Module core::core_arch::x86::avx512f

source ·
🔬This is a nightly-only experimental API. (stdsimd #48556)
Available on x86 or x86-64 only.

Macros

  • vpl 🔒 Experimental
  • vps 🔒 Experimental

Constants

Functions

  • _kand_mask16Experimentalavx512f
    Compute the bitwise AND of 16-bit masks a and b, and store the result in k.
  • _kandn_mask16Experimentalavx512f
    Compute the bitwise NOT of 16-bit masks a and then AND with b, and store the result in k.
  • _knot_mask16Experimentalavx512f
    Compute the bitwise NOT of 16-bit mask a, and store the result in k.
  • _kor_mask16Experimentalavx512f
    Compute the bitwise OR of 16-bit masks a and b, and store the result in k.
  • _kxnor_mask16Experimentalavx512f
    Compute the bitwise XNOR of 16-bit masks a and b, and store the result in k.
  • _kxor_mask16Experimentalavx512f
    Compute the bitwise XOR of 16-bit masks a and b, and store the result in k.
  • _mm256_abs_epi64Experimentalavx512f,avx512vl
    Compute the absolute value of packed signed 64-bit integers in a, and store the unsigned results in dst.
  • _mm256_alignr_epi32Experimentalavx512f,avx512vl
    Concatenate a and b into a 64-byte immediate result, shift the result right by imm8 32-bit elements, and store the low 32 bytes (8 elements) in dst.
  • _mm256_alignr_epi64Experimentalavx512f,avx512vl
    Concatenate a and b into a 64-byte immediate result, shift the result right by imm8 64-bit elements, and store the low 32 bytes (4 elements) in dst.
  • _mm256_broadcast_f32x4Experimentalavx512f,avx512vl
    Broadcast the 4 packed single-precision (32-bit) floating-point elements from a to all elements of dst.
  • _mm256_broadcast_i32x4Experimentalavx512f,avx512vl
    Broadcast the 4 packed 32-bit integers from a to all elements of dst.
  • _mm256_cmp_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm256_cmp_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm256_cmp_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm256_cmp_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm256_cmp_pd_maskExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm256_cmp_ps_maskExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm256_cmpeq_epi32_maskExperimentalavx512f,avx512vl
    Compare packed 32-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm256_cmpeq_epi64_maskExperimentalavx512f,avx512vl
    Compare packed 64-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm256_cmpeq_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm256_cmpeq_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm256_cmpge_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm256_cmpge_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm256_cmpge_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm256_cmpge_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm256_cmpgt_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm256_cmpgt_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm256_cmpgt_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm256_cmpgt_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm256_cmple_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm256_cmple_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm256_cmple_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm256_cmple_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm256_cmplt_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm256_cmplt_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm256_cmplt_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm256_cmplt_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm256_cmpneq_epi32_maskExperimentalavx512f,avx512vl
    Compare packed 32-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm256_cmpneq_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm256_cmpneq_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm256_cmpneq_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm256_cvtepi32_epi8Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst.
  • _mm256_cvtepi32_epi16Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst.
  • _mm256_cvtepi64_epi8Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst.
  • _mm256_cvtepi64_epi16Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst.
  • _mm256_cvtepi64_epi32Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst.
  • _mm256_cvtepu32_pdExperimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst.
  • _mm256_cvtpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst.
  • _mm256_cvtps_epu32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst.
  • _mm256_cvtsepi32_epi8Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst.
  • _mm256_cvtsepi32_epi16Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm256_cvtsepi64_epi8Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst.
  • _mm256_cvtsepi64_epi16Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm256_cvtsepi64_epi32Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst.
  • _mm256_cvttpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst.
  • _mm256_cvttps_epu32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst.
  • _mm256_cvtusepi32_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst.
  • _mm256_cvtusepi32_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst.
  • _mm256_cvtusepi64_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst.
  • _mm256_cvtusepi64_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst.
  • _mm256_cvtusepi64_epi32Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst.
  • _mm256_extractf32x4_psExperimentalavx512f,avx512vl
    Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from a, selected with imm8, and store the result in dst.
  • _mm256_extracti32x4_epi32Experimentalavx512f,avx512vl
    Extract 128 bits (composed of 4 packed 32-bit integers) from a, selected with IMM1, and store the result in dst.
  • _mm256_fixupimm_pdExperimentalavx512f,avx512vl
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst. imm8 is used to set the required flags reporting.
  • _mm256_fixupimm_psExperimentalavx512f,avx512vl
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst. imm8 is used to set the required flags reporting.
  • _mm256_getexp_pdExperimentalavx512f,avx512vl
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst. This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm256_getexp_psExperimentalavx512f,avx512vl
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst. This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm256_getmant_pdExperimentalavx512f,avx512vl
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm256_getmant_psExperimentalavx512f,avx512vl
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign. The mantissa is normalized to the interval specified by interv, which can take the following values: _MM_MANT_NORM_1_2 // interval [1, 2) _MM_MANT_NORM_p5_2 // interval [0.5, 2) _MM_MANT_NORM_p5_1 // interval [0.5, 1) _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5) The sign is determined by sc which can take the following values: _MM_MANT_SIGN_src // sign = sign(src) _MM_MANT_SIGN_zero // sign = 0 _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm256_i32scatter_epi64Experimentalavx512f,avx512vl
    Scatter 64-bit integers from a into memory using 32-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). scale should be 1, 2, 4 or 8.
  • _mm256_insertf32x4Experimentalavx512f,avx512vl
    Copy a to dst, then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from b into dst at the location specified by imm8.
  • _mm256_inserti32x4Experimentalavx512f,avx512vl
    Copy a to dst, then insert 128 bits (composed of 4 packed 32-bit integers) from b into dst at the location specified by imm8.
  • _mm256_load_epi32Experimentalavx512f,avx512vl
    Load 256-bits (composed of 8 packed 32-bit integers) from memory into dst. mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_load_epi64Experimentalavx512f,avx512vl
    Load 256-bits (composed of 4 packed 64-bit integers) from memory into dst. mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_loadu_epi32Experimentalavx512f,avx512vl
    Load 256-bits (composed of 8 packed 32-bit integers) from memory into dst. mem_addr does not need to be aligned on any particular boundary.
  • _mm256_loadu_epi64Experimentalavx512f,avx512vl
    Load 256-bits (composed of 4 packed 64-bit integers) from memory into dst. mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask2_permutex2var_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm256_mask2_permutex2var_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm256_mask2_permutex2var_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set)
  • _mm256_mask2_permutex2var_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm256_mask3_fmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fmaddsub_pdExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fmaddsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fmsubadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fmsubadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fnmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fnmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fnmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask3_fnmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm256_mask_abs_epi32Experimentalavx512f,avx512vl
    Compute the absolute value of packed signed 32-bit integers in a, and store the unsigned results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_abs_epi64Experimentalavx512f,avx512vl
    Compute the absolute value of packed signed 64-bit integers in a, and store the unsigned results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_add_epi32Experimentalavx512f,avx512vl
    Add packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_add_epi64Experimentalavx512f,avx512vl
    Add packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_add_pdExperimentalavx512f,avx512vl
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_add_psExperimentalavx512f,avx512vl
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_alignr_epi32Experimentalavx512f,avx512vl
    Concatenate a and b into a 64-byte immediate result, shift the result right by imm8 32-bit elements, and store the low 32 bytes (8 elements) in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_alignr_epi64Experimentalavx512f,avx512vl
    Concatenate a and b into a 64-byte immediate result, shift the result right by imm8 64-bit elements, and store the low 32 bytes (4 elements) in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_and_epi32Experimentalavx512f,avx512vl
    Performs element-by-element bitwise AND between packed 32-bit integer elements of a and b, storing the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_and_epi64Experimentalavx512f,avx512vl
    Compute the bitwise AND of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_andnot_epi32Experimentalavx512f,avx512vl
    Compute the bitwise NOT of packed 32-bit integers in a and then AND with b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_andnot_epi64Experimentalavx512f,avx512vl
    Compute the bitwise NOT of packed 64-bit integers in a and then AND with b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_blend_epi32Experimentalavx512f,avx512vl
    Blend packed 32-bit integers from a and b using control mask k, and store the results in dst.
  • _mm256_mask_blend_epi64Experimentalavx512f,avx512vl
    Blend packed 64-bit integers from a and b using control mask k, and store the results in dst.
  • _mm256_mask_blend_pdExperimentalavx512f,avx512vl
    Blend packed double-precision (64-bit) floating-point elements from a and b using control mask k, and store the results in dst.
  • _mm256_mask_blend_psExperimentalavx512f,avx512vl
    Blend packed single-precision (32-bit) floating-point elements from a and b using control mask k, and store the results in dst.
  • _mm256_mask_broadcast_f32x4Experimentalavx512f,avx512vl
    Broadcast the 4 packed single-precision (32-bit) floating-point elements from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_broadcast_i32x4Experimentalavx512f,avx512vl
    Broadcast the 4 packed 32-bit integers from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_broadcastd_epi32Experimentalavx512f,avx512vl
    Broadcast the low packed 32-bit integer from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_broadcastq_epi64Experimentalavx512f,avx512vl
    Broadcast the low packed 64-bit integer from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_broadcastsd_pdExperimentalavx512f,avx512vl
    Broadcast the low double-precision (64-bit) floating-point element from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_broadcastss_psExperimentalavx512f,avx512vl
    Broadcast the low single-precision (32-bit) floating-point element from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cmp_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmp_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmp_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmp_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmp_pd_maskExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmp_ps_maskExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpeq_epi32_maskExperimentalavx512f,avx512vl
    Compare packed 32-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpeq_epi64_maskExperimentalavx512f,avx512vl
    Compare packed 64-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpeq_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpeq_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpge_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpge_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpge_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpge_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpgt_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpgt_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpgt_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpgt_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmple_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmple_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmple_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmple_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmplt_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmplt_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmplt_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmplt_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpneq_epi32_maskExperimentalavx512f,avx512vl
    Compare packed 32-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpneq_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpneq_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_cmpneq_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_mask_compress_epi32Experimentalavx512f,avx512vl
    Contiguously store the active 32-bit integers in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm256_mask_compress_epi64Experimentalavx512f,avx512vl
    Contiguously store the active 64-bit integers in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm256_mask_compress_pdExperimentalavx512f,avx512vl
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm256_mask_compress_psExperimentalavx512f,avx512vl
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm256_mask_compressstoreu_epi32Experimentalavx512f,avx512vl
    Contiguously store the active 32-bit integers in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_compressstoreu_epi64Experimentalavx512f,avx512vl
    Contiguously store the active 64-bit integers in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_compressstoreu_pdExperimentalavx512f,avx512vl
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_compressstoreu_psExperimentalavx512f,avx512vl
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvt_roundps_phExperimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of: (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_mask_cvtepi8_epi32Experimentalavx512f,avx512vl
    Sign extend packed 8-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi8_epi64Experimentalavx512f,avx512vl
    Sign extend packed 8-bit integers in the low 4 bytes of a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi16_epi32Experimentalavx512f,avx512vl
    Sign extend packed 16-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi16_epi64Experimentalavx512f,avx512vl
    Sign extend packed 16-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi32_epi8Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi32_epi16Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi32_epi64Experimentalavx512f,avx512vl
    Sign extend packed 32-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi32_pdExperimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi32_psExperimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi32_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtepi32_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtepi64_epi8Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi64_epi16Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi64_epi32Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepi64_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtepi64_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtepi64_storeu_epi32Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtepu8_epi32Experimentalavx512f,avx512vl
    Zero extend packed unsigned 8-bit integers in the low 8 bytes of a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepu8_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 8-bit integers in the low 4 bytes of a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepu16_epi32Experimentalavx512f,avx512vl
    Zero extend packed unsigned 16-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepu16_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 16-bit integers in the low 8 bytes of a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepu32_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 32-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtepu32_pdExperimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtpd_epi32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtpd_psExperimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtph_psExperimentalavx512f,avx512vl
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtps_epi32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtps_epu32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtps_phExperimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_mask_cvtsepi32_epi8Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtsepi32_epi16Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtsepi32_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtsepi32_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtsepi64_epi8Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtsepi64_epi16Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtsepi64_epi32Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtsepi64_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtsepi64_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtsepi64_storeu_epi32Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvttpd_epi32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvttpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvttps_epi32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvttps_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtusepi32_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtusepi32_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtusepi32_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtusepi32_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtusepi64_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtusepi64_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtusepi64_epi32Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_cvtusepi64_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtusepi64_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_cvtusepi64_storeu_epi32Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm256_mask_div_pdExperimentalavx512f,avx512vl
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_div_psExperimentalavx512f,avx512vl
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_expand_epi32Experimentalavx512f,avx512vl
    Load contiguous active 32-bit integers from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_expand_epi64Experimentalavx512f,avx512vl
    Load contiguous active 64-bit integers from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_expand_pdExperimentalavx512f,avx512vl
    Load contiguous active double-precision (64-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_expand_psExperimentalavx512f,avx512vl
    Load contiguous active single-precision (32-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_expandloadu_epi32Experimentalavx512f,avx512vl,avx
    Load contiguous active 32-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_expandloadu_epi64Experimentalavx512f,avx512vl,avx
    Load contiguous active 64-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_expandloadu_pdExperimentalavx512f,avx512vl,avx
    Load contiguous active single-precision (64-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_expandloadu_psExperimentalavx512f,avx512vl,avx
    Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_extractf32x4_psExperimentalavx512f,avx512vl
    Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from a, selected with imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_extracti32x4_epi32Experimentalavx512f,avx512vl
    Extract 128 bits (composed of 4 packed 32-bit integers) from a, selected with IMM1, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_fixupimm_pdExperimentalavx512f,avx512vl
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm256_mask_fixupimm_psExperimentalavx512f,avx512vl
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm256_mask_fmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fmaddsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fmaddsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fmsubadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fmsubadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fnmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fnmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fnmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_fnmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_getexp_pdExperimentalavx512f,avx512vl
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm256_mask_getexp_psExperimentalavx512f,avx512vl
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm256_mask_getmant_pdExperimentalavx512f,avx512vl
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm256_mask_getmant_psExperimentalavx512f,avx512vl
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm256_mask_insertf32x4Experimentalavx512f,avx512vl
    Copy a to tmp, then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from b into tmp at the location specified by imm8. Store tmp to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_inserti32x4Experimentalavx512f,avx512vl
    Copy a to tmp, then insert 128 bits (composed of 4 packed 32-bit integers) from b into tmp at the location specified by imm8. Store tmp to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_load_epi32Experimentalavx512f,avx512vl,avx
    Load packed 32-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_mask_load_epi64Experimentalavx512f,avx512vl,avx
    Load packed 64-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_mask_load_pdExperimentalavx512f,avx512vl,avx
    Load packed double-precision (64-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_mask_load_psExperimentalavx512f,avx512vl,avx
    Load packed single-precision (32-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_mask_loadu_epi32Experimentalavx512f,avx512vl,avx
    Load packed 32-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask_loadu_epi64Experimentalavx512f,avx512vl,avx
    Load packed 64-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask_loadu_pdExperimentalavx512f,avx512vl,avx
    Load packed double-precision (64-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask_loadu_psExperimentalavx512f,avx512vl,avx
    Load packed single-precision (32-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask_max_epi32Experimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_max_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_max_epu32Experimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_max_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_max_pdExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_max_psExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_min_epi32Experimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_min_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_min_epu32Experimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_min_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_min_pdExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_min_psExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mov_epi32Experimentalavx512f,avx512vl
    Move packed 32-bit integers from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mov_epi64Experimentalavx512f,avx512vl
    Move packed 64-bit integers from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mov_pdExperimentalavx512f,avx512vl
    Move packed double-precision (64-bit) floating-point elements from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mov_psExperimentalavx512f,avx512vl
    Move packed single-precision (32-bit) floating-point elements from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_movedup_pdExperimentalavx512f,avx512vl
    Duplicate even-indexed double-precision (64-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_movehdup_psExperimentalavx512f,avx512vl
    Duplicate odd-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_moveldup_psExperimentalavx512f,avx512vl
    Duplicate even-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mul_epi32Experimentalavx512f,avx512vl
    Multiply the low signed 32-bit integers from each packed 64-bit element in a and b, and store the signed 64-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mul_epu32Experimentalavx512f,avx512vl
    Multiply the low unsigned 32-bit integers from each packed 64-bit element in a and b, and store the unsigned 64-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mul_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mul_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_mullo_epi32Experimentalavx512f,avx512vl
    Multiply the packed 32-bit integers in a and b, producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_or_epi32Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_or_epi64Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permute_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permute_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permutevar_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permutevar_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permutex2var_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_permutex2var_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_permutex2var_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_permutex2var_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm256_mask_permutex_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a within 256-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permutex_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 256-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permutexvar_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permutexvar_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permutexvar_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_permutexvar_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_rcp14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm256_mask_rcp14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm256_mask_rol_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_rol_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_rolv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_rolv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_ror_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_ror_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_rorv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_rorv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_roundscale_pdExperimentalavx512f,avx512vl
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_mask_roundscale_psExperimentalavx512f,avx512vl
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_mask_rsqrt14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm256_mask_rsqrt14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm256_mask_scalef_pdExperimentalavx512f,avx512vl
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_scalef_psExperimentalavx512f,avx512vl
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_set1_epi32Experimentalavx512f,avx512vl
    Broadcast 32-bit integer a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_set1_epi64Experimentalavx512f,avx512vl
    Broadcast 64-bit integer a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_shuffle_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_shuffle_f32x4Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_shuffle_f64x2Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_shuffle_i32x4Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 4 32-bit integers) selected by imm8 from a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_shuffle_i64x2Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 2 64-bit integers) selected by imm8 from a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_shuffle_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_shuffle_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sll_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sll_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_slli_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_slli_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sllv_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sllv_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sqrt_pdExperimentalavx512f,avx512vl
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sqrt_psExperimentalavx512f,avx512vl
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sra_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sra_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srai_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srai_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srav_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srav_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srl_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srl_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srli_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srli_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srlv_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_srlv_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_store_epi32Experimentalavx512f,avx512vl,avx
    Store packed 32-bit integers from a into memory using writemask k. mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_mask_store_epi64Experimentalavx512f,avx512vl,avx
    Store packed 64-bit integers from a into memory using writemask k. mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_mask_store_pdExperimentalavx512f,avx512vl,avx
    Store packed double-precision (64-bit) floating-point elements from a into memory using writemask k. mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_mask_store_psExperimentalavx512f,avx512vl,avx
    Store packed single-precision (32-bit) floating-point elements from a into memory using writemask k. mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_mask_storeu_epi32Experimentalavx512f,avx512vl,avx
    Store packed 32-bit integers from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask_storeu_epi64Experimentalavx512f,avx512vl,avx
    Store packed 64-bit integers from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask_storeu_pdExperimentalavx512f,avx512vl,avx
    Store packed double-precision (64-bit) floating-point elements from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask_storeu_psExperimentalavx512f,avx512vl,avx
    Store packed single-precision (32-bit) floating-point elements from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm256_mask_sub_epi32Experimentalavx512f,avx512vl
    Subtract packed 32-bit integers in b from packed 32-bit integers in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sub_epi64Experimentalavx512f,avx512vl
    Subtract packed 64-bit integers in b from packed 64-bit integers in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sub_pdExperimentalavx512f,avx512vl
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_sub_psExperimentalavx512f,avx512vl
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_ternarylogic_epi32Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from src, a, and b are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using writemask k at 32-bit granularity (32-bit elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_ternarylogic_epi64Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from src, a, and b are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using writemask k at 64-bit granularity (64-bit elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_test_epi32_maskExperimentalavx512f,avx512vl
    Compute the bitwise AND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is non-zero.
  • _mm256_mask_test_epi64_maskExperimentalavx512f,avx512vl
    Compute the bitwise AND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is non-zero.
  • _mm256_mask_testn_epi32_maskExperimentalavx512f,avx512vl
    Compute the bitwise NAND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is zero.
  • _mm256_mask_testn_epi64_maskExperimentalavx512f,avx512vl
    Compute the bitwise NAND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is zero.
  • _mm256_mask_unpackhi_epi32Experimentalavx512f,avx512vl
    Unpack and interleave 32-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_unpackhi_epi64Experimentalavx512f,avx512vl
    Unpack and interleave 64-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_unpackhi_pdExperimentalavx512f,avx512vl
    Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_unpackhi_psExperimentalavx512f,avx512vl
    Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_unpacklo_epi32Experimentalavx512f,avx512vl
    Unpack and interleave 32-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_unpacklo_epi64Experimentalavx512f,avx512vl
    Unpack and interleave 64-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_unpacklo_pdExperimentalavx512f,avx512vl
    Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_unpacklo_psExperimentalavx512f,avx512vl
    Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_xor_epi32Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_mask_xor_epi64Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm256_maskz_abs_epi32Experimentalavx512f,avx512vl
    Compute the absolute value of packed signed 32-bit integers in a, and store the unsigned results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_abs_epi64Experimentalavx512f,avx512vl
    Compute the absolute value of packed signed 64-bit integers in a, and store the unsigned results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_add_epi32Experimentalavx512f,avx512vl
    Add packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_add_epi64Experimentalavx512f,avx512vl
    Add packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_add_pdExperimentalavx512f,avx512vl
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_add_psExperimentalavx512f,avx512vl
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_alignr_epi32Experimentalavx512f,avx512vl
    Concatenate a and b into a 64-byte immediate result, shift the result right by imm8 32-bit elements, and store the low 32 bytes (8 elements) in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_alignr_epi64Experimentalavx512f,avx512vl
    Concatenate a and b into a 64-byte immediate result, shift the result right by imm8 64-bit elements, and store the low 32 bytes (4 elements) in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_and_epi32Experimentalavx512f,avx512vl
    Compute the bitwise AND of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_and_epi64Experimentalavx512f,avx512vl
    Compute the bitwise AND of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_andnot_epi32Experimentalavx512f,avx512vl
    Compute the bitwise NOT of packed 32-bit integers in a and then AND with b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_andnot_epi64Experimentalavx512f,avx512vl
    Compute the bitwise NOT of packed 64-bit integers in a and then AND with b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_broadcast_f32x4Experimentalavx512f,avx512vl
    Broadcast the 4 packed single-precision (32-bit) floating-point elements from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_broadcast_i32x4Experimentalavx512f,avx512vl
    Broadcast the 4 packed 32-bit integers from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_broadcastd_epi32Experimentalavx512f,avx512vl
    Broadcast the low packed 32-bit integer from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_broadcastq_epi64Experimentalavx512f,avx512vl
    Broadcast the low packed 64-bit integer from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_broadcastsd_pdExperimentalavx512f,avx512vl
    Broadcast the low double-precision (64-bit) floating-point element from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_broadcastss_psExperimentalavx512f,avx512vl
    Broadcast the low single-precision (32-bit) floating-point element from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_compress_epi32Experimentalavx512f,avx512vl
    Contiguously store the active 32-bit integers in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm256_maskz_compress_epi64Experimentalavx512f,avx512vl
    Contiguously store the active 64-bit integers in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm256_maskz_compress_pdExperimentalavx512f,avx512vl
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm256_maskz_compress_psExperimentalavx512f,avx512vl
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm256_maskz_cvt_roundps_phExperimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_maskz_cvtepi8_epi32Experimentalavx512f,avx512vl
    Sign extend packed 8-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi8_epi64Experimentalavx512f,avx512vl
    Sign extend packed 8-bit integers in the low 4 bytes of a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi16_epi32Experimentalavx512f,avx512vl
    Sign extend packed 16-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi16_epi64Experimentalavx512f,avx512vl
    Sign extend packed 16-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi32_epi8Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi32_epi16Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi32_epi64Experimentalavx512f,avx512vl
    Sign extend packed 32-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi32_pdExperimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi32_psExperimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi64_epi8Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi64_epi16Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepi64_epi32Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepu8_epi32Experimentalavx512f,avx512vl
    Zero extend packed unsigned 8-bit integers in the low 8 bytes of a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepu8_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 8-bit integers in the low 4 bytes of a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepu16_epi32Experimentalavx512f,avx512vl
    Zero extend packed unsigned 16-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepu16_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 16-bit integers in the low 8 bytes of a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepu32_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 32-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtepu32_pdExperimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtpd_epi32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtpd_psExperimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtph_psExperimentalavx512f,avx512vl
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtps_epi32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtps_epu32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtps_phExperimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_maskz_cvtsepi32_epi8Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtsepi32_epi16Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm256_maskz_cvtsepi64_epi8Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtsepi64_epi16Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtsepi64_epi32Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvttpd_epi32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvttpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvttps_epi32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvttps_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtusepi32_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtusepi32_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtusepi64_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtusepi64_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_cvtusepi64_epi32Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_div_pdExperimentalavx512f,avx512vl
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_div_psExperimentalavx512f,avx512vl
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_expand_epi32Experimentalavx512f,avx512vl
    Load contiguous active 32-bit integers from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_expand_epi64Experimentalavx512f,avx512vl
    Load contiguous active 64-bit integers from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_expand_pdExperimentalavx512f,avx512vl
    Load contiguous active double-precision (64-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_expand_psExperimentalavx512f,avx512vl
    Load contiguous active single-precision (32-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_expandloadu_epi32Experimentalavx512f,avx512vl,avx
    Load contiguous active 32-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_expandloadu_epi64Experimentalavx512f,avx512vl,avx
    Load contiguous active 64-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_expandloadu_pdExperimentalavx512f,avx512vl,avx
    Load contiguous active single-precision (64-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_expandloadu_psExperimentalavx512f,avx512vl,avx
    Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_extractf32x4_psExperimentalavx512f,avx512vl
    Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from a, selected with imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_extracti32x4_epi32Experimentalavx512f,avx512vl
    Extract 128 bits (composed of 4 packed 32-bit integers) from a, selected with IMM1, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fixupimm_pdExperimentalavx512f,avx512vl
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm256_maskz_fixupimm_psExperimentalavx512f,avx512vl
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm256_maskz_fmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fmaddsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fmaddsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fmsubadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fmsubadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fnmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fnmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fnmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_fnmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_getexp_pdExperimentalavx512f,avx512vl
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm256_maskz_getexp_psExperimentalavx512f,avx512vl
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm256_maskz_getmant_pdExperimentalavx512f,avx512vl
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm256_maskz_getmant_psExperimentalavx512f,avx512vl
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm256_maskz_insertf32x4Experimentalavx512f,avx512vl
    Copy a to tmp, then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from b into tmp at the location specified by imm8. Store tmp to dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_inserti32x4Experimentalavx512f,avx512vl
    Copy a to tmp, then insert 128 bits (composed of 4 packed 32-bit integers) from b into tmp at the location specified by imm8. Store tmp to dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_load_epi32Experimentalavx512f,avx512vl,avx
    Load packed 32-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_maskz_load_epi64Experimentalavx512f,avx512vl,avx
    Load packed 64-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_maskz_load_pdExperimentalavx512f,avx512vl,avx
    Load packed double-precision (64-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_maskz_load_psExperimentalavx512f,avx512vl,avx
    Load packed single-precision (32-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_maskz_loadu_epi32Experimentalavx512f,avx512vl,avx
    Load packed 32-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm256_maskz_loadu_epi64Experimentalavx512f,avx512vl,avx
    Load packed 64-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm256_maskz_loadu_pdExperimentalavx512f,avx512vl,avx
    Load packed double-precision (64-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm256_maskz_loadu_psExperimentalavx512f,avx512vl,avx
    Load packed single-precision (32-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm256_maskz_max_epi32Experimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_max_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_max_epu32Experimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_max_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_max_pdExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_max_psExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_min_epi32Experimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_min_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_min_epu32Experimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_min_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_min_pdExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_min_psExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mov_epi32Experimentalavx512f,avx512vl
    Move packed 32-bit integers from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mov_epi64Experimentalavx512f,avx512vl
    Move packed 64-bit integers from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mov_pdExperimentalavx512f,avx512vl
    Move packed double-precision (64-bit) floating-point elements from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mov_psExperimentalavx512f,avx512vl
    Move packed single-precision (32-bit) floating-point elements from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_movedup_pdExperimentalavx512f,avx512vl
    Duplicate even-indexed double-precision (64-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_movehdup_psExperimentalavx512f,avx512vl
    Duplicate odd-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_moveldup_psExperimentalavx512f,avx512vl
    Duplicate even-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mul_epi32Experimentalavx512f,avx512vl
    Multiply the low signed 32-bit integers from each packed 64-bit element in a and b, and store the signed 64-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mul_epu32Experimentalavx512f,avx512vl
    Multiply the low unsigned 32-bit integers from each packed 64-bit element in a and b, and store the unsigned 64-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mul_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mul_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_mullo_epi32Experimentalavx512f,avx512vl
    Multiply the packed 32-bit integers in a and b, producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_or_epi32Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_or_epi64Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permute_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permute_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutevar_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutevar_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutex2var_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutex2var_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutex2var_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutex2var_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutex_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a within 256-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutex_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 256-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutexvar_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutexvar_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutexvar_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_permutexvar_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_rcp14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm256_maskz_rcp14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm256_maskz_rol_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_rol_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_rolv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_rolv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_ror_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_ror_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_rorv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_rorv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_roundscale_pdExperimentalavx512f,avx512vl
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_maskz_roundscale_psExperimentalavx512f,avx512vl
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_maskz_rsqrt14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm256_maskz_rsqrt14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm256_maskz_scalef_pdExperimentalavx512f,avx512vl
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_scalef_psExperimentalavx512f,avx512vl
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_set1_epi32Experimentalavx512f,avx512vl
    Broadcast 32-bit integer a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_set1_epi64Experimentalavx512f,avx512vl
    Broadcast 64-bit integer a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_shuffle_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_shuffle_f32x4Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_shuffle_f64x2Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_shuffle_i32x4Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 4 32-bit integers) selected by imm8 from a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_shuffle_i64x2Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 2 64-bit integers) selected by imm8 from a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_shuffle_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_shuffle_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sll_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sll_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_slli_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_slli_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sllv_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sllv_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sqrt_pdExperimentalavx512f,avx512vl
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sqrt_psExperimentalavx512f,avx512vl
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sra_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sra_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srai_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srai_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srav_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srav_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srl_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srl_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srli_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srli_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srlv_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_srlv_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sub_epi32Experimentalavx512f,avx512vl
    Subtract packed 32-bit integers in b from packed 32-bit integers in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sub_epi64Experimentalavx512f,avx512vl
    Subtract packed 64-bit integers in b from packed 64-bit integers in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sub_pdExperimentalavx512f,avx512vl
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_sub_psExperimentalavx512f,avx512vl
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_ternarylogic_epi32Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using zeromask k at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_ternarylogic_epi64Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using zeromask k at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_unpackhi_epi32Experimentalavx512f,avx512vl
    Unpack and interleave 32-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_unpackhi_epi64Experimentalavx512f,avx512vl
    Unpack and interleave 64-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_unpackhi_pdExperimentalavx512f,avx512vl
    Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_unpackhi_psExperimentalavx512f,avx512vl
    Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_unpacklo_epi32Experimentalavx512f,avx512vl
    Unpack and interleave 32-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_unpacklo_epi64Experimentalavx512f,avx512vl
    Unpack and interleave 64-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_unpacklo_pdExperimentalavx512f,avx512vl
    Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_unpacklo_psExperimentalavx512f,avx512vl
    Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_xor_epi32Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_maskz_xor_epi64Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm256_max_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst.
  • _mm256_max_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst.
  • _mm256_min_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed minimum values in dst.
  • _mm256_min_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst.
  • _mm256_or_epi32Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst.
  • _mm256_or_epi64Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the resut in dst.
  • _mm256_permutex2var_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm256_permutex2var_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm256_permutex2var_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm256_permutex2var_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm256_permutex_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a within 256-bit lanes using the control in imm8, and store the results in dst.
  • _mm256_permutex_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 256-bit lanes using the control in imm8, and store the results in dst.
  • _mm256_permutexvar_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a across lanes using the corresponding index in idx, and store the results in dst.
  • _mm256_permutexvar_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a across lanes using the corresponding index in idx, and store the results in dst.
  • _mm256_permutexvar_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst.
  • _mm256_permutexvar_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a across lanes using the corresponding index in idx.
  • _mm256_rcp14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm256_rcp14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm256_rol_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst.
  • _mm256_rol_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst.
  • _mm256_rolv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm256_rolv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm256_ror_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst.
  • _mm256_ror_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst.
  • _mm256_rorv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm256_rorv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm256_roundscale_pdExperimentalavx512f,avx512vl
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_roundscale_psExperimentalavx512f,avx512vl
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm256_scalef_pdExperimentalavx512f,avx512vl
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst.
  • _mm256_scalef_psExperimentalavx512f,avx512vl
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst.
  • _mm256_shuffle_f32x4Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst.
  • _mm256_shuffle_f64x2Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst.
  • _mm256_shuffle_i32x4Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 4 32-bit integers) selected by imm8 from a and b, and store the results in dst.
  • _mm256_shuffle_i64x2Experimentalavx512f,avx512vl
    Shuffle 128-bits (composed of 2 64-bit integers) selected by imm8 from a and b, and store the results in dst.
  • _mm256_sra_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst.
  • _mm256_srai_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst.
  • _mm256_srav_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst.
  • _mm256_store_epi32Experimentalavx512f,avx512vl
    Store 256-bits (composed of 8 packed 32-bit integers) from a into memory. mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_store_epi64Experimentalavx512f,avx512vl
    Store 256-bits (composed of 4 packed 64-bit integers) from a into memory. mem_addr must be aligned on a 32-byte boundary or a general-protection exception may be generated.
  • _mm256_storeu_epi32Experimentalavx512f,avx512vl
    Store 256-bits (composed of 8 packed 32-bit integers) from a into memory. mem_addr does not need to be aligned on any particular boundary.
  • _mm256_storeu_epi64Experimentalavx512f,avx512vl
    Store 256-bits (composed of 4 packed 64-bit integers) from a into memory. mem_addr does not need to be aligned on any particular boundary.
  • _mm256_ternarylogic_epi32Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst.
  • _mm256_ternarylogic_epi64Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst.
  • _mm256_test_epi32_maskExperimentalavx512f,avx512vl
    Compute the bitwise AND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k if the intermediate value is non-zero.
  • _mm256_test_epi64_maskExperimentalavx512f,avx512vl
    Compute the bitwise AND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k if the intermediate value is non-zero.
  • _mm256_testn_epi32_maskExperimentalavx512f,avx512vl
    Compute the bitwise NAND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k if the intermediate value is zero.
  • _mm256_testn_epi64_maskExperimentalavx512f,avx512vl
    Compute the bitwise NAND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k if the intermediate value is zero.
  • _mm256_xor_epi32Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst.
  • _mm256_xor_epi64Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst.
  • _mm512_abs_epi32Experimentalavx512f
    Computes the absolute values of packed 32-bit integers in a.
  • _mm512_abs_epi64Experimentalavx512f
    Compute the absolute value of packed signed 64-bit integers in a, and store the unsigned results in dst.
  • _mm512_abs_pdExperimentalavx512f
    Finds the absolute value of each packed double-precision (64-bit) floating-point element in v2, storing the results in dst.
  • _mm512_abs_psExperimentalavx512f
    Finds the absolute value of each packed single-precision (32-bit) floating-point element in v2, storing the results in dst.
  • _mm512_add_epi32Experimentalavx512f
    Add packed 32-bit integers in a and b, and store the results in dst.
  • _mm512_add_epi64Experimentalavx512f
    Add packed 64-bit integers in a and b, and store the results in dst.
  • _mm512_add_pdExperimentalavx512f
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst.
  • _mm512_add_psExperimentalavx512f
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst.
  • _mm512_add_round_pdExperimentalavx512f
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst.
  • _mm512_add_round_psExperimentalavx512f
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst.
  • _mm512_alignr_epi32Experimentalavx512f
    Concatenate a and b into a 128-byte immediate result, shift the result right by imm8 32-bit elements, and store the low 64 bytes (16 elements) in dst.
  • _mm512_alignr_epi64Experimentalavx512f
    Concatenate a and b into a 128-byte immediate result, shift the result right by imm8 64-bit elements, and store the low 64 bytes (8 elements) in dst.
  • _mm512_and_epi32Experimentalavx512f
    Compute the bitwise AND of packed 32-bit integers in a and b, and store the results in dst.
  • _mm512_and_epi64Experimentalavx512f
    Compute the bitwise AND of 512 bits (composed of packed 64-bit integers) in a and b, and store the results in dst.
  • _mm512_and_si512Experimentalavx512f
    Compute the bitwise AND of 512 bits (representing integer data) in a and b, and store the result in dst.
  • _mm512_andnot_epi32Experimentalavx512f
    Compute the bitwise NOT of packed 32-bit integers in a and then AND with b, and store the results in dst.
  • _mm512_andnot_epi64Experimentalavx512f
    Compute the bitwise NOT of 512 bits (composed of packed 64-bit integers) in a and then AND with b, and store the results in dst.
  • _mm512_andnot_si512Experimentalavx512f
    Compute the bitwise NOT of 512 bits (representing integer data) in a and then AND with b, and store the result in dst.
  • _mm512_broadcast_f32x4Experimentalavx512f
    Broadcast the 4 packed single-precision (32-bit) floating-point elements from a to all elements of dst.
  • _mm512_broadcast_f64x4Experimentalavx512f
    Broadcast the 4 packed double-precision (64-bit) floating-point elements from a to all elements of dst.
  • _mm512_broadcast_i32x4Experimentalavx512f
    Broadcast the 4 packed 32-bit integers from a to all elements of dst.
  • _mm512_broadcast_i64x4Experimentalavx512f
    Broadcast the 4 packed 64-bit integers from a to all elements of dst.
  • _mm512_broadcastd_epi32Experimentalavx512f
    Broadcast the low packed 32-bit integer from a to all elements of dst.
  • _mm512_broadcastq_epi64Experimentalavx512f
    Broadcast the low packed 64-bit integer from a to all elements of dst.
  • _mm512_broadcastsd_pdExperimentalavx512f
    Broadcast the low double-precision (64-bit) floating-point element from a to all elements of dst.
  • _mm512_broadcastss_psExperimentalavx512f
    Broadcast the low single-precision (32-bit) floating-point element from a to all elements of dst.
  • _mm512_castpd128_pd512Experimentalavx512f
    Cast vector of type __m128d to type __m512d; the upper 384 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castpd256_pd512Experimentalavx512f
    Cast vector of type __m256d to type __m512d; the upper 256 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castpd512_pd128Experimentalavx512f
    Cast vector of type __m512d to type __m128d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castpd512_pd256Experimentalavx512f
    Cast vector of type __m512d to type __m256d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castpd_psExperimentalavx512f
    Cast vector of type __m512d to type __m512. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castpd_si512Experimentalavx512f
    Cast vector of type __m512d to type __m512i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castps128_ps512Experimentalavx512f
    Cast vector of type __m128 to type __m512; the upper 384 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castps256_ps512Experimentalavx512f
    Cast vector of type __m256 to type __m512; the upper 256 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castps512_ps128Experimentalavx512f
    Cast vector of type __m512 to type __m128. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castps512_ps256Experimentalavx512f
    Cast vector of type __m512 to type __m256. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castps_pdExperimentalavx512f
    Cast vector of type __m512 to type __m512d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castps_si512Experimentalavx512f
    Cast vector of type __m512 to type __m512i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castsi128_si512Experimentalavx512f
    Cast vector of type __m128i to type __m512i; the upper 384 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castsi256_si512Experimentalavx512f
    Cast vector of type __m256i to type __m512i; the upper 256 bits of the result are undefined. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castsi512_pdExperimentalavx512f
    Cast vector of type __m512i to type __m512d. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castsi512_psExperimentalavx512f
    Cast vector of type __m512i to type __m512. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castsi512_si128Experimentalavx512f
    Cast vector of type __m512i to type __m128i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_castsi512_si256Experimentalavx512f
    Cast vector of type __m512i to type __m256i. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_cmp_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm512_cmp_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm512_cmp_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm512_cmp_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm512_cmp_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm512_cmp_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm512_cmp_round_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cmp_round_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cmpeq_epi32_maskExperimentalavx512f
    Compare packed 32-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm512_cmpeq_epi64_maskExperimentalavx512f
    Compare packed 64-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm512_cmpeq_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm512_cmpeq_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm512_cmpeq_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for equality, and store the results in mask vector k.
  • _mm512_cmpeq_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for equality, and store the results in mask vector k.
  • _mm512_cmpge_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm512_cmpge_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm512_cmpge_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm512_cmpge_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm512_cmpgt_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm512_cmpgt_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm512_cmpgt_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm512_cmpgt_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm512_cmple_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm512_cmple_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm512_cmple_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm512_cmple_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm512_cmple_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm512_cmple_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm512_cmplt_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm512_cmplt_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm512_cmplt_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm512_cmplt_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm512_cmplt_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for less-than, and store the results in mask vector k.
  • _mm512_cmplt_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for less-than, and store the results in mask vector k.
  • _mm512_cmpneq_epi32_maskExperimentalavx512f
    Compare packed 32-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm512_cmpneq_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm512_cmpneq_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm512_cmpneq_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm512_cmpneq_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for not-equal, and store the results in mask vector k.
  • _mm512_cmpneq_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for not-equal, and store the results in mask vector k.
  • _mm512_cmpnle_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for not-less-than-or-equal, and store the results in mask vector k.
  • _mm512_cmpnle_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for not-less-than-or-equal, and store the results in mask vector k.
  • _mm512_cmpnlt_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for not-less-than, and store the results in mask vector k.
  • _mm512_cmpnlt_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for not-less-than, and store the results in mask vector k.
  • _mm512_cmpord_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b to see if neither is NaN, and store the results in mask vector k.
  • _mm512_cmpord_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b to see if neither is NaN, and store the results in mask vector k.
  • _mm512_cmpunord_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b to see if either is NaN, and store the results in mask vector k.
  • _mm512_cmpunord_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b to see if either is NaN, and store the results in mask vector k.
  • _mm512_cvt_roundepi32_psExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst.
  • _mm512_cvt_roundepu32_psExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst.
  • _mm512_cvt_roundpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst.
  • _mm512_cvt_roundpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst.
  • _mm512_cvt_roundpd_psExperimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst.
  • _mm512_cvt_roundph_psExperimentalavx512f
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cvt_roundps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst.
  • _mm512_cvt_roundps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst.
  • _mm512_cvt_roundps_pdExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed double-precision (64-bit) floating-point elements, and store the results in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cvt_roundps_phExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cvtepi8_epi32Experimentalavx512f
    Sign extend packed 8-bit integers in a to packed 32-bit integers, and store the results in dst.
  • _mm512_cvtepi8_epi64Experimentalavx512f
    Sign extend packed 8-bit integers in the low 8 bytes of a to packed 64-bit integers, and store the results in dst.
  • _mm512_cvtepi16_epi32Experimentalavx512f
    Sign extend packed 16-bit integers in a to packed 32-bit integers, and store the results in dst.
  • _mm512_cvtepi16_epi64Experimentalavx512f
    Sign extend packed 16-bit integers in a to packed 64-bit integers, and store the results in dst.
  • _mm512_cvtepi32_epi8Experimentalavx512f
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst.
  • _mm512_cvtepi32_epi16Experimentalavx512f
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst.
  • _mm512_cvtepi32_epi64Experimentalavx512f
    Sign extend packed 32-bit integers in a to packed 64-bit integers, and store the results in dst.
  • _mm512_cvtepi32_pdExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst.
  • _mm512_cvtepi32_psExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst.
  • _mm512_cvtepi32lo_pdExperimentalavx512f
    Performs element-by-element conversion of the lower half of packed 32-bit integer elements in v2 to packed double-precision (64-bit) floating-point elements, storing the results in dst.
  • _mm512_cvtepi64_epi8Experimentalavx512f
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst.
  • _mm512_cvtepi64_epi16Experimentalavx512f
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst.
  • _mm512_cvtepi64_epi32Experimentalavx512f
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst.
  • _mm512_cvtepu8_epi32Experimentalavx512f
    Zero extend packed unsigned 8-bit integers in a to packed 32-bit integers, and store the results in dst.
  • _mm512_cvtepu8_epi64Experimentalavx512f
    Zero extend packed unsigned 8-bit integers in the low 8 byte sof a to packed 64-bit integers, and store the results in dst.
  • _mm512_cvtepu16_epi32Experimentalavx512f
    Zero extend packed unsigned 16-bit integers in a to packed 32-bit integers, and store the results in dst.
  • _mm512_cvtepu16_epi64Experimentalavx512f
    Zero extend packed unsigned 16-bit integers in a to packed 64-bit integers, and store the results in dst.
  • _mm512_cvtepu32_epi64Experimentalavx512f
    Zero extend packed unsigned 32-bit integers in a to packed 64-bit integers, and store the results in dst.
  • _mm512_cvtepu32_pdExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst.
  • _mm512_cvtepu32_psExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst.
  • _mm512_cvtepu32lo_pdExperimentalavx512f
    Performs element-by-element conversion of the lower half of packed 32-bit unsigned integer elements in v2 to packed double-precision (64-bit) floating-point elements, storing the results in dst.
  • _mm512_cvtpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst.
  • _mm512_cvtpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst.
  • _mm512_cvtpd_psExperimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst.
  • _mm512_cvtpd_psloExperimentalavx512f
    Performs an element-by-element conversion of packed double-precision (64-bit) floating-point elements in v2 to single-precision (32-bit) floating-point elements and stores them in dst. The elements are stored in the lower half of the results vector, while the remaining upper half locations are set to 0.
  • _mm512_cvtph_psExperimentalavx512f
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst.
  • _mm512_cvtps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst.
  • _mm512_cvtps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst.
  • _mm512_cvtps_pdExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed double-precision (64-bit) floating-point elements, and store the results in dst.
  • _mm512_cvtps_phExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cvtpslo_pdExperimentalavx512f
    Performs element-by-element conversion of the lower half of packed single-precision (32-bit) floating-point elements in v2 to packed double-precision (64-bit) floating-point elements, storing the results in dst.
  • _mm512_cvtsepi32_epi8Experimentalavx512f
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst.
  • _mm512_cvtsepi32_epi16Experimentalavx512f
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm512_cvtsepi64_epi8Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst.
  • _mm512_cvtsepi64_epi16Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm512_cvtsepi64_epi32Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst.
  • _mm512_cvtsi512_si32Experimentalavx512f
    Copy the lower 32-bit integer in a to dst.
  • _mm512_cvtt_roundpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cvtt_roundpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cvtt_roundps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cvtt_roundps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_cvttpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst.
  • _mm512_cvttpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst.
  • _mm512_cvttps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst.
  • _mm512_cvttps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst.
  • _mm512_cvtusepi32_epi8Experimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst.
  • _mm512_cvtusepi32_epi16Experimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst.
  • _mm512_cvtusepi64_epi8Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst.
  • _mm512_cvtusepi64_epi16Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst.
  • _mm512_cvtusepi64_epi32Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst.
  • _mm512_div_pdExperimentalavx512f
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst.
  • _mm512_div_psExperimentalavx512f
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst.
  • _mm512_div_round_pdExperimentalavx512f
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, =and store the results in dst.
  • _mm512_div_round_psExperimentalavx512f
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst.
  • _mm512_extractf32x4_psExperimentalavx512f
    Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from a, selected with imm8, and store the result in dst.
  • _mm512_extractf64x4_pdExperimentalavx512f
    Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from a, selected with imm8, and store the result in dst.
  • _mm512_extracti32x4_epi32Experimentalavx512f
    Extract 128 bits (composed of 4 packed 32-bit integers) from a, selected with IMM2, and store the result in dst.
  • _mm512_extracti64x4_epi64Experimentalavx512f
    Extract 256 bits (composed of 4 packed 64-bit integers) from a, selected with IMM1, and store the result in dst.
  • _mm512_fixupimm_pdExperimentalavx512f
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst. imm8 is used to set the required flags reporting.
  • _mm512_fixupimm_psExperimentalavx512f
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst. imm8 is used to set the required flags reporting.
  • _mm512_fixupimm_round_pdExperimentalavx512f
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst. imm8 is used to set the required flags reporting.
  • _mm512_fixupimm_round_psExperimentalavx512f
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst. imm8 is used to set the required flags reporting.
  • _mm512_fmadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst.
  • _mm512_fmadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst.
  • _mm512_fmadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst.
  • _mm512_fmadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst.
  • _mm512_fmaddsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst.
  • _mm512_fmaddsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst.
  • _mm512_fmaddsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst.
  • _mm512_fmaddsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst.
  • _mm512_fmsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst.
  • _mm512_fmsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst.
  • _mm512_fmsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst.
  • _mm512_fmsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst.
  • _mm512_fmsubadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst.
  • _mm512_fmsubadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst.
  • _mm512_fmsubadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst.
  • _mm512_fmsubadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst.
  • _mm512_fnmadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst.
  • _mm512_fnmadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst.
  • _mm512_fnmadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst.
  • _mm512_fnmadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst.
  • _mm512_fnmsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst.
  • _mm512_fnmsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst.
  • _mm512_fnmsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst.
  • _mm512_fnmsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst.
  • _mm512_getexp_pdExperimentalavx512f
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst. This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm512_getexp_psExperimentalavx512f
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst. This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm512_getexp_round_pdExperimentalavx512f
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst. This intrinsic essentially calculates floor(log2(x)) for each element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_getexp_round_psExperimentalavx512f
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst. This intrinsic essentially calculates floor(log2(x)) for each element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_getmant_pdExperimentalavx512f
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm512_getmant_psExperimentalavx512f
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign. The mantissa is normalized to the interval specified by interv, which can take the following values: _MM_MANT_NORM_1_2 // interval [1, 2) _MM_MANT_NORM_p5_2 // interval [0.5, 2) _MM_MANT_NORM_p5_1 // interval [0.5, 1) _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5) The sign is determined by sc which can take the following values: _MM_MANT_SIGN_src // sign = sign(src) _MM_MANT_SIGN_zero // sign = 0 _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm512_getmant_round_pdExperimentalavx512f
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_getmant_round_psExperimentalavx512f
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_i32gather_epi32Experimentalavx512f
    Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst. scale should be 1, 2, 4 or 8.
  • _mm512_i32gather_epi64Experimentalavx512f
    Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst. scale should be 1, 2, 4 or 8.
  • _mm512_i32gather_pdExperimentalavx512f
    Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst. scale should be 1, 2, 4 or 8.
  • _mm512_i32gather_psExperimentalavx512f
    Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst. scale should be 1, 2, 4 or 8.
  • _mm512_i32scatter_epi32Experimentalavx512f
    Scatter 32-bit integers from a into memory using 32-bit indices. 32-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). scale should be 1, 2, 4 or 8.
  • _mm512_i32scatter_epi64Experimentalavx512f
    Scatter 64-bit integers from a into memory using 32-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). scale should be 1, 2, 4 or 8.
  • _mm512_i32scatter_pdExperimentalavx512f
    Scatter double-precision (64-bit) floating-point elements from a into memory using 32-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). scale should be 1, 2, 4 or 8.
  • _mm512_i32scatter_psExperimentalavx512f
    Scatter single-precision (32-bit) floating-point elements from a into memory using 32-bit indices. 32-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). scale should be 1, 2, 4 or 8.
  • _mm512_i64gather_epi32Experimentalavx512f
    Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst. scale should be 1, 2, 4 or 8.
  • _mm512_i64gather_epi64Experimentalavx512f
    Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst. scale should be 1, 2, 4 or 8.
  • _mm512_i64gather_pdExperimentalavx512f
    Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst. scale should be 1, 2, 4 or 8.
  • _mm512_i64gather_psExperimentalavx512f
    Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst. scale should be 1, 2, 4 or 8.
  • _mm512_i64scatter_epi32Experimentalavx512f
    Scatter 32-bit integers from a into memory using 64-bit indices. 32-bit elements are stored at addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). scale should be 1, 2, 4 or 8.
  • _mm512_i64scatter_epi64Experimentalavx512f
    Scatter 64-bit integers from a into memory using 64-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). scale should be 1, 2, 4 or 8.
  • _mm512_i64scatter_pdExperimentalavx512f
    Scatter double-precision (64-bit) floating-point elements from a into memory using 64-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). scale should be 1, 2, 4 or 8.
  • _mm512_i64scatter_psExperimentalavx512f
    Scatter single-precision (32-bit) floating-point elements from a into memory using 64-bit indices. 32-bit elements are stored at addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_insertf32x4Experimentalavx512f
    Copy a to dst, then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from b into dst at the location specified by imm8.
  • _mm512_insertf64x4Experimentalavx512f
    Copy a to dst, then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from b into dst at the location specified by imm8.
  • _mm512_inserti32x4Experimentalavx512f
    Copy a to dst, then insert 128 bits (composed of 4 packed 32-bit integers) from b into dst at the location specified by imm8.
  • _mm512_inserti64x4Experimentalavx512f
    Copy a to dst, then insert 256 bits (composed of 4 packed 64-bit integers) from b into dst at the location specified by imm8.
  • _mm512_int2maskExperimentalavx512f
    Converts integer mask into bitmask, storing the result in dst.
  • _mm512_kandExperimentalavx512f
    Compute the bitwise AND of 16-bit masks a and b, and store the result in k.
  • _mm512_kandnExperimentalavx512f
    Compute the bitwise NOT of 16-bit masks a and then AND with b, and store the result in k.
  • _mm512_kmovExperimentalavx512f
    Copy 16-bit mask a to k.
  • _mm512_knotExperimentalavx512f
    Compute the bitwise NOT of 16-bit mask a, and store the result in k.
  • _mm512_korExperimentalavx512f
    Compute the bitwise OR of 16-bit masks a and b, and store the result in k.
  • _mm512_kortestcExperimentalavx512f
    Performs bitwise OR between k1 and k2, storing the result in dst. CF flag is set if dst consists of all 1’s.
  • _mm512_kunpackbExperimentalavx512f
    Unpack and interleave 8 bits from masks a and b, and store the 16-bit result in k.
  • _mm512_kxnorExperimentalavx512f
    Compute the bitwise XNOR of 16-bit masks a and b, and store the result in k.
  • _mm512_kxorExperimentalavx512f
    Compute the bitwise XOR of 16-bit masks a and b, and store the result in k.
  • _mm512_load_epi32Experimentalavx512f
    Load 512-bits (composed of 16 packed 32-bit integers) from memory into dst. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_load_epi64Experimentalavx512f
    Load 512-bits (composed of 8 packed 64-bit integers) from memory into dst. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_load_pdExperimentalavx512f
    Load 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from memory into dst. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_load_psExperimentalavx512f
    Load 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from memory into dst. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_load_si512Experimentalavx512f
    Load 512-bits of integer data from memory into dst. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_loadu_epi32Experimentalavx512f
    Load 512-bits (composed of 16 packed 32-bit integers) from memory into dst. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_loadu_epi64Experimentalavx512f
    Load 512-bits (composed of 8 packed 64-bit integers) from memory into dst. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_loadu_pdExperimentalavx512f
    Loads 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from memory into result. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_loadu_psExperimentalavx512f
    Loads 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from memory into result. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_loadu_si512Experimentalavx512f
    Load 512-bits of integer data from memory into dst. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask2_permutex2var_epi32Experimentalavx512f
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm512_mask2_permutex2var_epi64Experimentalavx512f
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm512_mask2_permutex2var_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set)
  • _mm512_mask2_permutex2var_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm512_mask2intExperimentalavx512f
    Converts bit mask k1 into an integer value, storing the results in dst.
  • _mm512_mask3_fmadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmaddsub_pdExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmaddsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmaddsub_round_pdExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmaddsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmsubadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmsubadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmsubadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fmsubadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fnmadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fnmadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fnmadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fnmadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fnmsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fnmsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fnmsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask3_fnmsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm512_mask_abs_epi32Experimentalavx512f
    Computes the absolute value of packed 32-bit integers in a, and store the unsigned results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_abs_epi64Experimentalavx512f
    Compute the absolute value of packed signed 64-bit integers in a, and store the unsigned results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_abs_pdExperimentalavx512f
    Finds the absolute value of each packed double-precision (64-bit) floating-point element in v2, storing the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_abs_psExperimentalavx512f
    Finds the absolute value of each packed single-precision (32-bit) floating-point element in v2, storing the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_add_epi32Experimentalavx512f
    Add packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_add_epi64Experimentalavx512f
    Add packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_add_pdExperimentalavx512f
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_add_psExperimentalavx512f
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_add_round_pdExperimentalavx512f
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_add_round_psExperimentalavx512f
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_alignr_epi32Experimentalavx512f
    Concatenate a and b into a 128-byte immediate result, shift the result right by imm8 32-bit elements, and store the low 64 bytes (16 elements) in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_alignr_epi64Experimentalavx512f
    Concatenate a and b into a 128-byte immediate result, shift the result right by imm8 64-bit elements, and store the low 64 bytes (8 elements) in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_and_epi32Experimentalavx512f
    Performs element-by-element bitwise AND between packed 32-bit integer elements of a and b, storing the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_and_epi64Experimentalavx512f
    Compute the bitwise AND of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_andnot_epi32Experimentalavx512f
    Compute the bitwise NOT of packed 32-bit integers in a and then AND with b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_andnot_epi64Experimentalavx512f
    Compute the bitwise NOT of packed 64-bit integers in a and then AND with b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_blend_epi32Experimentalavx512f
    Blend packed 32-bit integers from a and b using control mask k, and store the results in dst.
  • _mm512_mask_blend_epi64Experimentalavx512f
    Blend packed 64-bit integers from a and b using control mask k, and store the results in dst.
  • _mm512_mask_blend_pdExperimentalavx512f
    Blend packed double-precision (64-bit) floating-point elements from a and b using control mask k, and store the results in dst.
  • _mm512_mask_blend_psExperimentalavx512f
    Blend packed single-precision (32-bit) floating-point elements from a and b using control mask k, and store the results in dst.
  • _mm512_mask_broadcast_f32x4Experimentalavx512f
    Broadcast the 4 packed single-precision (32-bit) floating-point elements from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_broadcast_f64x4Experimentalavx512f
    Broadcast the 4 packed double-precision (64-bit) floating-point elements from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_broadcast_i32x4Experimentalavx512f
    Broadcast the 4 packed 32-bit integers from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_broadcast_i64x4Experimentalavx512f
    Broadcast the 4 packed 64-bit integers from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_broadcastd_epi32Experimentalavx512f
    Broadcast the low packed 32-bit integer from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_broadcastq_epi64Experimentalavx512f
    Broadcast the low packed 64-bit integer from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_broadcastsd_pdExperimentalavx512f
    Broadcast the low double-precision (64-bit) floating-point element from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_broadcastss_psExperimentalavx512f
    Broadcast the low single-precision (32-bit) floating-point element from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cmp_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmp_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmp_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmp_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmp_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmp_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmp_round_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cmp_round_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cmpeq_epi32_maskExperimentalavx512f
    Compare packed 32-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpeq_epi64_maskExperimentalavx512f
    Compare packed 64-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpeq_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpeq_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpeq_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpeq_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpge_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpge_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpge_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpge_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpgt_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpgt_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpgt_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpgt_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmple_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmple_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmple_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmple_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmple_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmple_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmplt_epi32_maskExperimentalavx512f
    Compare packed signed 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmplt_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmplt_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmplt_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmplt_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmplt_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpneq_epi32_maskExperimentalavx512f
    Compare packed 32-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpneq_epi64_maskExperimentalavx512f
    Compare packed signed 64-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpneq_epu32_maskExperimentalavx512f
    Compare packed unsigned 32-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpneq_epu64_maskExperimentalavx512f
    Compare packed unsigned 64-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpneq_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpneq_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpnle_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for not-less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpnle_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for not-less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpnlt_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b for not-less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpnlt_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b for not-less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpord_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b to see if neither is NaN, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpord_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b to see if neither is NaN, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpunord_pd_maskExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b to see if either is NaN, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_cmpunord_ps_maskExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b to see if either is NaN, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_mask_compress_epi32Experimentalavx512f
    Contiguously store the active 32-bit integers in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm512_mask_compress_epi64Experimentalavx512f
    Contiguously store the active 64-bit integers in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm512_mask_compress_pdExperimentalavx512f
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm512_mask_compress_psExperimentalavx512f
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm512_mask_compressstoreu_epi32Experimentalavx512f
    Contiguously store the active 32-bit integers in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_compressstoreu_epi64Experimentalavx512f
    Contiguously store the active 64-bit integers in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_compressstoreu_pdExperimentalavx512f
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_compressstoreu_psExperimentalavx512f
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvt_roundepi32_psExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvt_roundepu32_psExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvt_roundpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvt_roundpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvt_roundpd_psExperimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvt_roundph_psExperimentalavx512f
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cvt_roundps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvt_roundps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvt_roundps_pdExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cvt_roundps_phExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cvtepi8_epi32Experimentalavx512f
    Sign extend packed 8-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi8_epi64Experimentalavx512f
    Sign extend packed 8-bit integers in the low 8 bytes of a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi16_epi32Experimentalavx512f
    Sign extend packed 16-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi16_epi64Experimentalavx512f
    Sign extend packed 16-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi32_epi8Experimentalavx512f
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi32_epi16Experimentalavx512f
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi32_epi64Experimentalavx512f
    Sign extend packed 32-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi32_pdExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi32_psExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi32_storeu_epi8Experimentalavx512f
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvtepi32_storeu_epi16Experimentalavx512f
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvtepi32lo_pdExperimentalavx512f
    Performs element-by-element conversion of the lower half of packed 32-bit integer elements in v2 to packed double-precision (64-bit) floating-point elements, storing the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi64_epi8Experimentalavx512f
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi64_epi16Experimentalavx512f
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi64_epi32Experimentalavx512f
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepi64_storeu_epi8Experimentalavx512f
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvtepi64_storeu_epi16Experimentalavx512f
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvtepi64_storeu_epi32Experimentalavx512f
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvtepu8_epi32Experimentalavx512f
    Zero extend packed unsigned 8-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepu8_epi64Experimentalavx512f
    Zero extend packed unsigned 8-bit integers in the low 8 bytes of a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepu16_epi32Experimentalavx512f
    Zero extend packed unsigned 16-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepu16_epi64Experimentalavx512f
    Zero extend packed unsigned 16-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepu32_epi64Experimentalavx512f
    Zero extend packed unsigned 32-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepu32_pdExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepu32_psExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtepu32lo_pdExperimentalavx512f
    Performs element-by-element conversion of the lower half of 32-bit unsigned integer elements in v2 to packed double-precision (64-bit) floating-point elements, storing the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtpd_psExperimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtpd_psloExperimentalavx512f
    Performs an element-by-element conversion of packed double-precision (64-bit) floating-point elements in v2 to single-precision (32-bit) floating-point elements and stores them in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The elements are stored in the lower half of the results vector, while the remaining upper half locations are set to 0.
  • _mm512_mask_cvtph_psExperimentalavx512f
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtps_pdExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtps_phExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cvtpslo_pdExperimentalavx512f
    Performs element-by-element conversion of the lower half of packed single-precision (32-bit) floating-point elements in v2 to packed double-precision (64-bit) floating-point elements, storing the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtsepi32_epi8Experimentalavx512f
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtsepi32_epi16Experimentalavx512f
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtsepi32_storeu_epi8Experimentalavx512f
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvtsepi64_epi8Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtsepi64_epi16Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtsepi64_epi32Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtsepi64_storeu_epi8Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvtt_roundpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cvtt_roundpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cvtt_roundps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cvtt_roundps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_cvttpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvttpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvttps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvttps_epu32Experimentalavx512f
    Convert packed double-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtusepi32_epi8Experimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtusepi32_epi16Experimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • Convert packed unsigned 32-bit integers in a to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • Convert packed unsigned 32-bit integers in a to packed 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_cvtusepi64_epi8Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtusepi64_epi16Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_cvtusepi64_epi32Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • Convert packed unsigned 64-bit integers in a to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • Convert packed unsigned 64-bit integers in a to packed 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • Convert packed unsigned 64-bit integers in a to packed 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm512_mask_div_pdExperimentalavx512f
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_div_psExperimentalavx512f
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_div_round_pdExperimentalavx512f
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_div_round_psExperimentalavx512f
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_expand_epi32Experimentalavx512f
    Load contiguous active 32-bit integers from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_expand_epi64Experimentalavx512f
    Load contiguous active 64-bit integers from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_expand_pdExperimentalavx512f
    Load contiguous active double-precision (64-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_expand_psExperimentalavx512f
    Load contiguous active single-precision (32-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_expandloadu_epi32Experimentalavx512f
    Load contiguous active 32-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_expandloadu_epi64Experimentalavx512f
    Load contiguous active 64-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_expandloadu_pdExperimentalavx512f
    Load contiguous active single-precision (64-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_expandloadu_psExperimentalavx512f
    Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_extractf32x4_psExperimentalavx512f
    Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from a, selected with imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_extractf64x4_pdExperimentalavx512f
    Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from a, selected with imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_extracti32x4_epi32Experimentalavx512f
    Extract 128 bits (composed of 4 packed 32-bit integers) from a, selected with IMM2, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_extracti64x4_epi64Experimentalavx512f
    Extract 256 bits (composed of 4 packed 64-bit integers) from a, selected with IMM1, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_fixupimm_pdExperimentalavx512f
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm512_mask_fixupimm_psExperimentalavx512f
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm512_mask_fixupimm_round_pdExperimentalavx512f
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm512_mask_fixupimm_round_psExperimentalavx512f
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm512_mask_fmadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmaddsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmaddsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmaddsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmaddsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmsubadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmsubadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmsubadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fmsubadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fnmadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fnmadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fnmadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fnmadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fnmsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fnmsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fnmsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_fnmsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_getexp_pdExperimentalavx512f
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm512_mask_getexp_psExperimentalavx512f
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm512_mask_getexp_round_pdExperimentalavx512f
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_getexp_round_psExperimentalavx512f
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_getmant_pdExperimentalavx512f
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm512_mask_getmant_psExperimentalavx512f
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm512_mask_getmant_round_pdExperimentalavx512f
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_getmant_round_psExperimentalavx512f
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_i32gather_epi32Experimentalavx512f
    Gather 32-bit integers from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i32gather_epi64Experimentalavx512f
    Gather 64-bit integers from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i32gather_pdExperimentalavx512f
    Gather double-precision (64-bit) floating-point elements from memory using 32-bit indices. 64-bit elements are loaded from addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i32gather_psExperimentalavx512f
    Gather single-precision (32-bit) floating-point elements from memory using 32-bit indices. 32-bit elements are loaded from addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i32scatter_epi32Experimentalavx512f
    Scatter 32-bit integers from a into memory using 32-bit indices. 32-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i32scatter_epi64Experimentalavx512f
    Scatter 64-bit integers from a into memory using 32-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i32scatter_pdExperimentalavx512f
    Scatter double-precision (64-bit) floating-point elements from a into memory using 32-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i32scatter_psExperimentalavx512f
    Scatter single-precision (32-bit) floating-point elements from a into memory using 32-bit indices. 32-bit elements are stored at addresses starting at base_addr and offset by each 32-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i64gather_epi32Experimentalavx512f
    Gather 32-bit integers from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i64gather_epi64Experimentalavx512f
    Gather 64-bit integers from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i64gather_pdExperimentalavx512f
    Gather double-precision (64-bit) floating-point elements from memory using 64-bit indices. 64-bit elements are loaded from addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i64gather_psExperimentalavx512f
    Gather single-precision (32-bit) floating-point elements from memory using 64-bit indices. 32-bit elements are loaded from addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale). Gathered elements are merged into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i64scatter_epi32Experimentalavx512f
    Scatter 32-bit integers from a into memory using 64-bit indices. 32-bit elements are stored at addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i64scatter_epi64Experimentalavx512f
    Scatter 64-bit integers from a into memory using 64-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i64scatter_pdExperimentalavx512f
    Scatter double-precision (64-bit) floating-point elements from a into memory using 64-bit indices. 64-bit elements are stored at addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_i64scatter_psExperimentalavx512f
    Scatter single-precision (32-bit) floating-point elements from a into memory using 64-bit indices. 32-bit elements are stored at addresses starting at base_addr and offset by each 64-bit element in vindex (each index is scaled by the factor in scale) subject to mask k (elements are not stored when the corresponding mask bit is not set). scale should be 1, 2, 4 or 8.
  • _mm512_mask_insertf32x4Experimentalavx512f
    Copy a to tmp, then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from b into tmp at the location specified by imm8. Store tmp to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_insertf64x4Experimentalavx512f
    Copy a to tmp, then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from b into tmp at the location specified by imm8. Store tmp to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_inserti32x4Experimentalavx512f
    Copy a to tmp, then insert 128 bits (composed of 4 packed 32-bit integers) from b into tmp at the location specified by imm8. Store tmp to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_inserti64x4Experimentalavx512f
    Copy a to tmp, then insert 256 bits (composed of 4 packed 64-bit integers) from b into tmp at the location specified by imm8. Store tmp to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_load_epi32Experimentalavx512f
    Load packed 32-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_mask_load_epi64Experimentalavx512f
    Load packed 64-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_mask_load_pdExperimentalavx512f
    Load packed double-precision (64-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_mask_load_psExperimentalavx512f
    Load packed single-precision (32-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_mask_loadu_epi32Experimentalavx512f
    Load packed 32-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask_loadu_epi64Experimentalavx512f
    Load packed 64-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask_loadu_pdExperimentalavx512f
    Load packed double-precision (64-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask_loadu_psExperimentalavx512f
    Load packed single-precision (32-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask_max_epi32Experimentalavx512f
    Compare packed signed 32-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_max_epi64Experimentalavx512f
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_max_epu32Experimentalavx512f
    Compare packed unsigned 32-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_max_epu64Experimentalavx512f
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_max_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_max_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_max_round_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_max_round_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_min_epi32Experimentalavx512f
    Compare packed signed 32-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_min_epi64Experimentalavx512f
    Compare packed signed 64-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_min_epu32Experimentalavx512f
    Compare packed unsigned 32-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_min_epu64Experimentalavx512f
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_min_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_min_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_min_round_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_min_round_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_mask_mov_epi32Experimentalavx512f
    Move packed 32-bit integers from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mov_epi64Experimentalavx512f
    Move packed 64-bit integers from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mov_pdExperimentalavx512f
    Move packed double-precision (64-bit) floating-point elements from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mov_psExperimentalavx512f
    Move packed single-precision (32-bit) floating-point elements from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_movedup_pdExperimentalavx512f
    Duplicate even-indexed double-precision (64-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_movehdup_psExperimentalavx512f
    Duplicate odd-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_moveldup_psExperimentalavx512f
    Duplicate even-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mul_epi32Experimentalavx512f
    Multiply the low signed 32-bit integers from each packed 64-bit element in a and b, and store the signed 64-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mul_epu32Experimentalavx512f
    Multiply the low unsigned 32-bit integers from each packed 64-bit element in a and b, and store the unsigned 64-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mul_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mul_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mul_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mul_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mullo_epi32Experimentalavx512f
    Multiply the packed 32-bit integers in a and b, producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_mullox_epi64Experimentalavx512f
    Multiplies elements in packed 64-bit integer vectors a and b together, storing the lower 64 bits of the result in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_or_epi32Experimentalavx512f
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_or_epi64Experimentalavx512f
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permute_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permute_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permutevar_epi32Experimentalavx512f
    Shuffle 32-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). Note that this intrinsic shuffles across 128-bit lanes, unlike past intrinsics that use the permutevar name. This intrinsic is identical to _mm512_mask_permutexvar_epi32, and it is recommended that you use that intrinsic name.
  • _mm512_mask_permutevar_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permutevar_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permutex2var_epi32Experimentalavx512f
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_permutex2var_epi64Experimentalavx512f
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_permutex2var_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_permutex2var_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm512_mask_permutex_epi64Experimentalavx512f
    Shuffle 64-bit integers in a within 256-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permutex_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 256-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permutexvar_epi32Experimentalavx512f
    Shuffle 32-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permutexvar_epi64Experimentalavx512f
    Shuffle 64-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permutexvar_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_permutexvar_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_rcp14_pdExperimentalavx512f
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm512_mask_rcp14_psExperimentalavx512f
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm512_mask_reduce_add_epi32Experimentalavx512f
    Reduce the packed 32-bit integers in a by addition using mask k. Returns the sum of all active elements in a.
  • _mm512_mask_reduce_add_epi64Experimentalavx512f
    Reduce the packed 64-bit integers in a by addition using mask k. Returns the sum of all active elements in a.
  • _mm512_mask_reduce_add_pdExperimentalavx512f
    Reduce the packed double-precision (64-bit) floating-point elements in a by addition using mask k. Returns the sum of all active elements in a.
  • _mm512_mask_reduce_add_psExperimentalavx512f
    Reduce the packed single-precision (32-bit) floating-point elements in a by addition using mask k. Returns the sum of all active elements in a.
  • _mm512_mask_reduce_and_epi32Experimentalavx512f
    Reduce the packed 32-bit integers in a by bitwise AND using mask k. Returns the bitwise AND of all active elements in a.
  • _mm512_mask_reduce_and_epi64Experimentalavx512f
    Reduce the packed 64-bit integers in a by addition using mask k. Returns the sum of all active elements in a.
  • _mm512_mask_reduce_max_epi32Experimentalavx512f
    Reduce the packed signed 32-bit integers in a by maximum using mask k. Returns the maximum of all active elements in a.
  • _mm512_mask_reduce_max_epi64Experimentalavx512f
    Reduce the packed signed 64-bit integers in a by maximum using mask k. Returns the maximum of all active elements in a.
  • _mm512_mask_reduce_max_epu32Experimentalavx512f
    Reduce the packed unsigned 32-bit integers in a by maximum using mask k. Returns the maximum of all active elements in a.
  • _mm512_mask_reduce_max_epu64Experimentalavx512f
    Reduce the packed unsigned 64-bit integers in a by maximum using mask k. Returns the maximum of all active elements in a.
  • _mm512_mask_reduce_max_pdExperimentalavx512f
    Reduce the packed double-precision (64-bit) floating-point elements in a by maximum using mask k. Returns the maximum of all active elements in a.
  • _mm512_mask_reduce_max_psExperimentalavx512f
    Reduce the packed single-precision (32-bit) floating-point elements in a by maximum using mask k. Returns the maximum of all active elements in a.
  • _mm512_mask_reduce_min_epi32Experimentalavx512f
    Reduce the packed signed 32-bit integers in a by maximum using mask k. Returns the minimum of all active elements in a.
  • _mm512_mask_reduce_min_epi64Experimentalavx512f
    Reduce the packed signed 64-bit integers in a by maximum using mask k. Returns the minimum of all active elements in a.
  • _mm512_mask_reduce_min_epu32Experimentalavx512f
    Reduce the packed unsigned 32-bit integers in a by maximum using mask k. Returns the minimum of all active elements in a.
  • _mm512_mask_reduce_min_epu64Experimentalavx512f
    Reduce the packed signed 64-bit integers in a by maximum using mask k. Returns the minimum of all active elements in a.
  • _mm512_mask_reduce_min_pdExperimentalavx512f
    Reduce the packed double-precision (64-bit) floating-point elements in a by maximum using mask k. Returns the minimum of all active elements in a.
  • _mm512_mask_reduce_min_psExperimentalavx512f
    Reduce the packed single-precision (32-bit) floating-point elements in a by maximum using mask k. Returns the minimum of all active elements in a.
  • _mm512_mask_reduce_mul_epi32Experimentalavx512f
    Reduce the packed 32-bit integers in a by multiplication using mask k. Returns the product of all active elements in a.
  • _mm512_mask_reduce_mul_epi64Experimentalavx512f
    Reduce the packed 64-bit integers in a by multiplication using mask k. Returns the product of all active elements in a.
  • _mm512_mask_reduce_mul_pdExperimentalavx512f
    Reduce the packed double-precision (64-bit) floating-point elements in a by multiplication using mask k. Returns the product of all active elements in a.
  • _mm512_mask_reduce_mul_psExperimentalavx512f
    Reduce the packed single-precision (32-bit) floating-point elements in a by multiplication using mask k. Returns the product of all active elements in a.
  • _mm512_mask_reduce_or_epi32Experimentalavx512f
    Reduce the packed 32-bit integers in a by bitwise OR using mask k. Returns the bitwise OR of all active elements in a.
  • _mm512_mask_reduce_or_epi64Experimentalavx512f
    Reduce the packed 64-bit integers in a by bitwise OR using mask k. Returns the bitwise OR of all active elements in a.
  • _mm512_mask_rol_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_rol_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_rolv_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_rolv_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_ror_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_ror_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_rorv_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_rorv_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_roundscale_pdExperimentalavx512f
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_mask_roundscale_psExperimentalavx512f
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_mask_roundscale_round_pdExperimentalavx512f
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_mask_roundscale_round_psExperimentalavx512f
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_mask_rsqrt14_pdExperimentalavx512f
    Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm512_mask_rsqrt14_psExperimentalavx512f
    Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm512_mask_scalef_pdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_scalef_psExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_scalef_round_pdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_scalef_round_psExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_set1_epi32Experimentalavx512f
    Broadcast 32-bit integer a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_set1_epi64Experimentalavx512f
    Broadcast 64-bit integer a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_shuffle_epi32Experimentalavx512f
    Shuffle 32-bit integers in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_shuffle_f32x4Experimentalavx512f
    Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_shuffle_f64x2Experimentalavx512f
    Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_shuffle_i32x4Experimentalavx512f
    Shuffle 128-bits (composed of 4 32-bit integers) selected by imm8 from a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_shuffle_i64x2Experimentalavx512f
    Shuffle 128-bits (composed of 2 64-bit integers) selected by imm8 from a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_shuffle_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_shuffle_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sll_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sll_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_slli_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_slli_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sllv_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sllv_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sqrt_pdExperimentalavx512f
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sqrt_psExperimentalavx512f
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sqrt_round_pdExperimentalavx512f
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sqrt_round_psExperimentalavx512f
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sra_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sra_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srai_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srai_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srav_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srav_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srl_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srl_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srli_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srli_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srlv_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_srlv_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_store_epi32Experimentalavx512f
    Store packed 32-bit integers from a into memory using writemask k. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_mask_store_epi64Experimentalavx512f
    Store packed 64-bit integers from a into memory using writemask k. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_mask_store_pdExperimentalavx512f
    Store packed double-precision (64-bit) floating-point elements from a into memory using writemask k. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_mask_store_psExperimentalavx512f
    Store packed single-precision (32-bit) floating-point elements from a into memory using writemask k. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_mask_storeu_epi32Experimentalavx512f
    Store packed 32-bit integers from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask_storeu_epi64Experimentalavx512f
    Store packed 64-bit integers from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask_storeu_pdExperimentalavx512f
    Store packed double-precision (64-bit) floating-point elements from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask_storeu_psExperimentalavx512f
    Store packed single-precision (32-bit) floating-point elements from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_mask_sub_epi32Experimentalavx512f
    Subtract packed 32-bit integers in b from packed 32-bit integers in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sub_epi64Experimentalavx512f
    Subtract packed 64-bit integers in b from packed 64-bit integers in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sub_pdExperimentalavx512f
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sub_psExperimentalavx512f
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sub_round_pdExperimentalavx512f
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_sub_round_psExperimentalavx512f
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_ternarylogic_epi32Experimentalavx512f
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from src, a, and b are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using writemask k at 32-bit granularity (32-bit elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_ternarylogic_epi64Experimentalavx512f
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from src, a, and b are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using writemask k at 64-bit granularity (64-bit elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_test_epi32_maskExperimentalavx512f
    Compute the bitwise AND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is non-zero.
  • _mm512_mask_test_epi64_maskExperimentalavx512f
    Compute the bitwise AND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is non-zero.
  • _mm512_mask_testn_epi32_maskExperimentalavx512f
    Compute the bitwise NAND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is zero.
  • _mm512_mask_testn_epi64_maskExperimentalavx512f
    Compute the bitwise NAND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is zero.
  • _mm512_mask_unpackhi_epi32Experimentalavx512f
    Unpack and interleave 32-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_unpackhi_epi64Experimentalavx512f
    Unpack and interleave 64-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_unpackhi_pdExperimentalavx512f
    Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_unpackhi_psExperimentalavx512f
    Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_unpacklo_epi32Experimentalavx512f
    Unpack and interleave 32-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_unpacklo_epi64Experimentalavx512f
    Unpack and interleave 64-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_unpacklo_pdExperimentalavx512f
    Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_unpacklo_psExperimentalavx512f
    Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_xor_epi32Experimentalavx512f
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_mask_xor_epi64Experimentalavx512f
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm512_maskz_abs_epi32Experimentalavx512f
    Computes the absolute value of packed 32-bit integers in a, and store the unsigned results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_abs_epi64Experimentalavx512f
    Compute the absolute value of packed signed 64-bit integers in a, and store the unsigned results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_add_epi32Experimentalavx512f
    Add packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_add_epi64Experimentalavx512f
    Add packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_add_pdExperimentalavx512f
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_add_psExperimentalavx512f
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_add_round_pdExperimentalavx512f
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_add_round_psExperimentalavx512f
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_alignr_epi32Experimentalavx512f
    Concatenate a and b into a 128-byte immediate result, shift the result right by imm8 32-bit elements, and stores the low 64 bytes (16 elements) in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_alignr_epi64Experimentalavx512f
    Concatenate a and b into a 128-byte immediate result, shift the result right by imm8 64-bit elements, and stores the low 64 bytes (8 elements) in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_and_epi32Experimentalavx512f
    Compute the bitwise AND of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_and_epi64Experimentalavx512f
    Compute the bitwise AND of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_andnot_epi32Experimentalavx512f
    Compute the bitwise NOT of packed 32-bit integers in a and then AND with b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_andnot_epi64Experimentalavx512f
    Compute the bitwise NOT of packed 64-bit integers in a and then AND with b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_broadcast_f32x4Experimentalavx512f
    Broadcast the 4 packed single-precision (32-bit) floating-point elements from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_broadcast_f64x4Experimentalavx512f
    Broadcast the 4 packed double-precision (64-bit) floating-point elements from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_broadcast_i32x4Experimentalavx512f
    Broadcast the 4 packed 32-bit integers from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_broadcast_i64x4Experimentalavx512f
    Broadcast the 4 packed 64-bit integers from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_broadcastd_epi32Experimentalavx512f
    Broadcast the low packed 32-bit integer from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_broadcastq_epi64Experimentalavx512f
    Broadcast the low packed 64-bit integer from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_broadcastsd_pdExperimentalavx512f
    Broadcast the low double-precision (64-bit) floating-point element from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_broadcastss_psExperimentalavx512f
    Broadcast the low single-precision (32-bit) floating-point element from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_compress_epi32Experimentalavx512f
    Contiguously store the active 32-bit integers in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm512_maskz_compress_epi64Experimentalavx512f
    Contiguously store the active 64-bit integers in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm512_maskz_compress_pdExperimentalavx512f
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm512_maskz_compress_psExperimentalavx512f
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm512_maskz_cvt_roundepi32_psExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvt_roundepu32_psExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvt_roundpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvt_roundpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvt_roundpd_psExperimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvt_roundph_psExperimentalavx512f
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_cvt_roundps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvt_roundps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvt_roundps_pdExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_cvt_roundps_phExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_cvtepi8_epi32Experimentalavx512f
    Sign extend packed 8-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi8_epi64Experimentalavx512f
    Sign extend packed 8-bit integers in the low 8 bytes of a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi16_epi32Experimentalavx512f
    Sign extend packed 16-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi16_epi64Experimentalavx512f
    Sign extend packed 16-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi32_epi8Experimentalavx512f
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi32_epi16Experimentalavx512f
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi32_epi64Experimentalavx512f
    Sign extend packed 32-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi32_pdExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi32_psExperimentalavx512f
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi64_epi8Experimentalavx512f
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi64_epi16Experimentalavx512f
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepi64_epi32Experimentalavx512f
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepu8_epi32Experimentalavx512f
    Zero extend packed unsigned 8-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepu8_epi64Experimentalavx512f
    Zero extend packed unsigned 8-bit integers in the low 8 bytes of a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepu16_epi32Experimentalavx512f
    Zero extend packed unsigned 16-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepu16_epi64Experimentalavx512f
    Zero extend packed unsigned 16-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepu32_epi64Experimentalavx512f
    Zero extend packed unsigned 32-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepu32_pdExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtepu32_psExperimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtpd_psExperimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtph_psExperimentalavx512f
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtps_pdExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtps_phExperimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_cvtsepi32_epi8Experimentalavx512f
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtsepi32_epi16Experimentalavx512f
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm512_maskz_cvtsepi64_epi8Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtsepi64_epi16Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtsepi64_epi32Experimentalavx512f
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtt_roundpd_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_cvtt_roundpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_cvtt_roundps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_cvtt_roundps_epu32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_cvttpd_epi32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvttpd_epu32Experimentalavx512f
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvttps_epi32Experimentalavx512f
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvttps_epu32Experimentalavx512f
    Convert packed double-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtusepi32_epi8Experimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtusepi32_epi16Experimentalavx512f
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtusepi64_epi8Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtusepi64_epi16Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_cvtusepi64_epi32Experimentalavx512f
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_div_pdExperimentalavx512f
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_div_psExperimentalavx512f
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_div_round_pdExperimentalavx512f
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_div_round_psExperimentalavx512f
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_expand_epi32Experimentalavx512f
    Load contiguous active 32-bit integers from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_expand_epi64Experimentalavx512f
    Load contiguous active 64-bit integers from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_expand_pdExperimentalavx512f
    Load contiguous active double-precision (64-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_expand_psExperimentalavx512f
    Load contiguous active single-precision (32-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_expandloadu_epi32Experimentalavx512f
    Load contiguous active 32-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_expandloadu_epi64Experimentalavx512f
    Load contiguous active 64-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_expandloadu_pdExperimentalavx512f
    Load contiguous active single-precision (64-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_expandloadu_psExperimentalavx512f
    Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_extractf32x4_psExperimentalavx512f
    Extract 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from a, selected with imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_extractf64x4_pdExperimentalavx512f
    Extract 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from a, selected with imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_extracti32x4_epi32Experimentalavx512f
    Extract 128 bits (composed of 4 packed 32-bit integers) from a, selected with IMM2, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_extracti64x4_epi64Experimentalavx512f
    Extract 256 bits (composed of 4 packed 64-bit integers) from a, selected with IMM1, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fixupimm_pdExperimentalavx512f
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm512_maskz_fixupimm_psExperimentalavx512f
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm512_maskz_fixupimm_round_pdExperimentalavx512f
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm512_maskz_fixupimm_round_psExperimentalavx512f
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm512_maskz_fmadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in a using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmaddsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmaddsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmaddsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmaddsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmsubadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmsubadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmsubadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fmsubadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fnmadd_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fnmadd_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fnmadd_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fnmadd_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fnmsub_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fnmsub_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fnmsub_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_fnmsub_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_getexp_pdExperimentalavx512f
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm512_maskz_getexp_psExperimentalavx512f
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm512_maskz_getexp_round_pdExperimentalavx512f
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_getexp_round_psExperimentalavx512f
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_getmant_pdExperimentalavx512f
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm512_maskz_getmant_psExperimentalavx512f
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm512_maskz_getmant_round_pdExperimentalavx512f
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_getmant_round_psExperimentalavx512f
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_insertf32x4Experimentalavx512f
    Copy a to tmp, then insert 128 bits (composed of 4 packed single-precision (32-bit) floating-point elements) from b into tmp at the location specified by imm8. Store tmp to dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_insertf64x4Experimentalavx512f
    Copy a to tmp, then insert 256 bits (composed of 4 packed double-precision (64-bit) floating-point elements) from b into tmp at the location specified by imm8. Store tmp to dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_inserti32x4Experimentalavx512f
    Copy a to tmp, then insert 128 bits (composed of 4 packed 32-bit integers) from b into tmp at the location specified by imm8. Store tmp to dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_inserti64x4Experimentalavx512f
    Copy a to tmp, then insert 256 bits (composed of 4 packed 64-bit integers) from b into tmp at the location specified by imm8. Store tmp to dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_load_epi32Experimentalavx512f
    Load packed 32-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_maskz_load_epi64Experimentalavx512f
    Load packed 64-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_maskz_load_pdExperimentalavx512f
    Load packed double-precision (64-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_maskz_load_psExperimentalavx512f
    Load packed single-precision (32-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_maskz_loadu_epi32Experimentalavx512f
    Load packed 32-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm512_maskz_loadu_epi64Experimentalavx512f
    Load packed 64-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm512_maskz_loadu_pdExperimentalavx512f
    Load packed double-precision (64-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm512_maskz_loadu_psExperimentalavx512f
    Load packed single-precision (32-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm512_maskz_max_epi32Experimentalavx512f
    Compare packed signed 32-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_max_epi64Experimentalavx512f
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_max_epu32Experimentalavx512f
    Compare packed unsigned 32-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_max_epu64Experimentalavx512f
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_max_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_max_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_max_round_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_max_round_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_min_epi32Experimentalavx512f
    Compare packed signed 32-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_min_epi64Experimentalavx512f
    Compare packed signed 64-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_min_epu32Experimentalavx512f
    Compare packed unsigned 32-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_min_epu64Experimentalavx512f
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_min_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_min_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_min_round_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_min_round_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_maskz_mov_epi32Experimentalavx512f
    Move packed 32-bit integers from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mov_epi64Experimentalavx512f
    Move packed 64-bit integers from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mov_pdExperimentalavx512f
    Move packed double-precision (64-bit) floating-point elements from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mov_psExperimentalavx512f
    Move packed single-precision (32-bit) floating-point elements from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_movedup_pdExperimentalavx512f
    Duplicate even-indexed double-precision (64-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_movehdup_psExperimentalavx512f
    Duplicate odd-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_moveldup_psExperimentalavx512f
    Duplicate even-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mul_epi32Experimentalavx512f
    Multiply the low signed 32-bit integers from each packed 64-bit element in a and b, and store the signed 64-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mul_epu32Experimentalavx512f
    Multiply the low unsigned 32-bit integers from each packed 64-bit element in a and b, and store the unsigned 64-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mul_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mul_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mul_round_pdExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mul_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_mullo_epi32Experimentalavx512f
    Multiply the packed 32-bit integers in a and b, producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_or_epi32Experimentalavx512f
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_or_epi64Experimentalavx512f
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permute_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permute_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutevar_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutevar_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutex2var_epi32Experimentalavx512f
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutex2var_epi64Experimentalavx512f
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutex2var_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutex2var_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutex_epi64Experimentalavx512f
    Shuffle 64-bit integers in a within 256-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutex_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 256-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutexvar_epi32Experimentalavx512f
    Shuffle 32-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutexvar_epi64Experimentalavx512f
    Shuffle 64-bit integers in a across lanes using the corresponding index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutexvar_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_permutexvar_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_rcp14_pdExperimentalavx512f
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm512_maskz_rcp14_psExperimentalavx512f
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm512_maskz_rol_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_rol_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_rolv_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_rolv_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_ror_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_ror_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_rorv_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_rorv_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_roundscale_pdExperimentalavx512f
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_maskz_roundscale_psExperimentalavx512f
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_maskz_roundscale_round_pdExperimentalavx512f
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_maskz_roundscale_round_psExperimentalavx512f
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_maskz_rsqrt14_pdExperimentalavx512f
    Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm512_maskz_rsqrt14_psExperimentalavx512f
    Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm512_maskz_scalef_pdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_scalef_psExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_scalef_round_pdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_scalef_round_psExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_set1_epi32Experimentalavx512f
    Broadcast 32-bit integer a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_set1_epi64Experimentalavx512f
    Broadcast 64-bit integer a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_shuffle_epi32Experimentalavx512f
    Shuffle 32-bit integers in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_shuffle_f32x4Experimentalavx512f
    Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_shuffle_f64x2Experimentalavx512f
    Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_shuffle_i32x4Experimentalavx512f
    Shuffle 128-bits (composed of 4 32-bit integers) selected by imm8 from a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_shuffle_i64x2Experimentalavx512f
    Shuffle 128-bits (composed of 2 64-bit integers) selected by imm8 from a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_shuffle_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_shuffle_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sll_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sll_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_slli_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_slli_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sllv_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sllv_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sqrt_pdExperimentalavx512f
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sqrt_psExperimentalavx512f
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sqrt_round_pdExperimentalavx512f
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sqrt_round_psExperimentalavx512f
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sra_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sra_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srai_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srai_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srav_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srav_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srl_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srl_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srli_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srli_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srlv_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_srlv_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sub_epi32Experimentalavx512f
    Subtract packed 32-bit integers in b from packed 32-bit integers in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sub_epi64Experimentalavx512f
    Subtract packed 64-bit integers in b from packed 64-bit integers in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sub_pdExperimentalavx512f
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sub_psExperimentalavx512f
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sub_round_pdExperimentalavx512f
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_sub_round_psExperimentalavx512f
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_ternarylogic_epi32Experimentalavx512f
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using zeromask k at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_ternarylogic_epi64Experimentalavx512f
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using zeromask k at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_unpackhi_epi32Experimentalavx512f
    Unpack and interleave 32-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_unpackhi_epi64Experimentalavx512f
    Unpack and interleave 64-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_unpackhi_pdExperimentalavx512f
    Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_unpackhi_psExperimentalavx512f
    Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_unpacklo_epi32Experimentalavx512f
    Unpack and interleave 32-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_unpacklo_epi64Experimentalavx512f
    Unpack and interleave 64-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_unpacklo_pdExperimentalavx512f
    Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_unpacklo_psExperimentalavx512f
    Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_xor_epi32Experimentalavx512f
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_maskz_xor_epi64Experimentalavx512f
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm512_max_epi32Experimentalavx512f
    Compare packed signed 32-bit integers in a and b, and store packed maximum values in dst.
  • _mm512_max_epi64Experimentalavx512f
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst.
  • _mm512_max_epu32Experimentalavx512f
    Compare packed unsigned 32-bit integers in a and b, and store packed maximum values in dst.
  • _mm512_max_epu64Experimentalavx512f
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst.
  • _mm512_max_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst.
  • _mm512_max_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst.
  • _mm512_max_round_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_max_round_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_min_epi32Experimentalavx512f
    Compare packed signed 32-bit integers in a and b, and store packed minimum values in dst.
  • _mm512_min_epi64Experimentalavx512f
    Compare packed signed 64-bit integers in a and b, and store packed minimum values in dst.
  • _mm512_min_epu32Experimentalavx512f
    Compare packed unsigned 32-bit integers in a and b, and store packed minimum values in dst.
  • _mm512_min_epu64Experimentalavx512f
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst.
  • _mm512_min_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst. Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst.
  • _mm512_min_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst.
  • _mm512_min_round_pdExperimentalavx512f
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_min_round_psExperimentalavx512f
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm512_movedup_pdExperimentalavx512f
    Duplicate even-indexed double-precision (64-bit) floating-point elements from a, and store the results in dst.
  • _mm512_movehdup_psExperimentalavx512f
    Duplicate odd-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst.
  • _mm512_moveldup_psExperimentalavx512f
    Duplicate even-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst.
  • _mm512_mul_epi32Experimentalavx512f
    Multiply the low signed 32-bit integers from each packed 64-bit element in a and b, and store the signed 64-bit results in dst.
  • _mm512_mul_epu32Experimentalavx512f
    Multiply the low unsigned 32-bit integers from each packed 64-bit element in a and b, and store the unsigned 64-bit results in dst.
  • _mm512_mul_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst.
  • _mm512_mul_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst.
  • _mm512_mul_round_pdExperimentalavx512f
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst.
  • _mm512_mul_round_psExperimentalavx512f
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst.
  • _mm512_mullo_epi32Experimentalavx512f
    Multiply the packed 32-bit integers in a and b, producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in dst.
  • _mm512_mullox_epi64Experimentalavx512f
    Multiplies elements in packed 64-bit integer vectors a and b together, storing the lower 64 bits of the result in dst.
  • _mm512_or_epi32Experimentalavx512f
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst.
  • _mm512_or_epi64Experimentalavx512f
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the resut in dst.
  • _mm512_or_si512Experimentalavx512f
    Compute the bitwise OR of 512 bits (representing integer data) in a and b, and store the result in dst.
  • _mm512_permute_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst.
  • _mm512_permute_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst.
  • _mm512_permutevar_epi32Experimentalavx512f
    Shuffle 32-bit integers in a across lanes using the corresponding index in idx, and store the results in dst. Note that this intrinsic shuffles across 128-bit lanes, unlike past intrinsics that use the permutevar name. This intrinsic is identical to _mm512_permutexvar_epi32, and it is recommended that you use that intrinsic name.
  • _mm512_permutevar_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst.
  • _mm512_permutevar_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst.
  • _mm512_permutex2var_epi32Experimentalavx512f
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm512_permutex2var_epi64Experimentalavx512f
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm512_permutex2var_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm512_permutex2var_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm512_permutex_epi64Experimentalavx512f
    Shuffle 64-bit integers in a within 256-bit lanes using the control in imm8, and store the results in dst.
  • _mm512_permutex_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a within 256-bit lanes using the control in imm8, and store the results in dst.
  • _mm512_permutexvar_epi32Experimentalavx512f
    Shuffle 32-bit integers in a across lanes using the corresponding index in idx, and store the results in dst.
  • _mm512_permutexvar_epi64Experimentalavx512f
    Shuffle 64-bit integers in a across lanes using the corresponding index in idx, and store the results in dst.
  • _mm512_permutexvar_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements in a across lanes using the corresponding index in idx, and store the results in dst.
  • _mm512_permutexvar_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a across lanes using the corresponding index in idx.
  • _mm512_rcp14_pdExperimentalavx512f
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm512_rcp14_psExperimentalavx512f
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm512_reduce_add_epi32Experimentalavx512f
    Reduce the packed 32-bit integers in a by addition. Returns the sum of all elements in a.
  • _mm512_reduce_add_epi64Experimentalavx512f
    Reduce the packed 64-bit integers in a by addition. Returns the sum of all elements in a.
  • _mm512_reduce_add_pdExperimentalavx512f
    Reduce the packed double-precision (64-bit) floating-point elements in a by addition. Returns the sum of all elements in a.
  • _mm512_reduce_add_psExperimentalavx512f
    Reduce the packed single-precision (32-bit) floating-point elements in a by addition. Returns the sum of all elements in a.
  • _mm512_reduce_and_epi32Experimentalavx512f
    Reduce the packed 32-bit integers in a by bitwise AND. Returns the bitwise AND of all elements in a.
  • _mm512_reduce_and_epi64Experimentalavx512f
    Reduce the packed 64-bit integers in a by bitwise AND. Returns the bitwise AND of all elements in a.
  • _mm512_reduce_max_epi32Experimentalavx512f
    Reduce the packed signed 32-bit integers in a by maximum. Returns the maximum of all elements in a.
  • _mm512_reduce_max_epi64Experimentalavx512f
    Reduce the packed signed 64-bit integers in a by maximum. Returns the maximum of all elements in a.
  • _mm512_reduce_max_epu32Experimentalavx512f
    Reduce the packed unsigned 32-bit integers in a by maximum. Returns the maximum of all elements in a.
  • _mm512_reduce_max_epu64Experimentalavx512f
    Reduce the packed unsigned 64-bit integers in a by maximum. Returns the maximum of all elements in a.
  • _mm512_reduce_max_pdExperimentalavx512f
    Reduce the packed double-precision (64-bit) floating-point elements in a by maximum. Returns the maximum of all elements in a.
  • _mm512_reduce_max_psExperimentalavx512f
    Reduce the packed single-precision (32-bit) floating-point elements in a by maximum. Returns the maximum of all elements in a.
  • _mm512_reduce_min_epi32Experimentalavx512f
    Reduce the packed signed 32-bit integers in a by minimum. Returns the minimum of all elements in a.
  • _mm512_reduce_min_epi64Experimentalavx512f
    Reduce the packed signed 64-bit integers in a by minimum. Returns the minimum of all elements in a.
  • _mm512_reduce_min_epu32Experimentalavx512f
    Reduce the packed unsigned 32-bit integers in a by minimum. Returns the minimum of all elements in a.
  • _mm512_reduce_min_epu64Experimentalavx512f
    Reduce the packed unsigned 64-bit integers in a by minimum. Returns the minimum of all elements in a.
  • _mm512_reduce_min_pdExperimentalavx512f
    Reduce the packed double-precision (64-bit) floating-point elements in a by minimum. Returns the minimum of all elements in a.
  • _mm512_reduce_min_psExperimentalavx512f
    Reduce the packed single-precision (32-bit) floating-point elements in a by minimum. Returns the minimum of all elements in a.
  • _mm512_reduce_mul_epi32Experimentalavx512f
    Reduce the packed 32-bit integers in a by multiplication. Returns the product of all elements in a.
  • _mm512_reduce_mul_epi64Experimentalavx512f
    Reduce the packed 64-bit integers in a by multiplication. Returns the product of all elements in a.
  • _mm512_reduce_mul_pdExperimentalavx512f
    Reduce the packed double-precision (64-bit) floating-point elements in a by multiplication. Returns the product of all elements in a.
  • _mm512_reduce_mul_psExperimentalavx512f
    Reduce the packed single-precision (32-bit) floating-point elements in a by multiplication. Returns the product of all elements in a.
  • _mm512_reduce_or_epi32Experimentalavx512f
    Reduce the packed 32-bit integers in a by bitwise OR. Returns the bitwise OR of all elements in a.
  • _mm512_reduce_or_epi64Experimentalavx512f
    Reduce the packed 64-bit integers in a by bitwise OR. Returns the bitwise OR of all elements in a.
  • _mm512_rol_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst.
  • _mm512_rol_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst.
  • _mm512_rolv_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm512_rolv_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm512_ror_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst.
  • _mm512_ror_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst.
  • _mm512_rorv_epi32Experimentalavx512f
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm512_rorv_epi64Experimentalavx512f
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm512_roundscale_pdExperimentalavx512f
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_roundscale_psExperimentalavx512f
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_roundscale_round_pdExperimentalavx512f
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_roundscale_round_psExperimentalavx512f
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm512_rsqrt14_pdExperimentalavx512f
    Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm512_rsqrt14_psExperimentalavx512f
    Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm512_scalef_pdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst.
  • _mm512_scalef_psExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst.
  • _mm512_scalef_round_pdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst.
  • _mm512_scalef_round_psExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst.
  • _mm512_set1_epi8Experimentalavx512f
    Broadcast 8-bit integer a to all elements of dst.
  • _mm512_set1_epi16Experimentalavx512f
    Broadcast the low packed 16-bit integer from a to all elements of dst.
  • _mm512_set1_epi32Experimentalavx512f
    Broadcast 32-bit integer a to all elements of dst.
  • _mm512_set1_epi64Experimentalavx512f
    Broadcast 64-bit integer a to all elements of dst.
  • _mm512_set1_pdExperimentalavx512f
    Broadcast 64-bit float a to all elements of dst.
  • _mm512_set1_psExperimentalavx512f
    Broadcast 32-bit float a to all elements of dst.
  • _mm512_set4_epi32Experimentalavx512f
    Set packed 32-bit integers in dst with the repeated 4 element sequence.
  • _mm512_set4_epi64Experimentalavx512f
    Set packed 64-bit integers in dst with the repeated 4 element sequence.
  • _mm512_set4_pdExperimentalavx512f
    Set packed double-precision (64-bit) floating-point elements in dst with the repeated 4 element sequence.
  • _mm512_set4_psExperimentalavx512f
    Set packed single-precision (32-bit) floating-point elements in dst with the repeated 4 element sequence.
  • _mm512_set_epi8Experimentalavx512f
    Set packed 8-bit integers in dst with the supplied values.
  • _mm512_set_epi16Experimentalavx512f
    Set packed 16-bit integers in dst with the supplied values.
  • _mm512_set_epi32Experimentalavx512f
    Sets packed 32-bit integers in dst with the supplied values.
  • _mm512_set_epi64Experimentalavx512f
    Set packed 64-bit integers in dst with the supplied values.
  • _mm512_set_pdExperimentalavx512f
    Set packed double-precision (64-bit) floating-point elements in dst with the supplied values.
  • _mm512_set_psExperimentalavx512f
    Sets packed 32-bit integers in dst with the supplied values.
  • _mm512_setr4_epi32Experimentalavx512f
    Set packed 32-bit integers in dst with the repeated 4 element sequence in reverse order.
  • _mm512_setr4_epi64Experimentalavx512f
    Set packed 64-bit integers in dst with the repeated 4 element sequence in reverse order.
  • _mm512_setr4_pdExperimentalavx512f
    Set packed double-precision (64-bit) floating-point elements in dst with the repeated 4 element sequence in reverse order.
  • _mm512_setr4_psExperimentalavx512f
    Set packed single-precision (32-bit) floating-point elements in dst with the repeated 4 element sequence in reverse order.
  • _mm512_setr_epi32Experimentalavx512f
    Sets packed 32-bit integers in dst with the supplied values in reverse order.
  • _mm512_setr_epi64Experimentalavx512f
    Set packed 64-bit integers in dst with the supplied values in reverse order.
  • _mm512_setr_pdExperimentalavx512f
    Set packed double-precision (64-bit) floating-point elements in dst with the supplied values in reverse order.
  • _mm512_setr_psExperimentalavx512f
    Sets packed 32-bit integers in dst with the supplied values in reverse order.
  • _mm512_setzeroExperimentalavx512f
    Return vector of type __m512 with all elements set to zero.
  • _mm512_setzero_epi32Experimentalavx512f
    Return vector of type __m512i with all elements set to zero.
  • _mm512_setzero_pdExperimentalavx512f
    Returns vector of type __m512d with all elements set to zero.
  • _mm512_setzero_psExperimentalavx512f
    Returns vector of type __m512d with all elements set to zero.
  • _mm512_setzero_si512Experimentalavx512f
    Returns vector of type __m512i with all elements set to zero.
  • _mm512_shuffle_epi32Experimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst.
  • _mm512_shuffle_f32x4Experimentalavx512f
    Shuffle 128-bits (composed of 4 single-precision (32-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst.
  • _mm512_shuffle_f64x2Experimentalavx512f
    Shuffle 128-bits (composed of 2 double-precision (64-bit) floating-point elements) selected by imm8 from a and b, and store the results in dst.
  • _mm512_shuffle_i32x4Experimentalavx512f
    Shuffle 128-bits (composed of 4 32-bit integers) selected by imm8 from a and b, and store the results in dst.
  • _mm512_shuffle_i64x2Experimentalavx512f
    Shuffle 128-bits (composed of 2 64-bit integers) selected by imm8 from a and b, and store the results in dst.
  • _mm512_shuffle_pdExperimentalavx512f
    Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in imm8, and store the results in dst.
  • _mm512_shuffle_psExperimentalavx512f
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst.
  • _mm512_sll_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by count while shifting in zeros, and store the results in dst.
  • _mm512_sll_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by count while shifting in zeros, and store the results in dst.
  • _mm512_slli_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by imm8 while shifting in zeros, and store the results in dst.
  • _mm512_slli_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by imm8 while shifting in zeros, and store the results in dst.
  • _mm512_sllv_epi32Experimentalavx512f
    Shift packed 32-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst.
  • _mm512_sllv_epi64Experimentalavx512f
    Shift packed 64-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst.
  • _mm512_sqrt_pdExperimentalavx512f
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst.
  • _mm512_sqrt_psExperimentalavx512f
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst.
  • _mm512_sqrt_round_pdExperimentalavx512f
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst.
  • _mm512_sqrt_round_psExperimentalavx512f
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst.
  • _mm512_sra_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by count while shifting in sign bits, and store the results in dst.
  • _mm512_sra_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst.
  • _mm512_srai_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst.
  • _mm512_srai_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst.
  • _mm512_srav_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst.
  • _mm512_srav_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst.
  • _mm512_srl_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by count while shifting in zeros, and store the results in dst.
  • _mm512_srl_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by count while shifting in zeros, and store the results in dst.
  • _mm512_srli_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by imm8 while shifting in zeros, and store the results in dst.
  • _mm512_srli_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by imm8 while shifting in zeros, and store the results in dst.
  • _mm512_srlv_epi32Experimentalavx512f
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst.
  • _mm512_srlv_epi64Experimentalavx512f
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst.
  • _mm512_store_epi32Experimentalavx512f
    Store 512-bits (composed of 16 packed 32-bit integers) from a into memory. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_store_epi64Experimentalavx512f
    Store 512-bits (composed of 8 packed 64-bit integers) from a into memory. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_store_pdExperimentalavx512f
    Store 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from a into memory. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_store_psExperimentalavx512f
    Store 512-bits of integer data from a into memory. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_store_si512Experimentalavx512f
    Store 512-bits of integer data from a into memory. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_storeu_epi32Experimentalavx512f
    Store 512-bits (composed of 16 packed 32-bit integers) from a into memory. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_storeu_epi64Experimentalavx512f
    Store 512-bits (composed of 8 packed 64-bit integers) from a into memory. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_storeu_pdExperimentalavx512f
    Stores 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from a into memory. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_storeu_si512Experimentalavx512f
    Store 512-bits of integer data from a into memory. mem_addr does not need to be aligned on any particular boundary.
  • _mm512_stream_pdExperimentalavx512f
    Store 512-bits (composed of 8 packed double-precision (64-bit) floating-point elements) from a into memory using a non-temporal memory hint. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_stream_psExperimentalavx512f
    Store 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from a into memory using a non-temporal memory hint. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_stream_si512Experimentalavx512f
    Store 512-bits of integer data from a into memory using a non-temporal memory hint. mem_addr must be aligned on a 64-byte boundary or a general-protection exception may be generated.
  • _mm512_sub_epi32Experimentalavx512f
    Subtract packed 32-bit integers in b from packed 32-bit integers in a, and store the results in dst.
  • _mm512_sub_epi64Experimentalavx512f
    Subtract packed 64-bit integers in b from packed 64-bit integers in a, and store the results in dst.
  • _mm512_sub_pdExperimentalavx512f
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst.
  • _mm512_sub_psExperimentalavx512f
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst.
  • _mm512_sub_round_pdExperimentalavx512f
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst.
  • _mm512_sub_round_psExperimentalavx512f
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst.
  • _mm512_ternarylogic_epi32Experimentalavx512f
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst.
  • _mm512_ternarylogic_epi64Experimentalavx512f
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst.
  • _mm512_test_epi32_maskExperimentalavx512f
    Compute the bitwise AND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k if the intermediate value is non-zero.
  • _mm512_test_epi64_maskExperimentalavx512f
    Compute the bitwise AND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k if the intermediate value is non-zero.
  • _mm512_testn_epi32_maskExperimentalavx512f
    Compute the bitwise NAND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k if the intermediate value is zero.
  • _mm512_testn_epi64_maskExperimentalavx512f
    Compute the bitwise NAND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k if the intermediate value is zero.
  • _mm512_undefinedExperimentalavx512f
    Return vector of type __m512 with indeterminate elements. Despite being “undefined”, this is some valid value and not equivalent to mem::MaybeUninit. In practice, this is equivalent to mem::zeroed.
  • _mm512_undefined_epi32Experimentalavx512f
    Return vector of type __m512i with indeterminate elements. Despite being “undefined”, this is some valid value and not equivalent to mem::MaybeUninit. In practice, this is equivalent to mem::zeroed.
  • _mm512_undefined_pdExperimentalavx512f
    Returns vector of type __m512d with indeterminate elements. Despite being “undefined”, this is some valid value and not equivalent to mem::MaybeUninit. In practice, this is equivalent to mem::zeroed.
  • _mm512_undefined_psExperimentalavx512f
    Returns vector of type __m512 with indeterminate elements. Despite being “undefined”, this is some valid value and not equivalent to mem::MaybeUninit. In practice, this is equivalent to mem::zeroed.
  • _mm512_unpackhi_epi32Experimentalavx512f
    Unpack and interleave 32-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst.
  • _mm512_unpackhi_epi64Experimentalavx512f
    Unpack and interleave 64-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst.
  • _mm512_unpackhi_pdExperimentalavx512f
    Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst.
  • _mm512_unpackhi_psExperimentalavx512f
    Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst.
  • _mm512_unpacklo_epi32Experimentalavx512f
    Unpack and interleave 32-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst.
  • _mm512_unpacklo_epi64Experimentalavx512f
    Unpack and interleave 64-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst.
  • _mm512_unpacklo_pdExperimentalavx512f
    Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst.
  • _mm512_unpacklo_psExperimentalavx512f
    Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst.
  • _mm512_xor_epi32Experimentalavx512f
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst.
  • _mm512_xor_epi64Experimentalavx512f
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst.
  • _mm512_xor_si512Experimentalavx512f
    Compute the bitwise XOR of 512 bits (representing integer data) in a and b, and store the result in dst.
  • _mm512_zextpd128_pd512Experimentalavx512f
    Cast vector of type __m128d to type __m512d; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_zextpd256_pd512Experimentalavx512f
    Cast vector of type __m256d to type __m512d; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_zextps128_ps512Experimentalavx512f
    Cast vector of type __m128 to type __m512; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_zextps256_ps512Experimentalavx512f
    Cast vector of type __m256 to type __m512; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_zextsi128_si512Experimentalavx512f
    Cast vector of type __m128i to type __m512i; the upper 384 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm512_zextsi256_si512Experimentalavx512f
    Cast vector of type __m256i to type __m512i; the upper 256 bits of the result are zeroed. This intrinsic is only used for compilation and does not generate any instructions, thus it has zero latency.
  • _mm_add_round_sdExperimentalavx512f
    Add the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_add_round_ssExperimentalavx512f
    Add the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_alignr_epi32Experimentalavx512f,avx512vl
    Concatenate a and b into a 32-byte immediate result, shift the result right by imm8 32-bit elements, and store the low 16 bytes (4 elements) in dst.
  • _mm_alignr_epi64Experimentalavx512f,avx512vl
    Concatenate a and b into a 32-byte immediate result, shift the result right by imm8 64-bit elements, and store the low 16 bytes (2 elements) in dst.
  • _mm_cmp_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm_cmp_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm_cmp_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm_cmp_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm_cmp_pd_maskExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm_cmp_ps_maskExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k.
  • _mm_cmp_round_sd_maskExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point element in a and b based on the comparison operand specified by imm8, and store the result in mask vector k.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cmp_round_ss_maskExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point element in a and b based on the comparison operand specified by imm8, and store the result in mask vector k.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cmp_sd_maskExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point element in a and b based on the comparison operand specified by imm8, and store the result in mask vector k.
  • _mm_cmp_ss_maskExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point element in a and b based on the comparison operand specified by imm8, and store the result in mask vector k.
  • _mm_cmpeq_epi32_maskExperimentalavx512f,avx512vl
    Compare packed 32-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm_cmpeq_epi64_maskExperimentalavx512f,avx512vl
    Compare packed 64-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm_cmpeq_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm_cmpeq_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for equality, and store the results in mask vector k.
  • _mm_cmpge_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm_cmpge_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm_cmpge_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm_cmpge_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k.
  • _mm_cmpgt_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm_cmpgt_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm_cmpgt_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm_cmpgt_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for greater-than, and store the results in mask vector k.
  • _mm_cmple_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm_cmple_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm_cmple_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm_cmple_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k.
  • _mm_cmplt_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm_cmplt_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm_cmplt_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm_cmplt_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for less-than, and store the results in mask vector k.
  • _mm_cmpneq_epi32_maskExperimentalavx512f,avx512vl
    Compare packed 32-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm_cmpneq_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm_cmpneq_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm_cmpneq_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for not-equal, and store the results in mask vector k.
  • _mm_comi_round_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point element in a and b based on the comparison operand specified by imm8, and return the boolean result (0 or 1).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_comi_round_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point element in a and b based on the comparison operand specified by imm8, and return the boolean result (0 or 1).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cvt_roundi32_ssExperimentalavx512f
    Convert the signed 32-bit integer b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_cvt_roundsd_i32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to a 32-bit integer, and store the result in dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_cvt_roundsd_si32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to a 32-bit integer, and store the result in dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_cvt_roundsd_ssExperimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_cvt_roundsd_u32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to an unsigned 32-bit integer, and store the result in dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_cvt_roundsi32_ssExperimentalavx512f
    Convert the signed 32-bit integer b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_cvt_roundss_i32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to a 32-bit integer, and store the result in dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_cvt_roundss_sdExperimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in b to a double-precision (64-bit) floating-point element, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cvt_roundss_si32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to a 32-bit integer, and store the result in dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_cvt_roundss_u32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to an unsigned 32-bit integer, and store the result in dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_cvt_roundu32_ssExperimentalavx512f
    Convert the unsigned 32-bit integer b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_cvtepi32_epi8Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst.
  • _mm_cvtepi32_epi16Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst.
  • _mm_cvtepi64_epi8Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst.
  • _mm_cvtepi64_epi16Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst.
  • _mm_cvtepi64_epi32Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst.
  • _mm_cvtepu32_pdExperimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst.
  • _mm_cvti32_sdExperimentalavx512f
    Convert the signed 32-bit integer b to a double-precision (64-bit) floating-point element, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_cvti32_ssExperimentalavx512f
    Convert the signed 32-bit integer b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_cvtpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst.
  • _mm_cvtps_epu32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst.
  • _mm_cvtsd_i32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to a 32-bit integer, and store the result in dst.
  • _mm_cvtsd_u32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to an unsigned 32-bit integer, and store the result in dst.
  • _mm_cvtsepi32_epi8Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst.
  • _mm_cvtsepi32_epi16Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm_cvtsepi64_epi8Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst.
  • _mm_cvtsepi64_epi16Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm_cvtsepi64_epi32Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst.
  • _mm_cvtss_i32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to a 32-bit integer, and store the result in dst.
  • _mm_cvtss_u32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to an unsigned 32-bit integer, and store the result in dst.
  • _mm_cvtt_roundsd_i32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to a 32-bit integer with truncation, and store the result in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cvtt_roundsd_si32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to a 32-bit integer with truncation, and store the result in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cvtt_roundsd_u32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to an unsigned 32-bit integer with truncation, and store the result in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cvtt_roundss_i32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to a 32-bit integer with truncation, and store the result in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cvtt_roundss_si32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to a 32-bit integer with truncation, and store the result in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cvtt_roundss_u32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to an unsigned 32-bit integer with truncation, and store the result in dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_cvttpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst.
  • _mm_cvttps_epu32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst.
  • _mm_cvttsd_i32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to a 32-bit integer with truncation, and store the result in dst.
  • _mm_cvttsd_u32Experimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in a to an unsigned 32-bit integer with truncation, and store the result in dst.
  • _mm_cvttss_i32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to a 32-bit integer with truncation, and store the result in dst.
  • _mm_cvttss_u32Experimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in a to an unsigned 32-bit integer with truncation, and store the result in dst.
  • _mm_cvtu32_sdExperimentalavx512f
    Convert the unsigned 32-bit integer b to a double-precision (64-bit) floating-point element, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_cvtu32_ssExperimentalavx512f
    Convert the unsigned 32-bit integer b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_cvtusepi32_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst.
  • _mm_cvtusepi32_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst.
  • _mm_cvtusepi64_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst.
  • _mm_cvtusepi64_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst.
  • _mm_cvtusepi64_epi32Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst.
  • _mm_div_round_sdExperimentalavx512f
    Divide the lower double-precision (64-bit) floating-point element in a by the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_div_round_ssExperimentalavx512f
    Divide the lower single-precision (32-bit) floating-point element in a by the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_fixupimm_pdExperimentalavx512f,avx512vl
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst. imm8 is used to set the required flags reporting.
  • _mm_fixupimm_psExperimentalavx512f,avx512vl
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst. imm8 is used to set the required flags reporting.
  • _mm_fixupimm_round_sdExperimentalavx512f
    Fix up the lower double-precision (64-bit) floating-point elements in a and b using the lower 64-bit integer in c, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst. imm8 is used to set the required flags reporting.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_fixupimm_round_ssExperimentalavx512f
    Fix up the lower single-precision (32-bit) floating-point elements in a and b using the lower 32-bit integer in c, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst. imm8 is used to set the required flags reporting.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_fixupimm_sdExperimentalavx512f
    Fix up the lower double-precision (64-bit) floating-point elements in a and b using the lower 64-bit integer in c, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst. imm8 is used to set the required flags reporting.
  • _mm_fixupimm_ssExperimentalavx512f
    Fix up the lower single-precision (32-bit) floating-point elements in a and b using the lower 32-bit integer in c, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst. imm8 is used to set the required flags reporting.
  • _mm_fmadd_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_fmadd_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_fmsub_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_fmsub_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_fnmadd_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_fnmadd_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_fnmsub_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_fnmsub_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, subtract the lower element in c from the negated intermediate result, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_getexp_pdExperimentalavx512f,avx512vl
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst. This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm_getexp_psExperimentalavx512f,avx512vl
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst. This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm_getexp_round_sdExperimentalavx512f
    Convert the exponent of the lower double-precision (64-bit) floating-point element in b to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_getexp_round_ssExperimentalavx512f
    Convert the exponent of the lower single-precision (32-bit) floating-point element in b to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_getexp_sdExperimentalavx512f
    Convert the exponent of the lower double-precision (64-bit) floating-point element in b to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
  • _mm_getexp_ssExperimentalavx512f
    Convert the exponent of the lower single-precision (32-bit) floating-point element in b to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
  • _mm_getmant_pdExperimentalavx512f,avx512vl
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm_getmant_psExperimentalavx512f,avx512vl
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign. The mantissa is normalized to the interval specified by interv, which can take the following values: _MM_MANT_NORM_1_2 // interval [1, 2) _MM_MANT_NORM_p5_2 // interval [0.5, 2) _MM_MANT_NORM_p5_1 // interval [0.5, 1) _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5) The sign is determined by sc which can take the following values: _MM_MANT_SIGN_src // sign = sign(src) _MM_MANT_SIGN_zero // sign = 0 _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm_getmant_round_sdExperimentalavx512f
    Normalize the mantissas of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_getmant_round_ssExperimentalavx512f
    Normalize the mantissas of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_getmant_sdExperimentalavx512f
    Normalize the mantissas of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_getmant_ssExperimentalavx512f
    Normalize the mantissas of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_load_epi32Experimentalavx512f,avx512vl
    Load 128-bits (composed of 4 packed 32-bit integers) from memory into dst. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_load_epi64Experimentalavx512f,avx512vl
    Load 128-bits (composed of 2 packed 64-bit integers) from memory into dst. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_loadu_epi32Experimentalavx512f,avx512vl
    Load 128-bits (composed of 4 packed 32-bit integers) from memory into dst. mem_addr does not need to be aligned on any particular boundary.
  • _mm_loadu_epi64Experimentalavx512f,avx512vl
    Load 128-bits (composed of 2 packed 64-bit integers) from memory into dst. mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask2_permutex2var_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm_mask2_permutex2var_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm_mask2_permutex2var_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set)
  • _mm_mask2_permutex2var_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from idx when the corresponding mask bit is not set).
  • _mm_mask3_fmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fmadd_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from c to the upper element of dst.
  • _mm_mask3_fmadd_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from c to the upper elements of dst.
  • _mm_mask3_fmadd_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from c to the upper element of dst.
  • _mm_mask3_fmadd_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from c to the upper elements of dst.
  • _mm_mask3_fmaddsub_pdExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fmaddsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fmsub_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from c to the upper element of dst.
  • _mm_mask3_fmsub_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from c to the upper elements of dst.
  • _mm_mask3_fmsub_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from c to the upper element of dst.
  • _mm_mask3_fmsub_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from c to the upper elements of dst.
  • _mm_mask3_fmsubadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fmsubadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fnmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fnmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fnmadd_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from c to the upper element of dst.
  • _mm_mask3_fnmadd_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from c to the upper elements of dst.
  • _mm_mask3_fnmadd_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from c to the upper element of dst.
  • _mm_mask3_fnmadd_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from c to the upper elements of dst.
  • _mm_mask3_fnmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fnmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from c when the corresponding mask bit is not set).
  • _mm_mask3_fnmsub_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from c to the upper element of dst.
  • _mm_mask3_fnmsub_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from c to the upper elements of dst.
  • _mm_mask3_fnmsub_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from c to the upper element of dst.
  • _mm_mask3_fnmsub_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from c to the upper elements of dst.
  • _mm_mask_abs_epi32Experimentalavx512f,avx512vl
    Compute the absolute value of packed signed 32-bit integers in a, and store the unsigned results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_add_epi32Experimentalavx512f,avx512vl
    Add packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_add_epi64Experimentalavx512f,avx512vl
    Add packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_add_pdExperimentalavx512f,avx512vl
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_add_psExperimentalavx512f,avx512vl
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_add_round_sdExperimentalavx512f
    Add the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_add_round_ssExperimentalavx512f
    Add the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_add_sdExperimentalavx512f
    Add the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_add_ssExperimentalavx512f
    Add the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_alignr_epi32Experimentalavx512f,avx512vl
    Concatenate a and b into a 32-byte immediate result, shift the result right by imm8 32-bit elements, and store the low 16 bytes (4 elements) in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_alignr_epi64Experimentalavx512f,avx512vl
    Concatenate a and b into a 32-byte immediate result, shift the result right by imm8 64-bit elements, and store the low 16 bytes (2 elements) in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_and_epi32Experimentalavx512f,avx512vl
    Performs element-by-element bitwise AND between packed 32-bit integer elements of a and b, storing the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_and_epi64Experimentalavx512f,avx512vl
    Compute the bitwise AND of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_andnot_epi32Experimentalavx512f,avx512vl
    Compute the bitwise NOT of packed 32-bit integers in a and then AND with b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_andnot_epi64Experimentalavx512f,avx512vl
    Compute the bitwise NOT of packed 64-bit integers in a and then AND with b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_blend_epi32Experimentalavx512f,avx512vl
    Blend packed 32-bit integers from a and b using control mask k, and store the results in dst.
  • _mm_mask_blend_epi64Experimentalavx512f,avx512vl
    Blend packed 64-bit integers from a and b using control mask k, and store the results in dst.
  • _mm_mask_blend_pdExperimentalavx512f,avx512vl
    Blend packed double-precision (64-bit) floating-point elements from a and b using control mask k, and store the results in dst.
  • _mm_mask_blend_psExperimentalavx512f,avx512vl
    Blend packed single-precision (32-bit) floating-point elements from a and b using control mask k, and store the results in dst.
  • _mm_mask_broadcastd_epi32Experimentalavx512f,avx512vl
    Broadcast the low packed 32-bit integer from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_broadcastq_epi64Experimentalavx512f,avx512vl
    Broadcast the low packed 64-bit integer from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_broadcastss_psExperimentalavx512f,avx512vl
    Broadcast the low single-precision (32-bit) floating-point element from a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cmp_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmp_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmp_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmp_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmp_pd_maskExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmp_ps_maskExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b based on the comparison operand specified by imm8, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmp_round_sd_maskExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point element in a and b based on the comparison operand specified by imm8, and store the result in mask vector k using zeromask k1 (the element is zeroed out when mask bit 0 is not set).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_cmp_round_ss_maskExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point element in a and b based on the comparison operand specified by imm8, and store the result in mask vector k using zeromask k1 (the element is zeroed out when mask bit 0 is not seti).
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_cmp_sd_maskExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point element in a and b based on the comparison operand specified by imm8, and store the result in mask vector k using zeromask k1 (the element is zeroed out when mask bit 0 is not set).
  • _mm_mask_cmp_ss_maskExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point element in a and b based on the comparison operand specified by imm8, and store the result in mask vector k using zeromask k1 (the element is zeroed out when mask bit 0 is not set).
  • _mm_mask_cmpeq_epi32_maskExperimentalavx512f,avx512vl
    Compare packed 32-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpeq_epi64_maskExperimentalavx512f,avx512vl
    Compare packed 64-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpeq_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpeq_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for equality, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpge_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpge_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpge_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpge_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for greater-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpgt_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpgt_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpgt_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpgt_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for greater-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmple_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmple_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmple_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmple_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmplt_epi32_maskExperimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmplt_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmplt_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for less-than-or-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmplt_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for less-than, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpneq_epi32_maskExperimentalavx512f,avx512vl
    Compare packed 32-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpneq_epi64_maskExperimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpneq_epu32_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_cmpneq_epu64_maskExperimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b for not-equal, and store the results in mask vector k using zeromask k1 (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_mask_compress_epi32Experimentalavx512f,avx512vl
    Contiguously store the active 32-bit integers in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm_mask_compress_epi64Experimentalavx512f,avx512vl
    Contiguously store the active 64-bit integers in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm_mask_compress_pdExperimentalavx512f,avx512vl
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm_mask_compress_psExperimentalavx512f,avx512vl
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in writemask k) to dst, and pass through the remaining elements from src.
  • _mm_mask_compressstoreu_epi32Experimentalavx512f,avx512vl
    Contiguously store the active 32-bit integers in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_compressstoreu_epi64Experimentalavx512f,avx512vl
    Contiguously store the active 64-bit integers in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_compressstoreu_pdExperimentalavx512f,avx512vl
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_compressstoreu_psExperimentalavx512f,avx512vl
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvt_roundps_phExperimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_cvt_roundsd_ssExperimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_cvt_roundss_sdExperimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in b to a double-precision (64-bit) floating-point element, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_cvtepi8_epi32Experimentalavx512f,avx512vl
    Sign extend packed 8-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi8_epi64Experimentalavx512f,avx512vl
    Sign extend packed 8-bit integers in the low 2 bytes of a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi16_epi32Experimentalavx512f,avx512vl
    Sign extend packed 16-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi16_epi64Experimentalavx512f,avx512vl
    Sign extend packed 16-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi32_epi8Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi32_epi16Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi32_epi64Experimentalavx512f,avx512vl
    Sign extend packed 32-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi32_pdExperimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi32_psExperimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi32_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtepi32_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtepi64_epi8Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi64_epi16Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi64_epi32Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepi64_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtepi64_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtepi64_storeu_epi32Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtepu8_epi32Experimentalavx512f,avx512vl
    Zero extend packed unsigned 8-bit integers in the low 4 bytes of a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepu8_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 8-bit integers in the low 2 bytes of a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepu16_epi32Experimentalavx512f,avx512vl
    Zero extend packed unsigned 16-bit integers in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepu16_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 16-bit integers in the low 4 bytes of a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepu32_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 32-bit integers in a to packed 64-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtepu32_pdExperimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtpd_epi32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtpd_psExperimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtph_psExperimentalavx512f,avx512vl
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtps_epi32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtps_epu32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtps_phExperimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_cvtsd_ssExperimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_cvtsepi32_epi8Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtsepi32_epi16Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtsepi32_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtsepi32_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtsepi64_epi8Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtsepi64_epi16Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtsepi64_epi32Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtsepi64_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtsepi64_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtsepi64_storeu_epi32Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtss_sdExperimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in b to a double-precision (64-bit) floating-point element, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_cvttpd_epi32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvttpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvttps_epi32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvttps_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtusepi32_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtusepi32_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtusepi32_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtusepi32_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtusepi64_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtusepi64_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtusepi64_epi32Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_cvtusepi64_storeu_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed 8-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtusepi64_storeu_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed 16-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_cvtusepi64_storeu_epi32Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed 32-bit integers with unsigned saturation, and store the active results (those with their respective bit set in writemask k) to unaligned memory at base_addr.
  • _mm_mask_div_pdExperimentalavx512f,avx512vl
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_div_psExperimentalavx512f,avx512vl
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_div_round_sdExperimentalavx512f
    Divide the lower double-precision (64-bit) floating-point element in a by the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_div_round_ssExperimentalavx512f
    Divide the lower single-precision (32-bit) floating-point element in a by the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_div_sdExperimentalavx512f
    Divide the lower double-precision (64-bit) floating-point element in a by the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_div_ssExperimentalavx512f
    Divide the lower single-precision (32-bit) floating-point element in a by the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_expand_epi32Experimentalavx512f,avx512vl
    Load contiguous active 32-bit integers from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_expand_epi64Experimentalavx512f,avx512vl
    Load contiguous active 64-bit integers from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_expand_pdExperimentalavx512f,avx512vl
    Load contiguous active double-precision (64-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_expand_psExperimentalavx512f,avx512vl
    Load contiguous active single-precision (32-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_expandloadu_epi32Experimentalavx512f,avx512vl,avx,sse
    Load contiguous active 32-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_expandloadu_epi64Experimentalavx512f,avx512vl,avx,sse
    Load contiguous active 64-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_expandloadu_pdExperimentalavx512f,avx512vl,avx,sse
    Load contiguous active single-precision (64-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_expandloadu_psExperimentalavx512f,avx512vl,avx,sse
    Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_fixupimm_pdExperimentalavx512f,avx512vl
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm_mask_fixupimm_psExperimentalavx512f,avx512vl
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm_mask_fixupimm_round_sdExperimentalavx512f
    Fix up the lower double-precision (64-bit) floating-point elements in a and b using the lower 64-bit integer in c, store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. imm8 is used to set the required flags reporting.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_fixupimm_round_ssExperimentalavx512f
    Fix up the lower single-precision (32-bit) floating-point elements in a and b using the lower 32-bit integer in c, store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. imm8 is used to set the required flags reporting.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_fixupimm_sdExperimentalavx512f
    Fix up the lower double-precision (64-bit) floating-point elements in a and b using the lower 64-bit integer in c, store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. imm8 is used to set the required flags reporting.
  • _mm_mask_fixupimm_ssExperimentalavx512f
    Fix up the lower single-precision (32-bit) floating-point elements in a and b using the lower 32-bit integer in c, store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. imm8 is used to set the required flags reporting.
  • _mm_mask_fmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fmadd_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_fmadd_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_fmadd_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_fmadd_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_fmaddsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fmaddsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fmsub_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_fmsub_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_fmsub_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_fmsub_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_fmsubadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fmsubadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fnmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fnmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fnmadd_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_fnmadd_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_fnmadd_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_fnmadd_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using writemask k (the element is copied from a when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_fnmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fnmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_fnmsub_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_fnmsub_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_fnmsub_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_fnmsub_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using writemask k (the element is copied from c when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_getexp_pdExperimentalavx512f,avx512vl
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm_mask_getexp_psExperimentalavx512f,avx512vl
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm_mask_getexp_round_sdExperimentalavx512f
    Convert the exponent of the lower double-precision (64-bit) floating-point element in b to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_getexp_round_ssExperimentalavx512f
    Convert the exponent of the lower single-precision (32-bit) floating-point element in b to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_getexp_sdExperimentalavx512f
    Convert the exponent of the lower double-precision (64-bit) floating-point element in b to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
  • _mm_mask_getexp_ssExperimentalavx512f
    Convert the exponent of the lower single-precision (32-bit) floating-point element in b to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
  • _mm_mask_getmant_pdExperimentalavx512f,avx512vl
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm_mask_getmant_psExperimentalavx512f,avx512vl
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm_mask_getmant_round_sdExperimentalavx512f
    Normalize the mantissas of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_getmant_round_ssExperimentalavx512f
    Normalize the mantissas of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_getmant_sdExperimentalavx512f
    Normalize the mantissas of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_getmant_ssExperimentalavx512f
    Normalize the mantissas of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_load_epi32Experimentalavx512f,avx512vl,avx,sse
    Load packed 32-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_mask_load_epi64Experimentalavx512f,avx512vl,avx,sse
    Load packed 64-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_mask_load_pdExperimentalavx512f,avx512vl,avx,sse
    Load packed double-precision (64-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_mask_load_psExperimentalavx512f,avx512vl,avx,sse
    Load packed single-precision (32-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_mask_loadu_epi32Experimentalavx512f,avx512vl,avx,sse
    Load packed 32-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask_loadu_epi64Experimentalavx512f,avx512vl,avx,sse
    Load packed 64-bit integers from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask_loadu_pdExperimentalavx512f,avx512vl,avx,sse
    Load packed double-precision (64-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask_loadu_psExperimentalavx512f,avx512vl,avx,sse
    Load packed single-precision (32-bit) floating-point elements from memory into dst using writemask k (elements are copied from src when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask_max_epi32Experimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_max_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_max_epu32Experimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_max_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_max_pdExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_max_psExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_max_round_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the maximum value in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_max_round_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the maximum value in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_max_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the maximum value in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_max_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the maximum value in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_min_epi32Experimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_min_epu32Experimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_min_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_min_pdExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_min_psExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_min_round_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the minimum value in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_min_round_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the minimum value in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mask_min_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the minimum value in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_min_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the minimum value in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_mov_epi32Experimentalavx512f,avx512vl
    Move packed 32-bit integers from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_mov_epi64Experimentalavx512f,avx512vl
    Move packed 64-bit integers from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_mov_pdExperimentalavx512f,avx512vl
    Move packed double-precision (64-bit) floating-point elements from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_mov_psExperimentalavx512f,avx512vl
    Move packed single-precision (32-bit) floating-point elements from a to dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_move_sdExperimentalavx512f
    Move the lower double-precision (64-bit) floating-point element from b to the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_move_ssExperimentalavx512f
    Move the lower single-precision (32-bit) floating-point element from b to the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_movedup_pdExperimentalavx512f,avx512vl
    Duplicate even-indexed double-precision (64-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_movehdup_psExperimentalavx512f,avx512vl
    Duplicate odd-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_moveldup_psExperimentalavx512f,avx512vl
    Duplicate even-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_mul_epi32Experimentalavx512f,avx512vl
    Multiply the low signed 32-bit integers from each packed 64-bit element in a and b, and store the signed 64-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_mul_epu32Experimentalavx512f,avx512vl
    Multiply the low unsigned 32-bit integers from each packed 64-bit element in a and b, and store the unsigned 64-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_mul_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_mul_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_mul_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_mul_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_mul_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_mul_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_mullo_epi32Experimentalavx512f,avx512vl
    Multiply the packed 32-bit integers in a and b, producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_or_epi32Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_or_epi64Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_permute_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_permute_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_permutevar_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_permutevar_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_permutex2var_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_permutex2var_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_permutex2var_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_permutex2var_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using writemask k (elements are copied from a when the corresponding mask bit is not set).
  • _mm_mask_rcp14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm_mask_rcp14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm_mask_rcp14_sdExperimentalavx512f
    Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_mask_rcp14_ssExperimentalavx512f
    Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_mask_rol_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_rol_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_rolv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_rolv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_ror_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_ror_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_rorv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_rorv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_roundscale_pdExperimentalavx512f,avx512vl
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_roundscale_psExperimentalavx512f,avx512vl
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_roundscale_round_sdExperimentalavx512f
    Round the lower double-precision (64-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_roundscale_round_ssExperimentalavx512f
    Round the lower single-precision (32-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_roundscale_sdExperimentalavx512f
    Round the lower double-precision (64-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_roundscale_ssExperimentalavx512f
    Round the lower single-precision (32-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_mask_rsqrt14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm_mask_rsqrt14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm_mask_rsqrt14_sdExperimentalavx512f
    Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_mask_rsqrt14_ssExperimentalavx512f
    Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_mask_scalef_pdExperimentalavx512f,avx512vl
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_scalef_psExperimentalavx512f,avx512vl
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_scalef_round_sdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_scalef_round_ssExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_scalef_sdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_scalef_ssExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_set1_epi32Experimentalavx512f,avx512vl
    Broadcast 32-bit integer a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_set1_epi64Experimentalavx512f,avx512vl
    Broadcast 64-bit integer a to all elements of dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_shuffle_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_shuffle_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_shuffle_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a using the control in imm8, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sll_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sll_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_slli_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_slli_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sllv_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sllv_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sqrt_pdExperimentalavx512f,avx512vl
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sqrt_psExperimentalavx512f,avx512vl
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sqrt_round_sdExperimentalavx512f
    Compute the square root of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_sqrt_round_ssExperimentalavx512f
    Compute the square root of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_sqrt_sdExperimentalavx512f
    Compute the square root of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_sqrt_ssExperimentalavx512f
    Compute the square root of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_sra_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sra_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srai_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srai_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srav_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srav_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srl_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srl_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srli_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srli_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srlv_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_srlv_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_store_epi32Experimentalavx512f,avx512vl,avx,sse
    Store packed 32-bit integers from a into memory using writemask k. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_mask_store_epi64Experimentalavx512f,avx512vl,avx,sse
    Store packed 64-bit integers from a into memory using writemask k. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_mask_store_pdExperimentalavx512f,avx512vl,avx,sse
    Store packed double-precision (64-bit) floating-point elements from a into memory using writemask k. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_mask_store_psExperimentalavx512f,avx512vl,avx,sse
    Store packed single-precision (32-bit) floating-point elements from a into memory using writemask k. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_mask_storeu_epi32Experimentalavx512f,avx512vl,avx,sse
    Store packed 32-bit integers from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask_storeu_epi64Experimentalavx512f,avx512vl,avx,sse
    Store packed 64-bit integers from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask_storeu_pdExperimentalavx512f,avx512vl,avx,sse
    Store packed double-precision (64-bit) floating-point elements from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask_storeu_psExperimentalavx512f,avx512vl,avx,sse
    Store packed single-precision (32-bit) floating-point elements from a into memory using writemask k. mem_addr does not need to be aligned on any particular boundary.
  • _mm_mask_sub_epi32Experimentalavx512f,avx512vl
    Subtract packed 32-bit integers in b from packed 32-bit integers in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sub_epi64Experimentalavx512f,avx512vl
    Subtract packed 64-bit integers in b from packed 64-bit integers in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sub_pdExperimentalavx512f,avx512vl
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sub_psExperimentalavx512f,avx512vl
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_sub_round_sdExperimentalavx512f
    Subtract the lower double-precision (64-bit) floating-point element in b from the lower double-precision (64-bit) floating-point element in a, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_sub_round_ssExperimentalavx512f
    Subtract the lower single-precision (32-bit) floating-point element in b from the lower single-precision (32-bit) floating-point element in a, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_sub_sdExperimentalavx512f
    Subtract the lower double-precision (64-bit) floating-point element in b from the lower double-precision (64-bit) floating-point element in a, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_mask_sub_ssExperimentalavx512f
    Subtract the lower single-precision (32-bit) floating-point element in b from the lower single-precision (32-bit) floating-point element in a, store the result in the lower element of dst using writemask k (the element is copied from src when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_mask_ternarylogic_epi32Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from src, a, and b are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using writemask k at 32-bit granularity (32-bit elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_ternarylogic_epi64Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from src, a, and b are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using writemask k at 64-bit granularity (64-bit elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_test_epi32_maskExperimentalavx512f,avx512vl
    Compute the bitwise AND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is non-zero.
  • _mm_mask_test_epi64_maskExperimentalavx512f,avx512vl
    Compute the bitwise AND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is non-zero.
  • _mm_mask_testn_epi32_maskExperimentalavx512f,avx512vl
    Compute the bitwise NAND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is zero.
  • _mm_mask_testn_epi64_maskExperimentalavx512f,avx512vl
    Compute the bitwise NAND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k (subject to writemask k) if the intermediate value is zero.
  • _mm_mask_unpackhi_epi32Experimentalavx512f,avx512vl
    Unpack and interleave 32-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_unpackhi_epi64Experimentalavx512f,avx512vl
    Unpack and interleave 64-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_unpackhi_pdExperimentalavx512f,avx512vl
    Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_unpackhi_psExperimentalavx512f,avx512vl
    Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_unpacklo_epi32Experimentalavx512f,avx512vl
    Unpack and interleave 32-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_unpacklo_epi64Experimentalavx512f,avx512vl
    Unpack and interleave 64-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_unpacklo_pdExperimentalavx512f,avx512vl
    Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_unpacklo_psExperimentalavx512f,avx512vl
    Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_xor_epi32Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_mask_xor_epi64Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
  • _mm_maskz_abs_epi32Experimentalavx512f,avx512vl
    Compute the absolute value of packed signed 32-bit integers in a, and store the unsigned results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_add_epi32Experimentalavx512f,avx512vl
    Add packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_add_epi64Experimentalavx512f,avx512vl
    Add packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_add_pdExperimentalavx512f,avx512vl
    Add packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_add_psExperimentalavx512f,avx512vl
    Add packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_add_round_sdExperimentalavx512f
    Add the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_add_round_ssExperimentalavx512f
    Add the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_add_sdExperimentalavx512f
    Add the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_add_ssExperimentalavx512f
    Add the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_alignr_epi32Experimentalavx512f,avx512vl
    Concatenate a and b into a 32-byte immediate result, shift the result right by imm8 32-bit elements, and store the low 16 bytes (4 elements) in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_alignr_epi64Experimentalavx512f,avx512vl
    Concatenate a and b into a 32-byte immediate result, shift the result right by imm8 64-bit elements, and store the low 16 bytes (2 elements) in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_and_epi32Experimentalavx512f,avx512vl
    Compute the bitwise AND of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_and_epi64Experimentalavx512f,avx512vl
    Compute the bitwise AND of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_andnot_epi32Experimentalavx512f,avx512vl
    Compute the bitwise NOT of packed 32-bit integers in a and then AND with b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_andnot_epi64Experimentalavx512f,avx512vl
    Compute the bitwise NOT of packed 64-bit integers in a and then AND with b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_broadcastd_epi32Experimentalavx512f,avx512vl
    Broadcast the low packed 32-bit integer from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_broadcastq_epi64Experimentalavx512f,avx512vl
    Broadcast the low packed 64-bit integer from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_broadcastss_psExperimentalavx512f,avx512vl
    Broadcast the low single-precision (32-bit) floating-point element from a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_compress_epi32Experimentalavx512f,avx512vl
    Contiguously store the active 32-bit integers in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm_maskz_compress_epi64Experimentalavx512f,avx512vl
    Contiguously store the active 64-bit integers in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm_maskz_compress_pdExperimentalavx512f,avx512vl
    Contiguously store the active double-precision (64-bit) floating-point elements in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm_maskz_compress_psExperimentalavx512f,avx512vl
    Contiguously store the active single-precision (32-bit) floating-point elements in a (those with their respective bit set in zeromask k) to dst, and set the remaining elements to zero.
  • _mm_maskz_cvt_roundps_phExperimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_cvt_roundsd_ssExperimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the rounding[3:0] parameter, which can be one of:
    (_MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC) // round to nearest, and suppress exceptions
    (_MM_FROUND_TO_NEG_INF |_MM_FROUND_NO_EXC) // round down, and suppress exceptions
    (_MM_FROUND_TO_POS_INF |_MM_FROUND_NO_EXC) // round up, and suppress exceptions
    (_MM_FROUND_TO_ZERO |_MM_FROUND_NO_EXC) // truncate, and suppress exceptions
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_cvt_roundss_sdExperimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in b to a double-precision (64-bit) floating-point element, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_cvtepi8_epi32Experimentalavx512f,avx512vl
    Sign extend packed 8-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi8_epi64Experimentalavx512f,avx512vl
    Sign extend packed 8-bit integers in the low 2 bytes of a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi16_epi32Experimentalavx512f,avx512vl
    Sign extend packed 16-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi16_epi64Experimentalavx512f,avx512vl
    Sign extend packed 16-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi32_epi8Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi32_epi16Experimentalavx512f,avx512vl
    Convert packed 32-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi32_epi64Experimentalavx512f,avx512vl
    Sign extend packed 32-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi32_pdExperimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi32_psExperimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi64_epi8Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 8-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi64_epi16Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 16-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepi64_epi32Experimentalavx512f,avx512vl
    Convert packed 64-bit integers in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepu8_epi32Experimentalavx512f,avx512vl
    Zero extend packed unsigned 8-bit integers in th elow 4 bytes of a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepu8_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 8-bit integers in the low 2 bytes of a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepu16_epi32Experimentalavx512f,avx512vl
    Zero extend packed unsigned 16-bit integers in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepu16_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 16-bit integers in the low 4 bytes of a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepu32_epi64Experimentalavx512f,avx512vl
    Zero extend packed unsigned 32-bit integers in a to packed 64-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtepu32_pdExperimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed double-precision (64-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtpd_epi32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtpd_psExperimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtph_psExperimentalavx512f,avx512vl
    Convert packed half-precision (16-bit) floating-point elements in a to packed single-precision (32-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtps_epi32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtps_epu32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtps_phExperimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed half-precision (16-bit) floating-point elements, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_cvtsd_ssExperimentalavx512f
    Convert the lower double-precision (64-bit) floating-point element in b to a single-precision (32-bit) floating-point element, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_cvtsepi32_epi8Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtsepi32_epi16Experimentalavx512f,avx512vl
    Convert packed signed 32-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst.
  • _mm_maskz_cvtsepi64_epi8Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 8-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtsepi64_epi16Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 16-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtsepi64_epi32Experimentalavx512f,avx512vl
    Convert packed signed 64-bit integers in a to packed 32-bit integers with signed saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtss_sdExperimentalavx512f
    Convert the lower single-precision (32-bit) floating-point element in b to a double-precision (64-bit) floating-point element, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_cvttpd_epi32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvttpd_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (64-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvttps_epi32Experimentalavx512f,avx512vl
    Convert packed single-precision (32-bit) floating-point elements in a to packed 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvttps_epu32Experimentalavx512f,avx512vl
    Convert packed double-precision (32-bit) floating-point elements in a to packed unsigned 32-bit integers with truncation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtusepi32_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtusepi32_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 32-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtusepi64_epi8Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 8-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtusepi64_epi16Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 16-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_cvtusepi64_epi32Experimentalavx512f,avx512vl
    Convert packed unsigned 64-bit integers in a to packed unsigned 32-bit integers with unsigned saturation, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_div_pdExperimentalavx512f,avx512vl
    Divide packed double-precision (64-bit) floating-point elements in a by packed elements in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_div_psExperimentalavx512f,avx512vl
    Divide packed single-precision (32-bit) floating-point elements in a by packed elements in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_div_round_sdExperimentalavx512f
    Divide the lower double-precision (64-bit) floating-point element in a by the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_div_round_ssExperimentalavx512f
    Divide the lower single-precision (32-bit) floating-point element in a by the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_div_sdExperimentalavx512f
    Divide the lower double-precision (64-bit) floating-point element in a by the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_div_ssExperimentalavx512f
    Divide the lower single-precision (32-bit) floating-point element in a by the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_expand_epi32Experimentalavx512f,avx512vl
    Load contiguous active 32-bit integers from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_expand_epi64Experimentalavx512f,avx512vl
    Load contiguous active 64-bit integers from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_expand_pdExperimentalavx512f,avx512vl
    Load contiguous active double-precision (64-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_expand_psExperimentalavx512f,avx512vl
    Load contiguous active single-precision (32-bit) floating-point elements from a (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_expandloadu_epi32Experimentalavx512f,avx512vl,avx,sse
    Load contiguous active 32-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_expandloadu_epi64Experimentalavx512f,avx512vl,avx,sse
    Load contiguous active 64-bit integers from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_expandloadu_pdExperimentalavx512f,avx512vl,avx,sse
    Load contiguous active single-precision (64-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_expandloadu_psExperimentalavx512f,avx512vl,avx,sse
    Load contiguous active single-precision (32-bit) floating-point elements from unaligned memory at mem_addr (those with their respective bit set in mask k), and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fixupimm_pdExperimentalavx512f,avx512vl
    Fix up packed double-precision (64-bit) floating-point elements in a and b using packed 64-bit integers in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm_maskz_fixupimm_psExperimentalavx512f,avx512vl
    Fix up packed single-precision (32-bit) floating-point elements in a and b using packed 32-bit integers in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). imm8 is used to set the required flags reporting.
  • _mm_maskz_fixupimm_round_sdExperimentalavx512f
    Fix up the lower double-precision (64-bit) floating-point elements in a and b using the lower 64-bit integer in c, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. imm8 is used to set the required flags reporting.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_fixupimm_round_ssExperimentalavx512f
    Fix up the lower single-precision (32-bit) floating-point elements in a and b using the lower 32-bit integer in c, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. imm8 is used to set the required flags reporting.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_fixupimm_sdExperimentalavx512f
    Fix up the lower double-precision (64-bit) floating-point elements in a and b using the lower 64-bit integer in c, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. imm8 is used to set the required flags reporting.
  • _mm_maskz_fixupimm_ssExperimentalavx512f
    Fix up the lower single-precision (32-bit) floating-point elements in a and b using the lower 32-bit integer in c, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. imm8 is used to set the required flags reporting.
  • _mm_maskz_fmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fmadd_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_fmadd_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_fmadd_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_fmadd_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the intermediate result to the lower element in c. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_fmaddsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fmaddsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fmsub_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_fmsub_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_fmsub_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_fmsub_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the intermediate result. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_fmsubadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, alternatively add and subtract packed elements in c to/from the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fmsubadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, alternatively subtract and add packed elements in c from/to the intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fnmadd_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fnmadd_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, add the negated intermediate result to packed elements in c, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fnmadd_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_fnmadd_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_fnmadd_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_fnmadd_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and add the negated intermediate result to the lower element in c. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_fnmsub_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fnmsub_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, subtract packed elements in c from the negated intermediate result, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_fnmsub_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_fnmsub_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_fnmsub_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_fnmsub_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point elements in a and b, and subtract the lower element in c from the negated intermediate result. Store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_getexp_pdExperimentalavx512f,avx512vl
    Convert the exponent of each packed double-precision (64-bit) floating-point element in a to a double-precision (64-bit) floating-point number representing the integer exponent, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm_maskz_getexp_psExperimentalavx512f,avx512vl
    Convert the exponent of each packed single-precision (32-bit) floating-point element in a to a single-precision (32-bit) floating-point number representing the integer exponent, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates floor(log2(x)) for each element.
  • _mm_maskz_getexp_round_sdExperimentalavx512f
    Convert the exponent of the lower double-precision (64-bit) floating-point element in b to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_getexp_round_ssExperimentalavx512f
    Convert the exponent of the lower single-precision (32-bit) floating-point element in b to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_getexp_sdExperimentalavx512f
    Convert the exponent of the lower double-precision (64-bit) floating-point element in b to a double-precision (64-bit) floating-point number representing the integer exponent, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
  • _mm_maskz_getexp_ssExperimentalavx512f
    Convert the exponent of the lower single-precision (32-bit) floating-point element in b to a single-precision (32-bit) floating-point number representing the integer exponent, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates floor(log2(x)) for the lower element.
  • _mm_maskz_getmant_pdExperimentalavx512f,avx512vl
    Normalize the mantissas of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm_maskz_getmant_psExperimentalavx512f,avx512vl
    Normalize the mantissas of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
  • _mm_maskz_getmant_round_sdExperimentalavx512f
    Normalize the mantissas of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_getmant_round_ssExperimentalavx512f
    Normalize the mantissas of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_getmant_sdExperimentalavx512f
    Normalize the mantissas of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_getmant_ssExperimentalavx512f
    Normalize the mantissas of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. This intrinsic essentially calculates ±(2^k)*|x.significand|, where k depends on the interval range defined by interv and the sign depends on sc and the source sign.
    The mantissa is normalized to the interval specified by interv, which can take the following values:
    _MM_MANT_NORM_1_2 // interval [1, 2)
    _MM_MANT_NORM_p5_2 // interval [0.5, 2)
    _MM_MANT_NORM_p5_1 // interval [0.5, 1)
    _MM_MANT_NORM_p75_1p5 // interval [0.75, 1.5)
    The sign is determined by sc which can take the following values:
    _MM_MANT_SIGN_src // sign = sign(src)
    _MM_MANT_SIGN_zero // sign = 0
    _MM_MANT_SIGN_nan // dst = NaN if sign(src) = 1
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_load_epi32Experimentalavx512f,avx512vl,avx,sse
    Load packed 32-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_maskz_load_epi64Experimentalavx512f,avx512vl,avx,sse
    Load packed 64-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_maskz_load_pdExperimentalavx512f,avx512vl,avx,sse
    Load packed double-precision (64-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_maskz_load_psExperimentalavx512f,avx512vl,avx,sse
    Load packed single-precision (32-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_maskz_loadu_epi32Experimentalavx512f,avx512vl,avx,sse
    Load packed 32-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm_maskz_loadu_epi64Experimentalavx512f,avx512vl,avx,sse
    Load packed 64-bit integers from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm_maskz_loadu_pdExperimentalavx512f,avx512vl,avx,sse
    Load packed double-precision (64-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm_maskz_loadu_psExperimentalavx512f,avx512vl,avx,sse
    Load packed single-precision (32-bit) floating-point elements from memory into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). mem_addr does not need to be aligned on any particular boundary.
  • _mm_maskz_max_epi32Experimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_max_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_max_epu32Experimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_max_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_max_pdExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_max_psExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed maximum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_max_round_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the maximum value in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_max_round_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the maximum value in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_max_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the maximum value in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_max_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the maximum value in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_min_epi32Experimentalavx512f,avx512vl
    Compare packed signed 32-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_min_epu32Experimentalavx512f,avx512vl
    Compare packed unsigned 32-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_min_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_min_pdExperimentalavx512f,avx512vl
    Compare packed double-precision (64-bit) floating-point elements in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_min_psExperimentalavx512f,avx512vl
    Compare packed single-precision (32-bit) floating-point elements in a and b, and store packed minimum values in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_min_round_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the minimum value in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_min_round_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the minimum value in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_maskz_min_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the minimum value in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_min_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the minimum value in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_mov_epi32Experimentalavx512f,avx512vl
    Move packed 32-bit integers from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_mov_epi64Experimentalavx512f,avx512vl
    Move packed 64-bit integers from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_mov_pdExperimentalavx512f,avx512vl
    Move packed double-precision (64-bit) floating-point elements from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_mov_psExperimentalavx512f,avx512vl
    Move packed single-precision (32-bit) floating-point elements from a into dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_move_sdExperimentalavx512f
    Move the lower double-precision (64-bit) floating-point element from b to the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_move_ssExperimentalavx512f
    Move the lower single-precision (32-bit) floating-point element from b to the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_movedup_pdExperimentalavx512f,avx512vl
    Duplicate even-indexed double-precision (64-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_movehdup_psExperimentalavx512f,avx512vl
    Duplicate odd-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_moveldup_psExperimentalavx512f,avx512vl
    Duplicate even-indexed single-precision (32-bit) floating-point elements from a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_mul_epi32Experimentalavx512f,avx512vl
    Multiply the low signed 32-bit integers from each packed 64-bit element in a and b, and store the signed 64-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_mul_epu32Experimentalavx512f,avx512vl
    Multiply the low unsigned 32-bit integers from each packed 64-bit element in a and b, and store the unsigned 64-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_mul_pdExperimentalavx512f,avx512vl
    Multiply packed double-precision (64-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_mul_psExperimentalavx512f,avx512vl
    Multiply packed single-precision (32-bit) floating-point elements in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_mul_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_mul_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_mul_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_mul_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_mullo_epi32Experimentalavx512f,avx512vl
    Multiply the packed 32-bit integers in a and b, producing intermediate 64-bit integers, and store the low 32 bits of the intermediate integers in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_or_epi32Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_or_epi64Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_permute_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_permute_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_permutevar_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_permutevar_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a within 128-bit lanes using the control in b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_permutex2var_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_permutex2var_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_permutex2var_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_permutex2var_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_rcp14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm_maskz_rcp14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm_maskz_rcp14_sdExperimentalavx512f
    Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_maskz_rcp14_ssExperimentalavx512f
    Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_maskz_rol_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_rol_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_rolv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_rolv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_ror_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_ror_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_rorv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_rorv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_roundscale_pdExperimentalavx512f,avx512vl
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_roundscale_psExperimentalavx512f,avx512vl
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_roundscale_round_sdExperimentalavx512f
    Round the lower double-precision (64-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_roundscale_round_ssExperimentalavx512f
    Round the lower single-precision (32-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_roundscale_sdExperimentalavx512f
    Round the lower double-precision (64-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_roundscale_ssExperimentalavx512f
    Round the lower single-precision (32-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_maskz_rsqrt14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm_maskz_rsqrt14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set). The maximum relative error for this approximation is less than 2^-14.
  • _mm_maskz_rsqrt14_sdExperimentalavx512f
    Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_maskz_rsqrt14_ssExperimentalavx512f
    Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_maskz_scalef_pdExperimentalavx512f,avx512vl
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_scalef_psExperimentalavx512f,avx512vl
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_scalef_round_sdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_scalef_round_ssExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_scalef_sdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_scalef_ssExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_set1_epi32Experimentalavx512f,avx512vl
    Broadcast 32-bit integer a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_set1_epi64Experimentalavx512f,avx512vl
    Broadcast 64-bit integer a to all elements of dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_shuffle_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_shuffle_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements within 128-bit lanes using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_shuffle_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a using the control in imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sll_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sll_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_slli_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_slli_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sllv_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sllv_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a left by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sqrt_pdExperimentalavx512f,avx512vl
    Compute the square root of packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sqrt_psExperimentalavx512f,avx512vl
    Compute the square root of packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sqrt_round_sdExperimentalavx512f
    Compute the square root of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_sqrt_round_ssExperimentalavx512f
    Compute the square root of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_sqrt_sdExperimentalavx512f
    Compute the square root of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_sqrt_ssExperimentalavx512f
    Compute the square root of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_sra_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sra_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srai_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srai_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srav_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srav_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srl_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srl_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srli_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srli_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srlv_epi32Experimentalavx512f,avx512vl
    Shift packed 32-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_srlv_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in zeros, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sub_epi32Experimentalavx512f,avx512vl
    Subtract packed 32-bit integers in b from packed 32-bit integers in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sub_epi64Experimentalavx512f,avx512vl
    Subtract packed 64-bit integers in b from packed 64-bit integers in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sub_pdExperimentalavx512f,avx512vl
    Subtract packed double-precision (64-bit) floating-point elements in b from packed double-precision (64-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sub_psExperimentalavx512f,avx512vl
    Subtract packed single-precision (32-bit) floating-point elements in b from packed single-precision (32-bit) floating-point elements in a, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_sub_round_sdExperimentalavx512f
    Subtract the lower double-precision (64-bit) floating-point element in b from the lower double-precision (64-bit) floating-point element in a, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_sub_round_ssExperimentalavx512f
    Subtract the lower single-precision (32-bit) floating-point element in b from the lower single-precision (32-bit) floating-point element in a, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_sub_sdExperimentalavx512f
    Subtract the lower double-precision (64-bit) floating-point element in b from the lower double-precision (64-bit) floating-point element in a, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper element from a to the upper element of dst.
  • _mm_maskz_sub_ssExperimentalavx512f
    Subtract the lower single-precision (32-bit) floating-point element in b from the lower single-precision (32-bit) floating-point element in a, store the result in the lower element of dst using zeromask k (the element is zeroed out when mask bit 0 is not set), and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_maskz_ternarylogic_epi32Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using zeromask k at 32-bit granularity (32-bit elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_ternarylogic_epi64Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst using zeromask k at 64-bit granularity (64-bit elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_unpackhi_epi32Experimentalavx512f,avx512vl
    Unpack and interleave 32-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_unpackhi_epi64Experimentalavx512f,avx512vl
    Unpack and interleave 64-bit integers from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_unpackhi_pdExperimentalavx512f,avx512vl
    Unpack and interleave double-precision (64-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_unpackhi_psExperimentalavx512f,avx512vl
    Unpack and interleave single-precision (32-bit) floating-point elements from the high half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_unpacklo_epi32Experimentalavx512f,avx512vl
    Unpack and interleave 32-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_unpacklo_epi64Experimentalavx512f,avx512vl
    Unpack and interleave 64-bit integers from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_unpacklo_pdExperimentalavx512f,avx512vl
    Unpack and interleave double-precision (64-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_unpacklo_psExperimentalavx512f,avx512vl
    Unpack and interleave single-precision (32-bit) floating-point elements from the low half of each 128-bit lane in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_xor_epi32Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_maskz_xor_epi64Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
  • _mm_max_epi64Experimentalavx512f,avx512vl
    Compare packed signed 64-bit integers in a and b, and store packed maximum values in dst.
  • _mm_max_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed maximum values in dst.
  • _mm_max_round_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the maximum value in the lower element of dst, and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_max_round_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the maximum value in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_min_epu64Experimentalavx512f,avx512vl
    Compare packed unsigned 64-bit integers in a and b, and store packed minimum values in dst.
  • _mm_min_round_sdExperimentalavx512f
    Compare the lower double-precision (64-bit) floating-point elements in a and b, store the minimum value in the lower element of dst , and copy the upper element from a to the upper element of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_min_round_ssExperimentalavx512f
    Compare the lower single-precision (32-bit) floating-point elements in a and b, store the minimum value in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
    Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.
  • _mm_mul_round_sdExperimentalavx512f
    Multiply the lower double-precision (64-bit) floating-point element in a and b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_mul_round_ssExperimentalavx512f
    Multiply the lower single-precision (32-bit) floating-point element in a and b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_or_epi32Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 32-bit integers in a and b, and store the results in dst.
  • _mm_or_epi64Experimentalavx512f,avx512vl
    Compute the bitwise OR of packed 64-bit integers in a and b, and store the resut in dst.
  • _mm_permutex2var_epi32Experimentalavx512f,avx512vl
    Shuffle 32-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm_permutex2var_epi64Experimentalavx512f,avx512vl
    Shuffle 64-bit integers in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm_permutex2var_pdExperimentalavx512f,avx512vl
    Shuffle double-precision (64-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm_permutex2var_psExperimentalavx512f,avx512vl
    Shuffle single-precision (32-bit) floating-point elements in a and b across lanes using the corresponding selector and index in idx, and store the results in dst.
  • _mm_rcp14_pdExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed double-precision (64-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_rcp14_psExperimentalavx512f,avx512vl
    Compute the approximate reciprocal of packed single-precision (32-bit) floating-point elements in a, and store the results in dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_rcp14_sdExperimentalavx512f
    Compute the approximate reciprocal of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_rcp14_ssExperimentalavx512f
    Compute the approximate reciprocal of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_rol_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst.
  • _mm_rol_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in imm8, and store the results in dst.
  • _mm_rolv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm_rolv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the left by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm_ror_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst.
  • _mm_ror_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in imm8, and store the results in dst.
  • _mm_rorv_epi32Experimentalavx512f,avx512vl
    Rotate the bits in each packed 32-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm_rorv_epi64Experimentalavx512f,avx512vl
    Rotate the bits in each packed 64-bit integer in a to the right by the number of bits specified in the corresponding element of b, and store the results in dst.
  • _mm_roundscale_pdExperimentalavx512f,avx512vl
    Round packed double-precision (64-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_roundscale_psExperimentalavx512f,avx512vl
    Round packed single-precision (32-bit) floating-point elements in a to the number of fraction bits specified by imm8, and store the results in dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_roundscale_round_sdExperimentalavx512f
    Round the lower double-precision (64-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_roundscale_round_ssExperimentalavx512f
    Round the lower single-precision (32-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_roundscale_sdExperimentalavx512f
    Round the lower double-precision (64-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_roundscale_ssExperimentalavx512f
    Round the lower single-precision (32-bit) floating-point element in b to the number of fraction bits specified by imm8, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
    Rounding is done according to the imm8[2:0] parameter, which can be one of:
    _MM_FROUND_TO_NEAREST_INT // round to nearest
    _MM_FROUND_TO_NEG_INF // round down
    _MM_FROUND_TO_POS_INF // round up
    _MM_FROUND_TO_ZERO // truncate
    _MM_FROUND_CUR_DIRECTION // use MXCSR.RC; see _MM_SET_ROUNDING_MODE
  • _mm_rsqrt14_sdExperimentalavx512f
    Compute the approximate reciprocal square root of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_rsqrt14_ssExperimentalavx512f
    Compute the approximate reciprocal square root of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst. The maximum relative error for this approximation is less than 2^-14.
  • _mm_scalef_pdExperimentalavx512f,avx512vl
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, and store the results in dst.
  • _mm_scalef_psExperimentalavx512f,avx512vl
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, and store the results in dst.
  • _mm_scalef_round_sdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_scalef_round_ssExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_scalef_sdExperimentalavx512f
    Scale the packed double-precision (64-bit) floating-point elements in a using values from b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_scalef_ssExperimentalavx512f
    Scale the packed single-precision (32-bit) floating-point elements in a using values from b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_sqrt_round_sdExperimentalavx512f
    Compute the square root of the lower double-precision (64-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_sqrt_round_ssExperimentalavx512f
    Compute the square root of the lower single-precision (32-bit) floating-point element in b, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_sra_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by count while shifting in sign bits, and store the results in dst.
  • _mm_srai_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by imm8 while shifting in sign bits, and store the results in dst.
  • _mm_srav_epi64Experimentalavx512f,avx512vl
    Shift packed 64-bit integers in a right by the amount specified by the corresponding element in count while shifting in sign bits, and store the results in dst.
  • _mm_store_epi32Experimentalavx512f,avx512vl
    Store 128-bits (composed of 4 packed 32-bit integers) from a into memory. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_store_epi64Experimentalavx512f,avx512vl
    Store 128-bits (composed of 2 packed 64-bit integers) from a into memory. mem_addr must be aligned on a 16-byte boundary or a general-protection exception may be generated.
  • _mm_storeu_epi32Experimentalavx512f,avx512vl
    Store 128-bits (composed of 4 packed 32-bit integers) from a into memory. mem_addr does not need to be aligned on any particular boundary.
  • _mm_storeu_epi64Experimentalavx512f,avx512vl
    Store 128-bits (composed of 2 packed 64-bit integers) from a into memory. mem_addr does not need to be aligned on any particular boundary.
  • _mm_sub_round_sdExperimentalavx512f
    Subtract the lower double-precision (64-bit) floating-point element in b from the lower double-precision (64-bit) floating-point element in a, store the result in the lower element of dst, and copy the upper element from a to the upper element of dst.
  • _mm_sub_round_ssExperimentalavx512f
    Subtract the lower single-precision (32-bit) floating-point element in b from the lower single-precision (32-bit) floating-point element in a, store the result in the lower element of dst, and copy the upper 3 packed elements from a to the upper elements of dst.
  • _mm_ternarylogic_epi32Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 32-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst.
  • _mm_ternarylogic_epi64Experimentalavx512f,avx512vl
    Bitwise ternary logic that provides the capability to implement any three-operand binary function; the specific binary function is specified by value in imm8. For each bit in each packed 64-bit integer, the corresponding bit from a, b, and c are used to form a 3 bit index into imm8, and the value at that bit in imm8 is written to the corresponding bit in dst.
  • _mm_test_epi32_maskExperimentalavx512f,avx512vl
    Compute the bitwise AND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k if the intermediate value is non-zero.
  • _mm_test_epi64_maskExperimentalavx512f,avx512vl
    Compute the bitwise AND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k if the intermediate value is non-zero.
  • _mm_testn_epi32_maskExperimentalavx512f,avx512vl
    Compute the bitwise NAND of packed 32-bit integers in a and b, producing intermediate 32-bit values, and set the corresponding bit in result mask k if the intermediate value is zero.
  • _mm_testn_epi64_maskExperimentalavx512f,avx512vl
    Compute the bitwise NAND of packed 64-bit integers in a and b, producing intermediate 64-bit values, and set the corresponding bit in result mask k if the intermediate value is zero.
  • _mm_xor_epi32Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 32-bit integers in a and b, and store the results in dst.
  • _mm_xor_epi64Experimentalavx512f,avx512vl
    Compute the bitwise XOR of packed 64-bit integers in a and b, and store the results in dst.
  • psllid128 🔒 Experimental
  • psllid256 🔒 Experimental
  • pslliq128 🔒 Experimental
  • pslliq256 🔒 Experimental
  • psraid128 🔒 Experimental
  • psraid256 🔒 Experimental
  • psrlid128 🔒 Experimental
  • psrlid256 🔒 Experimental
  • psrliq128 🔒 Experimental
  • psrliq256 🔒 Experimental
  • vaddpd 🔒 Experimental
  • vaddps 🔒 Experimental
  • vaddsd 🔒 Experimental
  • vaddss 🔒 Experimental
  • vcmppd 🔒 Experimental
  • vcmppd128 🔒 Experimental
  • vcmppd256 🔒 Experimental
  • vcmpps 🔒 Experimental
  • vcmpps128 🔒 Experimental
  • vcmpps256 🔒 Experimental
  • vcmpsd 🔒 Experimental
  • vcmpss 🔒 Experimental
  • vcomisd 🔒 Experimental
  • vcomiss 🔒 Experimental
  • vcompresspd 🔒 Experimental
  • vcompresspd128 🔒 Experimental
  • vcompresspd256 🔒 Experimental
  • vcompressps 🔒 Experimental
  • vcompressps128 🔒 Experimental
  • vcompressps256 🔒 Experimental
  • vcompressstored 🔒 Experimental
  • vcompressstored128 🔒 Experimental
  • vcompressstored256 🔒 Experimental
  • vcompressstorepd 🔒 Experimental
  • vcompressstorepd128 🔒 Experimental
  • vcompressstorepd256 🔒 Experimental
  • vcompressstoreps 🔒 Experimental
  • vcompressstoreps128 🔒 Experimental
  • vcompressstoreps256 🔒 Experimental
  • vcompressstoreq 🔒 Experimental
  • vcompressstoreq128 🔒 Experimental
  • vcompressstoreq256 🔒 Experimental
  • vcvtdq2ps 🔒 Experimental
  • vcvtpd2dq 🔒 Experimental
  • vcvtpd2ps 🔒 Experimental
  • vcvtpd2udq 🔒 Experimental
  • vcvtpd2udq128 🔒 Experimental
  • vcvtpd2udq256 🔒 Experimental
  • vcvtph2ps 🔒 Experimental
  • vcvtps2dq 🔒 Experimental
  • vcvtps2pd 🔒 Experimental
  • vcvtps2ph 🔒 Experimental
  • vcvtps2ph128 🔒 Experimental
  • vcvtps2ph256 🔒 Experimental
  • vcvtps2udq 🔒 Experimental
  • vcvtps2udq128 🔒 Experimental
  • vcvtps2udq256 🔒 Experimental
  • vcvtsd2si 🔒 Experimental
  • vcvtsd2ss 🔒 Experimental
  • vcvtsd2usi 🔒 Experimental
  • vcvtsi2sd 🔒 Experimental
  • vcvtsi2ss 🔒 Experimental
  • vcvtss2sd 🔒 Experimental
  • vcvtss2si 🔒 Experimental
  • vcvtss2usi 🔒 Experimental
  • vcvttpd2dq 🔒 Experimental
  • vcvttpd2dq128 🔒 Experimental
  • vcvttpd2dq256 🔒 Experimental
  • vcvttpd2udq 🔒 Experimental
  • vcvttpd2udq128 🔒 Experimental
  • vcvttpd2udq256 🔒 Experimental
  • vcvttps2dq 🔒 Experimental
  • vcvttps2dq128 🔒 Experimental
  • vcvttps2dq256 🔒 Experimental
  • vcvttps2udq 🔒 Experimental
  • vcvttps2udq128 🔒 Experimental
  • vcvttps2udq256 🔒 Experimental
  • vcvtudq2ps 🔒 Experimental
  • vcvtusi2sd 🔒 Experimental
  • vcvtusi2ss 🔒 Experimental
  • vdivpd 🔒 Experimental
  • vdivps 🔒 Experimental
  • vdivsd 🔒 Experimental
  • vdivss 🔒 Experimental
  • vexpandpd 🔒 Experimental
  • vexpandpd128 🔒 Experimental
  • vexpandpd256 🔒 Experimental
  • vexpandps 🔒 Experimental
  • vexpandps128 🔒 Experimental
  • vexpandps256 🔒 Experimental
  • vfixupimmpd 🔒 Experimental
  • vfixupimmpd128 🔒 Experimental
  • vfixupimmpd256 🔒 Experimental
  • vfixupimmpdz 🔒 Experimental
  • vfixupimmpdz128 🔒 Experimental
  • vfixupimmpdz256 🔒 Experimental
  • vfixupimmps 🔒 Experimental
  • vfixupimmps128 🔒 Experimental
  • vfixupimmps256 🔒 Experimental
  • vfixupimmpsz 🔒 Experimental
  • vfixupimmpsz128 🔒 Experimental
  • vfixupimmpsz256 🔒 Experimental
  • vfixupimmsd 🔒 Experimental
  • vfixupimmsdz 🔒 Experimental
  • vfixupimmss 🔒 Experimental
  • vfixupimmssz 🔒 Experimental
  • vfmadd132pd 🔒 Experimental
  • vfmadd132pdround 🔒 Experimental
  • vfmadd132ps 🔒 Experimental
  • vfmadd132psround 🔒 Experimental
  • vfmadd132sd 🔒 Experimental
  • vfmadd132ss 🔒 Experimental
  • vfmaddsub213pd 🔒 Experimental
  • vfmaddsub213ps 🔒 Experimental
  • vgatherdpd 🔒 Experimental
  • vgatherdps 🔒 Experimental
  • vgatherqpd 🔒 Experimental
  • vgatherqps 🔒 Experimental
  • vgetexppd 🔒 Experimental
  • vgetexppd128 🔒 Experimental
  • vgetexppd256 🔒 Experimental
  • vgetexpps 🔒 Experimental
  • vgetexpps128 🔒 Experimental
  • vgetexpps256 🔒 Experimental
  • vgetexpsd 🔒 Experimental
  • vgetexpss 🔒 Experimental
  • vgetmantpd 🔒 Experimental
  • vgetmantpd128 🔒 Experimental
  • vgetmantpd256 🔒 Experimental
  • vgetmantps 🔒 Experimental
  • vgetmantps128 🔒 Experimental
  • vgetmantps256 🔒 Experimental
  • vgetmantsd 🔒 Experimental
  • vgetmantss 🔒 Experimental
  • vmaxpd 🔒 Experimental
  • vmaxps 🔒 Experimental
  • vmaxsd 🔒 Experimental
  • vmaxss 🔒 Experimental
  • vminpd 🔒 Experimental
  • vminps 🔒 Experimental
  • vminsd 🔒 Experimental
  • vminss 🔒 Experimental
  • vmulpd 🔒 Experimental
  • vmulps 🔒 Experimental
  • vmulsd 🔒 Experimental
  • vmulss 🔒 Experimental
  • vpcmpd 🔒 Experimental
  • vpcmpd128 🔒 Experimental
  • vpcmpd256 🔒 Experimental
  • vpcmpq 🔒 Experimental
  • vpcmpq128 🔒 Experimental
  • vpcmpq256 🔒 Experimental
  • vpcmpud 🔒 Experimental
  • vpcmpud128 🔒 Experimental
  • vpcmpud256 🔒 Experimental
  • vpcmpuq 🔒 Experimental
  • vpcmpuq128 🔒 Experimental
  • vpcmpuq256 🔒 Experimental
  • vpcompressd 🔒 Experimental
  • vpcompressd128 🔒 Experimental
  • vpcompressd256 🔒 Experimental
  • vpcompressq 🔒 Experimental
  • vpcompressq128 🔒 Experimental
  • vpcompressq256 🔒 Experimental
  • vpermd 🔒 Experimental
  • vpermi2d 🔒 Experimental
  • vpermi2d128 🔒 Experimental
  • vpermi2d256 🔒 Experimental
  • vpermi2pd 🔒 Experimental
  • vpermi2pd128 🔒 Experimental
  • vpermi2pd256 🔒 Experimental
  • vpermi2ps 🔒 Experimental
  • vpermi2ps128 🔒 Experimental
  • vpermi2ps256 🔒 Experimental
  • vpermi2q 🔒 Experimental
  • vpermi2q128 🔒 Experimental
  • vpermi2q256 🔒 Experimental
  • vpermilpd 🔒 Experimental
  • vpermilps 🔒 Experimental
  • vpermpd 🔒 Experimental
  • vpermpd256 🔒 Experimental
  • vpermps 🔒 Experimental
  • vpermq 🔒 Experimental
  • vpermq256 🔒 Experimental
  • vpexpandd 🔒 Experimental
  • vpexpandd128 🔒 Experimental
  • vpexpandd256 🔒 Experimental
  • vpexpandq 🔒 Experimental
  • vpexpandq128 🔒 Experimental
  • vpexpandq256 🔒 Experimental
  • vpgatherdd 🔒 Experimental
  • vpgatherdq 🔒 Experimental
  • vpgatherqd 🔒 Experimental
  • vpgatherqq 🔒 Experimental
  • vpmaxsd 🔒 Experimental
  • vpmaxsq 🔒 Experimental
  • vpmaxsq128 🔒 Experimental
  • vpmaxsq256 🔒 Experimental
  • vpmaxud 🔒 Experimental
  • vpmaxuq 🔒 Experimental
  • vpmaxuq128 🔒 Experimental
  • vpmaxuq256 🔒 Experimental
  • vpminsd 🔒 Experimental
  • vpminsq 🔒 Experimental
  • vpminsq128 🔒 Experimental
  • vpminsq256 🔒 Experimental
  • vpminud 🔒 Experimental
  • vpminuq 🔒 Experimental
  • vpminuq128 🔒 Experimental
  • vpminuq256 🔒 Experimental
  • vpmovdb128 🔒 Experimental
  • vpmovdb256 🔒 Experimental
  • vpmovdbmem 🔒 Experimental
  • vpmovdbmem128 🔒 Experimental
  • vpmovdbmem256 🔒 Experimental
  • vpmovdw128 🔒 Experimental
  • vpmovdwmem 🔒 Experimental
  • vpmovdwmem128 🔒 Experimental
  • vpmovdwmem256 🔒 Experimental
  • vpmovqb 🔒 Experimental
  • vpmovqb128 🔒 Experimental
  • vpmovqb256 🔒 Experimental
  • vpmovqbmem 🔒 Experimental
  • vpmovqbmem128 🔒 Experimental
  • vpmovqbmem256 🔒 Experimental
  • vpmovqd128 🔒 Experimental
  • vpmovqdmem 🔒 Experimental
  • vpmovqdmem128 🔒 Experimental
  • vpmovqdmem256 🔒 Experimental
  • vpmovqw128 🔒 Experimental
  • vpmovqw256 🔒 Experimental
  • vpmovqwmem 🔒 Experimental
  • vpmovqwmem128 🔒 Experimental
  • vpmovqwmem256 🔒 Experimental
  • vpmovsdb 🔒 Experimental
  • vpmovsdb128 🔒 Experimental
  • vpmovsdb256 🔒 Experimental
  • vpmovsdbmem 🔒 Experimental
  • vpmovsdbmem128 🔒 Experimental
  • vpmovsdbmem256 🔒 Experimental
  • vpmovsdw 🔒 Experimental
  • vpmovsdw128 🔒 Experimental
  • vpmovsdw256 🔒 Experimental
  • vpmovsdwmem 🔒 Experimental
  • vpmovsdwmem128 🔒 Experimental
  • vpmovsdwmem256 🔒 Experimental
  • vpmovsqb 🔒 Experimental
  • vpmovsqb128 🔒 Experimental
  • vpmovsqb256 🔒 Experimental
  • vpmovsqbmem 🔒 Experimental
  • vpmovsqbmem128 🔒 Experimental
  • vpmovsqbmem256 🔒 Experimental
  • vpmovsqd 🔒 Experimental
  • vpmovsqd128 🔒 Experimental
  • vpmovsqd256 🔒 Experimental
  • vpmovsqdmem 🔒 Experimental
  • vpmovsqdmem128 🔒 Experimental
  • vpmovsqdmem256 🔒 Experimental
  • vpmovsqw 🔒 Experimental
  • vpmovsqw128 🔒 Experimental
  • vpmovsqw256 🔒 Experimental
  • vpmovsqwmem 🔒 Experimental
  • vpmovsqwmem128 🔒 Experimental
  • vpmovsqwmem256 🔒 Experimental
  • vpmovusdb 🔒 Experimental
  • vpmovusdb128 🔒 Experimental
  • vpmovusdb256 🔒 Experimental
  • vpmovusdbmem 🔒 Experimental
  • vpmovusdbmem128 🔒 Experimental
  • vpmovusdbmem256 🔒 Experimental
  • vpmovusdw 🔒 Experimental
  • vpmovusdw128 🔒 Experimental
  • vpmovusdw256 🔒 Experimental
  • vpmovusdwmem 🔒 Experimental
  • vpmovusdwmem128 🔒 Experimental
  • vpmovusdwmem256 🔒 Experimental
  • vpmovusqb 🔒 Experimental
  • vpmovusqb128 🔒 Experimental
  • vpmovusqb256 🔒 Experimental
  • vpmovusqbmem 🔒 Experimental
  • vpmovusqbmem128 🔒 Experimental
  • vpmovusqbmem256 🔒 Experimental
  • vpmovusqd 🔒 Experimental
  • vpmovusqd128 🔒 Experimental
  • vpmovusqd256 🔒 Experimental
  • vpmovusqdmem 🔒 Experimental
  • vpmovusqdmem128 🔒 Experimental
  • vpmovusqdmem256 🔒 Experimental
  • vpmovusqw 🔒 Experimental
  • vpmovusqw128 🔒 Experimental
  • vpmovusqw256 🔒 Experimental
  • vpmovusqwmem 🔒 Experimental
  • vpmovusqwmem128 🔒 Experimental
  • vpmovusqwmem256 🔒 Experimental
  • vpmuldq 🔒 Experimental
  • vpmuludq 🔒 Experimental
  • vprold 🔒 Experimental
  • vprold128 🔒 Experimental
  • vprold256 🔒 Experimental
  • vprolq 🔒 Experimental
  • vprolq128 🔒 Experimental
  • vprolq256 🔒 Experimental
  • vprolvd 🔒 Experimental
  • vprolvd128 🔒 Experimental
  • vprolvd256 🔒 Experimental
  • vprolvq 🔒 Experimental
  • vprolvq128 🔒 Experimental
  • vprolvq256 🔒 Experimental
  • vprord 🔒 Experimental
  • vprord128 🔒 Experimental
  • vprord256 🔒 Experimental
  • vprorq 🔒 Experimental
  • vprorq128 🔒 Experimental
  • vprorq256 🔒 Experimental
  • vprorvd 🔒 Experimental
  • vprorvd128 🔒 Experimental
  • vprorvd256 🔒 Experimental
  • vprorvq 🔒 Experimental
  • vprorvq128 🔒 Experimental
  • vprorvq256 🔒 Experimental
  • vpscatterdd 🔒 Experimental
  • vpscatterdq 🔒 Experimental
  • vpscatterdq256 🔒 Experimental
  • vpscatterqd 🔒 Experimental
  • vpscatterqq 🔒 Experimental
  • vpslld 🔒 Experimental
  • vpsllid 🔒 Experimental
  • vpslliq 🔒 Experimental
  • vpsllq 🔒 Experimental
  • vpsllvd 🔒 Experimental
  • vpsllvq 🔒 Experimental
  • vpsrad 🔒 Experimental
  • vpsraid512 🔒 Experimental
  • vpsraiq 🔒 Experimental
  • vpsraiq128 🔒 Experimental
  • vpsraiq256 🔒 Experimental
  • vpsraq 🔒 Experimental
  • vpsraq128 🔒 Experimental
  • vpsraq256 🔒 Experimental
  • vpsravd 🔒 Experimental
  • vpsravq 🔒 Experimental
  • vpsravq128 🔒 Experimental
  • vpsravq256 🔒 Experimental
  • vpsrld 🔒 Experimental
  • vpsrlid 🔒 Experimental
  • vpsrliq 🔒 Experimental
  • vpsrlq 🔒 Experimental
  • vpsrlvd 🔒 Experimental
  • vpsrlvq 🔒 Experimental
  • vpternlogd 🔒 Experimental
  • vpternlogd128 🔒 Experimental
  • vpternlogd256 🔒 Experimental
  • vpternlogq 🔒 Experimental
  • vpternlogq128 🔒 Experimental
  • vpternlogq256 🔒 Experimental
  • vrcp14pd 🔒 Experimental
  • vrcp14pd128 🔒 Experimental
  • vrcp14pd256 🔒 Experimental
  • vrcp14ps 🔒 Experimental
  • vrcp14ps128 🔒 Experimental
  • vrcp14ps256 🔒 Experimental
  • vrcp14sd 🔒 Experimental
  • vrcp14ss 🔒 Experimental
  • vrndscalepd 🔒 Experimental
  • vrndscalepd128 🔒 Experimental
  • vrndscalepd256 🔒 Experimental
  • vrndscaleps 🔒 Experimental
  • vrndscaleps128 🔒 Experimental
  • vrndscaleps256 🔒 Experimental
  • vrndscalesd 🔒 Experimental
  • vrndscaless 🔒 Experimental
  • vrsqrt14pd 🔒 Experimental
  • vrsqrt14pd128 🔒 Experimental
  • vrsqrt14pd256 🔒 Experimental
  • vrsqrt14ps 🔒 Experimental
  • vrsqrt14ps128 🔒 Experimental
  • vrsqrt14ps256 🔒 Experimental
  • vrsqrt14sd 🔒 Experimental
  • vrsqrt14ss 🔒 Experimental
  • vscalefpd 🔒 Experimental
  • vscalefpd128 🔒 Experimental
  • vscalefpd256 🔒 Experimental
  • vscalefps 🔒 Experimental
  • vscalefps128 🔒 Experimental
  • vscalefps256 🔒 Experimental
  • vscalefsd 🔒 Experimental
  • vscalefss 🔒 Experimental
  • vscatterdpd 🔒 Experimental
  • vscatterdps 🔒 Experimental
  • vscatterqpd 🔒 Experimental
  • vscatterqps 🔒 Experimental
  • vsqrtpd 🔒 Experimental
  • vsqrtps 🔒 Experimental
  • vsqrtsd 🔒 Experimental
  • vsqrtss 🔒 Experimental
  • vsubpd 🔒 Experimental
  • vsubps 🔒 Experimental
  • vsubsd 🔒 Experimental
  • vsubss 🔒 Experimental
  • Stores 512-bits (composed of 16 packed single-precision (32-bit) floating-point elements) from a into memory. mem_addr does not need to be aligned on any particular boundary.