Module core::core_arch::x86::avx512vnni

🔬This is a nightly-only experimental API. (stdsimd #48556)

Available on x86 or x86-64 only.

Functions

• _mm256_dpbusd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst.
• _mm256_dpbusds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst.
• _mm256_dpwssd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst.
• _mm256_dpwssds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst.
• _mm256_mask_dpbusd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm256_mask_dpbusds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm256_mask_dpwssd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm256_mask_dpwssds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm256_maskz_dpbusd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm256_maskz_dpbusds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm256_maskz_dpwssd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm256_maskz_dpwssds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm512_dpbusd_epi32^⚠Experimentalavx512vnni
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst.
• _mm512_dpbusds_epi32^⚠Experimentalavx512vnni
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst.
• _mm512_dpwssd_epi32^⚠Experimentalavx512vnni
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst.
• _mm512_dpwssds_epi32^⚠Experimentalavx512vnni
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst.
• _mm512_mask_dpbusd_epi32^⚠Experimentalavx512vnni
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm512_mask_dpbusds_epi32^⚠Experimentalavx512vnni
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm512_mask_dpwssd_epi32^⚠Experimentalavx512vnni
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm512_mask_dpwssds_epi32^⚠Experimentalavx512vnni
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm512_maskz_dpbusd_epi32^⚠Experimentalavx512vnni
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm512_maskz_dpbusds_epi32^⚠Experimentalavx512vnni
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm512_maskz_dpwssd_epi32^⚠Experimentalavx512vnni
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm512_maskz_dpwssds_epi32^⚠Experimentalavx512vnni
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm_dpbusd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst.
• _mm_dpbusds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst.
• _mm_dpwssd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst.
• _mm_dpwssds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst.
• _mm_mask_dpbusd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm_mask_dpbusds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm_mask_dpwssd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm_mask_dpwssds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using writemask k (elements are copied from src when the corresponding mask bit is not set).
• _mm_maskz_dpbusd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm_maskz_dpbusds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 4 adjacent pairs of unsigned 8-bit integers in a with corresponding signed 8-bit integers in b, producing 4 intermediate signed 16-bit results. Sum these 4 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm_maskz_dpwssd_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• _mm_maskz_dpwssds_epi32^⚠Experimentalavx512vnni,avx512vl
  Multiply groups of 2 adjacent pairs of signed 16-bit integers in a with corresponding 16-bit integers in b, producing 2 intermediate signed 32-bit results. Sum these 2 results with the corresponding 32-bit integer in src using signed saturation, and store the packed 32-bit results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).
• vpdpbusd 🔒 ^⚠Experimental
• vpdpbusd128 🔒 ^⚠Experimental
• vpdpbusd256 🔒 ^⚠Experimental
• vpdpbusds 🔒 ^⚠Experimental
• vpdpbusds128 🔒 ^⚠Experimental
• vpdpbusds256 🔒 ^⚠Experimental
• vpdpwssd 🔒 ^⚠Experimental
• vpdpwssd128 🔒 ^⚠Experimental
• vpdpwssd256 🔒 ^⚠Experimental
• vpdpwssds 🔒 ^⚠Experimental
• vpdpwssds128 🔒 ^⚠Experimental
• vpdpwssds256 🔒 ^⚠Experimental