_mm_com_epi64

Artigo
07/22/2011

[Note: This document describes a pre-release version of Visual Studio 2010 SP1 and may be revised in any later version.]

Visual Studio 2010 SP1 is required.

Microsoft Specific

Generates the XOP instruction vpcomq to return a bitmask generated by a quadword-by-quadword signed comparison of its first two sources using the condition given by the third source.

__m128i _mm_com_epi64 (
   __m128i src1,
   __m128i src2,
   int condition
);

Parameters

[in] src1
Packed 128-bit array of two 64-bit signed integers.
[in] src2
Packed 128-bit array of two 64-bit signed integers.
[in] condition
A 32-bit integer condition to be used in the comparison. Legal values are 0 through 7.

Return value

A packed 128-bit array r of two 64-bit unsigned integers, each of which is either 0x0000000000000000 or 0xFFFFFFFFFFFFFFFF. If cond is the logical operator implied by the value of condition, then

r[i] := src1[i] cond src2[i] ? 0xFFFFFFFFFFFFFFFF : 0x0000000000000000;

Requirements

Intrinsic	Architecture
_mm_com_epi64	XOP

Header file <intrin.h>

Remarks

This instruction compares each quadword of src1 to the corresponding quadword of src2, using the comparison operator described by condition. (Each pair of quadwords uses the same comparison operator.) The legal values condition are 0 through 7. By including <intrin.h> you will be able to use the following names rather than numerical values for condition:

Integer Value	Symbolic Name	Meaning
0	_MM_PCOMCTRL_LT	r[i] = (src1[i] < src2[i]) ? 0xFFFFFFFFFFFFFFFF : 0x0000000000000000
1	_MM_PCOMCTRL_LE	r[i] = (src1[i] <= src2[i]) ? 0xFFFFFFFFFFFFFFFF : 0x0000000000000000
2	_MM_PCOMCTRL_GT	r[i] = (src1[i] > src2[i]) ? 0xFFFFFFFFFFFFFFFF : 0x0000000000000000
3	_MM_PCOMCTRL_GE	r[i] = (src1[i] >= src2[i]) ? 0xFFFFFFFFFFFFFFFF : 0x0000000000000000
4	_MM_PCOMCTRL_EQ	r[i] = (src1[i] == src2[i]) ? 0xFFFFFFFFFFFFFFFF : 0x0000000000000000
5	_MM_PCOMCTRL_NEQ	r[i] = (src1[i] != src2[i]) ? 0xFFFFFFFFFFFFFFFF : 0x0000000000000000
6	_MM_PCOMCTRL_FALSE	r[i] = 0x0000000000000000 (mask of all zeros)
7	_MM_PCOMCTRL_TRUE	r[i] = 0xFFFFFFFFFFFFFFFF (mask of all ones)

If you prefer, you may use the following macros, which will be defined when <intrin.h> is included:

Macro	Meaning
_mm_comlt_epi64(src1, src2)	_mm_com_epi64(src1, src2, 0)
_mm_comle_epi64(src1, src2)	_mm_com_epi64(src1, src2, 1)
_mm_comgt_epi64(src1, src2)	_mm_com_epi64(src1, src2, 2)
_mm_comge_epi64(src1, src2)	_mm_com_epi64(src1, src2, 3)
_mm_comeq_epi64(src1, src2)	_mm_com_epi64(src1, src2, 4)
_mm_comneq_epi64(src1, src2)	_mm_com_epi64(src1, src2, 5)
_mm_comfalse_epi64(src1, src2)	_mm_com_epi64(src1, src2, 6)
_mm_comtrue_epi64(src1, src2)	_mm_com_epi64(src1, src2, 7)

The vpcomq instruction is part of the XOP family of instructions. Before you use this intrinsic, you must ensure that the processor supports this instruction. To determine hardware support for this instruction, call the __cpuid intrinsic with InfoType = 0x80000001 and check bit 11 of CPUInfo[2] (ECX). This bit is 1 when the instruction is supported, and 0 otherwise.

Example

#include <stdio.h>
#include <intrin.h>
main()
{
    __m128i a, b, d;
    int i;
    a.m128i_i64[0] = -10;
    a.m128i_i64[1] =  10;
    b.m128i_i64[0] =  22;
    b.m128i_i64[1] = -22;
    printf("src1:     %16I64d %16I64d\n",
            a.m128i_i64[0], a.m128i_i64[1]);
    printf("src2:     %16I64d %16I64d\n",
            b.m128i_i64[0], b.m128i_i64[1]);
    d = _mm_com_epi64(a, b, _MM_PCOMCTRL_LT);
    printf("< mask:   %016I64x %016I64x\n",
            d.m128i_u64[0], d.m128i_u64[1]);
    d = _mm_com_epi64(a, b, _MM_PCOMCTRL_GT);
    printf("> mask:   %016I64x %016I64x\n",
            d.m128i_u64[0], d.m128i_u64[1]);
}

src1:                  -10               10
src2:                   22              -22
< mask:   ffffffffffffffff 0000000000000000
> mask:   0000000000000000 ffffffffffffffff

Change History

Date	History	Reason
March 2011	Added this content.	SP1 feature change.

March 2011

Added this content.

SP1 feature change.

Compartilhar via