(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpgt_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpgt_epi16_mask() {
        let a = _mm256_set1_epi16(2);
        let b = _mm256_set1_epi16(-1);
        let m = _mm256_cmpgt_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpgt_epi16_mask() {
        let a = _mm256_set1_epi16(2);
        let b = _mm256_set1_epi16(-1);
        let mask = 0b001010101_01010101;
        let r = _mm256_mask_cmpgt_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpgt_epi16_mask() {
        let a = _mm_set1_epi16(2);
        let b = _mm_set1_epi16(-1);
        let m = _mm_cmpgt_epi16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpgt_epi16_mask() {
        let a = _mm_set1_epi16(2);
        let b = _mm_set1_epi16(-1);
        let mask = 0b01010101;
        let r = _mm_mask_cmpgt_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpgt_epi8_mask() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(-1);
        let m = _mm512_cmpgt_epi8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpgt_epi8_mask() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpgt_epi8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpgt_epi8_mask() {
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(-1);
        let m = _mm256_cmpgt_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpgt_epi8_mask() {
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmpgt_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpgt_epi8_mask() {
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(-1);
        let m = _mm_cmpgt_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpgt_epi8_mask() {
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(-1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmpgt_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmple_epu16_mask() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(-1);
        let m = _mm512_cmple_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmple_epu16_mask() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmple_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmple_epu16_mask() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(-1);
        let m = _mm256_cmple_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmple_epu16_mask() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(-1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmple_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmple_epu16_mask() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(-1);
        let m = _mm_cmple_epu16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmple_epu16_mask() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(-1);
        let mask = 0b01010101;
        let r = _mm_mask_cmple_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmple_epu8_mask() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(-1);
        let m = _mm512_cmple_epu8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmple_epu8_mask() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmple_epu8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmple_epu8_mask() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(-1);
        let m = _mm256_cmple_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmple_epu8_mask() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmple_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmple_epu8_mask() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(-1);
        let m = _mm_cmple_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmple_epu8_mask() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(-1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmple_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmple_epi16_mask() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(-1);
        let m = _mm512_cmple_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmple_epi16_mask() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmple_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmple_epi16_mask() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(-1);
        let m = _mm256_cmple_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmple_epi16_mask() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(-1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmple_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmple_epi16_mask() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(-1);
        let m = _mm_cmple_epi16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmple_epi16_mask() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(-1);
        let mask = 0b01010101;
        let r = _mm_mask_cmple_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmple_epi8_mask() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(-1);
        let m = _mm512_cmple_epi8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmple_epi8_mask() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmple_epi8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmple_epi8_mask() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(-1);
        let m = _mm256_cmple_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmple_epi8_mask() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmple_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmple_epi8_mask() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(-1);
        let m = _mm_cmple_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmple_epi8_mask() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(-1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmple_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpge_epu16_mask() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let m = _mm512_cmpge_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpge_epu16_mask() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpge_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpge_epu16_mask() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let m = _mm256_cmpge_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpge_epu16_mask() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmpge_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpge_epu16_mask() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let m = _mm_cmpge_epu16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpge_epu16_mask() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let mask = 0b01010101;
        let r = _mm_mask_cmpge_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpge_epu8_mask() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let m = _mm512_cmpge_epu8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpge_epu8_mask() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpge_epu8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpge_epu8_mask() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(1);
        let m = _mm256_cmpge_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpge_epu8_mask() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmpge_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpge_epu8_mask() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(1);
        let m = _mm_cmpge_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpge_epu8_mask() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmpge_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpge_epi16_mask() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(-1);
        let m = _mm512_cmpge_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpge_epi16_mask() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpge_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpge_epi16_mask() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(-1);
        let m = _mm256_cmpge_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpge_epi16_mask() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(-1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmpge_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpge_epi16_mask() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(-1);
        let m = _mm_cmpge_epi16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpge_epi16_mask() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(-1);
        let mask = 0b01010101;
        let r = _mm_mask_cmpge_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpge_epi8_mask() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(-1);
        let m = _mm512_cmpge_epi8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpge_epi8_mask() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpge_epi8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpge_epi8_mask() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(-1);
        let m = _mm256_cmpge_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpge_epi8_mask() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmpge_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpge_epi8_mask() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(-1);
        let m = _mm_cmpge_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpge_epi8_mask() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(-1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmpge_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpeq_epu16_mask() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let m = _mm512_cmpeq_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpeq_epu16_mask() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpeq_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpeq_epu16_mask() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let m = _mm256_cmpeq_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpeq_epu16_mask() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmpeq_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpeq_epu16_mask() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let m = _mm_cmpeq_epu16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpeq_epu16_mask() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let mask = 0b01010101;
        let r = _mm_mask_cmpeq_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpeq_epu8_mask() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let m = _mm512_cmpeq_epu8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpeq_epu8_mask() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpeq_epu8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpeq_epu8_mask() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(1);
        let m = _mm256_cmpeq_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpeq_epu8_mask() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmpeq_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpeq_epu8_mask() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(1);
        let m = _mm_cmpeq_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpeq_epu8_mask() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmpeq_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpeq_epi16_mask() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(-1);
        let m = _mm512_cmpeq_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpeq_epi16_mask() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpeq_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpeq_epi16_mask() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(-1);
        let m = _mm256_cmpeq_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpeq_epi16_mask() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(-1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmpeq_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpeq_epi16_mask() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(-1);
        let m = _mm_cmpeq_epi16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpeq_epi16_mask() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(-1);
        let mask = 0b01010101;
        let r = _mm_mask_cmpeq_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpeq_epi8_mask() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(-1);
        let m = _mm512_cmpeq_epi8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpeq_epi8_mask() {
        let a = _mm512_set1_epi8(-1);
        let b = _mm512_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpeq_epi8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpeq_epi8_mask() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(-1);
        let m = _mm256_cmpeq_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpeq_epi8_mask() {
        let a = _mm256_set1_epi8(-1);
        let b = _mm256_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmpeq_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpeq_epi8_mask() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(-1);
        let m = _mm_cmpeq_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpeq_epi8_mask() {
        let a = _mm_set1_epi8(-1);
        let b = _mm_set1_epi8(-1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmpeq_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpneq_epu16_mask() {
        let a = _mm512_set1_epi16(2);
        let b = _mm512_set1_epi16(1);
        let m = _mm512_cmpneq_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpneq_epu16_mask() {
        let a = _mm512_set1_epi16(2);
        let b = _mm512_set1_epi16(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpneq_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpneq_epu16_mask() {
        let a = _mm256_set1_epi16(2);
        let b = _mm256_set1_epi16(1);
        let m = _mm256_cmpneq_epu16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpneq_epu16_mask() {
        let a = _mm256_set1_epi16(2);
        let b = _mm256_set1_epi16(1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmpneq_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpneq_epu16_mask() {
        let a = _mm_set1_epi16(2);
        let b = _mm_set1_epi16(1);
        let m = _mm_cmpneq_epu16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpneq_epu16_mask() {
        let a = _mm_set1_epi16(2);
        let b = _mm_set1_epi16(1);
        let mask = 0b01010101;
        let r = _mm_mask_cmpneq_epu16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpneq_epu8_mask() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(1);
        let m = _mm512_cmpneq_epu8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpneq_epu8_mask() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpneq_epu8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpneq_epu8_mask() {
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(1);
        let m = _mm256_cmpneq_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpneq_epu8_mask() {
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmpneq_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpneq_epu8_mask() {
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(1);
        let m = _mm_cmpneq_epu8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpneq_epu8_mask() {
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmpneq_epu8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpneq_epi16_mask() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(-1);
        let m = _mm512_cmpneq_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpneq_epi16_mask() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpneq_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpneq_epi16_mask() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(-1);
        let m = _mm256_cmpneq_epi16_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpneq_epi16_mask() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(-1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmpneq_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpneq_epi16_mask() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(-1);
        let m = _mm_cmpneq_epi16_mask(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpneq_epi16_mask() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(-1);
        let mask = 0b01010101;
        let r = _mm_mask_cmpneq_epi16_mask(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmpneq_epi8_mask() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(-1);
        let m = _mm512_cmpneq_epi8_mask(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmpneq_epi8_mask() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmpneq_epi8_mask(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmpneq_epi8_mask() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(-1);
        let m = _mm256_cmpneq_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmpneq_epi8_mask() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(-1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmpneq_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmpneq_epi8_mask() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(-1);
        let m = _mm_cmpneq_epi8_mask(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmpneq_epi8_mask() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(-1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmpneq_epi8_mask(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmp_epu16_mask() {
        let a = _mm512_set1_epi16(0);
        let b = _mm512_set1_epi16(1);
        let m = _mm512_cmp_epu16_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmp_epu16_mask() {
        let a = _mm512_set1_epi16(0);
        let b = _mm512_set1_epi16(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmp_epu16_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmp_epu16_mask() {
        let a = _mm256_set1_epi16(0);
        let b = _mm256_set1_epi16(1);
        let m = _mm256_cmp_epu16_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmp_epu16_mask() {
        let a = _mm256_set1_epi16(0);
        let b = _mm256_set1_epi16(1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmp_epu16_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmp_epu16_mask() {
        let a = _mm_set1_epi16(0);
        let b = _mm_set1_epi16(1);
        let m = _mm_cmp_epu16_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmp_epu16_mask() {
        let a = _mm_set1_epi16(0);
        let b = _mm_set1_epi16(1);
        let mask = 0b01010101;
        let r = _mm_mask_cmp_epu16_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmp_epu8_mask() {
        let a = _mm512_set1_epi8(0);
        let b = _mm512_set1_epi8(1);
        let m = _mm512_cmp_epu8_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmp_epu8_mask() {
        let a = _mm512_set1_epi8(0);
        let b = _mm512_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmp_epu8_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmp_epu8_mask() {
        let a = _mm256_set1_epi8(0);
        let b = _mm256_set1_epi8(1);
        let m = _mm256_cmp_epu8_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmp_epu8_mask() {
        let a = _mm256_set1_epi8(0);
        let b = _mm256_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmp_epu8_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmp_epu8_mask() {
        let a = _mm_set1_epi8(0);
        let b = _mm_set1_epi8(1);
        let m = _mm_cmp_epu8_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmp_epu8_mask() {
        let a = _mm_set1_epi8(0);
        let b = _mm_set1_epi8(1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmp_epu8_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmp_epi16_mask() {
        let a = _mm512_set1_epi16(0);
        let b = _mm512_set1_epi16(1);
        let m = _mm512_cmp_epi16_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmp_epi16_mask() {
        let a = _mm512_set1_epi16(0);
        let b = _mm512_set1_epi16(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmp_epi16_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmp_epi16_mask() {
        let a = _mm256_set1_epi16(0);
        let b = _mm256_set1_epi16(1);
        let m = _mm256_cmp_epi16_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmp_epi16_mask() {
        let a = _mm256_set1_epi16(0);
        let b = _mm256_set1_epi16(1);
        let mask = 0b01010101_01010101;
        let r = _mm256_mask_cmp_epi16_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmp_epi16_mask() {
        let a = _mm_set1_epi16(0);
        let b = _mm_set1_epi16(1);
        let m = _mm_cmp_epi16_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmp_epi16_mask() {
        let a = _mm_set1_epi16(0);
        let b = _mm_set1_epi16(1);
        let mask = 0b01010101;
        let r = _mm_mask_cmp_epi16_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cmp_epi8_mask() {
        let a = _mm512_set1_epi8(0);
        let b = _mm512_set1_epi8(1);
        let m = _mm512_cmp_epi8_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(
            m,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111
        );
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cmp_epi8_mask() {
        let a = _mm512_set1_epi8(0);
        let b = _mm512_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101;
        let r = _mm512_mask_cmp_epi8_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(
            r,
            0b01010101_01010101_01010101_01010101_01010101_01010101_01010101_01010101
        );
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cmp_epi8_mask() {
        let a = _mm256_set1_epi8(0);
        let b = _mm256_set1_epi8(1);
        let m = _mm256_cmp_epi8_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111_11111111_11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cmp_epi8_mask() {
        let a = _mm256_set1_epi8(0);
        let b = _mm256_set1_epi8(1);
        let mask = 0b01010101_01010101_01010101_01010101;
        let r = _mm256_mask_cmp_epi8_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101_01010101_01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cmp_epi8_mask() {
        let a = _mm_set1_epi8(0);
        let b = _mm_set1_epi8(1);
        let m = _mm_cmp_epi8_mask::<_MM_CMPINT_LT>(a, b);
        assert_eq!(m, 0b11111111_11111111);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cmp_epi8_mask() {
        let a = _mm_set1_epi8(0);
        let b = _mm_set1_epi8(1);
        let mask = 0b01010101_01010101;
        let r = _mm_mask_cmp_epi8_mask::<_MM_CMPINT_LT>(mask, a, b);
        assert_eq!(r, 0b01010101_01010101);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_add_epi16() {
        let a = _mm256_set1_epi16(1);
        let e = _mm256_reduce_add_epi16(a);
        assert_eq!(16, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_add_epi16() {
        let a = _mm256_set1_epi16(1);
        let e = _mm256_mask_reduce_add_epi16(0b11111111_00000000, a);
        assert_eq!(8, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_add_epi16() {
        let a = _mm_set1_epi16(1);
        let e = _mm_reduce_add_epi16(a);
        assert_eq!(8, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_add_epi16() {
        let a = _mm_set1_epi16(1);
        let e = _mm_mask_reduce_add_epi16(0b11110000, a);
        assert_eq!(4, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_add_epi8() {
        let a = _mm256_set1_epi8(1);
        let e = _mm256_reduce_add_epi8(a);
        assert_eq!(32, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_add_epi8() {
        let a = _mm256_set1_epi8(1);
        let e = _mm256_mask_reduce_add_epi8(0b11111111_00000000_11111111_00000000, a);
        assert_eq!(16, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_add_epi8() {
        let a = _mm_set1_epi8(1);
        let e = _mm_reduce_add_epi8(a);
        assert_eq!(16, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_add_epi8() {
        let a = _mm_set1_epi8(1);
        let e = _mm_mask_reduce_add_epi8(0b11111111_00000000, a);
        assert_eq!(8, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_and_epi16() {
        let a = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm256_reduce_and_epi16(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_and_epi16() {
        let a = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm256_mask_reduce_and_epi16(0b11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_and_epi16() {
        let a = _mm_set_epi16(1, 1, 1, 1, 2, 2, 2, 2);
        let e = _mm_reduce_and_epi16(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_and_epi16() {
        let a = _mm_set_epi16(1, 1, 1, 1, 2, 2, 2, 2);
        let e = _mm_mask_reduce_and_epi16(0b11110000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_and_epi8() {
        let a = _mm256_set_epi8(
            1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2,
            2, 2, 2,
        );
        let e = _mm256_reduce_and_epi8(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_and_epi8() {
        let a = _mm256_set_epi8(
            1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2,
            2, 2, 2,
        );
        let e = _mm256_mask_reduce_and_epi8(0b11111111_00000000_11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_and_epi8() {
        let a = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm_reduce_and_epi8(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_and_epi8() {
        let a = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm_mask_reduce_and_epi8(0b11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_mul_epi16() {
        let a = _mm256_set_epi16(2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1);
        let e = _mm256_reduce_mul_epi16(a);
        assert_eq!(256, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_mul_epi16() {
        let a = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm256_mask_reduce_mul_epi16(0b11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_mul_epi16() {
        let a = _mm_set_epi16(2, 2, 2, 2, 1, 1, 1, 1);
        let e = _mm_reduce_mul_epi16(a);
        assert_eq!(16, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_mul_epi16() {
        let a = _mm_set_epi16(1, 1, 1, 1, 2, 2, 2, 2);
        let e = _mm_mask_reduce_mul_epi16(0b11110000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_mul_epi8() {
        let a = _mm256_set_epi8(
            1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
            2, 2, 2,
        );
        let e = _mm256_reduce_mul_epi8(a);
        assert_eq!(64, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_mul_epi8() {
        let a = _mm256_set_epi8(
            1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
            2, 2, 2,
        );
        let e = _mm256_mask_reduce_mul_epi8(0b11111111_00000000_11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_mul_epi8() {
        let a = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2);
        let e = _mm_reduce_mul_epi8(a);
        assert_eq!(8, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_mul_epi8() {
        let a = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2);
        let e = _mm_mask_reduce_mul_epi8(0b11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_max_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: i16 = _mm256_reduce_max_epi16(a);
        assert_eq!(15, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_max_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: i16 = _mm256_mask_reduce_max_epi16(0b11111111_00000000, a);
        assert_eq!(7, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_max_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let e: i16 = _mm_reduce_max_epi16(a);
        assert_eq!(7, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_max_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let e: i16 = _mm_mask_reduce_max_epi16(0b11110000, a);
        assert_eq!(3, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_max_epi8() {
        let a = _mm256_set_epi8(
            0, 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,
        );
        let e: i8 = _mm256_reduce_max_epi8(a);
        assert_eq!(31, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_max_epi8() {
        let a = _mm256_set_epi8(
            0, 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,
        );
        let e: i8 = _mm256_mask_reduce_max_epi8(0b1111111111111111_0000000000000000, a);
        assert_eq!(15, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_max_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: i8 = _mm_reduce_max_epi8(a);
        assert_eq!(15, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_max_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: i8 = _mm_mask_reduce_max_epi8(0b11111111_00000000, a);
        assert_eq!(7, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_max_epu16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: u16 = _mm256_reduce_max_epu16(a);
        assert_eq!(15, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_max_epu16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: u16 = _mm256_mask_reduce_max_epu16(0b11111111_00000000, a);
        assert_eq!(7, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_max_epu16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let e: u16 = _mm_reduce_max_epu16(a);
        assert_eq!(7, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_max_epu16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let e: u16 = _mm_mask_reduce_max_epu16(0b11110000, a);
        assert_eq!(3, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_max_epu8() {
        let a = _mm256_set_epi8(
            0, 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,
        );
        let e: u8 = _mm256_reduce_max_epu8(a);
        assert_eq!(31, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_max_epu8() {
        let a = _mm256_set_epi8(
            0, 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,
        );
        let e: u8 = _mm256_mask_reduce_max_epu8(0b1111111111111111_0000000000000000, a);
        assert_eq!(15, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_max_epu8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: u8 = _mm_reduce_max_epu8(a);
        assert_eq!(15, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_max_epu8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: u8 = _mm_mask_reduce_max_epu8(0b11111111_00000000, a);
        assert_eq!(7, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_min_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: i16 = _mm256_reduce_min_epi16(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_min_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: i16 = _mm256_mask_reduce_min_epi16(0b11111111_00000000, a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_min_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let e: i16 = _mm_reduce_min_epi16(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_min_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let e: i16 = _mm_mask_reduce_min_epi16(0b11110000, a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_min_epi8() {
        let a = _mm256_set_epi8(
            0, 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,
        );
        let e: i8 = _mm256_reduce_min_epi8(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_min_epi8() {
        let a = _mm256_set_epi8(
            0, 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,
        );
        let e: i8 = _mm256_mask_reduce_min_epi8(0b1111111111111111_0000000000000000, a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_min_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: i8 = _mm_reduce_min_epi8(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_min_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: i8 = _mm_mask_reduce_min_epi8(0b11111111_00000000, a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_min_epu16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: u16 = _mm256_reduce_min_epu16(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_min_epu16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: u16 = _mm256_mask_reduce_min_epu16(0b11111111_00000000, a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_min_epu16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let e: u16 = _mm_reduce_min_epu16(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_min_epu16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let e: u16 = _mm_mask_reduce_min_epu16(0b11110000, a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_min_epu8() {
        let a = _mm256_set_epi8(
            0, 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,
        );
        let e: u8 = _mm256_reduce_min_epu8(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_min_epu8() {
        let a = _mm256_set_epi8(
            0, 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,
        );
        let e: u8 = _mm256_mask_reduce_min_epu8(0b1111111111111111_0000000000000000, a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_min_epu8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: u8 = _mm_reduce_min_epu8(a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_min_epu8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let e: u8 = _mm_mask_reduce_min_epu8(0b11111111_00000000, a);
        assert_eq!(0, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_or_epi16() {
        let a = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm256_reduce_or_epi16(a);
        assert_eq!(3, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_or_epi16() {
        let a = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm256_mask_reduce_or_epi16(0b11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_or_epi16() {
        let a = _mm_set_epi16(1, 1, 1, 1, 2, 2, 2, 2);
        let e = _mm_reduce_or_epi16(a);
        assert_eq!(3, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_or_epi16() {
        let a = _mm_set_epi16(1, 1, 1, 1, 2, 2, 2, 2);
        let e = _mm_mask_reduce_or_epi16(0b11110000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_reduce_or_epi8() {
        let a = _mm256_set_epi8(
            1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2,
            2, 2, 2,
        );
        let e = _mm256_reduce_or_epi8(a);
        assert_eq!(3, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_reduce_or_epi8() {
        let a = _mm256_set_epi8(
            1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2,
            2, 2, 2,
        );
        let e = _mm256_mask_reduce_or_epi8(0b11111111_00000000_11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_reduce_or_epi8() {
        let a = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm_reduce_or_epi8(a);
        assert_eq!(3, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_reduce_or_epi8() {
        let a = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2);
        let e = _mm_mask_reduce_or_epi8(0b11111111_00000000, a);
        assert_eq!(1, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_loadu_epi16() {
        #[rustfmt::skip]
        let a: [i16; 32] = [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];
        let r = _mm512_loadu_epi16(&a[0]);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(32, 31, 30, 29, 28, 27, 26, 25, 24, 23, 22, 21, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_loadu_epi16() {
        let a: [i16; 16] = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
        let r = _mm256_loadu_epi16(&a[0]);
        let e = _mm256_set_epi16(16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_loadu_epi16() {
        let a: [i16; 8] = [1, 2, 3, 4, 5, 6, 7, 8];
        let r = _mm_loadu_epi16(&a[0]);
        let e = _mm_set_epi16(8, 7, 6, 5, 4, 3, 2, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_loadu_epi8() {
        #[rustfmt::skip]
        let a: [i8; 64] = [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,
                           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];
        let r = _mm512_loadu_epi8(&a[0]);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(32, 31, 30, 29, 28, 27, 26, 25, 24, 23, 22, 21, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1,
                                32, 31, 30, 29, 28, 27, 26, 25, 24, 23, 22, 21, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_loadu_epi8() {
        #[rustfmt::skip]
        let a: [i8; 32] = [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];
        let r = _mm256_loadu_epi8(&a[0]);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(32, 31, 30, 29, 28, 27, 26, 25, 24, 23, 22, 21, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_loadu_epi8() {
        let a: [i8; 16] = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
        let r = _mm_loadu_epi8(&a[0]);
        let e = _mm_set_epi8(16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_storeu_epi16() {
        let a = _mm512_set1_epi16(9);
        let mut r = _mm512_undefined_epi32();
        _mm512_storeu_epi16(&mut r as *mut _ as *mut i16, a);
        assert_eq_m512i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_storeu_epi16() {
        let a = _mm256_set1_epi16(9);
        let mut r = _mm256_set1_epi32(0);
        _mm256_storeu_epi16(&mut r as *mut _ as *mut i16, a);
        assert_eq_m256i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_storeu_epi16() {
        let a = _mm_set1_epi16(9);
        let mut r = _mm_set1_epi32(0);
        _mm_storeu_epi16(&mut r as *mut _ as *mut i16, a);
        assert_eq_m128i(r, a);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_storeu_epi8() {
        let a = _mm512_set1_epi8(9);
        let mut r = _mm512_undefined_epi32();
        _mm512_storeu_epi8(&mut r as *mut _ as *mut i8, a);
        assert_eq_m512i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_storeu_epi8() {
        let a = _mm256_set1_epi8(9);
        let mut r = _mm256_set1_epi32(0);
        _mm256_storeu_epi8(&mut r as *mut _ as *mut i8, a);
        assert_eq_m256i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_storeu_epi8() {
        let a = _mm_set1_epi8(9);
        let mut r = _mm_set1_epi32(0);
        _mm_storeu_epi8(&mut r as *mut _ as *mut i8, a);
        assert_eq_m128i(r, a);
    }
    #[simd_test(enable = "avx512f,avx512bw")]
    unsafe fn test_mm512_mask_loadu_epi16() {
        let src = _mm512_set1_epi16(42);
        let a = &[
            1_i16, 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,
        ];
        let p = a.as_ptr();
        let m = 0b10101010_11001100_11101000_11001010;
        let r = _mm512_mask_loadu_epi16(src, m, black_box(p));
        let e = &[
            42_i16, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16, 42, 42, 19, 20, 42, 42,
            23, 24, 42, 26, 42, 28, 42, 30, 42, 32,
        ];
        let e = _mm512_loadu_epi16(e.as_ptr());
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw")]
    unsafe fn test_mm512_maskz_loadu_epi16() {
        let a = &[
            1_i16, 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,
        ];
        let p = a.as_ptr();
        let m = 0b10101010_11001100_11101000_11001010;
        let r = _mm512_maskz_loadu_epi16(m, black_box(p));
        let e = &[
            0_i16, 2, 0, 4, 0, 0, 7, 8, 0, 0, 0, 12, 0, 14, 15, 16, 0, 0, 19, 20, 0, 0, 23, 24, 0,
            26, 0, 28, 0, 30, 0, 32,
        ];
        let e = _mm512_loadu_epi16(e.as_ptr());
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw")]
    unsafe fn test_mm512_mask_storeu_epi16() {
        let mut r = [42_i16; 32];
        let a = &[
            1_i16, 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,
        ];
        let a = _mm512_loadu_epi16(a.as_ptr());
        let m = 0b10101010_11001100_11101000_11001010;
        _mm512_mask_storeu_epi16(r.as_mut_ptr(), m, a);
        let e = &[
            42_i16, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16, 42, 42, 19, 20, 42, 42,
            23, 24, 42, 26, 42, 28, 42, 30, 42, 32,
        ];
        let e = _mm512_loadu_epi16(e.as_ptr());
        assert_eq_m512i(_mm512_loadu_epi16(r.as_ptr()), e);
    }
    #[simd_test(enable = "avx512f,avx512bw")]
    unsafe fn test_mm512_mask_loadu_epi8() {
        let src = _mm512_set1_epi8(42);
        let a = &[
            1_i8, 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, 40, 41, 42, 43, 44, 45,
            46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,
        ];
        let p = a.as_ptr();
        let m = 0b00000000_11111111_11111111_00000000_10101010_11001100_11101000_11001010;
        let r = _mm512_mask_loadu_epi8(src, m, black_box(p));
        let e = &[
            42_i8, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16, 42, 42, 19, 20, 42, 42,
            23, 24, 42, 26, 42, 28, 42, 30, 42, 32, 42, 42, 42, 42, 42, 42, 42, 42, 41, 42, 43, 44,
            45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 42, 42, 42, 42, 42, 42, 42, 42,
        ];
        let e = _mm512_loadu_epi8(e.as_ptr());
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw")]
    unsafe fn test_mm512_maskz_loadu_epi8() {
        let a = &[
            1_i8, 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, 40, 41, 42, 43, 44, 45,
            46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,
        ];
        let p = a.as_ptr();
        let m = 0b00000000_11111111_11111111_00000000_10101010_11001100_11101000_11001010;
        let r = _mm512_maskz_loadu_epi8(m, black_box(p));
        let e = &[
            0_i8, 2, 0, 4, 0, 0, 7, 8, 0, 0, 0, 12, 0, 14, 15, 16, 0, 0, 19, 20, 0, 0, 23, 24, 0,
            26, 0, 28, 0, 30, 0, 32, 0, 0, 0, 0, 0, 0, 0, 0, 41, 42, 43, 44, 45, 46, 47, 48, 49,
            50, 51, 52, 53, 54, 55, 56, 0, 0, 0, 0, 0, 0, 0, 0,
        ];
        let e = _mm512_loadu_epi8(e.as_ptr());
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw")]
    unsafe fn test_mm512_mask_storeu_epi8() {
        let mut r = [42_i8; 64];
        let a = &[
            1_i8, 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, 40, 41, 42, 43, 44, 45,
            46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,
        ];
        let a = _mm512_loadu_epi8(a.as_ptr());
        let m = 0b00000000_11111111_11111111_00000000_10101010_11001100_11101000_11001010;
        _mm512_mask_storeu_epi8(r.as_mut_ptr(), m, a);
        let e = &[
            42_i8, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16, 42, 42, 19, 20, 42, 42,
            23, 24, 42, 26, 42, 28, 42, 30, 42, 32, 42, 42, 42, 42, 42, 42, 42, 42, 41, 42, 43, 44,
            45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 42, 42, 42, 42, 42, 42, 42, 42,
        ];
        let e = _mm512_loadu_epi8(e.as_ptr());
        assert_eq_m512i(_mm512_loadu_epi8(r.as_ptr()), e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_loadu_epi16() {
        let src = _mm256_set1_epi16(42);
        let a = &[1_i16, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
        let p = a.as_ptr();
        let m = 0b11101000_11001010;
        let r = _mm256_mask_loadu_epi16(src, m, black_box(p));
        let e = &[
            42_i16, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16,
        ];
        let e = _mm256_loadu_epi16(e.as_ptr());
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_loadu_epi16() {
        let a = &[1_i16, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
        let p = a.as_ptr();
        let m = 0b11101000_11001010;
        let r = _mm256_maskz_loadu_epi16(m, black_box(p));
        let e = &[0_i16, 2, 0, 4, 0, 0, 7, 8, 0, 0, 0, 12, 0, 14, 15, 16];
        let e = _mm256_loadu_epi16(e.as_ptr());
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_storeu_epi16() {
        let mut r = [42_i16; 16];
        let a = &[1_i16, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
        let a = _mm256_loadu_epi16(a.as_ptr());
        let m = 0b11101000_11001010;
        _mm256_mask_storeu_epi16(r.as_mut_ptr(), m, a);
        let e = &[
            42_i16, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16,
        ];
        let e = _mm256_loadu_epi16(e.as_ptr());
        assert_eq_m256i(_mm256_loadu_epi16(r.as_ptr()), e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_loadu_epi8() {
        let src = _mm256_set1_epi8(42);
        let a = &[
            1_i8, 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,
        ];
        let p = a.as_ptr();
        let m = 0b10101010_11001100_11101000_11001010;
        let r = _mm256_mask_loadu_epi8(src, m, black_box(p));
        let e = &[
            42_i8, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16, 42, 42, 19, 20, 42, 42,
            23, 24, 42, 26, 42, 28, 42, 30, 42, 32,
        ];
        let e = _mm256_loadu_epi8(e.as_ptr());
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_loadu_epi8() {
        let a = &[
            1_i8, 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,
        ];
        let p = a.as_ptr();
        let m = 0b10101010_11001100_11101000_11001010;
        let r = _mm256_maskz_loadu_epi8(m, black_box(p));
        let e = &[
            0_i8, 2, 0, 4, 0, 0, 7, 8, 0, 0, 0, 12, 0, 14, 15, 16, 0, 0, 19, 20, 0, 0, 23, 24, 0,
            26, 0, 28, 0, 30, 0, 32,
        ];
        let e = _mm256_loadu_epi8(e.as_ptr());
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_storeu_epi8() {
        let mut r = [42_i8; 32];
        let a = &[
            1_i8, 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,
        ];
        let a = _mm256_loadu_epi8(a.as_ptr());
        let m = 0b10101010_11001100_11101000_11001010;
        _mm256_mask_storeu_epi8(r.as_mut_ptr(), m, a);
        let e = &[
            42_i8, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16, 42, 42, 19, 20, 42, 42,
            23, 24, 42, 26, 42, 28, 42, 30, 42, 32,
        ];
        let e = _mm256_loadu_epi8(e.as_ptr());
        assert_eq_m256i(_mm256_loadu_epi8(r.as_ptr()), e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm_mask_loadu_epi16() {
        let src = _mm_set1_epi16(42);
        let a = &[1_i16, 2, 3, 4, 5, 6, 7, 8];
        let p = a.as_ptr();
        let m = 0b11001010;
        let r = _mm_mask_loadu_epi16(src, m, black_box(p));
        let e = &[42_i16, 2, 42, 4, 42, 42, 7, 8];
        let e = _mm_loadu_epi16(e.as_ptr());
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_loadu_epi16() {
        let a = &[1_i16, 2, 3, 4, 5, 6, 7, 8];
        let p = a.as_ptr();
        let m = 0b11001010;
        let r = _mm_maskz_loadu_epi16(m, black_box(p));
        let e = &[0_i16, 2, 0, 4, 0, 0, 7, 8];
        let e = _mm_loadu_epi16(e.as_ptr());
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm_mask_storeu_epi16() {
        let mut r = [42_i16; 8];
        let a = &[1_i16, 2, 3, 4, 5, 6, 7, 8];
        let a = _mm_loadu_epi16(a.as_ptr());
        let m = 0b11001010;
        _mm_mask_storeu_epi16(r.as_mut_ptr(), m, a);
        let e = &[42_i16, 2, 42, 4, 42, 42, 7, 8];
        let e = _mm_loadu_epi16(e.as_ptr());
        assert_eq_m128i(_mm_loadu_epi16(r.as_ptr()), e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm_mask_loadu_epi8() {
        let src = _mm_set1_epi8(42);
        let a = &[1_i8, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
        let p = a.as_ptr();
        let m = 0b11101000_11001010;
        let r = _mm_mask_loadu_epi8(src, m, black_box(p));
        let e = &[
            42_i8, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16,
        ];
        let e = _mm_loadu_epi8(e.as_ptr());
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_loadu_epi8() {
        let a = &[1_i8, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
        let p = a.as_ptr();
        let m = 0b11101000_11001010;
        let r = _mm_maskz_loadu_epi8(m, black_box(p));
        let e = &[0_i8, 2, 0, 4, 0, 0, 7, 8, 0, 0, 0, 12, 0, 14, 15, 16];
        let e = _mm_loadu_epi8(e.as_ptr());
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512f,avx512bw,avx512vl")]
    unsafe fn test_mm_mask_storeu_epi8() {
        let mut r = [42_i8; 16];
        let a = &[1_i8, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16];
        let a = _mm_loadu_epi8(a.as_ptr());
        let m = 0b11101000_11001010;
        _mm_mask_storeu_epi8(r.as_mut_ptr(), m, a);
        let e = &[
            42_i8, 2, 42, 4, 42, 42, 7, 8, 42, 42, 42, 12, 42, 14, 15, 16,
        ];
        let e = _mm_loadu_epi8(e.as_ptr());
        assert_eq_m128i(_mm_loadu_epi8(r.as_ptr()), e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_madd_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_madd_epi16(a, b);
        let e = _mm512_set1_epi32(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_madd_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mask_madd_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_madd_epi16(a, 0b00000000_00001111, a, b);
        let e = _mm512_set_epi32(
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            2,
            2,
            2,
            2,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_madd_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_maskz_madd_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_madd_epi16(0b00000000_00001111, a, b);
        let e = _mm512_set_epi32(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_madd_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_mask_madd_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_madd_epi16(a, 0b00001111, a, b);
        let e = _mm256_set_epi32(
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            1 << 16 | 1,
            2,
            2,
            2,
            2,
        );
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_madd_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_maskz_madd_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_madd_epi16(0b00001111, a, b);
        let e = _mm256_set_epi32(0, 0, 0, 0, 2, 2, 2, 2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_madd_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_mask_madd_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_madd_epi16(a, 0b00001111, a, b);
        let e = _mm_set_epi32(2, 2, 2, 2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_madd_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_maskz_madd_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_madd_epi16(0b00001111, a, b);
        let e = _mm_set_epi32(2, 2, 2, 2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maddubs_epi16() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let r = _mm512_maddubs_epi16(a, b);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_maddubs_epi16() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let src = _mm512_set1_epi16(1);
        let r = _mm512_mask_maddubs_epi16(src, 0, a, b);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_add_epi16(src, 0b00000000_00000000_00000000_00000001, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1<<9|2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_maddubs_epi16() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let r = _mm512_maskz_maddubs_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_maddubs_epi16(0b00000000_11111111_00000000_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 2, 2, 2, 2,
                                 0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 2, 2, 2, 2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_maddubs_epi16() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(1);
        let src = _mm256_set1_epi16(1);
        let r = _mm256_mask_maddubs_epi16(src, 0, a, b);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_add_epi16(src, 0b00000000_00000001, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 << 9 | 2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_maddubs_epi16() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(1);
        let r = _mm256_maskz_maddubs_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_maddubs_epi16(0b00000000_11111111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 2, 2, 2, 2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_maddubs_epi16() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(1);
        let src = _mm_set1_epi16(1);
        let r = _mm_mask_maddubs_epi16(src, 0, a, b);
        assert_eq_m128i(r, src);
        let r = _mm_mask_add_epi16(src, 0b00000001, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 1, 1, 1, 1 << 9 | 2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_maddubs_epi16() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(1);
        let r = _mm_maskz_maddubs_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_maddubs_epi16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 2, 2, 2, 2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_packs_epi32() {
        let a = _mm512_set1_epi32(i32::MAX);
        let b = _mm512_set1_epi32(1);
        let r = _mm512_packs_epi32(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX, 1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX,
                                 1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX, 1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_packs_epi32() {
        let a = _mm512_set1_epi32(i32::MAX);
        let b = _mm512_set1_epi32(1 << 16 | 1);
        let r = _mm512_mask_packs_epi32(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_packs_epi32(b, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_packs_epi32() {
        let a = _mm512_set1_epi32(i32::MAX);
        let b = _mm512_set1_epi32(1);
        let r = _mm512_maskz_packs_epi32(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_packs_epi32(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_packs_epi32() {
        let a = _mm256_set1_epi32(i32::MAX);
        let b = _mm256_set1_epi32(1 << 16 | 1);
        let r = _mm256_mask_packs_epi32(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_packs_epi32(b, 0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_packs_epi32() {
        let a = _mm256_set1_epi32(i32::MAX);
        let b = _mm256_set1_epi32(1);
        let r = _mm256_maskz_packs_epi32(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_packs_epi32(0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_packs_epi32() {
        let a = _mm_set1_epi32(i32::MAX);
        let b = _mm_set1_epi32(1 << 16 | 1);
        let r = _mm_mask_packs_epi32(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_packs_epi32(b, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_packs_epi32() {
        let a = _mm_set1_epi32(i32::MAX);
        let b = _mm_set1_epi32(1);
        let r = _mm_maskz_packs_epi32(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_packs_epi32(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, i16::MAX, i16::MAX, i16::MAX, i16::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_packs_epi16() {
        let a = _mm512_set1_epi16(i16::MAX);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_packs_epi16(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX,
                                1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX,
                                1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX,
                                1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_packs_epi16() {
        let a = _mm512_set1_epi16(i16::MAX);
        let b = _mm512_set1_epi16(1 << 8 | 1);
        let r = _mm512_mask_packs_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_packs_epi16(
            b,
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_packs_epi16() {
        let a = _mm512_set1_epi16(i16::MAX);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_maskz_packs_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_packs_epi16(
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_packs_epi16() {
        let a = _mm256_set1_epi16(i16::MAX);
        let b = _mm256_set1_epi16(1 << 8 | 1);
        let r = _mm256_mask_packs_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_packs_epi16(b, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_packs_epi16() {
        let a = _mm256_set1_epi16(i16::MAX);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_maskz_packs_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_packs_epi16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_packs_epi16() {
        let a = _mm_set1_epi16(i16::MAX);
        let b = _mm_set1_epi16(1 << 8 | 1);
        let r = _mm_mask_packs_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_packs_epi16(b, 0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_packs_epi16() {
        let a = _mm_set1_epi16(i16::MAX);
        let b = _mm_set1_epi16(1);
        let r = _mm_maskz_packs_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_packs_epi16(0b00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_packus_epi32() {
        let a = _mm512_set1_epi32(-1);
        let b = _mm512_set1_epi32(1);
        let r = _mm512_packus_epi32(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0,
                                 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_packus_epi32() {
        let a = _mm512_set1_epi32(-1);
        let b = _mm512_set1_epi32(1 << 16 | 1);
        let r = _mm512_mask_packus_epi32(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_packus_epi32(b, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_packus_epi32() {
        let a = _mm512_set1_epi32(-1);
        let b = _mm512_set1_epi32(1);
        let r = _mm512_maskz_packus_epi32(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_packus_epi32(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_packus_epi32() {
        let a = _mm256_set1_epi32(-1);
        let b = _mm256_set1_epi32(1 << 16 | 1);
        let r = _mm256_mask_packus_epi32(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_packus_epi32(b, 0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_packus_epi32() {
        let a = _mm256_set1_epi32(-1);
        let b = _mm256_set1_epi32(1);
        let r = _mm256_maskz_packus_epi32(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_packus_epi32(0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_packus_epi32() {
        let a = _mm_set1_epi32(-1);
        let b = _mm_set1_epi32(1 << 16 | 1);
        let r = _mm_mask_packus_epi32(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_packus_epi32(b, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_packus_epi32() {
        let a = _mm_set1_epi32(-1);
        let b = _mm_set1_epi32(1);
        let r = _mm_maskz_packus_epi32(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_packus_epi32(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_packus_epi16() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_packus_epi16(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0,
                                1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0,
                                1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0,
                                1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_packus_epi16() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(1 << 8 | 1);
        let r = _mm512_mask_packus_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_packus_epi16(
            b,
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_packus_epi16() {
        let a = _mm512_set1_epi16(-1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_maskz_packus_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_packus_epi16(
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_packus_epi16() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(1 << 8 | 1);
        let r = _mm256_mask_packus_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_packus_epi16(b, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_packus_epi16() {
        let a = _mm256_set1_epi16(-1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_maskz_packus_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_packus_epi16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_packus_epi16() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(1 << 8 | 1);
        let r = _mm_mask_packus_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_packus_epi16(b, 0b00000000_00001111, a, b);
        let e = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_packus_epi16() {
        let a = _mm_set1_epi16(-1);
        let b = _mm_set1_epi16(1);
        let r = _mm_maskz_packus_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_packus_epi16(0b00000000_00001111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_avg_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_avg_epu16(a, b);
        let e = _mm512_set1_epi16(1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_avg_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_mask_avg_epu16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_avg_epu16(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_avg_epu16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(1);
        let r = _mm512_maskz_avg_epu16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_avg_epu16(0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_avg_epu16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_mask_avg_epu16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_avg_epu16(a, 0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_avg_epu16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(1);
        let r = _mm256_maskz_avg_epu16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_avg_epu16(0b00000000_00001111, a, b);
        let e = _mm256_set_epi16(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_avg_epu16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_mask_avg_epu16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_avg_epu16(a, 0b00001111, a, b);
        let e = _mm_set_epi16(1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_avg_epu16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(1);
        let r = _mm_maskz_avg_epu16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_avg_epu16(0b00001111, a, b);
        let e = _mm_set_epi16(0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_avg_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let r = _mm512_avg_epu8(a, b);
        let e = _mm512_set1_epi8(1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_avg_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let r = _mm512_mask_avg_epu8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_avg_epu8(
            a,
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_avg_epu8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(1);
        let r = _mm512_maskz_avg_epu8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_avg_epu8(
            0b00000000_000000000_00000000_00000000_00000000_0000000_00000000_00001111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_avg_epu8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(1);
        let r = _mm256_mask_avg_epu8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_avg_epu8(a, 0b00000000_00000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
                                1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_avg_epu8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(1);
        let r = _mm256_maskz_avg_epu8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_avg_epu8(0b00000000_0000000_00000000_00001111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                                0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_avg_epu8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(1);
        let r = _mm_mask_avg_epu8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_avg_epu8(a, 0b00000000_00001111, a, b);
        let e = _mm_set_epi8(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_avg_epu8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(1);
        let r = _mm_maskz_avg_epu8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_avg_epu8(0b00000000_00001111, a, b);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_sll_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm512_sll_epi16(a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_sll_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm512_mask_sll_epi16(a, 0, a, count);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_sll_epi16(a, 0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_sll_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm512_maskz_sll_epi16(0, a, count);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_sll_epi16(0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_sll_epi16() {
        let a = _mm256_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm256_mask_sll_epi16(a, 0, a, count);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_sll_epi16(a, 0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_sll_epi16() {
        let a = _mm256_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm256_maskz_sll_epi16(0, a, count);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_sll_epi16(0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_sll_epi16() {
        let a = _mm_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm_mask_sll_epi16(a, 0, a, count);
        assert_eq_m128i(r, a);
        let r = _mm_mask_sll_epi16(a, 0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_sll_epi16() {
        let a = _mm_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm_maskz_sll_epi16(0, a, count);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_sll_epi16(0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_slli_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let r = _mm512_slli_epi16::<1>(a);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_slli_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let r = _mm512_mask_slli_epi16::<1>(a, 0, a);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_slli_epi16::<1>(a, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_slli_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let r = _mm512_maskz_slli_epi16::<1>(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_slli_epi16::<1>(0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_slli_epi16() {
        let a = _mm256_set1_epi16(1 << 15);
        let r = _mm256_mask_slli_epi16::<1>(a, 0, a);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_slli_epi16::<1>(a, 0b11111111_11111111, a);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_slli_epi16() {
        let a = _mm256_set1_epi16(1 << 15);
        let r = _mm256_maskz_slli_epi16::<1>(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_slli_epi16::<1>(0b11111111_11111111, a);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_slli_epi16() {
        let a = _mm_set1_epi16(1 << 15);
        let r = _mm_mask_slli_epi16::<1>(a, 0, a);
        assert_eq_m128i(r, a);
        let r = _mm_mask_slli_epi16::<1>(a, 0b11111111, a);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_slli_epi16() {
        let a = _mm_set1_epi16(1 << 15);
        let r = _mm_maskz_slli_epi16::<1>(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_slli_epi16::<1>(0b11111111, a);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_sllv_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_sllv_epi16(a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_sllv_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_mask_sllv_epi16(a, 0, a, count);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_sllv_epi16(a, 0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_sllv_epi16() {
        let a = _mm512_set1_epi16(1 << 15);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_maskz_sllv_epi16(0, a, count);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_sllv_epi16(0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_sllv_epi16() {
        let a = _mm256_set1_epi16(1 << 15);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_sllv_epi16(a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_sllv_epi16() {
        let a = _mm256_set1_epi16(1 << 15);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_mask_sllv_epi16(a, 0, a, count);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_sllv_epi16(a, 0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_sllv_epi16() {
        let a = _mm256_set1_epi16(1 << 15);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_maskz_sllv_epi16(0, a, count);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_sllv_epi16(0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_sllv_epi16() {
        let a = _mm_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm_sllv_epi16(a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_sllv_epi16() {
        let a = _mm_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm_mask_sllv_epi16(a, 0, a, count);
        assert_eq_m128i(r, a);
        let r = _mm_mask_sllv_epi16(a, 0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_sllv_epi16() {
        let a = _mm_set1_epi16(1 << 15);
        let count = _mm_set1_epi16(2);
        let r = _mm_maskz_sllv_epi16(0, a, count);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_sllv_epi16(0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_srl_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm512_srl_epi16(a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_srl_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm512_mask_srl_epi16(a, 0, a, count);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_srl_epi16(a, 0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_srl_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm512_maskz_srl_epi16(0, a, count);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_srl_epi16(0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_srl_epi16() {
        let a = _mm256_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm256_mask_srl_epi16(a, 0, a, count);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_srl_epi16(a, 0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_srl_epi16() {
        let a = _mm256_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm256_maskz_srl_epi16(0, a, count);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_srl_epi16(0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_srl_epi16() {
        let a = _mm_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm_mask_srl_epi16(a, 0, a, count);
        assert_eq_m128i(r, a);
        let r = _mm_mask_srl_epi16(a, 0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_srl_epi16() {
        let a = _mm_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm_maskz_srl_epi16(0, a, count);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_srl_epi16(0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_srli_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let r = _mm512_srli_epi16::<2>(a);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_srli_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let r = _mm512_mask_srli_epi16::<2>(a, 0, a);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_srli_epi16::<2>(a, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_srli_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let r = _mm512_maskz_srli_epi16::<2>(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_srli_epi16::<2>(0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_srli_epi16() {
        let a = _mm256_set1_epi16(1 << 1);
        let r = _mm256_mask_srli_epi16::<2>(a, 0, a);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_srli_epi16::<2>(a, 0b11111111_11111111, a);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_srli_epi16() {
        let a = _mm256_set1_epi16(1 << 1);
        let r = _mm256_maskz_srli_epi16::<2>(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_srli_epi16::<2>(0b11111111_11111111, a);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_srli_epi16() {
        let a = _mm_set1_epi16(1 << 1);
        let r = _mm_mask_srli_epi16::<2>(a, 0, a);
        assert_eq_m128i(r, a);
        let r = _mm_mask_srli_epi16::<2>(a, 0b11111111, a);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_srli_epi16() {
        let a = _mm_set1_epi16(1 << 1);
        let r = _mm_maskz_srli_epi16::<2>(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_srli_epi16::<2>(0b11111111, a);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_srlv_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_srlv_epi16(a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_srlv_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_mask_srlv_epi16(a, 0, a, count);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_srlv_epi16(a, 0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_srlv_epi16() {
        let a = _mm512_set1_epi16(1 << 1);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_maskz_srlv_epi16(0, a, count);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_srlv_epi16(0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_srlv_epi16() {
        let a = _mm256_set1_epi16(1 << 1);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_srlv_epi16(a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_srlv_epi16() {
        let a = _mm256_set1_epi16(1 << 1);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_mask_srlv_epi16(a, 0, a, count);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_srlv_epi16(a, 0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_srlv_epi16() {
        let a = _mm256_set1_epi16(1 << 1);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_maskz_srlv_epi16(0, a, count);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_srlv_epi16(0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_srlv_epi16() {
        let a = _mm_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm_srlv_epi16(a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_srlv_epi16() {
        let a = _mm_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm_mask_srlv_epi16(a, 0, a, count);
        assert_eq_m128i(r, a);
        let r = _mm_mask_srlv_epi16(a, 0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_srlv_epi16() {
        let a = _mm_set1_epi16(1 << 1);
        let count = _mm_set1_epi16(2);
        let r = _mm_maskz_srlv_epi16(0, a, count);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_srlv_epi16(0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_sra_epi16() {
        let a = _mm512_set1_epi16(8);
        let count = _mm_set1_epi16(1);
        let r = _mm512_sra_epi16(a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_sra_epi16() {
        let a = _mm512_set1_epi16(8);
        let count = _mm_set1_epi16(1);
        let r = _mm512_mask_sra_epi16(a, 0, a, count);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_sra_epi16(a, 0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_sra_epi16() {
        let a = _mm512_set1_epi16(8);
        let count = _mm_set1_epi16(1);
        let r = _mm512_maskz_sra_epi16(0, a, count);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_sra_epi16(0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_sra_epi16() {
        let a = _mm256_set1_epi16(8);
        let count = _mm_set1_epi16(1);
        let r = _mm256_mask_sra_epi16(a, 0, a, count);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_sra_epi16(a, 0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_sra_epi16() {
        let a = _mm256_set1_epi16(8);
        let count = _mm_set1_epi16(1);
        let r = _mm256_maskz_sra_epi16(0, a, count);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_sra_epi16(0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_sra_epi16() {
        let a = _mm_set1_epi16(8);
        let count = _mm_set1_epi16(1);
        let r = _mm_mask_sra_epi16(a, 0, a, count);
        assert_eq_m128i(r, a);
        let r = _mm_mask_sra_epi16(a, 0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_sra_epi16() {
        let a = _mm_set1_epi16(8);
        let count = _mm_set1_epi16(1);
        let r = _mm_maskz_sra_epi16(0, a, count);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_sra_epi16(0b11111111, a, count);
        let e = _mm_set1_epi16(0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_srai_epi16() {
        let a = _mm512_set1_epi16(8);
        let r = _mm512_srai_epi16::<2>(a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_srai_epi16() {
        let a = _mm512_set1_epi16(8);
        let r = _mm512_mask_srai_epi16::<2>(a, 0, a);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_srai_epi16::<2>(a, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_srai_epi16() {
        let a = _mm512_set1_epi16(8);
        let r = _mm512_maskz_srai_epi16::<2>(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_srai_epi16::<2>(0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_srai_epi16() {
        let a = _mm256_set1_epi16(8);
        let r = _mm256_mask_srai_epi16::<2>(a, 0, a);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_srai_epi16::<2>(a, 0b11111111_11111111, a);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_srai_epi16() {
        let a = _mm256_set1_epi16(8);
        let r = _mm256_maskz_srai_epi16::<2>(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_srai_epi16::<2>(0b11111111_11111111, a);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_srai_epi16() {
        let a = _mm_set1_epi16(8);
        let r = _mm_mask_srai_epi16::<2>(a, 0, a);
        assert_eq_m128i(r, a);
        let r = _mm_mask_srai_epi16::<2>(a, 0b11111111, a);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_srai_epi16() {
        let a = _mm_set1_epi16(8);
        let r = _mm_maskz_srai_epi16::<2>(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_srai_epi16::<2>(0b11111111, a);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_srav_epi16() {
        let a = _mm512_set1_epi16(8);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_srav_epi16(a, count);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_srav_epi16() {
        let a = _mm512_set1_epi16(8);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_mask_srav_epi16(a, 0, a, count);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_srav_epi16(a, 0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_srav_epi16() {
        let a = _mm512_set1_epi16(8);
        let count = _mm512_set1_epi16(2);
        let r = _mm512_maskz_srav_epi16(0, a, count);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_srav_epi16(0b11111111_11111111_11111111_11111111, a, count);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_srav_epi16() {
        let a = _mm256_set1_epi16(8);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_srav_epi16(a, count);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_srav_epi16() {
        let a = _mm256_set1_epi16(8);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_mask_srav_epi16(a, 0, a, count);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_srav_epi16(a, 0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_srav_epi16() {
        let a = _mm256_set1_epi16(8);
        let count = _mm256_set1_epi16(2);
        let r = _mm256_maskz_srav_epi16(0, a, count);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_srav_epi16(0b11111111_11111111, a, count);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_srav_epi16() {
        let a = _mm_set1_epi16(8);
        let count = _mm_set1_epi16(2);
        let r = _mm_srav_epi16(a, count);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_srav_epi16() {
        let a = _mm_set1_epi16(8);
        let count = _mm_set1_epi16(2);
        let r = _mm_mask_srav_epi16(a, 0, a, count);
        assert_eq_m128i(r, a);
        let r = _mm_mask_srav_epi16(a, 0b11111111, a, count);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_srav_epi16() {
        let a = _mm_set1_epi16(8);
        let count = _mm_set1_epi16(2);
        let r = _mm_maskz_srav_epi16(0, a, count);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_srav_epi16(0b11111111, a, count);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_permutex2var_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0,  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);
        #[rustfmt::skip]
        let idx = _mm512_set_epi16(1, 1<<5, 2, 1<<5, 3, 1<<5, 4, 1<<5, 5, 1<<5, 6, 1<<5, 7, 1<<5, 8, 1<<5,
                                   9, 1<<5, 10, 1<<5, 11, 1<<5, 12, 1<<5, 13, 1<<5, 14, 1<<5, 15, 1<<5, 16, 1<<5);
        let b = _mm512_set1_epi16(100);
        let r = _mm512_permutex2var_epi16(a, idx, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            30, 100, 29, 100, 28, 100, 27, 100, 26, 100, 25, 100, 24, 100, 23, 100,
            22, 100, 21, 100, 20, 100, 19, 100, 18, 100, 17, 100, 16, 100, 15, 100,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_permutex2var_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0,  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);
        #[rustfmt::skip]
        let idx = _mm512_set_epi16(1, 1<<5, 2, 1<<5, 3, 1<<5, 4, 1<<5, 5, 1<<5, 6, 1<<5, 7, 1<<5, 8, 1<<5,
                                   9, 1<<5, 10, 1<<5, 11, 1<<5, 12, 1<<5, 13, 1<<5, 14, 1<<5, 15, 1<<5, 16, 1<<5);
        let b = _mm512_set1_epi16(100);
        let r = _mm512_mask_permutex2var_epi16(a, 0, idx, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_permutex2var_epi16(a, 0b11111111_11111111_11111111_11111111, idx, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            30, 100, 29, 100, 28, 100, 27, 100, 26, 100, 25, 100, 24, 100, 23, 100,
            22, 100, 21, 100, 20, 100, 19, 100, 18, 100, 17, 100, 16, 100, 15, 100,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_permutex2var_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0,  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);
        #[rustfmt::skip]
        let idx = _mm512_set_epi16(1, 1<<5, 2, 1<<5, 3, 1<<5, 4, 1<<5, 5, 1<<5, 6, 1<<5, 7, 1<<5, 8, 1<<5,
                                   9, 1<<5, 10, 1<<5, 11, 1<<5, 12, 1<<5, 13, 1<<5, 14, 1<<5, 15, 1<<5, 16, 1<<5);
        let b = _mm512_set1_epi16(100);
        let r = _mm512_maskz_permutex2var_epi16(0, a, idx, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_permutex2var_epi16(0b11111111_11111111_11111111_11111111, a, idx, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            30, 100, 29, 100, 28, 100, 27, 100, 26, 100, 25, 100, 24, 100, 23, 100,
            22, 100, 21, 100, 20, 100, 19, 100, 18, 100, 17, 100, 16, 100, 15, 100,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask2_permutex2var_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0,  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);
        #[rustfmt::skip]
        let idx = _mm512_set_epi16(1, 1<<5, 2, 1<<5, 3, 1<<5, 4, 1<<5, 5, 1<<5, 6, 1<<5, 7, 1<<5, 8, 1<<5,
                                   9, 1<<5, 10, 1<<5, 11, 1<<5, 12, 1<<5, 13, 1<<5, 14, 1<<5, 15, 1<<5, 16, 1<<5);
        let b = _mm512_set1_epi16(100);
        let r = _mm512_mask2_permutex2var_epi16(a, idx, 0, b);
        assert_eq_m512i(r, idx);
        let r = _mm512_mask2_permutex2var_epi16(a, idx, 0b11111111_11111111_11111111_11111111, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            30, 100, 29, 100, 28, 100, 27, 100, 26, 100, 25, 100, 24, 100, 23, 100,
            22, 100, 21, 100, 20, 100, 19, 100, 18, 100, 17, 100, 16, 100, 15, 100,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_permutex2var_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let idx = _mm256_set_epi16(1, 1<<4, 2, 1<<4, 3, 1<<4, 4, 1<<4, 5, 1<<4, 6, 1<<4, 7, 1<<4, 8, 1<<4);
        let b = _mm256_set1_epi16(100);
        let r = _mm256_permutex2var_epi16(a, idx, b);
        let e = _mm256_set_epi16(
            14, 100, 13, 100, 12, 100, 11, 100, 10, 100, 9, 100, 8, 100, 7, 100,
        );
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_permutex2var_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let idx = _mm256_set_epi16(1, 1<<4, 2, 1<<4, 3, 1<<4, 4, 1<<4, 5, 1<<4, 6, 1<<4, 7, 1<<4, 8, 1<<4);
        let b = _mm256_set1_epi16(100);
        let r = _mm256_mask_permutex2var_epi16(a, 0, idx, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_permutex2var_epi16(a, 0b11111111_11111111, idx, b);
        let e = _mm256_set_epi16(
            14, 100, 13, 100, 12, 100, 11, 100, 10, 100, 9, 100, 8, 100, 7, 100,
        );
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_permutex2var_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let idx = _mm256_set_epi16(1, 1<<4, 2, 1<<4, 3, 1<<4, 4, 1<<4, 5, 1<<4, 6, 1<<4, 7, 1<<4, 8, 1<<4);
        let b = _mm256_set1_epi16(100);
        let r = _mm256_maskz_permutex2var_epi16(0, a, idx, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_permutex2var_epi16(0b11111111_11111111, a, idx, b);
        let e = _mm256_set_epi16(
            14, 100, 13, 100, 12, 100, 11, 100, 10, 100, 9, 100, 8, 100, 7, 100,
        );
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask2_permutex2var_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        #[rustfmt::skip]
        let idx = _mm256_set_epi16(1, 1<<4, 2, 1<<4, 3, 1<<4, 4, 1<<4, 5, 1<<4, 6, 1<<4, 7, 1<<4, 8, 1<<4);
        let b = _mm256_set1_epi16(100);
        let r = _mm256_mask2_permutex2var_epi16(a, idx, 0, b);
        assert_eq_m256i(r, idx);
        let r = _mm256_mask2_permutex2var_epi16(a, idx, 0b11111111_11111111, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi16(
            14, 100, 13, 100, 12, 100, 11, 100, 10, 100, 9, 100, 8, 100, 7, 100,
        );
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_permutex2var_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let idx = _mm_set_epi16(1, 1 << 3, 2, 1 << 3, 3, 1 << 3, 4, 1 << 3);
        let b = _mm_set1_epi16(100);
        let r = _mm_permutex2var_epi16(a, idx, b);
        let e = _mm_set_epi16(6, 100, 5, 100, 4, 100, 3, 100);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_permutex2var_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let idx = _mm_set_epi16(1, 1 << 3, 2, 1 << 3, 3, 1 << 3, 4, 1 << 3);
        let b = _mm_set1_epi16(100);
        let r = _mm_mask_permutex2var_epi16(a, 0, idx, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_permutex2var_epi16(a, 0b11111111, idx, b);
        let e = _mm_set_epi16(6, 100, 5, 100, 4, 100, 3, 100);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_permutex2var_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let idx = _mm_set_epi16(1, 1 << 3, 2, 1 << 3, 3, 1 << 3, 4, 1 << 3);
        let b = _mm_set1_epi16(100);
        let r = _mm_maskz_permutex2var_epi16(0, a, idx, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_permutex2var_epi16(0b11111111, a, idx, b);
        let e = _mm_set_epi16(6, 100, 5, 100, 4, 100, 3, 100);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask2_permutex2var_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let idx = _mm_set_epi16(1, 1 << 3, 2, 1 << 3, 3, 1 << 3, 4, 1 << 3);
        let b = _mm_set1_epi16(100);
        let r = _mm_mask2_permutex2var_epi16(a, idx, 0, b);
        assert_eq_m128i(r, idx);
        let r = _mm_mask2_permutex2var_epi16(a, idx, 0b11111111, b);
        let e = _mm_set_epi16(6, 100, 5, 100, 4, 100, 3, 100);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_permutexvar_epi16() {
        let idx = _mm512_set1_epi16(1);
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0,  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);
        let r = _mm512_permutexvar_epi16(idx, a);
        let e = _mm512_set1_epi16(30);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_permutexvar_epi16() {
        let idx = _mm512_set1_epi16(1);
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0,  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);
        let r = _mm512_mask_permutexvar_epi16(a, 0, idx, a);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_permutexvar_epi16(a, 0b11111111_11111111_11111111_11111111, idx, a);
        let e = _mm512_set1_epi16(30);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_permutexvar_epi16() {
        let idx = _mm512_set1_epi16(1);
        #[rustfmt::skip]
        let a = _mm512_set_epi16(0,  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);
        let r = _mm512_maskz_permutexvar_epi16(0, idx, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_permutexvar_epi16(0b11111111_11111111_11111111_11111111, idx, a);
        let e = _mm512_set1_epi16(30);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_permutexvar_epi16() {
        let idx = _mm256_set1_epi16(1);
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let r = _mm256_permutexvar_epi16(idx, a);
        let e = _mm256_set1_epi16(14);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_permutexvar_epi16() {
        let idx = _mm256_set1_epi16(1);
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let r = _mm256_mask_permutexvar_epi16(a, 0, idx, a);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_permutexvar_epi16(a, 0b11111111_11111111, idx, a);
        let e = _mm256_set1_epi16(14);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_permutexvar_epi16() {
        let idx = _mm256_set1_epi16(1);
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let r = _mm256_maskz_permutexvar_epi16(0, idx, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_permutexvar_epi16(0b11111111_11111111, idx, a);
        let e = _mm256_set1_epi16(14);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_permutexvar_epi16() {
        let idx = _mm_set1_epi16(1);
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let r = _mm_permutexvar_epi16(idx, a);
        let e = _mm_set1_epi16(6);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_permutexvar_epi16() {
        let idx = _mm_set1_epi16(1);
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let r = _mm_mask_permutexvar_epi16(a, 0, idx, a);
        assert_eq_m128i(r, a);
        let r = _mm_mask_permutexvar_epi16(a, 0b11111111, idx, a);
        let e = _mm_set1_epi16(6);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_permutexvar_epi16() {
        let idx = _mm_set1_epi16(1);
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let r = _mm_maskz_permutexvar_epi16(0, idx, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_permutexvar_epi16(0b11111111, idx, a);
        let e = _mm_set1_epi16(6);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_blend_epi16() {
        let a = _mm512_set1_epi16(1);
        let b = _mm512_set1_epi16(2);
        let r = _mm512_mask_blend_epi16(0b11111111_00000000_11111111_00000000, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1,
                                 2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_blend_epi16() {
        let a = _mm256_set1_epi16(1);
        let b = _mm256_set1_epi16(2);
        let r = _mm256_mask_blend_epi16(0b11111111_00000000, a, b);
        let e = _mm256_set_epi16(2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_blend_epi16() {
        let a = _mm_set1_epi16(1);
        let b = _mm_set1_epi16(2);
        let r = _mm_mask_blend_epi16(0b11110000, a, b);
        let e = _mm_set_epi16(2, 2, 2, 2, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_blend_epi8() {
        let a = _mm512_set1_epi8(1);
        let b = _mm512_set1_epi8(2);
        let r = _mm512_mask_blend_epi8(
            0b11111111_00000000_11111111_00000000_11111111_00000000_11111111_00000000,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1,
                                2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1,
                                2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1,
                                2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_blend_epi8() {
        let a = _mm256_set1_epi8(1);
        let b = _mm256_set1_epi8(2);
        let r = _mm256_mask_blend_epi8(0b11111111_00000000_11111111_00000000, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1,
                                2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_blend_epi8() {
        let a = _mm_set1_epi8(1);
        let b = _mm_set1_epi8(2);
        let r = _mm_mask_blend_epi8(0b11111111_00000000, a, b);
        let e = _mm_set_epi8(2, 2, 2, 2, 2, 2, 2, 2, 1, 1, 1, 1, 1, 1, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_broadcastw_epi16() {
        let a = _mm_set_epi16(17, 18, 19, 20, 21, 22, 23, 24);
        let r = _mm512_broadcastw_epi16(a);
        let e = _mm512_set1_epi16(24);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_broadcastw_epi16() {
        let src = _mm512_set1_epi16(1);
        let a = _mm_set_epi16(17, 18, 19, 20, 21, 22, 23, 24);
        let r = _mm512_mask_broadcastw_epi16(src, 0, a);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_broadcastw_epi16(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(24);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_broadcastw_epi16() {
        let a = _mm_set_epi16(17, 18, 19, 20, 21, 22, 23, 24);
        let r = _mm512_maskz_broadcastw_epi16(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_broadcastw_epi16(0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(24);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_broadcastw_epi16() {
        let src = _mm256_set1_epi16(1);
        let a = _mm_set_epi16(17, 18, 19, 20, 21, 22, 23, 24);
        let r = _mm256_mask_broadcastw_epi16(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_broadcastw_epi16(src, 0b11111111_11111111, a);
        let e = _mm256_set1_epi16(24);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_broadcastw_epi16() {
        let a = _mm_set_epi16(17, 18, 19, 20, 21, 22, 23, 24);
        let r = _mm256_maskz_broadcastw_epi16(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_broadcastw_epi16(0b11111111_11111111, a);
        let e = _mm256_set1_epi16(24);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_broadcastw_epi16() {
        let src = _mm_set1_epi16(1);
        let a = _mm_set_epi16(17, 18, 19, 20, 21, 22, 23, 24);
        let r = _mm_mask_broadcastw_epi16(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_broadcastw_epi16(src, 0b11111111, a);
        let e = _mm_set1_epi16(24);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_broadcastw_epi16() {
        let a = _mm_set_epi16(17, 18, 19, 20, 21, 22, 23, 24);
        let r = _mm_maskz_broadcastw_epi16(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_broadcastw_epi16(0b11111111, a);
        let e = _mm_set1_epi16(24);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_broadcastb_epi8() {
        let a = _mm_set_epi8(
            17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32,
        );
        let r = _mm512_broadcastb_epi8(a);
        let e = _mm512_set1_epi8(32);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_broadcastb_epi8() {
        let src = _mm512_set1_epi8(1);
        let a = _mm_set_epi8(
            17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32,
        );
        let r = _mm512_mask_broadcastb_epi8(src, 0, a);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_broadcastb_epi8(
            src,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
        );
        let e = _mm512_set1_epi8(32);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_broadcastb_epi8() {
        let a = _mm_set_epi8(
            17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32,
        );
        let r = _mm512_maskz_broadcastb_epi8(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_broadcastb_epi8(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
        );
        let e = _mm512_set1_epi8(32);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_broadcastb_epi8() {
        let src = _mm256_set1_epi8(1);
        let a = _mm_set_epi8(
            17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32,
        );
        let r = _mm256_mask_broadcastb_epi8(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_broadcastb_epi8(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(32);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_broadcastb_epi8() {
        let a = _mm_set_epi8(
            17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32,
        );
        let r = _mm256_maskz_broadcastb_epi8(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_broadcastb_epi8(0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(32);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_broadcastb_epi8() {
        let src = _mm_set1_epi8(1);
        let a = _mm_set_epi8(
            17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32,
        );
        let r = _mm_mask_broadcastb_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_broadcastb_epi8(src, 0b11111111_11111111, a);
        let e = _mm_set1_epi8(32);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_broadcastb_epi8() {
        let a = _mm_set_epi8(
            17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32,
        );
        let r = _mm_maskz_broadcastb_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_broadcastb_epi8(0b11111111_11111111, a);
        let e = _mm_set1_epi8(32);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_unpackhi_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(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);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        let r = _mm512_unpackhi_epi16(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(33, 1,  34, 2,  35, 3,  36, 4,  41, 9,  42, 10, 43, 11, 44, 12,
                                 49, 17, 50, 18, 51, 19, 52, 20, 57, 25, 58, 26, 59, 27, 60, 28);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_unpackhi_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(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);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        let r = _mm512_mask_unpackhi_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_unpackhi_epi16(a, 0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(33, 1,  34, 2,  35, 3,  36, 4,  41, 9,  42, 10, 43, 11, 44, 12,
                                 49, 17, 50, 18, 51, 19, 52, 20, 57, 25, 58, 26, 59, 27, 60, 28);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_unpackhi_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(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);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        let r = _mm512_maskz_unpackhi_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_unpackhi_epi16(0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(33, 1,  34, 2,  35, 3,  36, 4,  41, 9,  42, 10, 43, 11, 44, 12,
                                 49, 17, 50, 18, 51, 19, 52, 20, 57, 25, 58, 26, 59, 27, 60, 28);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_unpackhi_epi16() {
        let a = _mm256_set_epi16(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
        let b = _mm256_set_epi16(
            33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
        );
        let r = _mm256_mask_unpackhi_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_unpackhi_epi16(a, 0b11111111_11111111, a, b);
        let e = _mm256_set_epi16(33, 1, 34, 2, 35, 3, 36, 4, 41, 9, 42, 10, 43, 11, 44, 12);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_unpackhi_epi16() {
        let a = _mm256_set_epi16(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
        let b = _mm256_set_epi16(
            33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
        );
        let r = _mm256_maskz_unpackhi_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_unpackhi_epi16(0b11111111_11111111, a, b);
        let e = _mm256_set_epi16(33, 1, 34, 2, 35, 3, 36, 4, 41, 9, 42, 10, 43, 11, 44, 12);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_unpackhi_epi16() {
        let a = _mm_set_epi16(1, 2, 3, 4, 5, 6, 7, 8);
        let b = _mm_set_epi16(33, 34, 35, 36, 37, 38, 39, 40);
        let r = _mm_mask_unpackhi_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_unpackhi_epi16(a, 0b11111111, a, b);
        let e = _mm_set_epi16(33, 1, 34, 2, 35, 3, 36, 4);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_unpackhi_epi16() {
        let a = _mm_set_epi16(1, 2, 3, 4, 5, 6, 7, 8);
        let b = _mm_set_epi16(33, 34, 35, 36, 37, 38, 39, 40);
        let r = _mm_maskz_unpackhi_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_unpackhi_epi16(0b11111111, a, b);
        let e = _mm_set_epi16(33, 1, 34, 2, 35, 3, 36, 4);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_unpackhi_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(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, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96,
                                97,  98,  99,  100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112,
                                113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 0);
        let r = _mm512_unpackhi_epi8(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(65, 1,  66, 2,  67, 3,  68, 4,  69, 5,  70, 6,  71, 7,  72, 8,
                                81, 17, 82, 18, 83, 19, 84, 20, 85, 21, 86, 22, 87, 23, 88, 24,
                                97, 33, 98, 34, 99, 35, 100, 36, 101, 37, 102, 38, 103, 39, 104, 40,
                                113, 49, 114, 50, 115, 51, 116, 52, 117, 53, 118, 54, 119, 55, 120, 56);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_unpackhi_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(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, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96,
                                97,  98,  99,  100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112,
                                113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 0);
        let r = _mm512_mask_unpackhi_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_unpackhi_epi8(
            a,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(65, 1,  66, 2,  67, 3,  68, 4,  69, 5,  70, 6,  71, 7,  72, 8,
                                81, 17, 82, 18, 83, 19, 84, 20, 85, 21, 86, 22, 87, 23, 88, 24,
                                97, 33, 98, 34, 99, 35, 100, 36, 101, 37, 102, 38, 103, 39, 104, 40,
                                113, 49, 114, 50, 115, 51, 116, 52, 117, 53, 118, 54, 119, 55, 120, 56);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_unpackhi_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(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, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96,
                                97,  98,  99,  100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112,
                                113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 0);
        let r = _mm512_maskz_unpackhi_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_unpackhi_epi8(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(65, 1,  66, 2,  67, 3,  68, 4,  69, 5,  70, 6,  71, 7,  72, 8,
                                81, 17, 82, 18, 83, 19, 84, 20, 85, 21, 86, 22, 87, 23, 88, 24,
                                97, 33, 98, 34, 99, 35, 100, 36, 101, 37, 102, 38, 103, 39, 104, 40,
                                113, 49, 114, 50, 115, 51, 116, 52, 117, 53, 118, 54, 119, 55, 120, 56);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_unpackhi_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(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);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96);
        let r = _mm256_mask_unpackhi_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_unpackhi_epi8(a, 0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(65, 1,  66, 2,  67, 3,  68, 4,  69, 5,  70, 6,  71, 7,  72, 8,
                                81, 17, 82, 18, 83, 19, 84, 20, 85, 21, 86, 22, 87, 23, 88, 24);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_unpackhi_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(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);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96);
        let r = _mm256_maskz_unpackhi_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_unpackhi_epi8(0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(65, 1,  66, 2,  67, 3,  68, 4,  69, 5,  70, 6,  71, 7,  72, 8,
                                81, 17, 82, 18, 83, 19, 84, 20, 85, 21, 86, 22, 87, 23, 88, 24);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_unpackhi_epi8() {
        let a = _mm_set_epi8(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
        let b = _mm_set_epi8(
            65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80,
        );
        let r = _mm_mask_unpackhi_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_unpackhi_epi8(a, 0b11111111_11111111, a, b);
        let e = _mm_set_epi8(65, 1, 66, 2, 67, 3, 68, 4, 69, 5, 70, 6, 71, 7, 72, 8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_unpackhi_epi8() {
        let a = _mm_set_epi8(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
        let b = _mm_set_epi8(
            65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80,
        );
        let r = _mm_maskz_unpackhi_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_unpackhi_epi8(0b11111111_11111111, a, b);
        let e = _mm_set_epi8(65, 1, 66, 2, 67, 3, 68, 4, 69, 5, 70, 6, 71, 7, 72, 8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_unpacklo_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(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);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        let r = _mm512_unpacklo_epi16(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(37, 5,  38, 6,  39, 7,  40, 8,  45, 13, 46, 14, 47, 15, 48, 16,
                                 53, 21, 54, 22, 55, 23, 56, 24, 61, 29, 62, 30, 63, 31, 64, 32);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_unpacklo_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(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);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        let r = _mm512_mask_unpacklo_epi16(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_unpacklo_epi16(a, 0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(37, 5,  38, 6,  39, 7,  40, 8,  45, 13, 46, 14, 47, 15, 48, 16,
                                 53, 21, 54, 22, 55, 23, 56, 24, 61, 29, 62, 30, 63, 31, 64, 32);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_unpacklo_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(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);
        #[rustfmt::skip]
        let b = _mm512_set_epi16(33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        let r = _mm512_maskz_unpacklo_epi16(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_unpacklo_epi16(0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(37, 5,  38, 6,  39, 7,  40, 8,  45, 13, 46, 14, 47, 15, 48, 16,
                                 53, 21, 54, 22, 55, 23, 56, 24, 61, 29, 62, 30, 63, 31, 64, 32);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_unpacklo_epi16() {
        let a = _mm256_set_epi16(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
        let b = _mm256_set_epi16(
            33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
        );
        let r = _mm256_mask_unpacklo_epi16(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_unpacklo_epi16(a, 0b11111111_11111111, a, b);
        let e = _mm256_set_epi16(37, 5, 38, 6, 39, 7, 40, 8, 45, 13, 46, 14, 47, 15, 48, 16);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_unpacklo_epi16() {
        let a = _mm256_set_epi16(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
        let b = _mm256_set_epi16(
            33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48,
        );
        let r = _mm256_maskz_unpacklo_epi16(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_unpacklo_epi16(0b11111111_11111111, a, b);
        let e = _mm256_set_epi16(37, 5, 38, 6, 39, 7, 40, 8, 45, 13, 46, 14, 47, 15, 48, 16);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_unpacklo_epi16() {
        let a = _mm_set_epi16(1, 2, 3, 4, 5, 6, 7, 8);
        let b = _mm_set_epi16(33, 34, 35, 36, 37, 38, 39, 40);
        let r = _mm_mask_unpacklo_epi16(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_unpacklo_epi16(a, 0b11111111, a, b);
        let e = _mm_set_epi16(37, 5, 38, 6, 39, 7, 40, 8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_unpacklo_epi16() {
        let a = _mm_set_epi16(1, 2, 3, 4, 5, 6, 7, 8);
        let b = _mm_set_epi16(33, 34, 35, 36, 37, 38, 39, 40);
        let r = _mm_maskz_unpacklo_epi16(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_unpacklo_epi16(0b11111111, a, b);
        let e = _mm_set_epi16(37, 5, 38, 6, 39, 7, 40, 8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_unpacklo_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(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, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96,
                                97,  98,  99,  100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112,
                                113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 0);
        let r = _mm512_unpacklo_epi8(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(73,  9,  74,  10, 75,  11, 76,  12, 77,  13, 78,  14, 79,  15, 80,  16,
                                89,  25, 90,  26, 91,  27, 92,  28, 93,  29, 94,  30, 95,  31, 96,  32,
                                105, 41, 106, 42, 107, 43, 108, 44, 109, 45, 110, 46, 111, 47, 112, 48,
                                121, 57, 122, 58, 123, 59, 124, 60, 125, 61, 126, 62, 127, 63, 0,   64);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_unpacklo_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(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, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96,
                                97,  98,  99,  100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112,
                                113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 0);
        let r = _mm512_mask_unpacklo_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_unpacklo_epi8(
            a,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(73,  9,  74,  10, 75,  11, 76,  12, 77,  13, 78,  14, 79,  15, 80,  16,
                                89,  25, 90,  26, 91,  27, 92,  28, 93,  29, 94,  30, 95,  31, 96,  32,
                                105, 41, 106, 42, 107, 43, 108, 44, 109, 45, 110, 46, 111, 47, 112, 48,
                                121, 57, 122, 58, 123, 59, 124, 60, 125, 61, 126, 62, 127, 63, 0,   64);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_unpacklo_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(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, 40, 41, 42, 43, 44, 45, 46, 47, 48,
                                49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64);
        #[rustfmt::skip]
        let b = _mm512_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96,
                                97,  98,  99,  100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112,
                                113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 0);
        let r = _mm512_maskz_unpacklo_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_unpacklo_epi8(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(73,  9,  74,  10, 75,  11, 76,  12, 77,  13, 78,  14, 79,  15, 80,  16,
                                89,  25, 90,  26, 91,  27, 92,  28, 93,  29, 94,  30, 95,  31, 96,  32,
                                105, 41, 106, 42, 107, 43, 108, 44, 109, 45, 110, 46, 111, 47, 112, 48,
                                121, 57, 122, 58, 123, 59, 124, 60, 125, 61, 126, 62, 127, 63, 0,   64);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_unpacklo_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(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);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96);
        let r = _mm256_mask_unpacklo_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_unpacklo_epi8(a, 0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(73,  9,  74,  10, 75,  11, 76,  12, 77,  13, 78,  14, 79,  15, 80,  16,
                                89,  25, 90,  26, 91,  27, 92,  28, 93,  29, 94,  30, 95,  31, 96,  32);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_unpacklo_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(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);
        #[rustfmt::skip]
        let b = _mm256_set_epi8(65,  66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,  79,  80,
                                81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,  92,  93,  94,  95,  96);
        let r = _mm256_maskz_unpacklo_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_unpacklo_epi8(0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(73,  9,  74,  10, 75,  11, 76,  12, 77,  13, 78,  14, 79,  15, 80,  16,
                                89,  25, 90,  26, 91,  27, 92,  28, 93,  29, 94,  30, 95,  31, 96,  32);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_unpacklo_epi8() {
        let a = _mm_set_epi8(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
        let b = _mm_set_epi8(
            65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80,
        );
        let r = _mm_mask_unpacklo_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_unpacklo_epi8(a, 0b11111111_11111111, a, b);
        let e = _mm_set_epi8(
            73, 9, 74, 10, 75, 11, 76, 12, 77, 13, 78, 14, 79, 15, 80, 16,
        );
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_unpacklo_epi8() {
        let a = _mm_set_epi8(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16);
        let b = _mm_set_epi8(
            65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80,
        );
        let r = _mm_maskz_unpacklo_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_unpacklo_epi8(0b11111111_11111111, a, b);
        let e = _mm_set_epi8(
            73, 9, 74, 10, 75, 11, 76, 12, 77, 13, 78, 14, 79, 15, 80, 16,
        );
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_mov_epi16() {
        let src = _mm512_set1_epi16(1);
        let a = _mm512_set1_epi16(2);
        let r = _mm512_mask_mov_epi16(src, 0, a);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_mov_epi16(src, 0b11111111_11111111_11111111_11111111, a);
        assert_eq_m512i(r, a);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_mov_epi16() {
        let a = _mm512_set1_epi16(2);
        let r = _mm512_maskz_mov_epi16(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_mov_epi16(0b11111111_11111111_11111111_11111111, a);
        assert_eq_m512i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_mov_epi16() {
        let src = _mm256_set1_epi16(1);
        let a = _mm256_set1_epi16(2);
        let r = _mm256_mask_mov_epi16(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_mov_epi16(src, 0b11111111_11111111, a);
        assert_eq_m256i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_mov_epi16() {
        let a = _mm256_set1_epi16(2);
        let r = _mm256_maskz_mov_epi16(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_mov_epi16(0b11111111_11111111, a);
        assert_eq_m256i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_mov_epi16() {
        let src = _mm_set1_epi16(1);
        let a = _mm_set1_epi16(2);
        let r = _mm_mask_mov_epi16(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_mov_epi16(src, 0b11111111, a);
        assert_eq_m128i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_mov_epi16() {
        let a = _mm_set1_epi16(2);
        let r = _mm_maskz_mov_epi16(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_mov_epi16(0b11111111, a);
        assert_eq_m128i(r, a);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_mov_epi8() {
        let src = _mm512_set1_epi8(1);
        let a = _mm512_set1_epi8(2);
        let r = _mm512_mask_mov_epi8(src, 0, a);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_mov_epi8(
            src,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
        );
        assert_eq_m512i(r, a);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_mov_epi8() {
        let a = _mm512_set1_epi8(2);
        let r = _mm512_maskz_mov_epi8(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_mov_epi8(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
        );
        assert_eq_m512i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_mov_epi8() {
        let src = _mm256_set1_epi8(1);
        let a = _mm256_set1_epi8(2);
        let r = _mm256_mask_mov_epi8(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_mov_epi8(src, 0b11111111_11111111_11111111_11111111, a);
        assert_eq_m256i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_mov_epi8() {
        let a = _mm256_set1_epi8(2);
        let r = _mm256_maskz_mov_epi8(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_mov_epi8(0b11111111_11111111_11111111_11111111, a);
        assert_eq_m256i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_mov_epi8() {
        let src = _mm_set1_epi8(1);
        let a = _mm_set1_epi8(2);
        let r = _mm_mask_mov_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_mov_epi8(src, 0b11111111_11111111, a);
        assert_eq_m128i(r, a);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_mov_epi8() {
        let a = _mm_set1_epi8(2);
        let r = _mm_maskz_mov_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_mov_epi8(0b11111111_11111111, a);
        assert_eq_m128i(r, a);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_set1_epi16() {
        let src = _mm512_set1_epi16(2);
        let a: i16 = 11;
        let r = _mm512_mask_set1_epi16(src, 0, a);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_set1_epi16(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(11);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_set1_epi16() {
        let a: i16 = 11;
        let r = _mm512_maskz_set1_epi16(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_set1_epi16(0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(11);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_set1_epi16() {
        let src = _mm256_set1_epi16(2);
        let a: i16 = 11;
        let r = _mm256_mask_set1_epi16(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_set1_epi16(src, 0b11111111_11111111, a);
        let e = _mm256_set1_epi16(11);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_set1_epi16() {
        let a: i16 = 11;
        let r = _mm256_maskz_set1_epi16(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_set1_epi16(0b11111111_11111111, a);
        let e = _mm256_set1_epi16(11);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_set1_epi16() {
        let src = _mm_set1_epi16(2);
        let a: i16 = 11;
        let r = _mm_mask_set1_epi16(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_set1_epi16(src, 0b11111111, a);
        let e = _mm_set1_epi16(11);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_set1_epi16() {
        let a: i16 = 11;
        let r = _mm_maskz_set1_epi16(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_set1_epi16(0b11111111, a);
        let e = _mm_set1_epi16(11);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_set1_epi8() {
        let src = _mm512_set1_epi8(2);
        let a: i8 = 11;
        let r = _mm512_mask_set1_epi8(src, 0, a);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_set1_epi8(
            src,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
        );
        let e = _mm512_set1_epi8(11);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_set1_epi8() {
        let a: i8 = 11;
        let r = _mm512_maskz_set1_epi8(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_set1_epi8(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
        );
        let e = _mm512_set1_epi8(11);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_set1_epi8() {
        let src = _mm256_set1_epi8(2);
        let a: i8 = 11;
        let r = _mm256_mask_set1_epi8(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_set1_epi8(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(11);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_set1_epi8() {
        let a: i8 = 11;
        let r = _mm256_maskz_set1_epi8(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_set1_epi8(0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(11);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_set1_epi8() {
        let src = _mm_set1_epi8(2);
        let a: i8 = 11;
        let r = _mm_mask_set1_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_set1_epi8(src, 0b11111111_11111111, a);
        let e = _mm_set1_epi8(11);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_set1_epi8() {
        let a: i8 = 11;
        let r = _mm_maskz_set1_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_set1_epi8(0b11111111_11111111, a);
        let e = _mm_set1_epi8(11);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_shufflelo_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(
            0, 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,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            0, 1, 2, 3, 7, 6, 6, 4, 8, 9, 10, 11, 15, 14, 14, 12,
            16, 17, 18, 19, 23, 22, 22, 20, 24, 25, 26, 27, 31, 30, 30, 28,
        );
        let r = _mm512_shufflelo_epi16::<0b00_01_01_11>(a);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_shufflelo_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(
            0, 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,
        );
        let r = _mm512_mask_shufflelo_epi16::<0b00_01_01_11>(a, 0, a);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_shufflelo_epi16::<0b00_01_01_11>(
            a,
            0b11111111_11111111_11111111_11111111,
            a,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            0, 1, 2, 3, 7, 6, 6, 4, 8, 9, 10, 11, 15, 14, 14, 12,
            16, 17, 18, 19, 23, 22, 22, 20, 24, 25, 26, 27, 31, 30, 30, 28,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_shufflelo_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(
            0, 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,
        );
        let r = _mm512_maskz_shufflelo_epi16::<0b00_01_01_11>(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r =
            _mm512_maskz_shufflelo_epi16::<0b00_01_01_11>(0b11111111_11111111_11111111_11111111, a);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            0, 1, 2, 3, 7, 6, 6, 4, 8, 9, 10, 11, 15, 14, 14, 12,
            16, 17, 18, 19, 23, 22, 22, 20, 24, 25, 26, 27, 31, 30, 30, 28,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_shufflelo_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let r = _mm256_mask_shufflelo_epi16::<0b00_01_01_11>(a, 0, a);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_shufflelo_epi16::<0b00_01_01_11>(a, 0b11111111_11111111, a);
        let e = _mm256_set_epi16(0, 1, 2, 3, 7, 6, 6, 4, 8, 9, 10, 11, 15, 14, 14, 12);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_shufflelo_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let r = _mm256_maskz_shufflelo_epi16::<0b00_01_01_11>(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_shufflelo_epi16::<0b00_01_01_11>(0b11111111_11111111, a);
        let e = _mm256_set_epi16(0, 1, 2, 3, 7, 6, 6, 4, 8, 9, 10, 11, 15, 14, 14, 12);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_shufflelo_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let r = _mm_mask_shufflelo_epi16::<0b00_01_01_11>(a, 0, a);
        assert_eq_m128i(r, a);
        let r = _mm_mask_shufflelo_epi16::<0b00_01_01_11>(a, 0b11111111, a);
        let e = _mm_set_epi16(0, 1, 2, 3, 7, 6, 6, 4);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_shufflelo_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let r = _mm_maskz_shufflelo_epi16::<0b00_01_01_11>(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_shufflelo_epi16::<0b00_01_01_11>(0b11111111, a);
        let e = _mm_set_epi16(0, 1, 2, 3, 7, 6, 6, 4);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_shufflehi_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(
            0, 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,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            3, 2, 2, 0, 4, 5, 6, 7, 11, 10, 10, 8, 12, 13, 14, 15,
            19, 18, 18, 16, 20, 21, 22, 23, 27, 26, 26, 24, 28, 29, 30, 31,
        );
        let r = _mm512_shufflehi_epi16::<0b00_01_01_11>(a);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_shufflehi_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(
            0, 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,
        );
        let r = _mm512_mask_shufflehi_epi16::<0b00_01_01_11>(a, 0, a);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_shufflehi_epi16::<0b00_01_01_11>(
            a,
            0b11111111_11111111_11111111_11111111,
            a,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            3, 2, 2, 0, 4, 5, 6, 7, 11, 10, 10, 8, 12, 13, 14, 15,
            19, 18, 18, 16, 20, 21, 22, 23, 27, 26, 26, 24, 28, 29, 30, 31,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_shufflehi_epi16() {
        #[rustfmt::skip]
        let a = _mm512_set_epi16(
            0, 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,
        );
        let r = _mm512_maskz_shufflehi_epi16::<0b00_01_01_11>(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r =
            _mm512_maskz_shufflehi_epi16::<0b00_01_01_11>(0b11111111_11111111_11111111_11111111, a);
        #[rustfmt::skip]
        let e = _mm512_set_epi16(
            3, 2, 2, 0, 4, 5, 6, 7, 11, 10, 10, 8, 12, 13, 14, 15,
            19, 18, 18, 16, 20, 21, 22, 23, 27, 26, 26, 24, 28, 29, 30, 31,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_shufflehi_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let r = _mm256_mask_shufflehi_epi16::<0b00_01_01_11>(a, 0, a);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_shufflehi_epi16::<0b00_01_01_11>(a, 0b11111111_11111111, a);
        let e = _mm256_set_epi16(3, 2, 2, 0, 4, 5, 6, 7, 11, 10, 10, 8, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_shufflehi_epi16() {
        let a = _mm256_set_epi16(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let r = _mm256_maskz_shufflehi_epi16::<0b00_01_01_11>(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_shufflehi_epi16::<0b00_01_01_11>(0b11111111_11111111, a);
        let e = _mm256_set_epi16(3, 2, 2, 0, 4, 5, 6, 7, 11, 10, 10, 8, 12, 13, 14, 15);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_shufflehi_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let r = _mm_mask_shufflehi_epi16::<0b00_01_01_11>(a, 0, a);
        assert_eq_m128i(r, a);
        let r = _mm_mask_shufflehi_epi16::<0b00_01_01_11>(a, 0b11111111, a);
        let e = _mm_set_epi16(3, 2, 2, 0, 4, 5, 6, 7);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_shufflehi_epi16() {
        let a = _mm_set_epi16(0, 1, 2, 3, 4, 5, 6, 7);
        let r = _mm_maskz_shufflehi_epi16::<0b00_01_01_11>(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_shufflehi_epi16::<0b00_01_01_11>(0b11111111, a);
        let e = _mm_set_epi16(3, 2, 2, 0, 4, 5, 6, 7);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_shuffle_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0,  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, 40, 41, 42, 43, 44, 45, 46, 47,
                                48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63);
        let b = _mm512_set1_epi8(1);
        let r = _mm512_shuffle_epi8(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
                                30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30,
                                46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46,
                                62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_shuffle_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0,  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, 40, 41, 42, 43, 44, 45, 46, 47,
                                48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63);
        let b = _mm512_set1_epi8(1);
        let r = _mm512_mask_shuffle_epi8(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_shuffle_epi8(
            a,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
                                30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30,
                                46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46,
                                62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_shuffle_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(0,  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, 40, 41, 42, 43, 44, 45, 46, 47,
                                48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63);
        let b = _mm512_set1_epi8(1);
        let r = _mm512_maskz_shuffle_epi8(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_shuffle_epi8(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
                                30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30,
                                46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46, 46,
                                62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62, 62);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_shuffle_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0,  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);
        let b = _mm256_set1_epi8(1);
        let r = _mm256_mask_shuffle_epi8(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_shuffle_epi8(a, 0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
                                30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_shuffle_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(0,  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);
        let b = _mm256_set1_epi8(1);
        let r = _mm256_maskz_shuffle_epi8(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_shuffle_epi8(0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
                                30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_shuffle_epi8() {
        let a = _mm_set_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
        let b = _mm_set1_epi8(1);
        let r = _mm_mask_shuffle_epi8(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_shuffle_epi8(a, 0b11111111_11111111, a, b);
        let e = _mm_set_epi8(
            14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
        );
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_shuffle_epi8() {
        #[rustfmt::skip]
        let a = _mm_set_epi8(0,  1,  2,  3,  4,  5,  6,  7,  8,  9,  10, 11, 12, 13, 14, 15);
        let b = _mm_set1_epi8(1);
        let r = _mm_maskz_shuffle_epi8(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_shuffle_epi8(0b11111111_11111111, a, b);
        let e = _mm_set_epi8(
            14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,
        );
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_test_epi16_mask() {
        let a = _mm512_set1_epi16(1 << 0);
        let b = _mm512_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm512_test_epi16_mask(a, b);
        let e: __mmask32 = 0b11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_test_epi16_mask() {
        let a = _mm512_set1_epi16(1 << 0);
        let b = _mm512_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm512_mask_test_epi16_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm512_mask_test_epi16_mask(0b11111111_11111111_11111111_11111111, a, b);
        let e: __mmask32 = 0b11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_test_epi16_mask() {
        let a = _mm256_set1_epi16(1 << 0);
        let b = _mm256_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm256_test_epi16_mask(a, b);
        let e: __mmask16 = 0b11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_test_epi16_mask() {
        let a = _mm256_set1_epi16(1 << 0);
        let b = _mm256_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm256_mask_test_epi16_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm256_mask_test_epi16_mask(0b11111111_11111111, a, b);
        let e: __mmask16 = 0b11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_test_epi16_mask() {
        let a = _mm_set1_epi16(1 << 0);
        let b = _mm_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm_test_epi16_mask(a, b);
        let e: __mmask8 = 0b11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_test_epi16_mask() {
        let a = _mm_set1_epi16(1 << 0);
        let b = _mm_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm_mask_test_epi16_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm_mask_test_epi16_mask(0b11111111, a, b);
        let e: __mmask8 = 0b11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_test_epi8_mask() {
        let a = _mm512_set1_epi8(1 << 0);
        let b = _mm512_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm512_test_epi8_mask(a, b);
        let e: __mmask64 =
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_test_epi8_mask() {
        let a = _mm512_set1_epi8(1 << 0);
        let b = _mm512_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm512_mask_test_epi8_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm512_mask_test_epi8_mask(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        let e: __mmask64 =
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_test_epi8_mask() {
        let a = _mm256_set1_epi8(1 << 0);
        let b = _mm256_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm256_test_epi8_mask(a, b);
        let e: __mmask32 = 0b11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_test_epi8_mask() {
        let a = _mm256_set1_epi8(1 << 0);
        let b = _mm256_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm256_mask_test_epi8_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm256_mask_test_epi8_mask(0b11111111_11111111_11111111_11111111, a, b);
        let e: __mmask32 = 0b11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_test_epi8_mask() {
        let a = _mm_set1_epi8(1 << 0);
        let b = _mm_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm_test_epi8_mask(a, b);
        let e: __mmask16 = 0b11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_test_epi8_mask() {
        let a = _mm_set1_epi8(1 << 0);
        let b = _mm_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm_mask_test_epi8_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm_mask_test_epi8_mask(0b11111111_11111111, a, b);
        let e: __mmask16 = 0b11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_testn_epi16_mask() {
        let a = _mm512_set1_epi16(1 << 0);
        let b = _mm512_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm512_testn_epi16_mask(a, b);
        let e: __mmask32 = 0b00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_testn_epi16_mask() {
        let a = _mm512_set1_epi16(1 << 0);
        let b = _mm512_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm512_mask_testn_epi16_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm512_mask_testn_epi16_mask(0b11111111_11111111_11111111_11111111, a, b);
        let e: __mmask32 = 0b00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_testn_epi16_mask() {
        let a = _mm256_set1_epi16(1 << 0);
        let b = _mm256_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm256_testn_epi16_mask(a, b);
        let e: __mmask16 = 0b00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_testn_epi16_mask() {
        let a = _mm256_set1_epi16(1 << 0);
        let b = _mm256_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm256_mask_testn_epi16_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm256_mask_testn_epi16_mask(0b11111111_11111111, a, b);
        let e: __mmask16 = 0b00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_testn_epi16_mask() {
        let a = _mm_set1_epi16(1 << 0);
        let b = _mm_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm_testn_epi16_mask(a, b);
        let e: __mmask8 = 0b00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_testn_epi16_mask() {
        let a = _mm_set1_epi16(1 << 0);
        let b = _mm_set1_epi16(1 << 0 | 1 << 1);
        let r = _mm_mask_testn_epi16_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm_mask_testn_epi16_mask(0b11111111, a, b);
        let e: __mmask8 = 0b00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_testn_epi8_mask() {
        let a = _mm512_set1_epi8(1 << 0);
        let b = _mm512_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm512_testn_epi8_mask(a, b);
        let e: __mmask64 =
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_testn_epi8_mask() {
        let a = _mm512_set1_epi8(1 << 0);
        let b = _mm512_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm512_mask_testn_epi8_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm512_mask_testn_epi8_mask(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        let e: __mmask64 =
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_testn_epi8_mask() {
        let a = _mm256_set1_epi8(1 << 0);
        let b = _mm256_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm256_testn_epi8_mask(a, b);
        let e: __mmask32 = 0b00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_testn_epi8_mask() {
        let a = _mm256_set1_epi8(1 << 0);
        let b = _mm256_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm256_mask_testn_epi8_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm256_mask_testn_epi8_mask(0b11111111_11111111_11111111_11111111, a, b);
        let e: __mmask32 = 0b00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_testn_epi8_mask() {
        let a = _mm_set1_epi8(1 << 0);
        let b = _mm_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm_testn_epi8_mask(a, b);
        let e: __mmask16 = 0b00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_testn_epi8_mask() {
        let a = _mm_set1_epi8(1 << 0);
        let b = _mm_set1_epi8(1 << 0 | 1 << 1);
        let r = _mm_mask_testn_epi8_mask(0, a, b);
        assert_eq!(r, 0);
        let r = _mm_mask_testn_epi8_mask(0b11111111_11111111, a, b);
        let e: __mmask16 = 0b00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_store_mask64() {
        let a: __mmask64 =
            0b11111111_00000000_11111111_00000000_11111111_00000000_11111111_00000000;
        let mut r = 0;
        _store_mask64(&mut r, a);
        assert_eq!(r, a);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_store_mask32() {
        let a: __mmask32 = 0b11111111_00000000_11111111_00000000;
        let mut r = 0;
        _store_mask32(&mut r, a);
        assert_eq!(r, a);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_load_mask64() {
        let p: __mmask64 =
            0b11111111_00000000_11111111_00000000_11111111_00000000_11111111_00000000;
        let r = _load_mask64(&p);
        let e: __mmask64 =
            0b11111111_00000000_11111111_00000000_11111111_00000000_11111111_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_load_mask32() {
        let p: __mmask32 = 0b11111111_00000000_11111111_00000000;
        let r = _load_mask32(&p);
        let e: __mmask32 = 0b11111111_00000000_11111111_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_sad_epu8() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(4);
        let r = _mm512_sad_epu8(a, b);
        let e = _mm512_set1_epi64(16);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_dbsad_epu8() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(4);
        let r = _mm512_dbsad_epu8::<0>(a, b);
        let e = _mm512_set1_epi16(8);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_dbsad_epu8() {
        let src = _mm512_set1_epi16(1);
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(4);
        let r = _mm512_mask_dbsad_epu8::<0>(src, 0, a, b);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_dbsad_epu8::<0>(src, 0b11111111_11111111_11111111_11111111, a, b);
        let e = _mm512_set1_epi16(8);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_dbsad_epu8() {
        let a = _mm512_set1_epi8(2);
        let b = _mm512_set1_epi8(4);
        let r = _mm512_maskz_dbsad_epu8::<0>(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_dbsad_epu8::<0>(0b11111111_11111111_11111111_11111111, a, b);
        let e = _mm512_set1_epi16(8);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_dbsad_epu8() {
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(4);
        let r = _mm256_dbsad_epu8::<0>(a, b);
        let e = _mm256_set1_epi16(8);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_dbsad_epu8() {
        let src = _mm256_set1_epi16(1);
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(4);
        let r = _mm256_mask_dbsad_epu8::<0>(src, 0, a, b);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_dbsad_epu8::<0>(src, 0b11111111_11111111, a, b);
        let e = _mm256_set1_epi16(8);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_dbsad_epu8() {
        let a = _mm256_set1_epi8(2);
        let b = _mm256_set1_epi8(4);
        let r = _mm256_maskz_dbsad_epu8::<0>(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_dbsad_epu8::<0>(0b11111111_11111111, a, b);
        let e = _mm256_set1_epi16(8);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_dbsad_epu8() {
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(4);
        let r = _mm_dbsad_epu8::<0>(a, b);
        let e = _mm_set1_epi16(8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_dbsad_epu8() {
        let src = _mm_set1_epi16(1);
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(4);
        let r = _mm_mask_dbsad_epu8::<0>(src, 0, a, b);
        assert_eq_m128i(r, src);
        let r = _mm_mask_dbsad_epu8::<0>(src, 0b11111111, a, b);
        let e = _mm_set1_epi16(8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_dbsad_epu8() {
        let a = _mm_set1_epi8(2);
        let b = _mm_set1_epi8(4);
        let r = _mm_maskz_dbsad_epu8::<0>(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_dbsad_epu8::<0>(0b11111111, a, b);
        let e = _mm_set1_epi16(8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_movepi16_mask() {
        let a = _mm512_set1_epi16(1 << 15);
        let r = _mm512_movepi16_mask(a);
        let e: __mmask32 = 0b11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_movepi16_mask() {
        let a = _mm256_set1_epi16(1 << 15);
        let r = _mm256_movepi16_mask(a);
        let e: __mmask16 = 0b11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_movepi16_mask() {
        let a = _mm_set1_epi16(1 << 15);
        let r = _mm_movepi16_mask(a);
        let e: __mmask8 = 0b11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_movepi8_mask() {
        let a = _mm512_set1_epi8(1 << 7);
        let r = _mm512_movepi8_mask(a);
        let e: __mmask64 =
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_movepi8_mask() {
        let a = _mm256_set1_epi8(1 << 7);
        let r = _mm256_movepi8_mask(a);
        let e: __mmask32 = 0b11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_movepi8_mask() {
        let a = _mm_set1_epi8(1 << 7);
        let r = _mm_movepi8_mask(a);
        let e: __mmask16 = 0b11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_movm_epi16() {
        let a: __mmask32 = 0b11111111_11111111_11111111_11111111;
        let r = _mm512_movm_epi16(a);
        let e = _mm512_set1_epi16(
            1 << 15
                | 1 << 14
                | 1 << 13
                | 1 << 12
                | 1 << 11
                | 1 << 10
                | 1 << 9
                | 1 << 8
                | 1 << 7
                | 1 << 6
                | 1 << 5
                | 1 << 4
                | 1 << 3
                | 1 << 2
                | 1 << 1
                | 1 << 0,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_movm_epi16() {
        let a: __mmask16 = 0b11111111_11111111;
        let r = _mm256_movm_epi16(a);
        let e = _mm256_set1_epi16(
            1 << 15
                | 1 << 14
                | 1 << 13
                | 1 << 12
                | 1 << 11
                | 1 << 10
                | 1 << 9
                | 1 << 8
                | 1 << 7
                | 1 << 6
                | 1 << 5
                | 1 << 4
                | 1 << 3
                | 1 << 2
                | 1 << 1
                | 1 << 0,
        );
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_movm_epi16() {
        let a: __mmask8 = 0b11111111;
        let r = _mm_movm_epi16(a);
        let e = _mm_set1_epi16(
            1 << 15
                | 1 << 14
                | 1 << 13
                | 1 << 12
                | 1 << 11
                | 1 << 10
                | 1 << 9
                | 1 << 8
                | 1 << 7
                | 1 << 6
                | 1 << 5
                | 1 << 4
                | 1 << 3
                | 1 << 2
                | 1 << 1
                | 1 << 0,
        );
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_movm_epi8() {
        let a: __mmask64 =
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111;
        let r = _mm512_movm_epi8(a);
        let e =
            _mm512_set1_epi8(1 << 7 | 1 << 6 | 1 << 5 | 1 << 4 | 1 << 3 | 1 << 2 | 1 << 1 | 1 << 0);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_movm_epi8() {
        let a: __mmask32 = 0b11111111_11111111_11111111_11111111;
        let r = _mm256_movm_epi8(a);
        let e =
            _mm256_set1_epi8(1 << 7 | 1 << 6 | 1 << 5 | 1 << 4 | 1 << 3 | 1 << 2 | 1 << 1 | 1 << 0);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_movm_epi8() {
        let a: __mmask16 = 0b11111111_11111111;
        let r = _mm_movm_epi8(a);
        let e =
            _mm_set1_epi8(1 << 7 | 1 << 6 | 1 << 5 | 1 << 4 | 1 << 3 | 1 << 2 | 1 << 1 | 1 << 0);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_cvtmask32_u32() {
        let a: __mmask32 = 0b11001100_00110011_01100110_10011001;
        let r = _cvtmask32_u32(a);
        let e: u32 = 0b11001100_00110011_01100110_10011001;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_cvtu32_mask32() {
        let a: u32 = 0b11001100_00110011_01100110_10011001;
        let r = _cvtu32_mask32(a);
        let e: __mmask32 = 0b11001100_00110011_01100110_10011001;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kadd_mask32() {
        let a: __mmask32 = 11;
        let b: __mmask32 = 22;
        let r = _kadd_mask32(a, b);
        let e: __mmask32 = 33;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kadd_mask64() {
        let a: __mmask64 = 11;
        let b: __mmask64 = 22;
        let r = _kadd_mask64(a, b);
        let e: __mmask64 = 33;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kand_mask32() {
        let a: __mmask32 = 0b11001100_00110011_11001100_00110011;
        let b: __mmask32 = 0b11001100_00110011_11001100_00110011;
        let r = _kand_mask32(a, b);
        let e: __mmask32 = 0b11001100_00110011_11001100_00110011;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kand_mask64() {
        let a: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        let b: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        let r = _kand_mask64(a, b);
        let e: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_knot_mask32() {
        let a: __mmask32 = 0b11001100_00110011_11001100_00110011;
        let r = _knot_mask32(a);
        let e: __mmask32 = 0b00110011_11001100_00110011_11001100;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_knot_mask64() {
        let a: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        let r = _knot_mask64(a);
        let e: __mmask64 =
            0b00110011_11001100_00110011_11001100_00110011_11001100_00110011_11001100;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kandn_mask32() {
        let a: __mmask32 = 0b11001100_00110011_11001100_00110011;
        let b: __mmask32 = 0b11001100_00110011_11001100_00110011;
        let r = _kandn_mask32(a, b);
        let e: __mmask32 = 0b00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kandn_mask64() {
        let a: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        let b: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        let r = _kandn_mask64(a, b);
        let e: __mmask64 =
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kor_mask32() {
        let a: __mmask32 = 0b00110011_11001100_00110011_11001100;
        let b: __mmask32 = 0b11001100_00110011_11001100_00110011;
        let r = _kor_mask32(a, b);
        let e: __mmask32 = 0b11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kor_mask64() {
        let a: __mmask64 =
            0b00110011_11001100_00110011_11001100_00110011_11001100_00110011_11001100;
        let b: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        let r = _kor_mask64(a, b);
        let e: __mmask64 =
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kxor_mask32() {
        let a: __mmask32 = 0b00110011_11001100_00110011_11001100;
        let b: __mmask32 = 0b11001100_00110011_11001100_00110011;
        let r = _kxor_mask32(a, b);
        let e: __mmask32 = 0b11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kxor_mask64() {
        let a: __mmask64 =
            0b00110011_11001100_00110011_11001100_00110011_11001100_00110011_11001100;
        let b: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        let r = _kxor_mask64(a, b);
        let e: __mmask64 =
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kxnor_mask32() {
        let a: __mmask32 = 0b00110011_11001100_00110011_11001100;
        let b: __mmask32 = 0b11001100_00110011_11001100_00110011;
        let r = _kxnor_mask32(a, b);
        let e: __mmask32 = 0b00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kxnor_mask64() {
        let a: __mmask64 =
            0b00110011_11001100_00110011_11001100_00110011_11001100_00110011_11001100;
        let b: __mmask64 =
            0b11001100_00110011_11001100_00110011_11001100_00110011_11001100_00110011;
        let r = _kxnor_mask64(a, b);
        let e: __mmask64 =
            0b00000000_00000000_00000000_00000000_00000000_00000000_00000000_00000000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kortest_mask32_u8() {
        let a: __mmask32 = 0b0110100101101001_0110100101101001;
        let b: __mmask32 = 0b1011011010110110_1011011010110110;
        let mut all_ones: u8 = 0;
        let r = _kortest_mask32_u8(a, b, &mut all_ones);
        assert_eq!(r, 0);
        assert_eq!(all_ones, 1);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kortest_mask64_u8() {
        let a: __mmask64 = 0b0110100101101001_0110100101101001;
        let b: __mmask64 = 0b1011011010110110_1011011010110110;
        let mut all_ones: u8 = 0;
        let r = _kortest_mask64_u8(a, b, &mut all_ones);
        assert_eq!(r, 0);
        assert_eq!(all_ones, 0);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kortestc_mask32_u8() {
        let a: __mmask32 = 0b0110100101101001_0110100101101001;
        let b: __mmask32 = 0b1011011010110110_1011011010110110;
        let r = _kortestc_mask32_u8(a, b);
        assert_eq!(r, 1);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kortestc_mask64_u8() {
        let a: __mmask64 = 0b0110100101101001_0110100101101001;
        let b: __mmask64 = 0b1011011010110110_1011011010110110;
        let r = _kortestc_mask64_u8(a, b);
        assert_eq!(r, 0);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kortestz_mask32_u8() {
        let a: __mmask32 = 0b0110100101101001_0110100101101001;
        let b: __mmask32 = 0b1011011010110110_1011011010110110;
        let r = _kortestz_mask32_u8(a, b);
        assert_eq!(r, 0);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kortestz_mask64_u8() {
        let a: __mmask64 = 0b0110100101101001_0110100101101001;
        let b: __mmask64 = 0b1011011010110110_1011011010110110;
        let r = _kortestz_mask64_u8(a, b);
        assert_eq!(r, 0);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kshiftli_mask32() {
        let a: __mmask32 = 0b0110100101101001_0110100101101001;
        let r = _kshiftli_mask32::<3>(a);
        let e: __mmask32 = 0b0100101101001011_0100101101001000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kshiftli_mask64() {
        let a: __mmask64 = 0b0110100101101001_0110100101101001;
        let r = _kshiftli_mask64::<3>(a);
        let e: __mmask64 = 0b0110100101101001011_0100101101001000;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kshiftri_mask32() {
        let a: __mmask32 = 0b0110100101101001_0110100101101001;
        let r = _kshiftri_mask32::<3>(a);
        let e: __mmask32 = 0b0000110100101101_0010110100101101;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_kshiftri_mask64() {
        let a: __mmask64 = 0b0110100101101001011_0100101101001000;
        let r = _kshiftri_mask64::<3>(a);
        let e: __mmask64 = 0b0110100101101001_0110100101101001;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_ktest_mask32_u8() {
        let a: __mmask32 = 0b0110100100111100_0110100100111100;
        let b: __mmask32 = 0b1001011011000011_1001011011000011;
        let mut and_not: u8 = 0;
        let r = _ktest_mask32_u8(a, b, &mut and_not);
        assert_eq!(r, 1);
        assert_eq!(and_not, 0);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_ktestc_mask32_u8() {
        let a: __mmask32 = 0b0110100100111100_0110100100111100;
        let b: __mmask32 = 0b1001011011000011_1001011011000011;
        let r = _ktestc_mask32_u8(a, b);
        assert_eq!(r, 0);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_ktestz_mask32_u8() {
        let a: __mmask32 = 0b0110100100111100_0110100100111100;
        let b: __mmask32 = 0b1001011011000011_1001011011000011;
        let r = _ktestz_mask32_u8(a, b);
        assert_eq!(r, 1);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_ktest_mask64_u8() {
        let a: __mmask64 = 0b0110100100111100_0110100100111100;
        let b: __mmask64 = 0b1001011011000011_1001011011000011;
        let mut and_not: u8 = 0;
        let r = _ktest_mask64_u8(a, b, &mut and_not);
        assert_eq!(r, 1);
        assert_eq!(and_not, 0);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_ktestc_mask64_u8() {
        let a: __mmask64 = 0b0110100100111100_0110100100111100;
        let b: __mmask64 = 0b1001011011000011_1001011011000011;
        let r = _ktestc_mask64_u8(a, b);
        assert_eq!(r, 0);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_ktestz_mask64_u8() {
        let a: __mmask64 = 0b0110100100111100_0110100100111100;
        let b: __mmask64 = 0b1001011011000011_1001011011000011;
        let r = _ktestz_mask64_u8(a, b);
        assert_eq!(r, 1);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_kunpackw() {
        let a: u32 = 0x00110011;
        let b: u32 = 0x00001011;
        let r = _mm512_kunpackw(a, b);
        let e: u32 = 0x00111011;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_kunpackd() {
        let a: u64 = 0x11001100_00110011;
        let b: u64 = 0x00101110_00001011;
        let r = _mm512_kunpackd(a, b);
        let e: u64 = 0x00110011_00001011;
        assert_eq!(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cvtepi16_epi8() {
        let a = _mm512_set1_epi16(2);
        let r = _mm512_cvtepi16_epi8(a);
        let e = _mm256_set1_epi8(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cvtepi16_epi8() {
        let src = _mm256_set1_epi8(1);
        let a = _mm512_set1_epi16(2);
        let r = _mm512_mask_cvtepi16_epi8(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm512_mask_cvtepi16_epi8(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_cvtepi16_epi8() {
        let a = _mm512_set1_epi16(2);
        let r = _mm512_maskz_cvtepi16_epi8(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm512_maskz_cvtepi16_epi8(0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cvtepi16_epi8() {
        let a = _mm256_set1_epi16(2);
        let r = _mm256_cvtepi16_epi8(a);
        let e = _mm_set1_epi8(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cvtepi16_epi8() {
        let src = _mm_set1_epi8(1);
        let a = _mm256_set1_epi16(2);
        let r = _mm256_mask_cvtepi16_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm256_mask_cvtepi16_epi8(src, 0b11111111_11111111, a);
        let e = _mm_set1_epi8(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_cvtepi16_epi8() {
        let a = _mm256_set1_epi16(2);
        let r = _mm256_maskz_cvtepi16_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm256_maskz_cvtepi16_epi8(0b11111111_11111111, a);
        let e = _mm_set1_epi8(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cvtepi16_epi8() {
        let a = _mm_set1_epi16(2);
        let r = _mm_cvtepi16_epi8(a);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 2, 2, 2, 2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cvtepi16_epi8() {
        let src = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1);
        let a = _mm_set1_epi16(2);
        let r = _mm_mask_cvtepi16_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_cvtepi16_epi8(src, 0b11111111, a);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 2, 2, 2, 2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_cvtepi16_epi8() {
        let a = _mm_set1_epi16(2);
        let r = _mm_maskz_cvtepi16_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_cvtepi16_epi8(0b11111111, a);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 2, 2, 2, 2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cvtsepi16_epi8() {
        let a = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_cvtsepi16_epi8(a);
        let e = _mm256_set1_epi8(i8::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cvtsepi16_epi8() {
        let src = _mm256_set1_epi8(1);
        let a = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_mask_cvtsepi16_epi8(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm512_mask_cvtsepi16_epi8(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(i8::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cvtsepi16_epi8() {
        let a = _mm256_set1_epi16(i16::MAX);
        let r = _mm256_cvtsepi16_epi8(a);
        let e = _mm_set1_epi8(i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cvtsepi16_epi8() {
        let src = _mm_set1_epi8(1);
        let a = _mm256_set1_epi16(i16::MAX);
        let r = _mm256_mask_cvtsepi16_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm256_mask_cvtsepi16_epi8(src, 0b11111111_11111111, a);
        let e = _mm_set1_epi8(i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_cvtsepi16_epi8() {
        let a = _mm256_set1_epi16(i16::MAX);
        let r = _mm256_maskz_cvtsepi16_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm256_maskz_cvtsepi16_epi8(0b11111111_11111111, a);
        let e = _mm_set1_epi8(i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cvtsepi16_epi8() {
        let a = _mm_set1_epi16(i16::MAX);
        let r = _mm_cvtsepi16_epi8(a);
        #[rustfmt::skip]
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cvtsepi16_epi8() {
        let src = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1);
        let a = _mm_set1_epi16(i16::MAX);
        let r = _mm_mask_cvtsepi16_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_cvtsepi16_epi8(src, 0b11111111, a);
        #[rustfmt::skip]
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_cvtsepi16_epi8() {
        let a = _mm_set1_epi16(i16::MAX);
        let r = _mm_maskz_cvtsepi16_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_cvtsepi16_epi8(0b11111111, a);
        #[rustfmt::skip]
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_cvtsepi16_epi8() {
        let a = _mm512_set1_epi16(i16::MAX);
        let r = _mm512_maskz_cvtsepi16_epi8(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm512_maskz_cvtsepi16_epi8(0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(i8::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cvtusepi16_epi8() {
        let a = _mm512_set1_epi16(i16::MIN);
        let r = _mm512_cvtusepi16_epi8(a);
        let e = _mm256_set1_epi8(-1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cvtusepi16_epi8() {
        let src = _mm256_set1_epi8(1);
        let a = _mm512_set1_epi16(i16::MIN);
        let r = _mm512_mask_cvtusepi16_epi8(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm512_mask_cvtusepi16_epi8(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(-1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_cvtusepi16_epi8() {
        let a = _mm512_set1_epi16(i16::MIN);
        let r = _mm512_maskz_cvtusepi16_epi8(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm512_maskz_cvtusepi16_epi8(0b11111111_11111111_11111111_11111111, a);
        let e = _mm256_set1_epi8(-1);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_cvtusepi16_epi8() {
        let a = _mm256_set1_epi16(i16::MIN);
        let r = _mm256_cvtusepi16_epi8(a);
        let e = _mm_set1_epi8(-1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cvtusepi16_epi8() {
        let src = _mm_set1_epi8(1);
        let a = _mm256_set1_epi16(i16::MIN);
        let r = _mm256_mask_cvtusepi16_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm256_mask_cvtusepi16_epi8(src, 0b11111111_11111111, a);
        let e = _mm_set1_epi8(-1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_cvtusepi16_epi8() {
        let a = _mm256_set1_epi16(i16::MIN);
        let r = _mm256_maskz_cvtusepi16_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm256_maskz_cvtusepi16_epi8(0b11111111_11111111, a);
        let e = _mm_set1_epi8(-1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_cvtusepi16_epi8() {
        let a = _mm_set1_epi16(i16::MIN);
        let r = _mm_cvtusepi16_epi8(a);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cvtusepi16_epi8() {
        let src = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1);
        let a = _mm_set1_epi16(i16::MIN);
        let r = _mm_mask_cvtusepi16_epi8(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_cvtusepi16_epi8(src, 0b11111111, a);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_cvtusepi16_epi8() {
        let a = _mm_set1_epi16(i16::MIN);
        let r = _mm_maskz_cvtusepi16_epi8(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_cvtusepi16_epi8(0b11111111, a);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1, -1, -1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cvtepi8_epi16() {
        let a = _mm256_set1_epi8(2);
        let r = _mm512_cvtepi8_epi16(a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cvtepi8_epi16() {
        let src = _mm512_set1_epi16(1);
        let a = _mm256_set1_epi8(2);
        let r = _mm512_mask_cvtepi8_epi16(src, 0, a);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_cvtepi8_epi16(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_cvtepi8_epi16() {
        let a = _mm256_set1_epi8(2);
        let r = _mm512_maskz_cvtepi8_epi16(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_cvtepi8_epi16(0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cvtepi8_epi16() {
        let src = _mm256_set1_epi16(1);
        let a = _mm_set1_epi8(2);
        let r = _mm256_mask_cvtepi8_epi16(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_cvtepi8_epi16(src, 0b11111111_11111111, a);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_cvtepi8_epi16() {
        let a = _mm_set1_epi8(2);
        let r = _mm256_maskz_cvtepi8_epi16(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_cvtepi8_epi16(0b11111111_11111111, a);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cvtepi8_epi16() {
        let src = _mm_set1_epi16(1);
        let a = _mm_set1_epi8(2);
        let r = _mm_mask_cvtepi8_epi16(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_cvtepi8_epi16(src, 0b11111111, a);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_cvtepi8_epi16() {
        let a = _mm_set1_epi8(2);
        let r = _mm_maskz_cvtepi8_epi16(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_cvtepi8_epi16(0b11111111, a);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_cvtepu8_epi16() {
        let a = _mm256_set1_epi8(2);
        let r = _mm512_cvtepu8_epi16(a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cvtepu8_epi16() {
        let src = _mm512_set1_epi16(1);
        let a = _mm256_set1_epi8(2);
        let r = _mm512_mask_cvtepu8_epi16(src, 0, a);
        assert_eq_m512i(r, src);
        let r = _mm512_mask_cvtepu8_epi16(src, 0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_cvtepu8_epi16() {
        let a = _mm256_set1_epi8(2);
        let r = _mm512_maskz_cvtepu8_epi16(0, a);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_cvtepu8_epi16(0b11111111_11111111_11111111_11111111, a);
        let e = _mm512_set1_epi16(2);
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cvtepu8_epi16() {
        let src = _mm256_set1_epi16(1);
        let a = _mm_set1_epi8(2);
        let r = _mm256_mask_cvtepu8_epi16(src, 0, a);
        assert_eq_m256i(r, src);
        let r = _mm256_mask_cvtepu8_epi16(src, 0b11111111_11111111, a);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_cvtepu8_epi16() {
        let a = _mm_set1_epi8(2);
        let r = _mm256_maskz_cvtepu8_epi16(0, a);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_cvtepu8_epi16(0b11111111_11111111, a);
        let e = _mm256_set1_epi16(2);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cvtepu8_epi16() {
        let src = _mm_set1_epi16(1);
        let a = _mm_set1_epi8(2);
        let r = _mm_mask_cvtepu8_epi16(src, 0, a);
        assert_eq_m128i(r, src);
        let r = _mm_mask_cvtepu8_epi16(src, 0b11111111, a);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_cvtepu8_epi16() {
        let a = _mm_set1_epi8(2);
        let r = _mm_maskz_cvtepu8_epi16(0, a);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_cvtepu8_epi16(0b11111111, a);
        let e = _mm_set1_epi16(2);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_bslli_epi128() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
        );
        let r = _mm512_bslli_epi128::<9>(a);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(
            0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
            0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
            0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
            0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_bsrli_epi128() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(
            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, 40, 41, 42, 43, 44, 45, 46, 47, 48,
            49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,
        );
        let r = _mm512_bsrli_epi128::<3>(a);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(
            0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,
            0, 0, 0, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29,
            0, 0, 0, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45,
            0, 0, 0, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_alignr_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
        );
        let b = _mm512_set1_epi8(1);
        let r = _mm512_alignr_epi8::<14>(a, b);
        #[rustfmt::skip]
        let e = _mm512_set_epi8(
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_alignr_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
        );
        let b = _mm512_set1_epi8(1);
        let r = _mm512_mask_alignr_epi8::<14>(a, 0, a, b);
        assert_eq_m512i(r, a);
        let r = _mm512_mask_alignr_epi8::<14>(
            a,
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_maskz_alignr_epi8() {
        #[rustfmt::skip]
        let a = _mm512_set_epi8(
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
        );
        let b = _mm512_set1_epi8(1);
        let r = _mm512_maskz_alignr_epi8::<14>(0, a, b);
        assert_eq_m512i(r, _mm512_setzero_si512());
        let r = _mm512_maskz_alignr_epi8::<14>(
            0b11111111_11111111_11111111_11111111_11111111_11111111_11111111_11111111,
            a,
            b,
        );
        #[rustfmt::skip]
        let e = _mm512_set_epi8(
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
        );
        assert_eq_m512i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_alignr_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
        );
        let b = _mm256_set1_epi8(1);
        let r = _mm256_mask_alignr_epi8::<14>(a, 0, a, b);
        assert_eq_m256i(r, a);
        let r = _mm256_mask_alignr_epi8::<14>(a, 0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
        );
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_maskz_alignr_epi8() {
        #[rustfmt::skip]
        let a = _mm256_set_epi8(
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
            1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,
        );
        let b = _mm256_set1_epi8(1);
        let r = _mm256_maskz_alignr_epi8::<14>(0, a, b);
        assert_eq_m256i(r, _mm256_setzero_si256());
        let r = _mm256_maskz_alignr_epi8::<14>(0b11111111_11111111_11111111_11111111, a, b);
        #[rustfmt::skip]
        let e = _mm256_set_epi8(
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
            0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1,
        );
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_alignr_epi8() {
        let a = _mm_set_epi8(1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0);
        let b = _mm_set1_epi8(1);
        let r = _mm_mask_alignr_epi8::<14>(a, 0, a, b);
        assert_eq_m128i(r, a);
        let r = _mm_mask_alignr_epi8::<14>(a, 0b11111111_11111111, a, b);
        let e = _mm_set_epi8(0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_maskz_alignr_epi8() {
        let a = _mm_set_epi8(1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0);
        let b = _mm_set1_epi8(1);
        let r = _mm_maskz_alignr_epi8::<14>(0, a, b);
        assert_eq_m128i(r, _mm_setzero_si128());
        let r = _mm_maskz_alignr_epi8::<14>(0b11111111_11111111, a, b);
        let e = _mm_set_epi8(0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cvtsepi16_storeu_epi8() {
        let a = _mm512_set1_epi16(i16::MAX);
        let mut r = _mm256_undefined_si256();
        _mm512_mask_cvtsepi16_storeu_epi8(
            &mut r as *mut _ as *mut i8,
            0b11111111_11111111_11111111_11111111,
            a,
        );
        let e = _mm256_set1_epi8(i8::MAX);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cvtsepi16_storeu_epi8() {
        let a = _mm256_set1_epi16(i16::MAX);
        let mut r = _mm_undefined_si128();
        _mm256_mask_cvtsepi16_storeu_epi8(&mut r as *mut _ as *mut i8, 0b11111111_11111111, a);
        let e = _mm_set1_epi8(i8::MAX);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cvtsepi16_storeu_epi8() {
        let a = _mm_set1_epi16(i16::MAX);
        let mut r = _mm_set1_epi8(0);
        _mm_mask_cvtsepi16_storeu_epi8(&mut r as *mut _ as *mut i8, 0b11111111, a);
        #[rustfmt::skip]
        let e = _mm_set_epi8(
            0, 0, 0, 0, 0, 0, 0, 0,
            i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX, i8::MAX,
        );
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cvtepi16_storeu_epi8() {
        let a = _mm512_set1_epi16(8);
        let mut r = _mm256_undefined_si256();
        _mm512_mask_cvtepi16_storeu_epi8(
            &mut r as *mut _ as *mut i8,
            0b11111111_11111111_11111111_11111111,
            a,
        );
        let e = _mm256_set1_epi8(8);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cvtepi16_storeu_epi8() {
        let a = _mm256_set1_epi16(8);
        let mut r = _mm_undefined_si128();
        _mm256_mask_cvtepi16_storeu_epi8(&mut r as *mut _ as *mut i8, 0b11111111_11111111, a);
        let e = _mm_set1_epi8(8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cvtepi16_storeu_epi8() {
        let a = _mm_set1_epi16(8);
        let mut r = _mm_set1_epi8(0);
        _mm_mask_cvtepi16_storeu_epi8(&mut r as *mut _ as *mut i8, 0b11111111, a);
        let e = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0, 8, 8, 8, 8, 8, 8, 8, 8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw")]
    unsafe fn test_mm512_mask_cvtusepi16_storeu_epi8() {
        let a = _mm512_set1_epi16(i16::MAX);
        let mut r = _mm256_undefined_si256();
        _mm512_mask_cvtusepi16_storeu_epi8(
            &mut r as *mut _ as *mut i8,
            0b11111111_11111111_11111111_11111111,
            a,
        );
        let e = _mm256_set1_epi8(u8::MAX as i8);
        assert_eq_m256i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm256_mask_cvtusepi16_storeu_epi8() {
        let a = _mm256_set1_epi16(i16::MAX);
        let mut r = _mm_undefined_si128();
        _mm256_mask_cvtusepi16_storeu_epi8(&mut r as *mut _ as *mut i8, 0b11111111_11111111, a);
        let e = _mm_set1_epi8(u8::MAX as i8);
        assert_eq_m128i(r, e);
    }
    #[simd_test(enable = "avx512bw,avx512vl")]
    unsafe fn test_mm_mask_cvtusepi16_storeu_epi8() {
        let a = _mm_set1_epi16(i16::MAX);
        let mut r = _mm_set1_epi8(0);
        _mm_mask_cvtusepi16_storeu_epi8(&mut r as *mut _ as *mut i8, 0b11111111, a);
        #[rustfmt::skip]
        let e = _mm_set_epi8(
            0, 0, 0, 0,
            0, 0, 0, 0,
            u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, 
            u8::MAX as i8, u8::MAX as i8, u8::MAX as i8, u8::MAX as i8,
        );
        assert_eq_m128i(r, e);
    }
}