Re: Implementation of some string.h function using SSE2 instructions

[Paweł Sikora <pluto at agmk dot net>]

Wojciech Muła pisze:
> Hi all,
>
> I just subscribed to the list.
>
> Does anybody work on SSE2 implementation?

Hi,

few days ago i've started implementation of strcmp()
for aligned data using <emmintrin.h> builtins.
gcc42/43 optimizes such code pretty well
and raw asm. implementation isn't required.

$ cat sse2_strings.h
#ifndef sse2_strings_h
#define sse2_strings_h

typedef char __attribute__(( aligned( 16 ) )) sse2_aligned_byte;
typedef sse2_aligned_byte const* sse2_byte_buffer;

int sse2_strcmp( sse2_byte_buffer s1, sse2_byte_buffer s2 );

#endif

$ cat sse2_strings.c
#include "sse2_strings.h"
#include <emmintrin.h>

static inline __m128i not( __m128i x )
{
        __m128i zero = { 0 };
        __m128i ones = _mm_cmpeq_epi8( zero, zero );
        return _mm_xor_si128( x, ones );
}

int sse2_strcmp( sse2_byte_buffer s1, sse2_byte_buffer s2 )
{
        for ( int mask = 0; ; s1 += sizeof( __m128i ), s2 += sizeof( 
__m128i ) )
        {
                __m128i m1 = *( __m128i* )( s1 );
                __m128i m2 = *( __m128i* )( s2 );
                __m128i r1 = not( _mm_cmpeq_epi8( m1, m2 ) );
                __m128i zero = { 0 };
                __m128i r2 = _mm_cmpeq_epi8( m1, zero );
                __m128i r3 = _mm_cmpeq_epi8( m2, zero );
                __m128i r = _mm_or_si128( r1, _mm_or_si128( r2, r3 ) );
                mask = _mm_movemask_epi8( r );
                if ( mask )
                {
                        unsigned index = __builtin_ffs( mask ) - 1;
                        return ( s1[ index ] - s2[ index ] );
                }
        }
}