_mm_mask_i32gather_epi32, _mm256_mask_i32gather_epi32

Gathers 2/4 doubleword values from memory referenced by the given base address, dword indices, and scale, using the given dword mask values. The corresponding Intel® AVX2 instruction is VPGATHERDD.

Syntax

extern __m128i _mm_mask_i32gather_epi32(__m128i def_vals, int const * base, __m128i vindex, __m128i vmask, const int scale);

extern __m256i _mm256_mask_i32gather_epi32(__m256i def_vals, int const * base, __m256i vindex, __m256i vmask, const int scale);

Arguments

def_val

the vector of dword values copied to the destination when the corresponding element of the vector mask is '0'.

base

the base address used to reference the loaded dword elements.

vindex

the vector of dword indices used to reference the loaded dword elements.

vmask

the vector of dword elements used as a vector mask; only the most significant bit of each dword is used as a mask.

scale

The compilation time literal constant, which is used as the vector indices scale to address the loaded elements. Possible values are one of the following: 1, 2, 4, 8.

Description

The intrinsics conditionally loads 2/4 doubleword values from memory referenced by the given base address, dword indices and scale, and using the given dword mask values.

Below is the pseudo-code for the intrinsics:

_mm_mask_i32gather_epi32():

result[31:0] = (vmask[31]==1) ? (mem[base+vindex[31:0]*scale]) : (def_vals[31:0]);
result[63:32] = (vmask[63]==1) ? (mem[base+vindex[63:32]*scale]) : (def_vals[63:32]);
result[95:64] = (vmask[95]==1) ? (mem[base+vindex[95:64]*scale]) : (def_vals[95:64]);
result127:96] = (vmask[127]==1) ? (mem[base+vindex[127:96]*scale]) : (def_vals[127:96]);

_mm256_mask_i32gather_epi32():

result[31:0] = (vmask[31]==1) ? (mem[base+vindex[31:0]*scale]) : (def_vals[31:0]);
result[63:32] = (vmask[63]==1) ? (mem[base+vindex[63:32]*scale]) : (def_vals[63:32]);
result[95:64] = (vmask[95]==1) ? (mem[base+vindex[95:64]*scale]) : (def_vals[95:64]);
result127:96] = (vmask[127]==1) ? (mem[base+vindex[127:96]*scale]) : (def_vals[127:96]);
result[159:128] = (vmask[159]==1) ? (mem[base+vindex[159:128]*scale]) : (def_vals[159:128]);
result[191:160] = (vmask[191]==1) ? (mem[base+vindex[191:160]*scale]) : (def_vals[191:160]);
result[223:192] = (vmask[223]==1) ? (mem[base+vindex[223:192]*scale]) : (def_vals[223:192]);
result[255:224] = (vmask[255]==1) ? (mem[base+vindex[255:224]*scale]) : (def_vals[255:224]);

Returns

A 128/256 vector with conditionally gathered integer32 values.

Para obtener información más completa sobre las optimizaciones del compilador, consulte nuestro Aviso de optimización.