_mm_cvtepi8_epi32

Microsoft Specific

Emits the Streaming SIMD Extensions 4 (SSE4) instruction pmovsxbd. This instruction performs a conversion of signed integers from 8-bit to 32-bit.

__m128i _mm_cvtepi8_epi32( 
   __m128i a
);

Parameters

• [in] a
  A 128-bit parameter that contains four signed 8-bit integers in the lower 32 bits.

Return value

A 128-bit result that contains four 32-bit integers. These integers are sign-extended representations of the 8-bit integers that are supplied by a.

Requirements

Header file <smmintrin.h>

Remarks

The result is defined as follows:

r0 := a0
r1 := (a0 < 0) ? 0xff : 0
r2 := (a0 < 0) ? 0xff : 0
r3 := (a0 < 0) ? 0xff : 0

r4 := a1
r5 := (a1 < 0) ? 0xff : 0
r6 := (a1 < 0) ? 0xff : 0
r7 := (a1 < 0) ? 0xff : 0

r8 := a2
r9 := (a2 < 0) ? 0xff : 0
r10 := (a2 < 0) ? 0xff : 0
r11 := (a2 < 0) ? 0xff : 0

r12 := a3
r13 := (a3 < 0) ? 0xff : 0
r14 := (a3 < 0) ? 0xff : 0
r15 := (a3 < 0) ? 0xff : 0

r0-r15 and a0-a15 are the sequentially ordered 8-bit components of return value r and parameter a, respectively. r0 and a0 are the least significant 8 bits.

Before you use this intrinsic, software must ensure that the processor supports the instruction.

Example

#include <stdio.h>
#include <smmintrin.h>

int main ()
{
    __m128i a;

    a.m128i_i8[0] = 1;
    a.m128i_i8[1] = -1;
    a.m128i_i8[2] = 127;
    a.m128i_i8[3] = -128;

    __m128i res = _mm_cvtepi8_epi32(a);

    printf_s("Original lowest 8 bit integers:\n%5i, %5i\n%5i, %5i\n\n",
        a.m128i_i8[3], a.m128i_i8[2], a.m128i_i8[1], a.m128i_i8[0]);

    printf_s("Resulting 32 bit integers:\n%5i, %5i\n%5i, %5i\n",
        res.m128i_i32[3], res.m128i_i32[2], res.m128i_i32[1], res.m128i_i32[0]);

    return 0;
}

Original lowest 8 bit integers:
 -128,   127
   -1,     1

Resulting 32 bit integers:
 -128,   127
   -1,     1

See Also

Reference

Compiler Intrinsics