1 /*

     2  * Copyright (c) 2005

     3  *	Eric Anholt.  All rights reserved.

     4  *

     5  * Redistribution and use in source and binary forms, with or without

     6  * modification, are permitted provided that the following conditions

     7  * are met:

     8  * 1. Redistributions of source code must retain the above copyright

     9  *    notice, this list of conditions and the following disclaimer.

    10  * 2. Redistributions in binary form must reproduce the above copyright

    11  *    notice, this list of conditions and the following disclaimer in the

    12  *    documentation and/or other materials provided with the distribution.

    13  *

    14  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND

    15  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE

    16  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE

    17  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR BE LIABLE

    18  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL

    19  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS

    20  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)

    21  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT

    22  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY

    23  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF

    24  * SUCH DAMAGE.

    25  */

    26 //Portions Copyright (c)  2008-2009 Nokia Corporation and/or its subsidiary(-ies). All rights reserved. 

    27 

    28 #ifdef HAVE_CONFIG_H

    29 #include "config.h"

    30 #endif

    31 #include <liboil/liboilclasses.h>

    32 #include <liboil/liboilfunction.h>

    33 #include <emmintrin.h>

    34 

    35 #define SSE_FUNCTION __attribute__((force_align_arg_pointer))

    36 

    37 SSE_FUNCTION static void

    38 copy_u8_sse (uint8_t *dest, const uint8_t *src, int n)

    39 {

    40   for (; ((long)dest & 15) && (n > 0); n--) {

    41     *dest++ = *src++;

    42   }

    43   for (; n >= 16; n -= 16) {

    44     _mm_store_si128((__m128i *)dest, _mm_loadu_si128((__m128i *)src));

    45     src += 16;

    46     dest += 16;

    47   }

    48   for (; n > 0; n--) {

    49     *dest++ = *src++;

    50   }

    51 }

    52 OIL_DEFINE_IMPL_FULL (copy_u8_sse, copy_u8, OIL_IMPL_FLAG_SSE2);

    53 

    54 SSE_FUNCTION static void

    55 copy_u8_sse_unroll2 (uint8_t *dest, const uint8_t *src, int n)

    56 {

    57   for (; ((long)dest & 15) && (n > 0); n--) {

    58     *dest++ = *src++;

    59   }

    60   for (; n >= 32; n -= 32) {

    61     _mm_store_si128((__m128i *)dest, _mm_loadu_si128((__m128i *)src));

    62     _mm_store_si128((__m128i *)(dest + 16), _mm_loadu_si128((__m128i *)(src + 16)));

    63     src += 32;

    64     dest += 32;

    65   }

    66   if (n >= 16) {

    67     _mm_store_si128((__m128i *)dest, _mm_loadu_si128((__m128i *)src));

    68     src += 16;

    69     dest += 16;

    70     n -= 16;

    71   }

    72   for (; n > 0; n--) {

    73     *dest++ = *src++;

    74   }

    75 }

    76 OIL_DEFINE_IMPL_FULL (copy_u8_sse_unroll2, copy_u8, OIL_IMPL_FLAG_SSE2);

    77 

    78 

    79 #ifdef	__SYMBIAN32__

    80  

    81 OilFunctionImpl* __oil_function_impl_copy_u8_sse, copy_u8() {

    82 		return &_oil_function_impl_copy_u8_sse, copy_u8;

    83 }

    84 #endif

    85 

    86 #ifdef	__SYMBIAN32__

    87  

    88 OilFunctionImpl* __oil_function_impl_copy_u8_sse_unroll2, copy_u8() {

    89 		return &_oil_function_impl_copy_u8_sse_unroll2, copy_u8;

    90 }

    91 #endif

    92