src/video/SDL_blit_N.c
author Sam Lantinga <slouken@libsdl.org>
Thu, 16 Feb 2006 10:11:48 +0000
changeset 1361 19418e4422cb
parent 1358 c71e05b4dc2e
child 1402 d910939febfa
permissions -rw-r--r--
New configure-based build system. Still work in progress, but much improved
slouken@0
     1
/*
slouken@0
     2
    SDL - Simple DirectMedia Layer
slouken@1312
     3
    Copyright (C) 1997-2006 Sam Lantinga
slouken@0
     4
slouken@0
     5
    This library is free software; you can redistribute it and/or
slouken@1312
     6
    modify it under the terms of the GNU Lesser General Public
slouken@0
     7
    License as published by the Free Software Foundation; either
slouken@1312
     8
    version 2.1 of the License, or (at your option) any later version.
slouken@0
     9
slouken@0
    10
    This library is distributed in the hope that it will be useful,
slouken@0
    11
    but WITHOUT ANY WARRANTY; without even the implied warranty of
slouken@0
    12
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
slouken@1312
    13
    Lesser General Public License for more details.
slouken@0
    14
slouken@1312
    15
    You should have received a copy of the GNU Lesser General Public
slouken@1312
    16
    License along with this library; if not, write to the Free Software
slouken@1312
    17
    Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
slouken@0
    18
slouken@0
    19
    Sam Lantinga
slouken@252
    20
    slouken@libsdl.org
slouken@0
    21
*/
slouken@0
    22
slouken@0
    23
#include "SDL_video.h"
slouken@1358
    24
#include "SDL_endian.h"
slouken@1358
    25
#include "SDL_cpuinfo.h"
slouken@0
    26
#include "SDL_blit.h"
slouken@0
    27
slouken@0
    28
/* Functions to blit from N-bit surfaces to other surfaces */
slouken@0
    29
slouken@1361
    30
#if SDL_ALTIVEC_BLITTERS
icculus@1175
    31
#ifdef HAVE_ALTIVEC_H
icculus@1162
    32
#include <altivec.h>
icculus@1175
    33
#endif
slouken@1338
    34
#define assert(X)
icculus@1047
    35
#ifdef MACOSX
icculus@1047
    36
#include <sys/sysctl.h>
icculus@1047
    37
static size_t GetL3CacheSize( void )
icculus@1047
    38
{
icculus@1047
    39
    const char key[] = "hw.l3cachesize";
icculus@1047
    40
    u_int64_t result = 0;
icculus@1047
    41
    size_t typeSize = sizeof( result );
icculus@1047
    42
icculus@1047
    43
icculus@1047
    44
    int err = sysctlbyname( key, &result, &typeSize, NULL, 0 );
icculus@1047
    45
    if( 0 != err ) return 0;
icculus@1047
    46
icculus@1047
    47
    return result;
icculus@1047
    48
}
icculus@1047
    49
#else
icculus@1047
    50
static size_t GetL3CacheSize( void )
icculus@1047
    51
{
icculus@1047
    52
    /* XXX: Just guess G4 */
icculus@1047
    53
    return 2097152;
icculus@1047
    54
}
icculus@1047
    55
#endif /* MACOSX */
icculus@1047
    56
icculus@1162
    57
#if ((defined MACOSX) && (__GNUC__ < 4))
icculus@1162
    58
    #define VECUINT8_LITERAL(a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p) \
icculus@1162
    59
        (vector unsigned char) ( a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p )
icculus@1162
    60
    #define VECUINT16_LITERAL(a,b,c,d,e,f,g,h) \
icculus@1162
    61
        (vector unsigned short) ( a,b,c,d,e,f,g,h )
icculus@1162
    62
#else
icculus@1162
    63
    #define VECUINT8_LITERAL(a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p) \
icculus@1162
    64
        (vector unsigned char) { a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p }
icculus@1162
    65
    #define VECUINT16_LITERAL(a,b,c,d,e,f,g,h) \
icculus@1162
    66
        (vector unsigned short) { a,b,c,d,e,f,g,h }
icculus@1162
    67
#endif
icculus@1162
    68
icculus@1047
    69
#define UNALIGNED_PTR(x) (((size_t) x) & 0x0000000F)
icculus@1047
    70
#define VSWIZZLE32(a,b,c,d) (vector unsigned char) \
icculus@1047
    71
                               ( 0x00+a, 0x00+b, 0x00+c, 0x00+d, \
icculus@1047
    72
                                 0x04+a, 0x04+b, 0x04+c, 0x04+d, \
icculus@1047
    73
                                 0x08+a, 0x08+b, 0x08+c, 0x08+d, \
icculus@1047
    74
                                 0x0C+a, 0x0C+b, 0x0C+c, 0x0C+d )
icculus@1047
    75
icculus@1047
    76
#define MAKE8888(dstfmt, r, g, b, a)  \
icculus@1047
    77
    ( ((r<<dstfmt->Rshift)&dstfmt->Rmask) | \
icculus@1047
    78
      ((g<<dstfmt->Gshift)&dstfmt->Gmask) | \
icculus@1047
    79
      ((b<<dstfmt->Bshift)&dstfmt->Bmask) | \
icculus@1047
    80
      ((a<<dstfmt->Ashift)&dstfmt->Amask) )
icculus@1047
    81
icculus@1047
    82
/*
icculus@1047
    83
 * Data Stream Touch...Altivec cache prefetching.
icculus@1047
    84
 *
icculus@1047
    85
 *  Don't use this on a G5...however, the speed boost is very significant
icculus@1047
    86
 *   on a G4.
icculus@1047
    87
 */
icculus@1047
    88
#define DST_CHAN_SRC 1
icculus@1047
    89
#define DST_CHAN_DEST 2
icculus@1047
    90
icculus@1047
    91
/* macro to set DST control word value... */
icculus@1047
    92
#define DST_CTRL(size, count, stride) \
icculus@1047
    93
    (((size) << 24) | ((count) << 16) | (stride))
icculus@1047
    94
icculus@1047
    95
#define VEC_ALIGNER(src) ((UNALIGNED_PTR(src)) \
icculus@1047
    96
    ? vec_lvsl(0, src) \
icculus@1047
    97
    : vec_add(vec_lvsl(8, src), vec_splat_u8(8)))
icculus@1047
    98
icculus@1047
    99
/* Calculate the permute vector used for 32->32 swizzling */
icculus@1047
   100
static vector unsigned char calc_swizzle32(const SDL_PixelFormat *srcfmt,
icculus@1047
   101
                                  const SDL_PixelFormat *dstfmt)
icculus@1047
   102
{
icculus@1047
   103
    /*
icculus@1047
   104
    * We have to assume that the bits that aren't used by other
icculus@1047
   105
     *  colors is alpha, and it's one complete byte, since some formats
icculus@1047
   106
     *  leave alpha with a zero mask, but we should still swizzle the bits.
icculus@1047
   107
     */
icculus@1047
   108
    /* ARGB */
icculus@1047
   109
    const static struct SDL_PixelFormat default_pixel_format = {
icculus@1047
   110
        NULL, 0, 0,
icculus@1047
   111
        0, 0, 0, 0,
icculus@1047
   112
        16, 8, 0, 24,
icculus@1047
   113
        0x00FF0000, 0x0000FF00, 0x000000FF, 0xFF000000,
icculus@1047
   114
        0, 0};
icculus@1047
   115
    if (!srcfmt) {
icculus@1047
   116
        srcfmt = &default_pixel_format;
icculus@1047
   117
    }
icculus@1047
   118
    if (!dstfmt) {
icculus@1047
   119
        dstfmt = &default_pixel_format;
icculus@1047
   120
    }
icculus@1162
   121
    vector unsigned char plus = VECUINT8_LITERAL(
icculus@1162
   122
                                      0x00, 0x00, 0x00, 0x00,
icculus@1047
   123
                                      0x04, 0x04, 0x04, 0x04,
icculus@1047
   124
                                      0x08, 0x08, 0x08, 0x08,
icculus@1047
   125
                                      0x0C, 0x0C, 0x0C, 0x0C );
icculus@1047
   126
    vector unsigned char vswiz;
icculus@1047
   127
    vector unsigned int srcvec;
icculus@1047
   128
#define RESHIFT(X) (3 - ((X) >> 3))
icculus@1047
   129
    Uint32 rmask = RESHIFT(srcfmt->Rshift) << (dstfmt->Rshift);
icculus@1047
   130
    Uint32 gmask = RESHIFT(srcfmt->Gshift) << (dstfmt->Gshift);
icculus@1047
   131
    Uint32 bmask = RESHIFT(srcfmt->Bshift) << (dstfmt->Bshift);
icculus@1047
   132
    Uint32 amask;
icculus@1047
   133
    /* Use zero for alpha if either surface doesn't have alpha */
icculus@1047
   134
    if (dstfmt->Amask) {
icculus@1047
   135
        amask = ((srcfmt->Amask) ? RESHIFT(srcfmt->Ashift) : 0x10) << (dstfmt->Ashift);
icculus@1047
   136
    } else {    
icculus@1047
   137
        amask = 0x10101010 & ((dstfmt->Rmask | dstfmt->Gmask | dstfmt->Bmask) ^ 0xFFFFFFFF);
icculus@1047
   138
    }           
icculus@1047
   139
#undef RESHIFT  
icculus@1162
   140
    ((unsigned int *)(char*)&srcvec)[0] = (rmask | gmask | bmask | amask);
icculus@1047
   141
    vswiz = vec_add(plus, (vector unsigned char)vec_splat(srcvec, 0));
icculus@1047
   142
    return(vswiz);
icculus@1047
   143
}
icculus@1047
   144
icculus@1047
   145
static void Blit_RGB888_RGB565(SDL_BlitInfo *info);
icculus@1047
   146
static void Blit_RGB888_RGB565Altivec(SDL_BlitInfo *info) {
icculus@1047
   147
    int height = info->d_height;
icculus@1047
   148
    Uint8 *src = (Uint8 *) info->s_pixels;
icculus@1047
   149
    int srcskip = info->s_skip;
icculus@1047
   150
    Uint8 *dst = (Uint8 *) info->d_pixels;
icculus@1047
   151
    int dstskip = info->d_skip;
icculus@1047
   152
    SDL_PixelFormat *srcfmt = info->src;
icculus@1047
   153
    vector unsigned char valpha = vec_splat_u8(0);
icculus@1047
   154
    vector unsigned char vpermute = calc_swizzle32(srcfmt, NULL);
icculus@1162
   155
    vector unsigned char vgmerge = VECUINT8_LITERAL(
icculus@1047
   156
        0x00, 0x02, 0x00, 0x06,
icculus@1047
   157
        0x00, 0x0a, 0x00, 0x0e,
icculus@1047
   158
        0x00, 0x12, 0x00, 0x16,
icculus@1047
   159
        0x00, 0x1a, 0x00, 0x1e);
icculus@1047
   160
    vector unsigned short v1 = vec_splat_u16(1);
icculus@1047
   161
    vector unsigned short v3 = vec_splat_u16(3);
icculus@1162
   162
    vector unsigned short v3f = VECUINT16_LITERAL(
icculus@1047
   163
        0x003f, 0x003f, 0x003f, 0x003f,
icculus@1047
   164
        0x003f, 0x003f, 0x003f, 0x003f);
icculus@1162
   165
    vector unsigned short vfc = VECUINT16_LITERAL(
icculus@1047
   166
        0x00fc, 0x00fc, 0x00fc, 0x00fc,
icculus@1047
   167
        0x00fc, 0x00fc, 0x00fc, 0x00fc);
icculus@1047
   168
    vector unsigned short vf800 = (vector unsigned short)vec_splat_u8(-7);
icculus@1047
   169
    vf800 = vec_sl(vf800, vec_splat_u16(8));
icculus@1047
   170
icculus@1047
   171
    while (height--) {
icculus@1047
   172
        vector unsigned char valigner;
icculus@1047
   173
        vector unsigned char voverflow;
icculus@1047
   174
        vector unsigned char vsrc;
icculus@1047
   175
icculus@1047
   176
        int width = info->d_width;
icculus@1047
   177
        int extrawidth;
icculus@1047
   178
icculus@1047
   179
        /* do scalar until we can align... */
icculus@1047
   180
#define ONE_PIXEL_BLEND(condition, widthvar) \
icculus@1047
   181
        while (condition) { \
icculus@1162
   182
            Uint32 Pixel; \
icculus@1047
   183
            unsigned sR, sG, sB, sA; \
icculus@1162
   184
            DISEMBLE_RGBA((Uint8 *)src, 4, srcfmt, Pixel, \
icculus@1047
   185
                          sR, sG, sB, sA); \
icculus@1047
   186
            *(Uint16 *)(dst) = (((sR << 8) & 0x0000F800) | \
icculus@1047
   187
                                ((sG << 3) & 0x000007E0) | \
icculus@1047
   188
                                ((sB >> 3) & 0x0000001F)); \
icculus@1047
   189
            dst += 2; \
icculus@1047
   190
            src += 4; \
icculus@1047
   191
            widthvar--; \
icculus@1047
   192
        }
icculus@1047
   193
icculus@1047
   194
        ONE_PIXEL_BLEND(((UNALIGNED_PTR(dst)) && (width)), width);
icculus@1047
   195
icculus@1047
   196
        /* After all that work, here's the vector part! */
icculus@1047
   197
        extrawidth = (width % 8);  /* trailing unaligned stores */
icculus@1047
   198
        width -= extrawidth;
icculus@1047
   199
        vsrc = vec_ld(0, src);
icculus@1047
   200
        valigner = VEC_ALIGNER(src);
icculus@1047
   201
icculus@1047
   202
        while (width) {
icculus@1047
   203
            vector unsigned short vpixel, vrpixel, vgpixel, vbpixel;
icculus@1047
   204
            vector unsigned int vsrc1, vsrc2;
icculus@1047
   205
            vector unsigned char vdst;
icculus@1047
   206
icculus@1047
   207
            voverflow = vec_ld(15, src);
icculus@1047
   208
            vsrc = vec_perm(vsrc, voverflow, valigner);
icculus@1047
   209
            vsrc1 = (vector unsigned int)vec_perm(vsrc, valpha, vpermute);
icculus@1047
   210
            src += 16;
icculus@1047
   211
            vsrc = voverflow;
icculus@1047
   212
            voverflow = vec_ld(15, src);
icculus@1047
   213
            vsrc = vec_perm(vsrc, voverflow, valigner);
icculus@1047
   214
            vsrc2 = (vector unsigned int)vec_perm(vsrc, valpha, vpermute);
icculus@1047
   215
            /* 1555 */
icculus@1047
   216
            vpixel = (vector unsigned short)vec_packpx(vsrc1, vsrc2);
icculus@1047
   217
            vgpixel = (vector unsigned short)vec_perm(vsrc1, vsrc2, vgmerge);
icculus@1047
   218
            vgpixel = vec_and(vgpixel, vfc);
icculus@1047
   219
            vgpixel = vec_sl(vgpixel, v3);
icculus@1047
   220
            vrpixel = vec_sl(vpixel, v1);
icculus@1047
   221
            vrpixel = vec_and(vrpixel, vf800);
icculus@1047
   222
            vbpixel = vec_and(vpixel, v3f);
icculus@1047
   223
            vdst = vec_or((vector unsigned char)vrpixel, (vector unsigned char)vgpixel);
icculus@1047
   224
            /* 565 */
icculus@1047
   225
            vdst = vec_or(vdst, (vector unsigned char)vbpixel);
icculus@1047
   226
            vec_st(vdst, 0, dst);
icculus@1047
   227
icculus@1047
   228
            width -= 8;
icculus@1047
   229
            src += 16;
icculus@1047
   230
            dst += 16;
icculus@1047
   231
            vsrc = voverflow;
icculus@1047
   232
        }
icculus@1047
   233
icculus@1047
   234
        assert(width == 0);
icculus@1047
   235
icculus@1047
   236
        /* do scalar until we can align... */
icculus@1047
   237
        ONE_PIXEL_BLEND((extrawidth), extrawidth);
icculus@1047
   238
#undef ONE_PIXEL_BLEND
icculus@1047
   239
icculus@1047
   240
        src += srcskip;  /* move to next row, accounting for pitch. */
icculus@1047
   241
        dst += dstskip;
icculus@1047
   242
    }
icculus@1047
   243
icculus@1047
   244
icculus@1047
   245
}
icculus@1047
   246
icculus@1047
   247
static void Blit_RGB565_32Altivec(SDL_BlitInfo *info) {
icculus@1047
   248
    int height = info->d_height;
icculus@1047
   249
    Uint8 *src = (Uint8 *) info->s_pixels;
icculus@1047
   250
    int srcskip = info->s_skip;
icculus@1047
   251
    Uint8 *dst = (Uint8 *) info->d_pixels;
icculus@1047
   252
    int dstskip = info->d_skip;
icculus@1047
   253
    SDL_PixelFormat *srcfmt = info->src;
icculus@1047
   254
    SDL_PixelFormat *dstfmt = info->dst;
icculus@1047
   255
    unsigned alpha;
icculus@1047
   256
    vector unsigned char valpha;
icculus@1047
   257
    vector unsigned char vpermute;
icculus@1047
   258
    vector unsigned short vf800;
icculus@1047
   259
    vector unsigned int v8 = vec_splat_u32(8);
icculus@1047
   260
    vector unsigned int v16 = vec_add(v8, v8);
icculus@1047
   261
    vector unsigned short v2 = vec_splat_u16(2);
icculus@1047
   262
    vector unsigned short v3 = vec_splat_u16(3);
icculus@1047
   263
    /* 
icculus@1047
   264
        0x10 - 0x1f is the alpha
icculus@1047
   265
        0x00 - 0x0e evens are the red
icculus@1047
   266
        0x01 - 0x0f odds are zero
icculus@1047
   267
    */
icculus@1162
   268
    vector unsigned char vredalpha1 = VECUINT8_LITERAL(
icculus@1047
   269
        0x10, 0x00, 0x01, 0x01,
icculus@1047
   270
        0x10, 0x02, 0x01, 0x01,
icculus@1047
   271
        0x10, 0x04, 0x01, 0x01,
icculus@1047
   272
        0x10, 0x06, 0x01, 0x01
icculus@1047
   273
    );
icculus@1162
   274
    vector unsigned char vredalpha2 = (vector unsigned char) (
icculus@1047
   275
        vec_add((vector unsigned int)vredalpha1, vec_sl(v8, v16))
icculus@1047
   276
    );
icculus@1047
   277
    /*
icculus@1047
   278
        0x00 - 0x0f is ARxx ARxx ARxx ARxx
icculus@1047
   279
        0x11 - 0x0f odds are blue
icculus@1047
   280
    */
icculus@1162
   281
    vector unsigned char vblue1 = VECUINT8_LITERAL(
icculus@1047
   282
        0x00, 0x01, 0x02, 0x11,
icculus@1047
   283
        0x04, 0x05, 0x06, 0x13,
icculus@1047
   284
        0x08, 0x09, 0x0a, 0x15,
icculus@1047
   285
        0x0c, 0x0d, 0x0e, 0x17
icculus@1047
   286
    );
icculus@1047
   287
    vector unsigned char vblue2 = (vector unsigned char)(
icculus@1047
   288
        vec_add((vector unsigned int)vblue1, v8)
icculus@1047
   289
    );
icculus@1047
   290
    /*
icculus@1047
   291
        0x00 - 0x0f is ARxB ARxB ARxB ARxB
icculus@1047
   292
        0x10 - 0x0e evens are green
icculus@1047
   293
    */
icculus@1162
   294
    vector unsigned char vgreen1 = VECUINT8_LITERAL(
icculus@1047
   295
        0x00, 0x01, 0x10, 0x03,
icculus@1047
   296
        0x04, 0x05, 0x12, 0x07,
icculus@1047
   297
        0x08, 0x09, 0x14, 0x0b,
icculus@1047
   298
        0x0c, 0x0d, 0x16, 0x0f
icculus@1047
   299
    );
icculus@1047
   300
    vector unsigned char vgreen2 = (vector unsigned char)(
icculus@1047
   301
        vec_add((vector unsigned int)vgreen1, vec_sl(v8, v8))
icculus@1047
   302
    );
icculus@1047
   303
    
icculus@1047
   304
icculus@1047
   305
    assert(srcfmt->BytesPerPixel == 2);
icculus@1047
   306
    assert(dstfmt->BytesPerPixel == 4);
icculus@1047
   307
icculus@1047
   308
    vf800 = (vector unsigned short)vec_splat_u8(-7);
icculus@1047
   309
    vf800 = vec_sl(vf800, vec_splat_u16(8));
icculus@1047
   310
icculus@1047
   311
    if (dstfmt->Amask && srcfmt->alpha) {
icculus@1047
   312
        ((unsigned char *)&valpha)[0] = alpha = srcfmt->alpha;
icculus@1047
   313
        valpha = vec_splat(valpha, 0);
icculus@1047
   314
    } else {
icculus@1047
   315
        alpha = 0;
icculus@1047
   316
        valpha = vec_splat_u8(0);
icculus@1047
   317
    }
icculus@1047
   318
icculus@1047
   319
    vpermute = calc_swizzle32(NULL, dstfmt);
icculus@1047
   320
    while (height--) {
icculus@1047
   321
        vector unsigned char valigner;
icculus@1047
   322
        vector unsigned char voverflow;
icculus@1047
   323
        vector unsigned char vsrc;
icculus@1047
   324
icculus@1047
   325
        int width = info->d_width;
icculus@1047
   326
        int extrawidth;
icculus@1047
   327
icculus@1047
   328
        /* do scalar until we can align... */
icculus@1047
   329
#define ONE_PIXEL_BLEND(condition, widthvar) \
icculus@1047
   330
        while (condition) { \
icculus@1047
   331
            unsigned sR, sG, sB; \
icculus@1162
   332
            unsigned short Pixel = *((unsigned short *)src); \
icculus@1162
   333
            sR = (Pixel >> 8) & 0xf8; \
icculus@1162
   334
            sG = (Pixel >> 3) & 0xfc; \
icculus@1162
   335
            sB = (Pixel << 3) & 0xf8; \
icculus@1047
   336
            ASSEMBLE_RGBA(dst, 4, dstfmt, sR, sG, sB, alpha); \
icculus@1047
   337
            src += 2; \
icculus@1047
   338
            dst += 4; \
icculus@1047
   339
            widthvar--; \
icculus@1047
   340
        }
icculus@1047
   341
        ONE_PIXEL_BLEND(((UNALIGNED_PTR(dst)) && (width)), width);
icculus@1047
   342
icculus@1047
   343
        /* After all that work, here's the vector part! */
icculus@1047
   344
        extrawidth = (width % 8);  /* trailing unaligned stores */
icculus@1047
   345
        width -= extrawidth;
icculus@1047
   346
        vsrc = vec_ld(0, src);
icculus@1047
   347
        valigner = VEC_ALIGNER(src);
icculus@1047
   348
icculus@1047
   349
        while (width) {
icculus@1047
   350
            vector unsigned short vR, vG, vB;
icculus@1047
   351
            vector unsigned char vdst1, vdst2;
icculus@1047
   352
icculus@1047
   353
            voverflow = vec_ld(15, src);
icculus@1047
   354
            vsrc = vec_perm(vsrc, voverflow, valigner);
icculus@1047
   355
icculus@1047
   356
            vR = vec_and((vector unsigned short)vsrc, vf800);
icculus@1047
   357
            vB = vec_sl((vector unsigned short)vsrc, v3);
icculus@1047
   358
            vG = vec_sl(vB, v2);
icculus@1047
   359
icculus@1047
   360
            vdst1 = (vector unsigned char)vec_perm((vector unsigned char)vR, valpha, vredalpha1);
icculus@1047
   361
            vdst1 = vec_perm(vdst1, (vector unsigned char)vB, vblue1);
icculus@1047
   362
            vdst1 = vec_perm(vdst1, (vector unsigned char)vG, vgreen1);
icculus@1047
   363
            vdst1 = vec_perm(vdst1, valpha, vpermute);
icculus@1047
   364
            vec_st(vdst1, 0, dst);
icculus@1047
   365
icculus@1047
   366
            vdst2 = (vector unsigned char)vec_perm((vector unsigned char)vR, valpha, vredalpha2);
icculus@1047
   367
            vdst2 = vec_perm(vdst2, (vector unsigned char)vB, vblue2);
icculus@1047
   368
            vdst2 = vec_perm(vdst2, (vector unsigned char)vG, vgreen2);
icculus@1047
   369
            vdst2 = vec_perm(vdst2, valpha, vpermute);
icculus@1047
   370
            vec_st(vdst2, 16, dst);
icculus@1047
   371
            
icculus@1047
   372
            width -= 8;
icculus@1047
   373
            dst += 32;
icculus@1047
   374
            src += 16;
icculus@1047
   375
            vsrc = voverflow;
icculus@1047
   376
        }
icculus@1047
   377
icculus@1047
   378
        assert(width == 0);
icculus@1047
   379
icculus@1047
   380
icculus@1047
   381
        /* do scalar until we can align... */
icculus@1047
   382
        ONE_PIXEL_BLEND((extrawidth), extrawidth);
icculus@1047
   383
#undef ONE_PIXEL_BLEND
icculus@1047
   384
icculus@1047
   385
        src += srcskip;  /* move to next row, accounting for pitch. */
icculus@1047
   386
        dst += dstskip;
icculus@1047
   387
    }
icculus@1047
   388
icculus@1047
   389
}
icculus@1047
   390
icculus@1139
   391
icculus@1139
   392
static void Blit_RGB555_32Altivec(SDL_BlitInfo *info) {
icculus@1139
   393
    int height = info->d_height;
icculus@1139
   394
    Uint8 *src = (Uint8 *) info->s_pixels;
icculus@1139
   395
    int srcskip = info->s_skip;
icculus@1139
   396
    Uint8 *dst = (Uint8 *) info->d_pixels;
icculus@1139
   397
    int dstskip = info->d_skip;
icculus@1139
   398
    SDL_PixelFormat *srcfmt = info->src;
icculus@1139
   399
    SDL_PixelFormat *dstfmt = info->dst;
icculus@1139
   400
    unsigned alpha;
icculus@1139
   401
    vector unsigned char valpha;
icculus@1139
   402
    vector unsigned char vpermute;
icculus@1139
   403
    vector unsigned short vf800;
icculus@1139
   404
    vector unsigned int v8 = vec_splat_u32(8);
icculus@1139
   405
    vector unsigned int v16 = vec_add(v8, v8);
icculus@1139
   406
    vector unsigned short v1 = vec_splat_u16(1);
icculus@1139
   407
    vector unsigned short v3 = vec_splat_u16(3);
icculus@1139
   408
    /* 
icculus@1139
   409
        0x10 - 0x1f is the alpha
icculus@1139
   410
        0x00 - 0x0e evens are the red
icculus@1139
   411
        0x01 - 0x0f odds are zero
icculus@1139
   412
    */
icculus@1162
   413
    vector unsigned char vredalpha1 = VECUINT8_LITERAL(
icculus@1139
   414
        0x10, 0x00, 0x01, 0x01,
icculus@1139
   415
        0x10, 0x02, 0x01, 0x01,
icculus@1139
   416
        0x10, 0x04, 0x01, 0x01,
icculus@1139
   417
        0x10, 0x06, 0x01, 0x01
icculus@1139
   418
    );
icculus@1139
   419
    vector unsigned char vredalpha2 = (vector unsigned char)(
icculus@1139
   420
        vec_add((vector unsigned int)vredalpha1, vec_sl(v8, v16))
icculus@1139
   421
    );
icculus@1139
   422
    /*
icculus@1139
   423
        0x00 - 0x0f is ARxx ARxx ARxx ARxx
icculus@1139
   424
        0x11 - 0x0f odds are blue
icculus@1139
   425
    */
icculus@1162
   426
    vector unsigned char vblue1 = VECUINT8_LITERAL(
icculus@1139
   427
        0x00, 0x01, 0x02, 0x11,
icculus@1139
   428
        0x04, 0x05, 0x06, 0x13,
icculus@1139
   429
        0x08, 0x09, 0x0a, 0x15,
icculus@1139
   430
        0x0c, 0x0d, 0x0e, 0x17
icculus@1139
   431
    );
icculus@1139
   432
    vector unsigned char vblue2 = (vector unsigned char)(
icculus@1139
   433
        vec_add((vector unsigned int)vblue1, v8)
icculus@1139
   434
    );
icculus@1139
   435
    /*
icculus@1139
   436
        0x00 - 0x0f is ARxB ARxB ARxB ARxB
icculus@1139
   437
        0x10 - 0x0e evens are green
icculus@1139
   438
    */
icculus@1162
   439
    vector unsigned char vgreen1 = VECUINT8_LITERAL(
icculus@1139
   440
        0x00, 0x01, 0x10, 0x03,
icculus@1139
   441
        0x04, 0x05, 0x12, 0x07,
icculus@1139
   442
        0x08, 0x09, 0x14, 0x0b,
icculus@1139
   443
        0x0c, 0x0d, 0x16, 0x0f
icculus@1139
   444
    );
icculus@1139
   445
    vector unsigned char vgreen2 = (vector unsigned char)(
icculus@1139
   446
        vec_add((vector unsigned int)vgreen1, vec_sl(v8, v8))
icculus@1139
   447
    );
icculus@1139
   448
    
icculus@1139
   449
icculus@1139
   450
    assert(srcfmt->BytesPerPixel == 2);
icculus@1139
   451
    assert(dstfmt->BytesPerPixel == 4);
icculus@1139
   452
icculus@1139
   453
    vf800 = (vector unsigned short)vec_splat_u8(-7);
icculus@1139
   454
    vf800 = vec_sl(vf800, vec_splat_u16(8));
icculus@1139
   455
icculus@1139
   456
    if (dstfmt->Amask && srcfmt->alpha) {
icculus@1139
   457
        ((unsigned char *)&valpha)[0] = alpha = srcfmt->alpha;
icculus@1139
   458
        valpha = vec_splat(valpha, 0);
icculus@1139
   459
    } else {
icculus@1139
   460
        alpha = 0;
icculus@1139
   461
        valpha = vec_splat_u8(0);
icculus@1139
   462
    }
icculus@1139
   463
icculus@1139
   464
    vpermute = calc_swizzle32(NULL, dstfmt);
icculus@1139
   465
    while (height--) {
icculus@1139
   466
        vector unsigned char valigner;
icculus@1139
   467
        vector unsigned char voverflow;
icculus@1139
   468
        vector unsigned char vsrc;
icculus@1139
   469
icculus@1139
   470
        int width = info->d_width;
icculus@1139
   471
        int extrawidth;
icculus@1139
   472
icculus@1139
   473
        /* do scalar until we can align... */
icculus@1139
   474
#define ONE_PIXEL_BLEND(condition, widthvar) \
icculus@1139
   475
        while (condition) { \
icculus@1139
   476
            unsigned sR, sG, sB; \
icculus@1162
   477
            unsigned short Pixel = *((unsigned short *)src); \
icculus@1162
   478
            sR = (Pixel >> 7) & 0xf8; \
icculus@1162
   479
            sG = (Pixel >> 2) & 0xf8; \
icculus@1162
   480
            sB = (Pixel << 3) & 0xf8; \
icculus@1139
   481
            ASSEMBLE_RGBA(dst, 4, dstfmt, sR, sG, sB, alpha); \
icculus@1139
   482
            src += 2; \
icculus@1139
   483
            dst += 4; \
icculus@1139
   484
            widthvar--; \
icculus@1139
   485
        }
icculus@1139
   486
        ONE_PIXEL_BLEND(((UNALIGNED_PTR(dst)) && (width)), width);
icculus@1139
   487
icculus@1139
   488
        /* After all that work, here's the vector part! */
icculus@1139
   489
        extrawidth = (width % 8);  /* trailing unaligned stores */
icculus@1139
   490
        width -= extrawidth;
icculus@1139
   491
        vsrc = vec_ld(0, src);
icculus@1139
   492
        valigner = VEC_ALIGNER(src);
icculus@1139
   493
icculus@1139
   494
        while (width) {
icculus@1139
   495
            vector unsigned short vR, vG, vB;
icculus@1139
   496
            vector unsigned char vdst1, vdst2;
icculus@1139
   497
icculus@1139
   498
            voverflow = vec_ld(15, src);
icculus@1139
   499
            vsrc = vec_perm(vsrc, voverflow, valigner);
icculus@1139
   500
icculus@1139
   501
            vR = vec_and(vec_sl((vector unsigned short)vsrc,v1), vf800);
icculus@1139
   502
            vB = vec_sl((vector unsigned short)vsrc, v3);
icculus@1139
   503
            vG = vec_sl(vB, v3);
icculus@1139
   504
icculus@1139
   505
            vdst1 = (vector unsigned char)vec_perm((vector unsigned char)vR, valpha, vredalpha1);
icculus@1139
   506
            vdst1 = vec_perm(vdst1, (vector unsigned char)vB, vblue1);
icculus@1139
   507
            vdst1 = vec_perm(vdst1, (vector unsigned char)vG, vgreen1);
icculus@1139
   508
            vdst1 = vec_perm(vdst1, valpha, vpermute);
icculus@1139
   509
            vec_st(vdst1, 0, dst);
icculus@1139
   510
icculus@1139
   511
            vdst2 = (vector unsigned char)vec_perm((vector unsigned char)vR, valpha, vredalpha2);
icculus@1139
   512
            vdst2 = vec_perm(vdst2, (vector unsigned char)vB, vblue2);
icculus@1139
   513
            vdst2 = vec_perm(vdst2, (vector unsigned char)vG, vgreen2);
icculus@1139
   514
            vdst2 = vec_perm(vdst2, valpha, vpermute);
icculus@1139
   515
            vec_st(vdst2, 16, dst);
icculus@1139
   516
            
icculus@1139
   517
            width -= 8;
icculus@1139
   518
            dst += 32;
icculus@1139
   519
            src += 16;
icculus@1139
   520
            vsrc = voverflow;
icculus@1139
   521
        }
icculus@1139
   522
icculus@1139
   523
        assert(width == 0);
icculus@1139
   524
icculus@1139
   525
icculus@1139
   526
        /* do scalar until we can align... */
icculus@1139
   527
        ONE_PIXEL_BLEND((extrawidth), extrawidth);
icculus@1139
   528
#undef ONE_PIXEL_BLEND
icculus@1139
   529
icculus@1139
   530
        src += srcskip;  /* move to next row, accounting for pitch. */
icculus@1139
   531
        dst += dstskip;
icculus@1139
   532
    }
icculus@1139
   533
icculus@1139
   534
}
icculus@1139
   535
icculus@1047
   536
static void BlitNtoNKey(SDL_BlitInfo *info);
icculus@1047
   537
static void BlitNtoNKeyCopyAlpha(SDL_BlitInfo *info);
icculus@1047
   538
static void Blit32to32KeyAltivec(SDL_BlitInfo *info)
icculus@1047
   539
{
icculus@1047
   540
    int height = info->d_height;
icculus@1047
   541
    Uint32 *srcp = (Uint32 *) info->s_pixels;
icculus@1047
   542
    int srcskip = info->s_skip;
icculus@1047
   543
    Uint32 *dstp = (Uint32 *) info->d_pixels;
icculus@1047
   544
    int dstskip = info->d_skip;
icculus@1047
   545
    SDL_PixelFormat *srcfmt = info->src;
icculus@1047
   546
    int srcbpp = srcfmt->BytesPerPixel;
icculus@1047
   547
    SDL_PixelFormat *dstfmt = info->dst;
icculus@1047
   548
    int dstbpp = dstfmt->BytesPerPixel;
icculus@1047
   549
    int copy_alpha = (srcfmt->Amask && dstfmt->Amask);
icculus@1047
   550
	unsigned alpha = dstfmt->Amask ? srcfmt->alpha : 0;
icculus@1047
   551
    Uint32 rgbmask = srcfmt->Rmask | srcfmt->Gmask | srcfmt->Bmask;
icculus@1047
   552
	Uint32 ckey = info->src->colorkey;
icculus@1047
   553
    vector unsigned int valpha;
icculus@1047
   554
    vector unsigned char vpermute;
icculus@1047
   555
    vector unsigned char vzero;
icculus@1047
   556
    vector unsigned int vckey;
icculus@1047
   557
    vector unsigned int vrgbmask;
icculus@1047
   558
    vpermute = calc_swizzle32(srcfmt, dstfmt);
icculus@1047
   559
    if (info->d_width < 16) {
icculus@1047
   560
        if(copy_alpha) {
icculus@1047
   561
            return BlitNtoNKeyCopyAlpha(info);
icculus@1047
   562
        } else {
icculus@1047
   563
            return BlitNtoNKey(info);
icculus@1047
   564
        }
icculus@1047
   565
    }
icculus@1047
   566
    vzero = vec_splat_u8(0);
icculus@1047
   567
    if (alpha) {
icculus@1047
   568
        ((unsigned char *)&valpha)[0] = (unsigned char)alpha;
icculus@1047
   569
        valpha = (vector unsigned int)vec_splat((vector unsigned char)valpha, 0);
icculus@1047
   570
    } else {
icculus@1047
   571
        valpha = (vector unsigned int)vzero;
icculus@1047
   572
    }
icculus@1047
   573
    ckey &= rgbmask;
icculus@1162
   574
    ((unsigned int *)(char*)&vckey)[0] = ckey;
icculus@1047
   575
    vckey = vec_splat(vckey, 0);
icculus@1162
   576
    ((unsigned int *)(char*)&vrgbmask)[0] = rgbmask;
icculus@1047
   577
    vrgbmask = vec_splat(vrgbmask, 0);
icculus@1047
   578
icculus@1047
   579
    while (height--) {
icculus@1047
   580
#define ONE_PIXEL_BLEND(condition, widthvar) \
icculus@1047
   581
        if (copy_alpha) { \
icculus@1047
   582
            while (condition) { \
icculus@1162
   583
                Uint32 Pixel; \
icculus@1047
   584
                unsigned sR, sG, sB, sA; \
icculus@1162
   585
                DISEMBLE_RGBA((Uint8 *)srcp, srcbpp, srcfmt, Pixel, \
icculus@1047
   586
                          sR, sG, sB, sA); \
icculus@1162
   587
                if ( (Pixel & rgbmask) != ckey ) { \
icculus@1047
   588
                      ASSEMBLE_RGBA((Uint8 *)dstp, dstbpp, dstfmt, \
icculus@1047
   589
                            sR, sG, sB, sA); \
icculus@1047
   590
                } \
icculus@1162
   591
                dstp = (Uint32 *) (((Uint8 *) dstp) + dstbpp); \
icculus@1162
   592
                srcp = (Uint32 *) (((Uint8 *) srcp) + srcbpp); \
icculus@1047
   593
                widthvar--; \
icculus@1047
   594
            } \
icculus@1047
   595
        } else { \
icculus@1047
   596
            while (condition) { \
icculus@1162
   597
                Uint32 Pixel; \
icculus@1047
   598
                unsigned sR, sG, sB; \
icculus@1162
   599
                RETRIEVE_RGB_PIXEL((Uint8 *)srcp, srcbpp, Pixel); \
icculus@1162
   600
                if ( Pixel != ckey ) { \
icculus@1162
   601
                    RGB_FROM_PIXEL(Pixel, srcfmt, sR, sG, sB); \
icculus@1047
   602
                    ASSEMBLE_RGBA((Uint8 *)dstp, dstbpp, dstfmt, \
icculus@1047
   603
                              sR, sG, sB, alpha); \
icculus@1047
   604
                } \
icculus@1162
   605
                dstp = (Uint32 *) (((Uint8 *)dstp) + dstbpp); \
icculus@1162
   606
                srcp = (Uint32 *) (((Uint8 *)srcp) + srcbpp); \
icculus@1047
   607
                widthvar--; \
icculus@1047
   608
            } \
icculus@1047
   609
        }
icculus@1047
   610
        int width = info->d_width;
icculus@1047
   611
        ONE_PIXEL_BLEND((UNALIGNED_PTR(dstp)) && (width), width);
icculus@1047
   612
        assert(width > 0);
icculus@1047
   613
        if (width > 0) {
icculus@1047
   614
            int extrawidth = (width % 4);
icculus@1047
   615
            vector unsigned char valigner = VEC_ALIGNER(srcp);
icculus@1047
   616
            vector unsigned int vs = vec_ld(0, srcp);
icculus@1047
   617
            width -= extrawidth;
icculus@1047
   618
            assert(width >= 4);
icculus@1047
   619
            while (width) {
icculus@1047
   620
                vector unsigned char vsel;
icculus@1047
   621
                vector unsigned int vd;
icculus@1047
   622
                vector unsigned int voverflow = vec_ld(15, srcp);
icculus@1047
   623
                /* load the source vec */
icculus@1047
   624
                vs = vec_perm(vs, voverflow, valigner);
icculus@1047
   625
                /* vsel is set for items that match the key */
icculus@1047
   626
                vsel = (vector unsigned char)vec_and(vs, vrgbmask);
icculus@1047
   627
                vsel = (vector unsigned char)vec_cmpeq(vs, vckey);
icculus@1047
   628
                /* permute the src vec to the dest format */
icculus@1047
   629
                vs = vec_perm(vs, valpha, vpermute);
icculus@1047
   630
                /* load the destination vec */
icculus@1047
   631
                vd = vec_ld(0, dstp);
icculus@1047
   632
                /* select the source and dest into vs */
icculus@1047
   633
                vd = (vector unsigned int)vec_sel((vector unsigned char)vs, (vector unsigned char)vd, vsel);
icculus@1047
   634
                
icculus@1047
   635
                vec_st(vd, 0, dstp);
icculus@1047
   636
                srcp += 4;
icculus@1047
   637
                width -= 4;
icculus@1047
   638
                dstp += 4;
icculus@1047
   639
                vs = voverflow;
icculus@1047
   640
            }
icculus@1047
   641
            ONE_PIXEL_BLEND((extrawidth), extrawidth);
icculus@1047
   642
#undef ONE_PIXEL_BLEND
icculus@1047
   643
            srcp += srcskip >> 2;
icculus@1047
   644
            dstp += dstskip >> 2;
icculus@1047
   645
        }
icculus@1047
   646
    }
icculus@1047
   647
}
icculus@1047
   648
icculus@1047
   649
/* Altivec code to swizzle one 32-bit surface to a different 32-bit format. */
icculus@1047
   650
/* Use this on a G5 */
icculus@1047
   651
static void ConvertAltivec32to32_noprefetch(SDL_BlitInfo *info)
icculus@1047
   652
{
icculus@1047
   653
    int height = info->d_height;
icculus@1047
   654
    Uint32 *src = (Uint32 *) info->s_pixels;
icculus@1047
   655
    int srcskip = info->s_skip;
icculus@1047
   656
    Uint32 *dst = (Uint32 *) info->d_pixels;
icculus@1047
   657
    int dstskip = info->d_skip;
icculus@1047
   658
    SDL_PixelFormat *srcfmt = info->src;
icculus@1047
   659
    SDL_PixelFormat *dstfmt = info->dst;
icculus@1047
   660
    vector unsigned int vzero = vec_splat_u32(0);
icculus@1047
   661
    vector unsigned char vpermute = calc_swizzle32(srcfmt, dstfmt);
icculus@1047
   662
    if (dstfmt->Amask && !srcfmt->Amask) {
icculus@1047
   663
        if (srcfmt->alpha) {
icculus@1047
   664
            vector unsigned char valpha;
icculus@1047
   665
            ((unsigned char *)&valpha)[0] = srcfmt->alpha;
icculus@1047
   666
            vzero = (vector unsigned int)vec_splat(valpha, 0);
icculus@1047
   667
        }
icculus@1047
   668
    }
icculus@1047
   669
slouken@1339
   670
    assert(srcfmt->BytesPerPixel == 4);
slouken@1339
   671
    assert(dstfmt->BytesPerPixel == 4);
icculus@1047
   672
icculus@1047
   673
    while (height--) {
icculus@1047
   674
        vector unsigned char valigner;
icculus@1047
   675
        vector unsigned int vbits;
icculus@1047
   676
        vector unsigned int voverflow;
icculus@1047
   677
        Uint32 bits;
icculus@1047
   678
        Uint8 r, g, b, a;
icculus@1047
   679
icculus@1047
   680
        int width = info->d_width;
icculus@1047
   681
        int extrawidth;
icculus@1047
   682
icculus@1047
   683
        /* do scalar until we can align... */
icculus@1047
   684
        while ((UNALIGNED_PTR(dst)) && (width)) {
icculus@1047
   685
            bits = *(src++);
icculus@1047
   686
            RGBA_FROM_8888(bits, srcfmt, r, g, b, a);
icculus@1047
   687
            *(dst++) = MAKE8888(dstfmt, r, g, b, a);
icculus@1047
   688
            width--;
icculus@1047
   689
        }
icculus@1047
   690
icculus@1047
   691
        /* After all that work, here's the vector part! */
icculus@1047
   692
        extrawidth = (width % 4);
icculus@1047
   693
        width -= extrawidth;
icculus@1047
   694
        valigner = VEC_ALIGNER(src);
icculus@1047
   695
        vbits = vec_ld(0, src);
icculus@1047
   696
icculus@1047
   697
       while (width) {
icculus@1047
   698
            voverflow = vec_ld(15, src);
icculus@1047
   699
            src += 4;
icculus@1047
   700
            width -= 4;
icculus@1047
   701
            vbits = vec_perm(vbits, voverflow, valigner);  /* src is ready. */
icculus@1047
   702
            vbits = vec_perm(vbits, vzero, vpermute);  /* swizzle it. */
icculus@1047
   703
            vec_st(vbits, 0, dst);  /* store it back out. */
icculus@1047
   704
            dst += 4;
icculus@1047
   705
            vbits = voverflow;
icculus@1047
   706
        }
icculus@1047
   707
icculus@1047
   708
        assert(width == 0);
icculus@1047
   709
icculus@1047
   710
        /* cover pixels at the end of the row that didn't fit in 16 bytes. */
icculus@1047
   711
        while (extrawidth) {
icculus@1047
   712
            bits = *(src++);  /* max 7 pixels, don't bother with prefetch. */
icculus@1047
   713
            RGBA_FROM_8888(bits, srcfmt, r, g, b, a);
icculus@1047
   714
            *(dst++) = MAKE8888(dstfmt, r, g, b, a);
icculus@1047
   715
            extrawidth--;
icculus@1047
   716
        }
icculus@1047
   717
icculus@1047
   718
        src += srcskip >> 2;  /* move to next row, accounting for pitch. */
icculus@1047
   719
        dst += dstskip >> 2;
icculus@1047
   720
    }
icculus@1047
   721
icculus@1047
   722
}
icculus@1047
   723
icculus@1047
   724
/* Altivec code to swizzle one 32-bit surface to a different 32-bit format. */
icculus@1047
   725
/* Use this on a G4 */
icculus@1047
   726
static void ConvertAltivec32to32_prefetch(SDL_BlitInfo *info)
icculus@1047
   727
{
icculus@1047
   728
    const int scalar_dst_lead = sizeof (Uint32) * 4;
icculus@1047
   729
    const int vector_dst_lead = sizeof (Uint32) * 16;
icculus@1047
   730
icculus@1047
   731
    int height = info->d_height;
icculus@1047
   732
    Uint32 *src = (Uint32 *) info->s_pixels;
icculus@1047
   733
    int srcskip = info->s_skip;
icculus@1047
   734
    Uint32 *dst = (Uint32 *) info->d_pixels;
icculus@1047
   735
    int dstskip = info->d_skip;
icculus@1047
   736
    SDL_PixelFormat *srcfmt = info->src;
icculus@1047
   737
    SDL_PixelFormat *dstfmt = info->dst;
icculus@1047
   738
    vector unsigned int vzero = vec_splat_u32(0);
icculus@1047
   739
    vector unsigned char vpermute = calc_swizzle32(srcfmt, dstfmt);
icculus@1047
   740
    if (dstfmt->Amask && !srcfmt->Amask) {
icculus@1047
   741
        if (srcfmt->alpha) {
icculus@1047
   742
            vector unsigned char valpha;
icculus@1047
   743
            ((unsigned char *)&valpha)[0] = srcfmt->alpha;
icculus@1047
   744
            vzero = (vector unsigned int)vec_splat(valpha, 0);
icculus@1047
   745
        }
icculus@1047
   746
    }
icculus@1047
   747
slouken@1339
   748
    assert(srcfmt->BytesPerPixel == 4);
slouken@1339
   749
    assert(dstfmt->BytesPerPixel == 4);
icculus@1047
   750
icculus@1047
   751
    while (height--) {
icculus@1047
   752
        vector unsigned char valigner;
icculus@1047
   753
        vector unsigned int vbits;
icculus@1047
   754
        vector unsigned int voverflow;
icculus@1047
   755
        Uint32 bits;
icculus@1047
   756
        Uint8 r, g, b, a;
icculus@1047
   757
icculus@1047
   758
        int width = info->d_width;
icculus@1047
   759
        int extrawidth;
icculus@1047
   760
icculus@1047
   761
        /* do scalar until we can align... */
icculus@1047
   762
        while ((UNALIGNED_PTR(dst)) && (width)) {
icculus@1047
   763
            vec_dstt(src+scalar_dst_lead, DST_CTRL(2,32,1024), DST_CHAN_SRC);
icculus@1047
   764
            vec_dstst(dst+scalar_dst_lead, DST_CTRL(2,32,1024), DST_CHAN_DEST);
icculus@1047
   765
            bits = *(src++);
icculus@1047
   766
            RGBA_FROM_8888(bits, srcfmt, r, g, b, a);
icculus@1047
   767
            *(dst++) = MAKE8888(dstfmt, r, g, b, a);
icculus@1047
   768
            width--;
icculus@1047
   769
        }
icculus@1047
   770
icculus@1047
   771
        /* After all that work, here's the vector part! */
icculus@1047
   772
        extrawidth = (width % 4);
icculus@1047
   773
        width -= extrawidth;
icculus@1047
   774
        valigner = VEC_ALIGNER(src);
icculus@1047
   775
        vbits = vec_ld(0, src);
icculus@1047
   776
icculus@1047
   777
        while (width) {
icculus@1047
   778
            vec_dstt(src+vector_dst_lead, DST_CTRL(2,32,1024), DST_CHAN_SRC);
icculus@1047
   779
            vec_dstst(dst+vector_dst_lead, DST_CTRL(2,32,1024), DST_CHAN_DEST);
icculus@1047
   780
            voverflow = vec_ld(15, src);
icculus@1047
   781
            src += 4;
icculus@1047
   782
            width -= 4;
icculus@1047
   783
            vbits = vec_perm(vbits, voverflow, valigner);  /* src is ready. */
icculus@1047
   784
            vbits = vec_perm(vbits, vzero, vpermute);  /* swizzle it. */
icculus@1047
   785
            vec_st(vbits, 0, dst);  /* store it back out. */
icculus@1047
   786
            dst += 4;
icculus@1047
   787
            vbits = voverflow;
icculus@1047
   788
        }
icculus@1047
   789
        
icculus@1047
   790
        assert(width == 0);
icculus@1047
   791
icculus@1047
   792
        /* cover pixels at the end of the row that didn't fit in 16 bytes. */
icculus@1047
   793
        while (extrawidth) {
icculus@1047
   794
            bits = *(src++);  /* max 7 pixels, don't bother with prefetch. */
icculus@1047
   795
            RGBA_FROM_8888(bits, srcfmt, r, g, b, a);
icculus@1047
   796
            *(dst++) = MAKE8888(dstfmt, r, g, b, a);
icculus@1047
   797
            extrawidth--;
icculus@1047
   798
        }
icculus@1047
   799
icculus@1047
   800
        src += srcskip >> 2;  /* move to next row, accounting for pitch. */
icculus@1047
   801
        dst += dstskip >> 2;
icculus@1047
   802
    }
icculus@1047
   803
icculus@1047
   804
    vec_dss(DST_CHAN_SRC);
icculus@1047
   805
    vec_dss(DST_CHAN_DEST);
icculus@1047
   806
}
icculus@1047
   807
icculus@1047
   808
static Uint32 GetBlitFeatures( void )
icculus@1047
   809
{
icculus@1047
   810
    static Uint32 features = 0xffffffff;
icculus@1047
   811
    if (features == 0xffffffff) {
icculus@1047
   812
        /* Provide an override for testing .. */
slouken@1336
   813
        char *override = SDL_getenv("SDL_ALTIVEC_BLIT_FEATURES");
icculus@1047
   814
        if (override) {
icculus@1047
   815
            features = 0;
slouken@1336
   816
            SDL_sscanf(override, "%u", &features);
icculus@1047
   817
        } else {
icculus@1047
   818
            features = ( 0
icculus@1047
   819
                /* Feature 1 is has-MMX */
icculus@1047
   820
                | ((SDL_HasMMX()) ? 1 : 0)
icculus@1047
   821
                /* Feature 2 is has-AltiVec */
icculus@1047
   822
                | ((SDL_HasAltiVec()) ? 2 : 0)
icculus@1047
   823
                /* Feature 4 is dont-use-prefetch */
icculus@1162
   824
                /* !!!! FIXME: Check for G5 or later, not the cache size! Always prefetch on a G4. */
icculus@1047
   825
                | ((GetL3CacheSize() == 0) ? 4 : 0)
icculus@1047
   826
            );
icculus@1047
   827
        }
icculus@1047
   828
    }
icculus@1047
   829
    return features;
icculus@1047
   830
}
icculus@1047
   831
#else
icculus@1047
   832
/* Feature 1 is has-MMX */
icculus@1047
   833
#define GetBlitFeatures() ((Uint32)(SDL_HasMMX() ? 1 : 0))
icculus@1047
   834
#endif
icculus@1047
   835
icculus@1053
   836
/* This is now endian dependent */
icculus@1053
   837
#if ( SDL_BYTEORDER == SDL_LIL_ENDIAN )
icculus@1053
   838
#define HI	1
icculus@1053
   839
#define LO	0
icculus@1053
   840
#else /* ( SDL_BYTEORDER == SDL_BIG_ENDIAN ) */
icculus@1053
   841
#define HI	0
icculus@1053
   842
#define LO	1
icculus@1053
   843
#endif
icculus@1053
   844
slouken@1361
   845
#if SDL_HERMES_BLITTERS
slouken@0
   846
slouken@0
   847
/* Heheheh, we coerce Hermes into using SDL blit information */
slouken@0
   848
#define X86_ASSEMBLER
slouken@0
   849
#define HermesConverterInterface	SDL_BlitInfo
slouken@0
   850
#define HermesClearInterface		void
slouken@0
   851
#define STACKCALL
slouken@0
   852
slouken@1361
   853
#include "../hermes/HeadMMX.h"
slouken@1361
   854
#include "../hermes/HeadX86.h"
slouken@0
   855
slouken@0
   856
#else
slouken@0
   857
slouken@0
   858
/* Special optimized blit for RGB 8-8-8 --> RGB 3-3-2 */
slouken@0
   859
#define RGB888_RGB332(dst, src) { \
slouken@0
   860
	dst = (((src)&0x00E00000)>>16)| \
slouken@0
   861
	      (((src)&0x0000E000)>>11)| \
slouken@0
   862
	      (((src)&0x000000C0)>>6); \
slouken@0
   863
}
slouken@0
   864
static void Blit_RGB888_index8(SDL_BlitInfo *info)
slouken@0
   865
{
slouken@0
   866
#ifndef USE_DUFFS_LOOP
slouken@0
   867
	int c;
slouken@0
   868
#endif
slouken@0
   869
	int width, height;
slouken@0
   870
	Uint32 *src;
slouken@91
   871
	const Uint8 *map;
slouken@91
   872
	Uint8 *dst;
slouken@0
   873
	int srcskip, dstskip;
slouken@0
   874
slouken@0
   875
	/* Set up some basic variables */
slouken@0
   876
	width = info->d_width;
slouken@0
   877
	height = info->d_height;
slouken@0
   878
	src = (Uint32 *)info->s_pixels;
slouken@0
   879
	srcskip = info->s_skip/4;
slouken@0
   880
	dst = info->d_pixels;
slouken@0
   881
	dstskip = info->d_skip;
slouken@0
   882
	map = info->table;
slouken@0
   883
slouken@0
   884
	if ( map == NULL ) {
slouken@0
   885
		while ( height-- ) {
slouken@0
   886
#ifdef USE_DUFFS_LOOP
slouken@0
   887
			DUFFS_LOOP(
slouken@0
   888
				RGB888_RGB332(*dst++, *src);
slouken@0
   889
			, width);
slouken@0
   890
#else
slouken@0
   891
			for ( c=width/4; c; --c ) {
slouken@0
   892
				/* Pack RGB into 8bit pixel */
slouken@0
   893
				++src;
slouken@0
   894
				RGB888_RGB332(*dst++, *src);
slouken@0
   895
				++src;
slouken@0
   896
				RGB888_RGB332(*dst++, *src);
slouken@0
   897
				++src;
slouken@0
   898
				RGB888_RGB332(*dst++, *src);
slouken@0
   899
				++src;
slouken@0
   900
			}
slouken@91
   901
			switch ( width & 3 ) {
slouken@0
   902
				case 3:
slouken@0
   903
					RGB888_RGB332(*dst++, *src);
slouken@0
   904
					++src;
slouken@0
   905
				case 2:
slouken@0
   906
					RGB888_RGB332(*dst++, *src);
slouken@0
   907
					++src;
slouken@0
   908
				case 1:
slouken@0
   909
					RGB888_RGB332(*dst++, *src);
slouken@0
   910
					++src;
slouken@0
   911
			}
slouken@0
   912
#endif /* USE_DUFFS_LOOP */
slouken@0
   913
			src += srcskip;
slouken@0
   914
			dst += dstskip;
slouken@0
   915
		}
slouken@0
   916
	} else {
icculus@1162
   917
		int Pixel;
slouken@0
   918
slouken@0
   919
		while ( height-- ) {
slouken@0
   920
#ifdef USE_DUFFS_LOOP
slouken@0
   921
			DUFFS_LOOP(
icculus@1162
   922
				RGB888_RGB332(Pixel, *src);
icculus@1162
   923
				*dst++ = map[Pixel];
slouken@0
   924
				++src;
slouken@0
   925
			, width);
slouken@0
   926
#else
slouken@0
   927
			for ( c=width/4; c; --c ) {
slouken@0
   928
				/* Pack RGB into 8bit pixel */
icculus@1162
   929
				RGB888_RGB332(Pixel, *src);
icculus@1162
   930
				*dst++ = map[Pixel];
slouken@0
   931
				++src;
icculus@1162
   932
				RGB888_RGB332(Pixel, *src);
icculus@1162
   933
				*dst++ = map[Pixel];
slouken@0
   934
				++src;
icculus@1162
   935
				RGB888_RGB332(Pixel, *src);
icculus@1162
   936
				*dst++ = map[Pixel];
slouken@0
   937
				++src;
icculus@1162
   938
				RGB888_RGB332(Pixel, *src);
icculus@1162
   939
				*dst++ = map[Pixel];
slouken@0
   940
				++src;
slouken@0
   941
			}
slouken@91
   942
			switch ( width & 3 ) {
slouken@0
   943
				case 3:
icculus@1162
   944
					RGB888_RGB332(Pixel, *src);
icculus@1162
   945
					*dst++ = map[Pixel];
slouken@0
   946
					++src;
slouken@0
   947
				case 2:
icculus@1162
   948
					RGB888_RGB332(Pixel, *src);
icculus@1162
   949
					*dst++ = map[Pixel];
slouken@0
   950
					++src;
slouken@0
   951
				case 1:
icculus@1162
   952
					RGB888_RGB332(Pixel, *src);
icculus@1162
   953
					*dst++ = map[Pixel];
slouken@0
   954
					++src;
slouken@0
   955
			}
slouken@0
   956
#endif /* USE_DUFFS_LOOP */
slouken@0
   957
			src += srcskip;
slouken@0
   958
			dst += dstskip;
slouken@0
   959
		}
slouken@0
   960
	}
slouken@0
   961
}
slouken@0
   962
/* Special optimized blit for RGB 8-8-8 --> RGB 5-5-5 */
slouken@0
   963
#define RGB888_RGB555(dst, src) { \
slouken@0
   964
	*(Uint16 *)(dst) = (((*src)&0x00F80000)>>9)| \
slouken@0
   965
	                   (((*src)&0x0000F800)>>6)| \
slouken@0
   966
	                   (((*src)&0x000000F8)>>3); \
slouken@0
   967
}
slouken@0
   968
#define RGB888_RGB555_TWO(dst, src) { \
slouken@0
   969
	*(Uint32 *)(dst) = (((((src[HI])&0x00F80000)>>9)| \
slouken@0
   970
	                     (((src[HI])&0x0000F800)>>6)| \
slouken@0
   971
	                     (((src[HI])&0x000000F8)>>3))<<16)| \
slouken@0
   972
	                     (((src[LO])&0x00F80000)>>9)| \
slouken@0
   973
	                     (((src[LO])&0x0000F800)>>6)| \
slouken@0
   974
	                     (((src[LO])&0x000000F8)>>3); \
slouken@0
   975
}
slouken@0
   976
static void Blit_RGB888_RGB555(SDL_BlitInfo *info)
slouken@0
   977
{
slouken@0
   978
#ifndef USE_DUFFS_LOOP
slouken@0
   979
	int c;
slouken@0
   980
#endif
slouken@0
   981
	int width, height;
slouken@0
   982
	Uint32 *src;
slouken@0
   983
	Uint16 *dst;
slouken@0
   984
	int srcskip, dstskip;
slouken@0
   985
slouken@0
   986
	/* Set up some basic variables */
slouken@0
   987
	width = info->d_width;
slouken@0
   988
	height = info->d_height;
slouken@0
   989
	src = (Uint32 *)info->s_pixels;
slouken@0
   990
	srcskip = info->s_skip/4;
slouken@0
   991
	dst = (Uint16 *)info->d_pixels;
slouken@0
   992
	dstskip = info->d_skip/2;
slouken@0
   993
slouken@0
   994
#ifdef USE_DUFFS_LOOP
slouken@0
   995
	while ( height-- ) {
slouken@0
   996
		DUFFS_LOOP(
slouken@0
   997
			RGB888_RGB555(dst, src);
slouken@0
   998
			++src;
slouken@0
   999
			++dst;
slouken@0
  1000
		, width);
slouken@0
  1001
		src += srcskip;
slouken@0
  1002
		dst += dstskip;
slouken@0
  1003
	}
slouken@0
  1004
#else
slouken@0
  1005
	/* Memory align at 4-byte boundary, if necessary */
slouken@0
  1006
	if ( (long)dst & 0x03 ) {
slouken@0
  1007
		/* Don't do anything if width is 0 */
slouken@0
  1008
		if ( width == 0 ) {
slouken@0
  1009
			return;
slouken@0
  1010
		}
slouken@0
  1011
		--width;
slouken@0
  1012
slouken@0
  1013
		while ( height-- ) {
slouken@0
  1014
			/* Perform copy alignment */
slouken@0
  1015
			RGB888_RGB555(dst, src);
slouken@0
  1016
			++src;
slouken@0
  1017
			++dst;
slouken@0
  1018
slouken@0
  1019
			/* Copy in 4 pixel chunks */
slouken@0
  1020
			for ( c=width/4; c; --c ) {
slouken@0
  1021
				RGB888_RGB555_TWO(dst, src);
slouken@0
  1022
				src += 2;
slouken@0
  1023
				dst += 2;
slouken@0
  1024
				RGB888_RGB555_TWO(dst, src);
slouken@0
  1025
				src += 2;
slouken@0
  1026
				dst += 2;
slouken@0
  1027
			}
slouken@0
  1028
			/* Get any leftovers */
slouken@91
  1029
			switch (width & 3) {
slouken@0
  1030
				case 3:
slouken@0
  1031
					RGB888_RGB555(dst, src);
slouken@0
  1032
					++src;
slouken@0
  1033
					++dst;
slouken@0
  1034
				case 2:
slouken@0
  1035
					RGB888_RGB555_TWO(dst, src);
slouken@0
  1036
					src += 2;
slouken@0
  1037
					dst += 2;
slouken@0
  1038
					break;
slouken@0
  1039
				case 1:
slouken@0
  1040
					RGB888_RGB555(dst, src);
slouken@0
  1041
					++src;
slouken@0
  1042
					++dst;
slouken@0
  1043
					break;
slouken@0
  1044
			}
slouken@0
  1045
			src += srcskip;
slouken@0
  1046
			dst += dstskip;
slouken@0
  1047
		}
slouken@0
  1048
	} else { 
slouken@0
  1049
		while ( height-- ) {
slouken@0
  1050
			/* Copy in 4 pixel chunks */
slouken@0
  1051
			for ( c=width/4; c; --c ) {
slouken@0
  1052
				RGB888_RGB555_TWO(dst, src);
slouken@0
  1053
				src += 2;
slouken@0
  1054
				dst += 2;
slouken@0
  1055
				RGB888_RGB555_TWO(dst, src);
slouken@0
  1056
				src += 2;
slouken@0
  1057
				dst += 2;
slouken@0
  1058
			}
slouken@0
  1059
			/* Get any leftovers */
slouken@91
  1060
			switch (width & 3) {
slouken@0
  1061
				case 3:
slouken@0
  1062
					RGB888_RGB555(dst, src);
slouken@0
  1063
					++src;
slouken@0
  1064
					++dst;
slouken@0
  1065
				case 2:
slouken@0
  1066
					RGB888_RGB555_TWO(dst, src);
slouken@0
  1067
					src += 2;
slouken@0
  1068
					dst += 2;
slouken@0
  1069
					break;
slouken@0
  1070
				case 1:
slouken@0
  1071
					RGB888_RGB555(dst, src);
slouken@0
  1072
					++src;
slouken@0
  1073
					++dst;
slouken@0
  1074
					break;
slouken@0
  1075
			}
slouken@0
  1076
			src += srcskip;
slouken@0
  1077
			dst += dstskip;
slouken@0
  1078
		}
slouken@0
  1079
	}
slouken@0
  1080
#endif /* USE_DUFFS_LOOP */
slouken@0
  1081
}
slouken@0
  1082
/* Special optimized blit for RGB 8-8-8 --> RGB 5-6-5 */
slouken@0
  1083
#define RGB888_RGB565(dst, src) { \
slouken@0
  1084
	*(Uint16 *)(dst) = (((*src)&0x00F80000)>>8)| \
slouken@0
  1085
	                   (((*src)&0x0000FC00)>>5)| \
slouken@0
  1086
	                   (((*src)&0x000000F8)>>3); \
slouken@0
  1087
}
slouken@0
  1088
#define RGB888_RGB565_TWO(dst, src) { \
slouken@0
  1089
	*(Uint32 *)(dst) = (((((src[HI])&0x00F80000)>>8)| \
slouken@0
  1090
	                     (((src[HI])&0x0000FC00)>>5)| \
slouken@0
  1091
	                     (((src[HI])&0x000000F8)>>3))<<16)| \
slouken@0
  1092
	                     (((src[LO])&0x00F80000)>>8)| \
slouken@0
  1093
	                     (((src[LO])&0x0000FC00)>>5)| \
slouken@0
  1094
	                     (((src[LO])&0x000000F8)>>3); \
slouken@0
  1095
}
slouken@0
  1096
static void Blit_RGB888_RGB565(SDL_BlitInfo *info)
slouken@0
  1097
{
slouken@0
  1098
#ifndef USE_DUFFS_LOOP
slouken@0
  1099
	int c;
slouken@0
  1100
#endif
slouken@0
  1101
	int width, height;
slouken@0
  1102
	Uint32 *src;
slouken@0
  1103
	Uint16 *dst;
slouken@0
  1104
	int srcskip, dstskip;
slouken@0
  1105
slouken@0
  1106
	/* Set up some basic variables */
slouken@0
  1107
	width = info->d_width;
slouken@0
  1108
	height = info->d_height;
slouken@0
  1109
	src = (Uint32 *)info->s_pixels;
slouken@0
  1110
	srcskip = info->s_skip/4;
slouken@0
  1111
	dst = (Uint16 *)info->d_pixels;
slouken@0
  1112
	dstskip = info->d_skip/2;
slouken@0
  1113
slouken@0
  1114
#ifdef USE_DUFFS_LOOP
slouken@0
  1115
	while ( height-- ) {
slouken@0
  1116
		DUFFS_LOOP(
slouken@0
  1117
			RGB888_RGB565(dst, src);
slouken@0
  1118
			++src;
slouken@0
  1119
			++dst;
slouken@0
  1120
		, width);
slouken@0
  1121
		src += srcskip;
slouken@0
  1122
		dst += dstskip;
slouken@0
  1123
	}
slouken@0
  1124
#else
slouken@0
  1125
	/* Memory align at 4-byte boundary, if necessary */
slouken@0
  1126
	if ( (long)dst & 0x03 ) {
slouken@0
  1127
		/* Don't do anything if width is 0 */
slouken@0
  1128
		if ( width == 0 ) {
slouken@0
  1129
			return;
slouken@0
  1130
		}
slouken@0
  1131
		--width;
slouken@0
  1132
slouken@0
  1133
		while ( height-- ) {
slouken@0
  1134
			/* Perform copy alignment */
slouken@0
  1135
			RGB888_RGB565(dst, src);
slouken@0
  1136
			++src;
slouken@0
  1137
			++dst;
slouken@0
  1138
slouken@0
  1139
			/* Copy in 4 pixel chunks */
slouken@0
  1140
			for ( c=width/4; c; --c ) {
slouken@0
  1141
				RGB888_RGB565_TWO(dst, src);
slouken@0
  1142
				src += 2;
slouken@0
  1143
				dst += 2;
slouken@0
  1144
				RGB888_RGB565_TWO(dst, src);
slouken@0
  1145
				src += 2;
slouken@0
  1146
				dst += 2;
slouken@0
  1147
			}
slouken@0
  1148
			/* Get any leftovers */
slouken@91
  1149
			switch (width & 3) {
slouken@0
  1150
				case 3:
slouken@0
  1151
					RGB888_RGB565(dst, src);
slouken@0
  1152
					++src;
slouken@0
  1153
					++dst;
slouken@0
  1154
				case 2:
slouken@0
  1155
					RGB888_RGB565_TWO(dst, src);
slouken@0
  1156
					src += 2;
slouken@0
  1157
					dst += 2;
slouken@0
  1158
					break;
slouken@0
  1159
				case 1:
slouken@0
  1160
					RGB888_RGB565(dst, src);
slouken@0
  1161
					++src;
slouken@0
  1162
					++dst;
slouken@0
  1163
					break;
slouken@0
  1164
			}
slouken@0
  1165
			src += srcskip;
slouken@0
  1166
			dst += dstskip;
slouken@0
  1167
		}
slouken@0
  1168
	} else { 
slouken@0
  1169
		while ( height-- ) {
slouken@0
  1170
			/* Copy in 4 pixel chunks */
slouken@0
  1171
			for ( c=width/4; c; --c ) {
slouken@0
  1172
				RGB888_RGB565_TWO(dst, src);
slouken@0
  1173
				src += 2;
slouken@0
  1174
				dst += 2;
slouken@0
  1175
				RGB888_RGB565_TWO(dst, src);
slouken@0
  1176
				src += 2;
slouken@0
  1177
				dst += 2;
slouken@0
  1178
			}
slouken@0
  1179
			/* Get any leftovers */
slouken@91
  1180
			switch (width & 3) {
slouken@0
  1181
				case 3:
slouken@0
  1182
					RGB888_RGB565(dst, src);
slouken@0
  1183
					++src;
slouken@0
  1184
					++dst;
slouken@0
  1185
				case 2:
slouken@0
  1186
					RGB888_RGB565_TWO(dst, src);
slouken@0
  1187
					src += 2;
slouken@0
  1188
					dst += 2;
slouken@0
  1189
					break;
slouken@0
  1190
				case 1:
slouken@0
  1191
					RGB888_RGB565(dst, src);
slouken@0
  1192
					++src;
slouken@0
  1193
					++dst;
slouken@0
  1194
					break;
slouken@0
  1195
			}
slouken@0
  1196
			src += srcskip;
slouken@0
  1197
			dst += dstskip;
slouken@0
  1198
		}
slouken@0
  1199
	}
slouken@0
  1200
#endif /* USE_DUFFS_LOOP */
slouken@0
  1201
}
slouken@0
  1202
slouken@1361
  1203
#endif /* SDL_HERMES_BLITTERS */
slouken@0
  1204
slouken@0
  1205
slouken@0
  1206
/* Special optimized blit for RGB 5-6-5 --> 32-bit RGB surfaces */
icculus@1047
  1207
#define RGB565_32(dst, src, map) (map[src[LO]*2] + map[src[HI]*2+1])
slouken@91
  1208
static void Blit_RGB565_32(SDL_BlitInfo *info, const Uint32 *map)
slouken@0
  1209
{
slouken@0
  1210
#ifndef USE_DUFFS_LOOP
slouken@0
  1211
	int c;
slouken@0
  1212
#endif
slouken@0
  1213
	int width, height;
slouken@0
  1214
	Uint8 *src;
slouken@0
  1215
	Uint32 *dst;
slouken@0
  1216
	int srcskip, dstskip;
slouken@0
  1217
slouken@0
  1218
	/* Set up some basic variables */
slouken@0
  1219
	width = info->d_width;
slouken@0
  1220
	height = info->d_height;
slouken@0
  1221
	src = (Uint8 *)info->s_pixels;
slouken@0
  1222
	srcskip = info->s_skip;
slouken@0
  1223
	dst = (Uint32 *)info->d_pixels;
slouken@0
  1224
	dstskip = info->d_skip/4;
slouken@0
  1225
slouken@0
  1226
#ifdef USE_DUFFS_LOOP
slouken@0
  1227
	while ( height-- ) {
slouken@0
  1228
		DUFFS_LOOP(
slouken@0
  1229
		{
slouken@0
  1230
			*dst++ = RGB565_32(dst, src, map);
slouken@0
  1231
			src += 2;
slouken@0
  1232
		},
slouken@0
  1233
		width);
slouken@0
  1234
		src += srcskip;
slouken@0
  1235
		dst += dstskip;
slouken@0
  1236
	}
slouken@0
  1237
#else
slouken@0
  1238
	while ( height-- ) {
slouken@0
  1239
		/* Copy in 4 pixel chunks */
slouken@0
  1240
		for ( c=width/4; c; --c ) {
slouken@0
  1241
			*dst++ = RGB565_32(dst, src, map);
slouken@0
  1242
			src += 2;
slouken@0
  1243
			*dst++ = RGB565_32(dst, src, map);
slouken@0
  1244
			src += 2;
slouken@0
  1245
			*dst++ = RGB565_32(dst, src, map);
slouken@0
  1246
			src += 2;
slouken@0
  1247
			*dst++ = RGB565_32(dst, src, map);
slouken@0
  1248
			src += 2;
slouken@0
  1249
		}
slouken@0
  1250
		/* Get any leftovers */
slouken@91
  1251
		switch (width & 3) {
slouken@0
  1252
			case 3:
slouken@0
  1253
				*dst++ = RGB565_32(dst, src, map);
slouken@0
  1254
				src += 2;
slouken@0
  1255
			case 2:
slouken@0
  1256
				*dst++ = RGB565_32(dst, src, map);
slouken@0
  1257
				src += 2;
slouken@0
  1258
			case 1:
slouken@0
  1259
				*dst++ = RGB565_32(dst, src, map);
slouken@0
  1260
				src += 2;
slouken@0
  1261
				break;
slouken@0
  1262
		}
slouken@0
  1263
		src += srcskip;
slouken@0
  1264
		dst += dstskip;
slouken@0
  1265
	}
slouken@0
  1266
#endif /* USE_DUFFS_LOOP */
slouken@0
  1267
}
slouken@0
  1268
slouken@0
  1269
/* Special optimized blit for RGB 5-6-5 --> ARGB 8-8-8-8 */
slouken@91
  1270
static const Uint32 RGB565_ARGB8888_LUT[512] = {
slouken@0
  1271
		0x00000000, 0xff000000, 0x00000008, 0xff002000,
slouken@0
  1272
		0x00000010, 0xff004000, 0x00000018, 0xff006100,
slouken@0
  1273
		0x00000020, 0xff008100, 0x00000029, 0xff00a100,
slouken@0
  1274
		0x00000031, 0xff00c200, 0x00000039, 0xff00e200,
slouken@0
  1275
		0x00000041, 0xff080000, 0x0000004a, 0xff082000,
slouken@0
  1276
		0x00000052, 0xff084000, 0x0000005a, 0xff086100,
slouken@0
  1277
		0x00000062, 0xff088100, 0x0000006a, 0xff08a100,
slouken@0
  1278
		0x00000073, 0xff08c200, 0x0000007b, 0xff08e200,
slouken@0
  1279
		0x00000083, 0xff100000, 0x0000008b, 0xff102000,
slouken@0
  1280
		0x00000094, 0xff104000, 0x0000009c, 0xff106100,
slouken@0
  1281
		0x000000a4, 0xff108100, 0x000000ac, 0xff10a100,
slouken@0
  1282
		0x000000b4, 0xff10c200, 0x000000bd, 0xff10e200,
slouken@0
  1283
		0x000000c5, 0xff180000, 0x000000cd, 0xff182000,
slouken@0
  1284
		0x000000d5, 0xff184000, 0x000000de, 0xff186100,
slouken@0
  1285
		0x000000e6, 0xff188100, 0x000000ee, 0xff18a100,
slouken@0
  1286
		0x000000f6, 0xff18c200, 0x000000ff, 0xff18e200,
slouken@0
  1287
		0x00000400, 0xff200000, 0x00000408, 0xff202000,
slouken@0
  1288
		0x00000410, 0xff204000, 0x00000418, 0xff206100,
slouken@0
  1289
		0x00000420, 0xff208100, 0x00000429, 0xff20a100,
slouken@0
  1290
		0x00000431, 0xff20c200, 0x00000439, 0xff20e200,
slouken@0
  1291
		0x00000441, 0xff290000, 0x0000044a, 0xff292000,
slouken@0
  1292
		0x00000452, 0xff294000, 0x0000045a, 0xff296100,
slouken@0
  1293
		0x00000462, 0xff298100, 0x0000046a, 0xff29a100,
slouken@0
  1294
		0x00000473, 0xff29c200, 0x0000047b, 0xff29e200,
slouken@0
  1295
		0x00000483, 0xff310000, 0x0000048b, 0xff312000,
slouken@0
  1296
		0x00000494, 0xff314000, 0x0000049c, 0xff316100,
slouken@0
  1297
		0x000004a4, 0xff318100, 0x000004ac, 0xff31a100,
slouken@0
  1298
		0x000004b4, 0xff31c200, 0x000004bd, 0xff31e200,
slouken@0
  1299
		0x000004c5, 0xff390000, 0x000004cd, 0xff392000,
slouken@0
  1300
		0x000004d5, 0xff394000, 0x000004de, 0xff396100,
slouken@0
  1301
		0x000004e6, 0xff398100, 0x000004ee, 0xff39a100,
slouken@0
  1302
		0x000004f6, 0xff39c200, 0x000004ff, 0xff39e200,
slouken@0
  1303
		0x00000800, 0xff410000, 0x00000808, 0xff412000,
slouken@0
  1304
		0x00000810, 0xff414000, 0x00000818, 0xff416100,
slouken@0
  1305
		0x00000820, 0xff418100, 0x00000829, 0xff41a100,
slouken@0
  1306
		0x00000831, 0xff41c200, 0x00000839, 0xff41e200,
slouken@0
  1307
		0x00000841, 0xff4a0000, 0x0000084a, 0xff4a2000,
slouken@0
  1308
		0x00000852, 0xff4a4000, 0x0000085a, 0xff4a6100,
slouken@0
  1309
		0x00000862, 0xff4a8100, 0x0000086a, 0xff4aa100,
slouken@0
  1310
		0x00000873, 0xff4ac200, 0x0000087b, 0xff4ae200,
slouken@0
  1311
		0x00000883, 0xff520000, 0x0000088b, 0xff522000,
slouken@0
  1312
		0x00000894, 0xff524000, 0x0000089c, 0xff526100,
slouken@0
  1313
		0x000008a4, 0xff528100, 0x000008ac, 0xff52a100,
slouken@0
  1314
		0x000008b4, 0xff52c200, 0x000008bd, 0xff52e200,
slouken@0
  1315
		0x000008c5, 0xff5a0000, 0x000008cd, 0xff5a2000,
slouken@0
  1316
		0x000008d5, 0xff5a4000, 0x000008de, 0xff5a6100,
slouken@0
  1317
		0x000008e6, 0xff5a8100, 0x000008ee, 0xff5aa100,
slouken@0
  1318
		0x000008f6, 0xff5ac200, 0x000008ff, 0xff5ae200,
slouken@0
  1319
		0x00000c00, 0xff620000, 0x00000c08, 0xff622000,
slouken@0
  1320
		0x00000c10, 0xff624000, 0x00000c18, 0xff626100,
slouken@0
  1321
		0x00000c20, 0xff628100, 0x00000c29, 0xff62a100,
slouken@0
  1322
		0x00000c31, 0xff62c200, 0x00000c39, 0xff62e200,
slouken@0
  1323
		0x00000c41, 0xff6a0000, 0x00000c4a, 0xff6a2000,
slouken@0
  1324
		0x00000c52, 0xff6a4000, 0x00000c5a, 0xff6a6100,
slouken@0
  1325
		0x00000c62, 0xff6a8100, 0x00000c6a, 0xff6aa100,
slouken@0
  1326
		0x00000c73, 0xff6ac200, 0x00000c7b, 0xff6ae200,
slouken@0
  1327
		0x00000c83, 0xff730000, 0x00000c8b, 0xff732000,
slouken@0
  1328
		0x00000c94, 0xff734000, 0x00000c9c, 0xff736100,
slouken@0
  1329
		0x00000ca4, 0xff738100, 0x00000cac, 0xff73a100,
slouken@0
  1330
		0x00000cb4, 0xff73c200, 0x00000cbd, 0xff73e200,
slouken@0
  1331
		0x00000cc5, 0xff7b0000, 0x00000ccd, 0xff7b2000,
slouken@0
  1332
		0x00000cd5, 0xff7b4000, 0x00000cde, 0xff7b6100,
slouken@0
  1333
		0x00000ce6, 0xff7b8100, 0x00000cee, 0xff7ba100,
slouken@0
  1334
		0x00000cf6, 0xff7bc200, 0x00000cff, 0xff7be200,
slouken@0
  1335
		0x00001000, 0xff830000, 0x00001008, 0xff832000,
slouken@0
  1336
		0x00001010, 0xff834000, 0x00001018, 0xff836100,
slouken@0
  1337
		0x00001020, 0xff838100, 0x00001029, 0xff83a100,
slouken@0
  1338
		0x00001031, 0xff83c200, 0x00001039, 0xff83e200,
slouken@0
  1339
		0x00001041, 0xff8b0000, 0x0000104a, 0xff8b2000,
slouken@0
  1340
		0x00001052, 0xff8b4000, 0x0000105a, 0xff8b6100,
slouken@0
  1341
		0x00001062, 0xff8b8100, 0x0000106a, 0xff8ba100,
slouken@0
  1342
		0x00001073, 0xff8bc200, 0x0000107b, 0xff8be200,
slouken@0
  1343
		0x00001083, 0xff940000, 0x0000108b, 0xff942000,
slouken@0
  1344
		0x00001094, 0xff944000, 0x0000109c, 0xff946100,
slouken@0
  1345
		0x000010a4, 0xff948100, 0x000010ac, 0xff94a100,
slouken@0
  1346
		0x000010b4, 0xff94c200, 0x000010bd, 0xff94e200,
slouken@0
  1347
		0x000010c5, 0xff9c0000, 0x000010cd, 0xff9c2000,
slouken@0
  1348
		0x000010d5, 0xff9c4000, 0x000010de, 0xff9c6100,
slouken@0
  1349
		0x000010e6, 0xff9c8100, 0x000010ee, 0xff9ca100,
slouken@0
  1350
		0x000010f6, 0xff9cc200, 0x000010ff, 0xff9ce200,
slouken@0
  1351
		0x00001400, 0xffa40000, 0x00001408, 0xffa42000,
slouken@0
  1352
		0x00001410, 0xffa44000, 0x00001418, 0xffa46100,
slouken@0
  1353
		0x00001420, 0xffa48100, 0x00001429, 0xffa4a100,
slouken@0
  1354
		0x00001431, 0xffa4c200, 0x00001439, 0xffa4e200,
slouken@0
  1355
		0x00001441, 0xffac0000, 0x0000144a, 0xffac2000,
slouken@0
  1356
		0x00001452, 0xffac4000, 0x0000145a, 0xffac6100,
slouken@0
  1357
		0x00001462, 0xffac8100, 0x0000146a, 0xffaca100,
slouken@0
  1358
		0x00001473, 0xffacc200, 0x0000147b, 0xfface200,
slouken@0
  1359
		0x00001483, 0xffb40000, 0x0000148b, 0xffb42000,
slouken@0
  1360
		0x00001494, 0xffb44000, 0x0000149c, 0xffb46100,
slouken@0
  1361
		0x000014a4, 0xffb48100, 0x000014ac, 0xffb4a100,
slouken@0
  1362
		0x000014b4, 0xffb4c200, 0x000014bd, 0xffb4e200,
slouken@0
  1363
		0x000014c5, 0xffbd0000, 0x000014cd, 0xffbd2000,
slouken@0
  1364
		0x000014d5, 0xffbd4000, 0x000014de, 0xffbd6100,
slouken@0
  1365
		0x000014e6, 0xffbd8100, 0x000014ee, 0xffbda100,
slouken@0
  1366
		0x000014f6, 0xffbdc200, 0x000014ff, 0xffbde200,
slouken@0
  1367
		0x00001800, 0xffc50000, 0x00001808, 0xffc52000,
slouken@0
  1368
		0x00001810, 0xffc54000, 0x00001818, 0xffc56100,
slouken@0
  1369
		0x00001820, 0xffc58100, 0x00001829, 0xffc5a100,
slouken@0
  1370
		0x00001831, 0xffc5c200, 0x00001839, 0xffc5e200,
slouken@0
  1371
		0x00001841, 0xffcd0000, 0x0000184a, 0xffcd2000,
slouken@0
  1372
		0x00001852, 0xffcd4000, 0x0000185a, 0xffcd6100,
slouken@0
  1373
		0x00001862, 0xffcd8100, 0x0000186a, 0xffcda100,
slouken@0
  1374
		0x00001873, 0xffcdc200, 0x0000187b, 0xffcde200,
slouken@0
  1375
		0x00001883, 0xffd50000, 0x0000188b, 0xffd52000,
slouken@0
  1376
		0x00001894, 0xffd54000, 0x0000189c, 0xffd56100,
slouken@0
  1377
		0x000018a4, 0xffd58100, 0x000018ac, 0xffd5a100,
slouken@0
  1378
		0x000018b4, 0xffd5c200, 0x000018bd, 0xffd5e200,
slouken@0
  1379
		0x000018c5, 0xffde0000, 0x000018cd, 0xffde2000,
slouken@0
  1380
		0x000018d5, 0xffde4000, 0x000018de, 0xffde6100,
slouken@0
  1381
		0x000018e6, 0xffde8100, 0x000018ee, 0xffdea100,
slouken@0
  1382
		0x000018f6, 0xffdec200, 0x000018ff, 0xffdee200,
slouken@0
  1383
		0x00001c00, 0xffe60000, 0x00001c08, 0xffe62000,
slouken@0
  1384
		0x00001c10, 0xffe64000, 0x00001c18, 0xffe66100,
slouken@0
  1385
		0x00001c20, 0xffe68100, 0x00001c29, 0xffe6a100,
slouken@0
  1386
		0x00001c31, 0xffe6c200, 0x00001c39, 0xffe6e200,
slouken@0
  1387
		0x00001c41, 0xffee0000, 0x00001c4a, 0xffee2000,
slouken@0
  1388
		0x00001c52, 0xffee4000, 0x00001c5a, 0xffee6100,
slouken@0
  1389
		0x00001c62, 0xffee8100, 0x00001c6a, 0xffeea100,
slouken@0
  1390
		0x00001c73, 0xffeec200, 0x00001c7b, 0xffeee200,
slouken@0
  1391
		0x00001c83, 0xfff60000, 0x00001c8b, 0xfff62000,
slouken@0
  1392
		0x00001c94, 0xfff64000, 0x00001c9c, 0xfff66100,
slouken@0
  1393
		0x00001ca4, 0xfff68100, 0x00001cac, 0xfff6a100,
slouken@0
  1394
		0x00001cb4, 0xfff6c200, 0x00001cbd, 0xfff6e200,
slouken@0
  1395
		0x00001cc5, 0xffff0000, 0x00001ccd, 0xffff2000,
slouken@0
  1396
		0x00001cd5, 0xffff4000, 0x00001cde, 0xffff6100,
slouken@0
  1397
		0x00001ce6, 0xffff8100, 0x00001cee, 0xffffa100,
slouken@0
  1398
		0x00001cf6, 0xffffc200, 0x00001cff, 0xffffe200
slouken@0
  1399
};
slouken@0
  1400
static void Blit_RGB565_ARGB8888(SDL_BlitInfo *info)
slouken@0
  1401
{
slouken@0
  1402
    Blit_RGB565_32(info, RGB565_ARGB8888_LUT);
slouken@0
  1403
}
slouken@0
  1404
slouken@0
  1405
/* Special optimized blit for RGB 5-6-5 --> ABGR 8-8-8-8 */
slouken@91
  1406
static const Uint32 RGB565_ABGR8888_LUT[512] = {
slouken@0
  1407
		0xff000000, 0x00000000, 0xff080000, 0x00002000,
slouken@0
  1408
		0xff100000, 0x00004000, 0xff180000, 0x00006100,
slouken@0
  1409
		0xff200000, 0x00008100, 0xff290000, 0x0000a100,
slouken@0
  1410
		0xff310000, 0x0000c200, 0xff390000, 0x0000e200,
slouken@0
  1411
		0xff410000, 0x00000008, 0xff4a0000, 0x00002008,
slouken@0
  1412
		0xff520000, 0x00004008, 0xff5a0000, 0x00006108,
slouken@0
  1413
		0xff620000, 0x00008108, 0xff6a0000, 0x0000a108,
slouken@0
  1414
		0xff730000, 0x0000c208, 0xff7b0000, 0x0000e208,
slouken@0
  1415
		0xff830000, 0x00000010, 0xff8b0000, 0x00002010,
slouken@0
  1416
		0xff940000, 0x00004010, 0xff9c0000, 0x00006110,
slouken@0
  1417
		0xffa40000, 0x00008110, 0xffac0000, 0x0000a110,
slouken@0
  1418
		0xffb40000, 0x0000c210, 0xffbd0000, 0x0000e210,
slouken@0
  1419
		0xffc50000, 0x00000018, 0xffcd0000, 0x00002018,
slouken@0
  1420
		0xffd50000, 0x00004018, 0xffde0000, 0x00006118,
slouken@0
  1421
		0xffe60000, 0x00008118, 0xffee0000, 0x0000a118,
slouken@0
  1422
		0xfff60000, 0x0000c218, 0xffff0000, 0x0000e218,
slouken@0
  1423
		0xff000400, 0x00000020, 0xff080400, 0x00002020,
slouken@0
  1424
		0xff100400, 0x00004020, 0xff180400, 0x00006120,
slouken@0
  1425
		0xff200400, 0x00008120, 0xff290400, 0x0000a120,
slouken@0
  1426
		0xff310400, 0x0000c220, 0xff390400, 0x0000e220,
slouken@0
  1427
		0xff410400, 0x00000029, 0xff4a0400, 0x00002029,
slouken@0
  1428
		0xff520400, 0x00004029, 0xff5a0400, 0x00006129,
slouken@0
  1429
		0xff620400, 0x00008129, 0xff6a0400, 0x0000a129,
slouken@0
  1430
		0xff730400, 0x0000c229, 0xff7b0400, 0x0000e229,
slouken@0
  1431
		0xff830400, 0x00000031, 0xff8b0400, 0x00002031,
slouken@0
  1432
		0xff940400, 0x00004031, 0xff9c0400, 0x00006131,
slouken@0
  1433
		0xffa40400, 0x00008131, 0xffac0400, 0x0000a131,
slouken@0
  1434
		0xffb40400, 0x0000c231, 0xffbd0400, 0x0000e231,
slouken@0
  1435
		0xffc50400, 0x00000039, 0xffcd0400, 0x00002039,
slouken@0
  1436
		0xffd50400, 0x00004039, 0xffde0400, 0x00006139,
slouken@0
  1437
		0xffe60400, 0x00008139, 0xffee0400, 0x0000a139,
slouken@0
  1438
		0xfff60400, 0x0000c239, 0xffff0400, 0x0000e239,
slouken@0
  1439
		0xff000800, 0x00000041, 0xff080800, 0x00002041,
slouken@0
  1440
		0xff100800, 0x00004041, 0xff180800, 0x00006141,
slouken@0
  1441
		0xff200800, 0x00008141, 0xff290800, 0x0000a141,
slouken@0
  1442
		0xff310800, 0x0000c241, 0xff390800, 0x0000e241,
slouken@0
  1443
		0xff410800, 0x0000004a, 0xff4a0800, 0x0000204a,
slouken@0
  1444
		0xff520800, 0x0000404a, 0xff5a0800, 0x0000614a,
slouken@0
  1445
		0xff620800, 0x0000814a, 0xff6a0800, 0x0000a14a,
slouken@0
  1446
		0xff730800, 0x0000c24a, 0xff7b0800, 0x0000e24a,
slouken@0
  1447
		0xff830800, 0x00000052, 0xff8b0800, 0x00002052,
slouken@0
  1448
		0xff940800, 0x00004052, 0xff9c0800, 0x00006152,
slouken@0
  1449
		0xffa40800, 0x00008152, 0xffac0800, 0x0000a152,
slouken@0
  1450
		0xffb40800, 0x0000c252, 0xffbd0800, 0x0000e252,
slouken@0
  1451
		0xffc50800, 0x0000005a, 0xffcd0800, 0x0000205a,
slouken@0
  1452
		0xffd50800, 0x0000405a, 0xffde0800, 0x0000615a,
slouken@0
  1453
		0xffe60800, 0x0000815a, 0xffee0800, 0x0000a15a,
slouken@0
  1454
		0xfff60800, 0x0000c25a, 0xffff0800, 0x0000e25a,
slouken@0
  1455
		0xff000c00, 0x00000062, 0xff080c00, 0x00002062,
slouken@0
  1456
		0xff100c00, 0x00004062, 0xff180c00, 0x00006162,
slouken@0
  1457
		0xff200c00, 0x00008162, 0xff290c00, 0x0000a162,
slouken@0
  1458
		0xff310c00, 0x0000c262, 0xff390c00, 0x0000e262,
slouken@0
  1459
		0xff410c00, 0x0000006a, 0xff4a0c00, 0x0000206a,
slouken@0
  1460
		0xff520c00, 0x0000406a, 0xff5a0c00, 0x0000616a,
slouken@0
  1461
		0xff620c00, 0x0000816a, 0xff6a0c00, 0x0000a16a,
slouken@0
  1462
		0xff730c00, 0x0000c26a, 0xff7b0c00, 0x0000e26a,
slouken@0
  1463
		0xff830c00, 0x00000073, 0xff8b0c00, 0x00002073,
slouken@0
  1464
		0xff940c00, 0x00004073, 0xff9c0c00, 0x00006173,
slouken@0
  1465
		0xffa40c00, 0x00008173, 0xffac0c00, 0x0000a173,
slouken@0
  1466
		0xffb40c00, 0x0000c273, 0xffbd0c00, 0x0000e273,
slouken@0
  1467
		0xffc50c00, 0x0000007b, 0xffcd0c00, 0x0000207b,
slouken@0
  1468
		0xffd50c00, 0x0000407b, 0xffde0c00, 0x0000617b,
slouken@0
  1469
		0xffe60c00, 0x0000817b, 0xffee0c00, 0x0000a17b,
slouken@0
  1470
		0xfff60c00, 0x0000c27b, 0xffff0c00, 0x0000e27b,
slouken@0
  1471
		0xff001000, 0x00000083, 0xff081000, 0x00002083,
slouken@0
  1472
		0xff101000, 0x00004083, 0xff181000, 0x00006183,
slouken@0
  1473
		0xff201000, 0x00008183, 0xff291000, 0x0000a183,
slouken@0
  1474
		0xff311000, 0x0000c283, 0xff391000, 0x0000e283,
slouken@0
  1475
		0xff411000, 0x0000008b, 0xff4a1000, 0x0000208b,
slouken@0
  1476
		0xff521000, 0x0000408b, 0xff5a1000, 0x0000618b,
slouken@0
  1477
		0xff621000, 0x0000818b, 0xff6a1000, 0x0000a18b,
slouken@0
  1478
		0xff731000, 0x0000c28b, 0xff7b1000, 0x0000e28b,
slouken@0
  1479
		0xff831000, 0x00000094, 0xff8b1000, 0x00002094,
slouken@0
  1480
		0xff941000, 0x00004094, 0xff9c1000, 0x00006194,
slouken@0
  1481
		0xffa41000, 0x00008194, 0xffac1000, 0x0000a194,
slouken@0
  1482
		0xffb41000, 0x0000c294, 0xffbd1000, 0x0000e294,
slouken@0
  1483
		0xffc51000, 0x0000009c, 0xffcd1000, 0x0000209c,
slouken@0
  1484
		0xffd51000, 0x0000409c, 0xffde1000, 0x0000619c,
slouken@0
  1485
		0xffe61000, 0x0000819c, 0xffee1000, 0x0000a19c,
slouken@0
  1486
		0xfff61000, 0x0000c29c, 0xffff1000, 0x0000e29c,
slouken@0
  1487
		0xff001400, 0x000000a4, 0xff081400, 0x000020a4,
slouken@0
  1488
		0xff101400, 0x000040a4, 0xff181400, 0x000061a4,
slouken@0
  1489
		0xff201400, 0x000081a4, 0xff291400, 0x0000a1a4,
slouken@0
  1490
		0xff311400, 0x0000c2a4, 0xff391400, 0x0000e2a4,
slouken@0
  1491
		0xff411400, 0x000000ac, 0xff4a1400, 0x000020ac,
slouken@0
  1492
		0xff521400, 0x000040ac, 0xff5a1400, 0x000061ac,
slouken@0
  1493
		0xff621400, 0x000081ac, 0xff6a1400, 0x0000a1ac,
slouken@0
  1494
		0xff731400, 0x0000c2ac, 0xff7b1400, 0x0000e2ac,
slouken@0
  1495
		0xff831400, 0x000000b4, 0xff8b1400, 0x000020b4,
slouken@0
  1496
		0xff941400, 0x000040b4, 0xff9c1400, 0x000061b4,
slouken@0
  1497
		0xffa41400, 0x000081b4, 0xffac1400, 0x0000a1b4,
slouken@0
  1498
		0xffb41400, 0x0000c2b4, 0xffbd1400, 0x0000e2b4,
slouken@0
  1499
		0xffc51400, 0x000000bd, 0xffcd1400, 0x000020bd,
slouken@0
  1500
		0xffd51400, 0x000040bd, 0xffde1400, 0x000061bd,
slouken@0
  1501
		0xffe61400, 0x000081bd, 0xffee1400, 0x0000a1bd,
slouken@0
  1502
		0xfff61400, 0x0000c2bd, 0xffff1400, 0x0000e2bd,
slouken@0
  1503
		0xff001800, 0x000000c5, 0xff081800, 0x000020c5,
slouken@0
  1504
		0xff101800, 0x000040c5, 0xff181800, 0x000061c5,
slouken@0
  1505
		0xff201800, 0x000081c5, 0xff291800, 0x0000a1c5,
slouken@0
  1506
		0xff311800, 0x0000c2c5, 0xff391800, 0x0000e2c5,
slouken@0
  1507
		0xff411800, 0x000000cd, 0xff4a1800, 0x000020cd,
slouken@0
  1508
		0xff521800, 0x000040cd, 0xff5a1800, 0x000061cd,
slouken@0
  1509
		0xff621800, 0x000081cd, 0xff6a1800, 0x0000a1cd,
slouken@0
  1510
		0xff731800, 0x0000c2cd, 0xff7b1800, 0x0000e2cd,
slouken@0
  1511
		0xff831800, 0x000000d5, 0xff8b1800, 0x000020d5,
slouken@0
  1512
		0xff941800, 0x000040d5, 0xff9c1800, 0x000061d5,
slouken@0
  1513
		0xffa41800, 0x000081d5, 0xffac1800, 0x0000a1d5,
slouken@0
  1514
		0xffb41800, 0x0000c2d5, 0xffbd1800, 0x0000e2d5,
slouken@0
  1515
		0xffc51800, 0x000000de, 0xffcd1800, 0x000020de,
slouken@0
  1516
		0xffd51800, 0x000040de, 0xffde1800, 0x000061de,
slouken@0
  1517
		0xffe61800, 0x000081de, 0xffee1800, 0x0000a1de,
slouken@0
  1518
		0xfff61800, 0x0000c2de, 0xffff1800, 0x0000e2de,
slouken@0
  1519
		0xff001c00, 0x000000e6, 0xff081c00, 0x000020e6,
slouken@0
  1520
		0xff101c00, 0x000040e6, 0xff181c00, 0x000061e6,
slouken@0
  1521
		0xff201c00, 0x000081e6, 0xff291c00, 0x0000a1e6,
slouken@0
  1522
		0xff311c00, 0x0000c2e6, 0xff391c00, 0x0000e2e6,
slouken@0
  1523
		0xff411c00, 0x000000ee, 0xff4a1c00, 0x000020ee,
slouken@0
  1524
		0xff521c00, 0x000040ee, 0xff5a1c00, 0x000061ee,
slouken@0
  1525
		0xff621c00, 0x000081ee, 0xff6a1c00, 0x0000a1ee,
slouken@0
  1526
		0xff731c00, 0x0000c2ee, 0xff7b1c00, 0x0000e2ee,
slouken@0
  1527
		0xff831c00, 0x000000f6, 0xff8b1c00, 0x000020f6,
slouken@0
  1528
		0xff941c00, 0x000040f6, 0xff9c1c00, 0x000061f6,
slouken@0
  1529
		0xffa41c00, 0x000081f6, 0xffac1c00, 0x0000a1f6,
slouken@0
  1530
		0xffb41c00, 0x0000c2f6, 0xffbd1c00, 0x0000e2f6,
slouken@0
  1531
		0xffc51c00, 0x000000ff, 0xffcd1c00, 0x000020ff,
slouken@0
  1532
		0xffd51c00, 0x000040ff, 0xffde1c00, 0x000061ff,
slouken@0
  1533
		0xffe61c00, 0x000081ff, 0xffee1c00, 0x0000a1ff,
slouken@0
  1534
		0xfff61c00, 0x0000c2ff, 0xffff1c00, 0x0000e2ff
slouken@0
  1535
};
slouken@0
  1536
static void Blit_RGB565_ABGR8888(SDL_BlitInfo *info)
slouken@0
  1537
{
slouken@0
  1538
    Blit_RGB565_32(info, RGB565_ABGR8888_LUT);
slouken@0
  1539
}
slouken@0
  1540
slouken@0
  1541
/* Special optimized blit for RGB 5-6-5 --> RGBA 8-8-8-8 */
slouken@91
  1542
static const Uint32 RGB565_RGBA8888_LUT[512] = {
slouken@0
  1543
		0x000000ff, 0x00000000, 0x000008ff, 0x00200000,
slouken@0
  1544
		0x000010ff, 0x00400000, 0x000018ff, 0x00610000,
slouken@0
  1545
		0x000020ff, 0x00810000, 0x000029ff, 0x00a10000,
slouken@0
  1546
		0x000031ff, 0x00c20000, 0x000039ff, 0x00e20000,
slouken@0
  1547
		0x000041ff, 0x08000000, 0x00004aff, 0x08200000,
slouken@0
  1548
		0x000052ff, 0x08400000, 0x00005aff, 0x08610000,
slouken@0
  1549
		0x000062ff, 0x08810000, 0x00006aff, 0x08a10000,
slouken@0
  1550
		0x000073ff, 0x08c20000, 0x00007bff, 0x08e20000,
slouken@0
  1551
		0x000083ff, 0x10000000, 0x00008bff, 0x10200000,
slouken@0
  1552
		0x000094ff, 0x10400000, 0x00009cff, 0x10610000,
slouken@0
  1553
		0x0000a4ff, 0x10810000, 0x0000acff, 0x10a10000,
slouken@0
  1554
		0x0000b4ff, 0x10c20000, 0x0000bdff, 0x10e20000,
slouken@0
  1555
		0x0000c5ff, 0x18000000, 0x0000cdff, 0x18200000,
slouken@0
  1556
		0x0000d5ff, 0x18400000, 0x0000deff, 0x18610000,
slouken@0
  1557
		0x0000e6ff, 0x18810000, 0x0000eeff, 0x18a10000,
slouken@0
  1558
		0x0000f6ff, 0x18c20000, 0x0000ffff, 0x18e20000,
slouken@0
  1559
		0x000400ff, 0x20000000, 0x000408ff, 0x20200000,
slouken@0
  1560
		0x000410ff, 0x20400000, 0x000418ff, 0x20610000,
slouken@0
  1561
		0x000420ff, 0x20810000, 0x000429ff, 0x20a10000,
slouken@0
  1562
		0x000431ff, 0x20c20000, 0x000439ff, 0x20e20000,
slouken@0
  1563
		0x000441ff, 0x29000000, 0x00044aff, 0x29200000,
slouken@0
  1564
		0x000452ff, 0x29400000, 0x00045aff, 0x29610000,
slouken@0
  1565
		0x000462ff, 0x29810000, 0x00046aff, 0x29a10000,
slouken@0
  1566
		0x000473ff, 0x29c20000, 0x00047bff, 0x29e20000,
slouken@0
  1567
		0x000483ff, 0x31000000, 0x00048bff, 0x31200000,
slouken@0
  1568
		0x000494ff, 0x31400000, 0x00049cff, 0x31610000,
slouken@0
  1569
		0x0004a4ff, 0x31810000, 0x0004acff, 0x31a10000,
slouken@0
  1570
		0x0004b4ff, 0x31c20000, 0x0004bdff, 0x31e20000,
slouken@0
  1571
		0x0004c5ff, 0x39000000, 0x0004cdff, 0x39200000,
slouken@0
  1572
		0x0004d5ff, 0x39400000, 0x0004deff, 0x39610000,
slouken@0
  1573
		0x0004e6ff, 0x39810000, 0x0004eeff, 0x39a10000,
slouken@0
  1574
		0x0004f6ff, 0x39c20000, 0x0004ffff, 0x39e20000,
slouken@0
  1575
		0x000800ff, 0x41000000, 0x000808ff, 0x41200000,
slouken@0
  1576
		0x000810ff, 0x41400000, 0x000818ff, 0x41610000,
slouken@0
  1577
		0x000820ff, 0x41810000, 0x000829ff, 0x41a10000,
slouken@0
  1578
		0x000831ff, 0x41c20000, 0x000839ff, 0x41e20000,
slouken@0
  1579
		0x000841ff, 0x4a000000, 0x00084aff, 0x4a200000,
slouken@0
  1580
		0x000852ff, 0x4a400000, 0x00085aff, 0x4a610000,
slouken@0
  1581
		0x000862ff, 0x4a810000, 0x00086aff, 0x4aa10000,
slouken@0
  1582
		0x000873ff, 0x4ac20000, 0x00087bff, 0x4ae20000,
slouken@0
  1583
		0x000883ff, 0x52000000, 0x00088bff, 0x52200000,
slouken@0
  1584
		0x000894ff, 0x52400000, 0x00089cff, 0x52610000,
slouken@0
  1585
		0x0008a4ff, 0x52810000, 0x0008acff, 0x52a10000,
slouken@0
  1586
		0x0008b4ff, 0x52c20000, 0x0008bdff, 0x52e20000,
slouken@0
  1587
		0x0008c5ff, 0x5a000000, 0x0008cdff, 0x5a200000,
slouken@0
  1588
		0x0008d5ff, 0x5a400000, 0x0008deff, 0x5a610000,
slouken@0
  1589
		0x0008e6ff, 0x5a810000, 0x0008eeff, 0x5aa10000,
slouken@0
  1590
		0x0008f6ff, 0x5ac20000, 0x0008ffff, 0x5ae20000,
slouken@0
  1591
		0x000c00ff, 0x62000000, 0x000c08ff, 0x62200000,
slouken@0
  1592
		0x000c10ff, 0x62400000, 0x000c18ff, 0x62610000,
slouken@0
  1593
		0x000c20ff, 0x62810000, 0x000c29ff, 0x62a10000,
slouken@0
  1594
		0x000c31ff, 0x62c20000, 0x000c39ff, 0x62e20000,
slouken@0
  1595
		0x000c41ff, 0x6a000000, 0x000c4aff, 0x6a200000,
slouken@0
  1596
		0x000c52ff, 0x6a400000, 0x000c5aff, 0x6a610000,
slouken@0
  1597
		0x000c62ff, 0x6a810000, 0x000c6aff, 0x6aa10000,
slouken@0
  1598
		0x000c73ff, 0x6ac20000, 0x000c7bff, 0x6ae20000,
slouken@0
  1599
		0x000c83ff, 0x73000000, 0x000c8bff, 0x73200000,
slouken@0
  1600
		0x000c94ff, 0x73400000, 0x000c9cff, 0x73610000,
slouken@0
  1601
		0x000ca4ff, 0x73810000, 0x000cacff, 0x73a10000,
slouken@0
  1602
		0x000cb4ff, 0x73c20000, 0x000cbdff, 0x73e20000,
slouken@0
  1603
		0x000cc5ff, 0x7b000000, 0x000ccdff, 0x7b200000,
slouken@0
  1604
		0x000cd5ff, 0x7b400000, 0x000cdeff, 0x7b610000,
slouken@0
  1605
		0x000ce6ff, 0x7b810000, 0x000ceeff, 0x7ba10000,
slouken@0
  1606
		0x000cf6ff, 0x7bc20000, 0x000cffff, 0x7be20000,
slouken@0
  1607
		0x001000ff, 0x83000000, 0x001008ff, 0x83200000,
slouken@0
  1608
		0x001010ff, 0x83400000, 0x001018ff, 0x83610000,
slouken@0
  1609
		0x001020ff, 0x83810000, 0x001029ff, 0x83a10000,
slouken@0
  1610
		0x001031ff, 0x83c20000, 0x001039ff, 0x83e20000,
slouken@0
  1611
		0x001041ff, 0x8b000000, 0x00104aff, 0x8b200000,
slouken@0
  1612
		0x001052ff, 0x8b400000, 0x00105aff, 0x8b610000,
slouken@0
  1613
		0x001062ff, 0x8b810000, 0x00106aff, 0x8ba10000,
slouken@0
  1614
		0x001073ff, 0x8bc20000, 0x00107bff, 0x8be20000,
slouken@0
  1615
		0x001083ff, 0x94000000, 0x00108bff, 0x94200000,
slouken@0
  1616
		0x001094ff, 0x94400000, 0x00109cff, 0x94610000,
slouken@0
  1617
		0x0010a4ff, 0x94810000, 0x0010acff, 0x94a10000,
slouken@0
  1618
		0x0010b4ff, 0x94c20000, 0x0010bdff, 0x94e20000,
slouken@0
  1619
		0x0010c5ff, 0x9c000000, 0x0010cdff, 0x9c200000,
slouken@0
  1620
		0x0010d5ff, 0x9c400000, 0x0010deff, 0x9c610000,
slouken@0
  1621
		0x0010e6ff, 0x9c810000, 0x0010eeff, 0x9ca10000,
slouken@0
  1622
		0x0010f6ff, 0x9cc20000, 0x0010ffff, 0x9ce20000,
slouken@0
  1623
		0x001400ff, 0xa4000000, 0x001408ff, 0xa4200000,
slouken@0
  1624
		0x001410ff, 0xa4400000, 0x001418ff, 0xa4610000,
slouken@0
  1625
		0x001420ff, 0xa4810000, 0x001429ff, 0xa4a10000,
slouken@0
  1626
		0x001431ff, 0xa4c20000, 0x001439ff, 0xa4e20000,
slouken@0
  1627
		0x001441ff, 0xac000000, 0x00144aff, 0xac200000,
slouken@0
  1628
		0x001452ff, 0xac400000, 0x00145aff, 0xac610000,
slouken@0
  1629
		0x001462ff, 0xac810000, 0x00146aff, 0xaca10000,
slouken@0
  1630
		0x001473ff, 0xacc20000, 0x00147bff, 0xace20000,
slouken@0
  1631
		0x001483ff, 0xb4000000, 0x00148bff, 0xb4200000,
slouken@0
  1632
		0x001494ff, 0xb4400000, 0x00149cff, 0xb4610000,
slouken@0
  1633
		0x0014a4ff, 0xb4810000, 0x0014acff, 0xb4a10000,
slouken@0
  1634
		0x0014b4ff, 0xb4c20000, 0x0014bdff, 0xb4e20000,
slouken@0
  1635
		0x0014c5ff, 0xbd000000, 0x0014cdff, 0xbd200000,
slouken@0
  1636
		0x0014d5ff, 0xbd400000, 0x0014deff, 0xbd610000,
slouken@0
  1637
		0x0014e6ff, 0xbd810000, 0x0014eeff, 0xbda10000,
slouken@0
  1638
		0x0014f6ff, 0xbdc20000, 0x0014ffff, 0xbde20000,
slouken@0
  1639
		0x001800ff, 0xc5000000, 0x001808ff, 0xc5200000,
slouken@0
  1640
		0x001810ff, 0xc5400000, 0x001818ff, 0xc5610000,
slouken@0
  1641
		0x001820ff, 0xc5810000, 0x001829ff, 0xc5a10000,
slouken@0
  1642
		0x001831ff, 0xc5c20000, 0x001839ff, 0xc5e20000,
slouken@0
  1643
		0x001841ff, 0xcd000000, 0x00184aff, 0xcd200000,
slouken@0
  1644
		0x001852ff, 0xcd400000, 0x00185aff, 0xcd610000,
slouken@0
  1645
		0x001862ff, 0xcd810000, 0x00186aff, 0xcda10000,
slouken@0
  1646
		0x001873ff, 0xcdc20000, 0x00187bff, 0xcde20000,
slouken@0
  1647
		0x001883ff, 0xd5000000, 0x00188bff, 0xd5200000,
slouken@0
  1648
		0x001894ff, 0xd5400000, 0x00189cff, 0xd5610000,
slouken@0
  1649
		0x0018a4ff, 0xd5810000, 0x0018acff, 0xd5a10000,
slouken@0
  1650
		0x0018b4ff, 0xd5c20000, 0x0018bdff, 0xd5e20000,
slouken@0
  1651
		0x0018c5ff, 0xde000000, 0x0018cdff, 0xde200000,
slouken@0
  1652
		0x0018d5ff, 0xde400000, 0x0018deff, 0xde610000,
slouken@0
  1653
		0x0018e6ff, 0xde810000, 0x0018eeff, 0xdea10000,
slouken@0
  1654
		0x0018f6ff, 0xdec20000, 0x0018ffff, 0xdee20000,
slouken@0
  1655
		0x001c00ff, 0xe6000000, 0x001c08ff, 0xe6200000,
slouken@0
  1656
		0x001c10ff, 0xe6400000, 0x001c18ff, 0xe6610000,
slouken@0
  1657
		0x001c20ff, 0xe6810000, 0x001c29ff, 0xe6a10000,
slouken@0
  1658
		0x001c31ff, 0xe6c20000, 0x001c39ff, 0xe6e20000,
slouken@0
  1659
		0x001c41ff, 0xee000000, 0x001c4aff, 0xee200000,
slouken@0
  1660
		0x001c52ff, 0xee400000, 0x001c5aff, 0xee610000,
slouken@0
  1661
		0x001c62ff, 0xee810000, 0x001c6aff, 0xeea10000,
slouken@0
  1662
		0x001c73ff, 0xeec20000, 0x001c7bff, 0xeee20000,
slouken@0
  1663
		0x001c83ff, 0xf6000000, 0x001c8bff, 0xf6200000,
slouken@0
  1664
		0x001c94ff, 0xf6400000, 0x001c9cff, 0xf6610000,
slouken@0
  1665
		0x001ca4ff, 0xf6810000, 0x001cacff, 0xf6a10000,
slouken@0
  1666
		0x001cb4ff, 0xf6c20000, 0x001cbdff, 0xf6e20000,
slouken@0
  1667
		0x001cc5ff, 0xff000000, 0x001ccdff, 0xff200000,
slouken@0
  1668
		0x001cd5ff, 0xff400000, 0x001cdeff, 0xff610000,
slouken@0
  1669
		0x001ce6ff, 0xff810000, 0x001ceeff, 0xffa10000,
slouken@0
  1670
		0x001cf6ff, 0xffc20000, 0x001cffff, 0xffe20000,
slouken@0
  1671
};
slouken@0
  1672
static void Blit_RGB565_RGBA8888(SDL_BlitInfo *info)
slouken@0
  1673
{
slouken@0
  1674
    Blit_RGB565_32(info, RGB565_RGBA8888_LUT);
slouken@0
  1675
}
slouken@0
  1676
slouken@0
  1677
/* Special optimized blit for RGB 5-6-5 --> BGRA 8-8-8-8 */
slouken@91
  1678
static const Uint32 RGB565_BGRA8888_LUT[512] = {
slouken@0
  1679
		0x00000000, 0x000000ff, 0x08000000, 0x002000ff,
slouken@0
  1680
		0x10000000, 0x004000ff, 0x18000000, 0x006100ff,
slouken@0
  1681
		0x20000000, 0x008100ff, 0x29000000, 0x00a100ff,
slouken@0
  1682
		0x31000000, 0x00c200ff, 0x39000000, 0x00e200ff,
slouken@0
  1683
		0x41000000, 0x000008ff, 0x4a000000, 0x002008ff,
slouken@0
  1684
		0x52000000, 0x004008ff, 0x5a000000, 0x006108ff,
slouken@0
  1685
		0x62000000, 0x008108ff, 0x6a000000, 0x00a108ff,
slouken@0
  1686
		0x73000000, 0x00c208ff, 0x7b000000, 0x00e208ff,
slouken@0
  1687
		0x83000000, 0x000010ff, 0x8b000000, 0x002010ff,
slouken@0
  1688
		0x94000000, 0x004010ff, 0x9c000000, 0x006110ff,
slouken@0
  1689
		0xa4000000, 0x008110ff, 0xac000000, 0x00a110ff,
slouken@0
  1690
		0xb4000000, 0x00c210ff, 0xbd000000, 0x00e210ff,
slouken@0
  1691
		0xc5000000, 0x000018ff, 0xcd000000, 0x002018ff,
slouken@0
  1692
		0xd5000000, 0x004018ff, 0xde000000, 0x006118ff,
slouken@0
  1693
		0xe6000000, 0x008118ff, 0xee000000, 0x00a118ff,
slouken@0
  1694
		0xf6000000, 0x00c218ff, 0xff000000, 0x00e218ff,
slouken@0
  1695
		0x00040000, 0x000020ff, 0x08040000, 0x002020ff,
slouken@0
  1696
		0x10040000, 0x004020ff, 0x18040000, 0x006120ff,
slouken@0
  1697
		0x20040000, 0x008120ff, 0x29040000, 0x00a120ff,
slouken@0
  1698
		0x31040000, 0x00c220ff, 0x39040000, 0x00e220ff,
slouken@0
  1699
		0x41040000, 0x000029ff, 0x4a040000, 0x002029ff,
slouken@0
  1700
		0x52040000, 0x004029ff, 0x5a040000, 0x006129ff,
slouken@0
  1701
		0x62040000, 0x008129ff, 0x6a040000, 0x00a129ff,
slouken@0
  1702
		0x73040000, 0x00c229ff, 0x7b040000, 0x00e229ff,
slouken@0
  1703
		0x83040000, 0x000031ff, 0x8b040000, 0x002031ff,
slouken@0
  1704
		0x94040000, 0x004031ff, 0x9c040000, 0x006131ff,
slouken@0
  1705
		0xa4040000, 0x008131ff, 0xac040000, 0x00a131ff,
slouken@0
  1706
		0xb4040000, 0x00c231ff, 0xbd040000, 0x00e231ff,
slouken@0
  1707
		0xc5040000, 0x000039ff, 0xcd040000, 0x002039ff,
slouken@0
  1708
		0xd5040000, 0x004039ff, 0xde040000, 0x006139ff,
slouken@0
  1709
		0xe6040000, 0x008139ff, 0xee040000, 0x00a139ff,
slouken@0
  1710
		0xf6040000, 0x00c239ff, 0xff040000, 0x00e239ff,
slouken@0
  1711
		0x00080000, 0x000041ff, 0x08080000, 0x002041ff,
slouken@0
  1712
		0x10080000, 0x004041ff, 0x18080000, 0x006141ff,
slouken@0
  1713
		0x20080000, 0x008141ff, 0x29080000, 0x00a141ff,
slouken@0
  1714
		0x31080000, 0x00c241ff, 0x39080000, 0x00e241ff,
slouken@0
  1715
		0x41080000, 0x00004aff, 0x4a080000, 0x00204aff,
slouken@0
  1716
		0x52080000, 0x00404aff, 0x5a080000, 0x00614aff,
slouken@0
  1717
		0x62080000, 0x00814aff, 0x6a080000, 0x00a14aff,
slouken@0
  1718
		0x73080000, 0x00c24aff, 0x7b080000, 0x00e24aff,
slouken@0
  1719
		0x83080000, 0x000052ff, 0x8b080000, 0x002052ff,
slouken@0
  1720
		0x94080000, 0x004052ff, 0x9c080000, 0x006152ff,
slouken@0
  1721
		0xa4080000, 0x008152ff, 0xac080000, 0x00a152ff,
slouken@0
  1722
		0xb4080000, 0x00c252ff, 0xbd080000, 0x00e252ff,
slouken@0
  1723
		0xc5080000, 0x00005aff, 0xcd080000, 0x00205aff,
slouken@0
  1724
		0xd5080000, 0x00405aff, 0xde080000, 0x00615aff,
slouken@0
  1725
		0xe6080000, 0x00815aff, 0xee080000, 0x00a15aff,
slouken@0
  1726
		0xf6080000, 0x00c25aff, 0xff080000, 0x00e25aff,
slouken@0
  1727
		0x000c0000, 0x000062ff, 0x080c0000, 0x002062ff,
slouken@0
  1728
		0x100c0000, 0x004062ff, 0x180c0000, 0x006162ff,
slouken@0
  1729
		0x200c0000, 0x008162ff, 0x290c0000, 0x00a162ff,
slouken@0
  1730
		0x310c0000, 0x00c262ff, 0x390c0000, 0x00e262ff,
slouken@0
  1731
		0x410c0000, 0x00006aff, 0x4a0c0000, 0x00206aff,
slouken@0
  1732
		0x520c0000, 0x00406aff, 0x5a0c0000, 0x00616aff,
slouken@0
  1733
		0x620c0000, 0x00816aff, 0x6a0c0000, 0x00a16aff,
slouken@0
  1734
		0x730c0000, 0x00c26aff, 0x7b0c0000, 0x00e26aff,
slouken@0
  1735
		0x830c0000, 0x000073ff, 0x8b0c0000, 0x002073ff,
slouken@0
  1736
		0x940c0000, 0x004073ff, 0x9c0c0000, 0x006173ff,
slouken@0
  1737
		0xa40c0000, 0x008173ff, 0xac0c0000, 0x00a173ff,
slouken@0
  1738
		0xb40c0000, 0x00c273ff, 0xbd0c0000, 0x00e273ff,
slouken@0
  1739
		0xc50c0000, 0x00007bff, 0xcd0c0000, 0x00207bff,
slouken@0
  1740
		0xd50c0000, 0x00407bff, 0xde0c0000, 0x00617bff,
slouken@0
  1741
		0xe60c0000, 0x00817bff, 0xee0c0000, 0x00a17bff,
slouken@0
  1742
		0xf60c0000, 0x00c27bff, 0xff0c0000, 0x00e27bff,
slouken@0
  1743
		0x00100000, 0x000083ff, 0x08100000, 0x002083ff,
slouken@0
  1744
		0x10100000, 0x004083ff, 0x18100000, 0x006183ff,
slouken@0
  1745
		0x20100000, 0x008183ff, 0x29100000, 0x00a183ff,
slouken@0
  1746
		0x31100000, 0x00c283ff, 0x39100000, 0x00e283ff,
slouken@0
  1747
		0x41100000, 0x00008bff, 0x4a100000, 0x00208bff,
slouken@0
  1748
		0x52100000, 0x00408bff, 0x5a100000, 0x00618bff,
slouken@0
  1749
		0x62100000, 0x00818bff, 0x6a100000, 0x00a18bff,
slouken@0
  1750
		0x73100000, 0x00c28bff, 0x7b100000, 0x00e28bff,
slouken@0
  1751
		0x83100000, 0x000094ff, 0x8b100000, 0x002094ff,
slouken@0
  1752
		0x94100000, 0x004094ff, 0x9c100000, 0x006194ff,
slouken@0
  1753
		0xa4100000, 0x008194ff, 0xac100000, 0x00a194ff,
slouken@0
  1754
		0xb4100000, 0x00c294ff, 0xbd100000, 0x00e294ff,
slouken@0
  1755
		0xc5100000, 0x00009cff, 0xcd100000, 0x00209cff,
slouken@0
  1756
		0xd5100000, 0x00409cff, 0xde100000, 0x00619cff,
slouken@0
  1757
		0xe6100000, 0x00819cff, 0xee100000, 0x00a19cff,
slouken@0
  1758
		0xf6100000, 0x00c29cff, 0xff100000, 0x00e29cff,
slouken@0
  1759
		0x00140000, 0x0000a4ff, 0x08140000, 0x0020a4ff,
slouken@0
  1760
		0x10140000, 0x0040a4ff, 0x18140000, 0x0061a4ff,
slouken@0
  1761
		0x20140000, 0x0081a4ff, 0x29140000, 0x00a1a4ff,
slouken@0
  1762
		0x31140000, 0x00c2a4ff, 0x39140000, 0x00e2a4ff,
slouken@0
  1763
		0x41140000, 0x0000acff, 0x4a140000, 0x0020acff,
slouken@0
  1764
		0x52140000, 0x0040acff, 0x5a140000, 0x0061acff,
slouken@0
  1765
		0x62140000, 0x0081acff, 0x6a140000, 0x00a1acff,
slouken@0
  1766
		0x73140000, 0x00c2acff, 0x7b140000, 0x00e2acff,
slouken@0
  1767
		0x83140000, 0x0000b4ff, 0x8b140000, 0x0020b4ff,
slouken@0
  1768
		0x94140000, 0x0040b4ff, 0x9c140000, 0x0061b4ff,
slouken@0
  1769
		0xa4140000, 0x0081b4ff, 0xac140000, 0x00a1b4ff,
slouken@0
  1770
		0xb4140000, 0x00c2b4ff, 0xbd140000, 0x00e2b4ff,
slouken@0
  1771
		0xc5140000, 0x0000bdff, 0xcd140000, 0x0020bdff,
slouken@0
  1772
		0xd5140000, 0x0040bdff, 0xde140000, 0x0061bdff,
slouken@0
  1773
		0xe6140000, 0x0081bdff, 0xee140000, 0x00a1bdff,
slouken@0
  1774
		0xf6140000, 0x00c2bdff, 0xff140000, 0x00e2bdff,
slouken@0
  1775
		0x00180000, 0x0000c5ff, 0x08180000, 0x0020c5ff,
slouken@0
  1776
		0x10180000, 0x0040c5ff, 0x18180000, 0x0061c5ff,
slouken@0
  1777
		0x20180000, 0x0081c5ff, 0x29180000, 0x00a1c5ff,
slouken@0
  1778
		0x31180000, 0x00c2c5ff, 0x39180000, 0x00e2c5ff,
slouken@0
  1779
		0x41180000, 0x0000cdff, 0x4a180000, 0x0020cdff,
slouken@0
  1780
		0x52180000, 0x0040cdff, 0x5a180000, 0x0061cdff,
slouken@0
  1781
		0x62180000, 0x0081cdff, 0x6a180000, 0x00a1cdff,
slouken@0
  1782
		0x73180000, 0x00c2cdff, 0x7b180000, 0x00e2cdff,
slouken@0
  1783
		0x83180000, 0x0000d5ff, 0x8b180000, 0x0020d5ff,
slouken@0
  1784
		0x94180000, 0x0040d5ff, 0x9c180000, 0x0061d5ff,
slouken@0
  1785
		0xa4180000, 0x0081d5ff, 0xac180000, 0x00a1d5ff,
slouken@0
  1786
		0xb4180000, 0x00c2d5ff, 0xbd180000, 0x00e2d5ff,
slouken@0
  1787
		0xc5180000, 0x0000deff, 0xcd180000, 0x0020deff,
slouken@0
  1788
		0xd5180000, 0x0040deff, 0xde180000, 0x0061deff,
slouken@0
  1789
		0xe6180000, 0x0081deff, 0xee180000, 0x00a1deff,
slouken@0
  1790
		0xf6180000, 0x00c2deff, 0xff180000, 0x00e2deff,
slouken@0
  1791
		0x001c0000, 0x0000e6ff, 0x081c0000, 0x0020e6ff,
slouken@0
  1792
		0x101c0000, 0x0040e6ff, 0x181c0000, 0x0061e6ff,
slouken@0
  1793
		0x201c0000, 0x0081e6ff, 0x291c0000, 0x00a1e6ff,
slouken@0
  1794
		0x311c0000, 0x00c2e6ff, 0x391c0000, 0x00e2e6ff,
slouken@0
  1795
		0x411c0000, 0x0000eeff, 0x4a1c0000, 0x0020eeff,
slouken@0
  1796
		0x521c0000, 0x0040eeff, 0x5a1c0000, 0x0061eeff,
slouken@0
  1797
		0x621c0000, 0x0081eeff, 0x6a1c0000, 0x00a1eeff,
slouken@0
  1798
		0x731c0000, 0x00c2eeff, 0x7b1c0000, 0x00e2eeff,
slouken@0
  1799
		0x831c0000, 0x0000f6ff, 0x8b1c0000, 0x0020f6ff,
slouken@0
  1800
		0x941c0000, 0x0040f6ff, 0x9c1c0000, 0x0061f6ff,
slouken@0
  1801
		0xa41c0000, 0x0081f6ff, 0xac1c0000, 0x00a1f6ff,
slouken@0
  1802
		0xb41c0000, 0x00c2f6ff, 0xbd1c0000, 0x00e2f6ff,
slouken@0
  1803
		0xc51c0000, 0x0000ffff, 0xcd1c0000, 0x0020ffff,
slouken@0
  1804
		0xd51c0000, 0x0040ffff, 0xde1c0000, 0x0061ffff,
slouken@0
  1805
		0xe61c0000, 0x0081ffff, 0xee1c0000, 0x00a1ffff,
slouken@0
  1806
		0xf61c0000, 0x00c2ffff, 0xff1c0000, 0x00e2ffff
slouken@0
  1807
};
slouken@0
  1808
static void Blit_RGB565_BGRA8888(SDL_BlitInfo *info)
slouken@0
  1809
{
slouken@0
  1810
    Blit_RGB565_32(info, RGB565_BGRA8888_LUT);
slouken@0
  1811
}
slouken@0
  1812
slouken@0
  1813
/* Special optimized blit for RGB 8-8-8 --> RGB 3-3-2 */
slouken@0
  1814
#ifndef RGB888_RGB332
slouken@0
  1815
#define RGB888_RGB332(dst, src) { \
slouken@0
  1816
	dst = (((src)&0x00E00000)>>16)| \
slouken@0
  1817
	      (((src)&0x0000E000)>>11)| \
slouken@0
  1818
	      (((src)&0x000000C0)>>6); \
slouken@0
  1819
}
slouken@0
  1820
#endif
slouken@0
  1821
static void Blit_RGB888_index8_map(SDL_BlitInfo *info)
slouken@0
  1822
{
slouken@0
  1823
#ifndef USE_DUFFS_LOOP
slouken@0
  1824
	int c;
slouken@0
  1825
#endif
icculus@1162
  1826
	int Pixel;
slouken@0
  1827
	int width, height;
slouken@0
  1828
	Uint32 *src;
slouken@91
  1829
	const Uint8 *map;
slouken@91
  1830
	Uint8 *dst;
slouken@0
  1831
	int srcskip, dstskip;
slouken@0
  1832
slouken@0
  1833
	/* Set up some basic variables */
slouken@0
  1834
	width = info->d_width;
slouken@0
  1835
	height = info->d_height;
slouken@0
  1836
	src = (Uint32 *)info->s_pixels;
slouken@0
  1837
	srcskip = info->s_skip/4;
slouken@0
  1838
	dst = info->d_pixels;
slouken@0
  1839
	dstskip = info->d_skip;
slouken@0
  1840
	map = info->table;
slouken@0
  1841
slouken@0
  1842
#ifdef USE_DUFFS_LOOP
slouken@0
  1843
	while ( height-- ) {
slouken@0
  1844
		DUFFS_LOOP(
icculus@1162
  1845
			RGB888_RGB332(Pixel, *src);
icculus@1162
  1846
			*dst++ = map[Pixel];
slouken@0
  1847
			++src;
slouken@0
  1848
		, width);
slouken@0
  1849
		src += srcskip;
slouken@0
  1850
		dst += dstskip;
slouken@0
  1851
	}
slouken@0
  1852
#else
slouken@0
  1853
	while ( height-- ) {
slouken@0
  1854
		for ( c=width/4; c; --c ) {
slouken@0
  1855
			/* Pack RGB into 8bit pixel */
icculus@1162
  1856
			RGB888_RGB332(Pixel, *src);
icculus@1162
  1857
			*dst++ = map[Pixel];
slouken@0
  1858
			++src;
icculus@1162
  1859
			RGB888_RGB332(Pixel, *src);
icculus@1162
  1860
			*dst++ = map[Pixel];
slouken@0
  1861
			++src;
icculus@1162
  1862
			RGB888_RGB332(Pixel, *src);
icculus@1162
  1863
			*dst++ = map[Pixel];
slouken@0
  1864
			++src;
icculus@1162
  1865
			RGB888_RGB332(Pixel, *src);
icculus@1162
  1866
			*dst++ = map[Pixel];
slouken@0
  1867
			++src;
slouken@0
  1868
		}
slouken@91
  1869
		switch ( width & 3 ) {
slouken@0
  1870
			case 3:
icculus@1162
  1871
				RGB888_RGB332(Pixel, *src);
icculus@1162
  1872
				*dst++ = map[Pixel];
slouken@0
  1873
				++src;
slouken@0
  1874
			case 2:
icculus@1162
  1875
				RGB888_RGB332(Pixel, *src);
icculus@1162
  1876
				*dst++ = map[Pixel];
slouken@0
  1877
				++src;
slouken@0
  1878
			case 1:
icculus@1162
  1879
				RGB888_RGB332(Pixel, *src);
icculus@1162
  1880
				*dst++ = map[Pixel];
slouken@0
  1881
				++src;
slouken@0
  1882
		}
slouken@0
  1883
		src += srcskip;
slouken@0
  1884
		dst += dstskip;
slouken@0
  1885
	}
slouken@0
  1886
#endif /* USE_DUFFS_LOOP */
slouken@0
  1887
}
slouken@0
  1888
static void BlitNto1(SDL_BlitInfo *info)
slouken@0
  1889
{
slouken@0
  1890
#ifndef USE_DUFFS_LOOP
slouken@0
  1891
	int c;
slouken@0
  1892
#endif
slouken@0
  1893
	int width, height;
slouken@91
  1894
	Uint8 *src;
slouken@91
  1895
	const Uint8 *map;
slouken@91
  1896
	Uint8 *dst;
slouken@0
  1897
	int srcskip, dstskip;
slouken@0
  1898
	int srcbpp;
icculus@1162
  1899
	Uint32 Pixel;
slouken@0
  1900
	int  sR, sG, sB;
slouken@0
  1901
	SDL_PixelFormat *srcfmt;
slouken@0
  1902
slouken@0
  1903
	/* Set up some basic variables */
slouken@0
  1904
	width = info->d_width;
slouken@0
  1905
	height = info->d_height;
slouken@0
  1906
	src = info->s_pixels;
slouken@0
  1907
	srcskip = info->s_skip;
slouken@0
  1908
	dst = info->d_pixels;
slouken@0
  1909
	dstskip = info->d_skip;
slouken@0
  1910
	map = info->table;
slouken@0
  1911
	srcfmt = info->src;
slouken@0
  1912
	srcbpp = srcfmt->BytesPerPixel;
slouken@0
  1913
slouken@0
  1914
	if ( map == NULL ) {
slouken@0
  1915
		while ( height-- ) {
slouken@0
  1916
#ifdef USE_DUFFS_LOOP
slouken@0
  1917
			DUFFS_LOOP(
icculus@1162
  1918
				DISEMBLE_RGB(src, srcbpp, srcfmt, Pixel,
slouken@0
  1919
								sR, sG, sB);
slouken@0
  1920
				if ( 1 ) {
slouken@0
  1921
				  	/* Pack RGB into 8bit pixel */
slouken@0
  1922
				  	*dst = ((sR>>5)<<(3+2))|
slouken@0
  1923
					        ((sG>>5)<<(2)) |
slouken@0
  1924
					        ((sB>>6)<<(0)) ;
slouken@0
  1925
				}
slouken@0
  1926
				dst++;
slouken@0
  1927
				src += srcbpp;
slouken@0
  1928
			, width);
slouken@0
  1929
#else
slouken@0
  1930
			for ( c=width; c; --c ) {
icculus@1162
  1931
				DISEMBLE_RGB(src, srcbpp, srcfmt, Pixel,
slouken@0
  1932
								sR, sG, sB);
slouken@0
  1933
				if ( 1 ) {
slouken@0
  1934
				  	/* Pack RGB into 8bit pixel */
slouken@0
  1935
				  	*dst = ((sR>>5)<<(3+2))|
slouken@0
  1936
					        ((sG>>5)<<(2)) |
slouken@0
  1937
					        ((sB>>6)<<(0)) ;
slouken@0
  1938
				}
slouken@0
  1939
				dst++;
slouken@0
  1940
				src += srcbpp;
slouken@0
  1941
			}
slouken@0
  1942
#endif
slouken@0
  1943
			src += srcskip;
slouken@0
  1944
			dst += dstskip;
slouken@0
  1945
		}
slouken@0
  1946
	} else {
slouken@0
  1947
		while ( height-- ) {
slouken@0
  1948
#ifdef USE_DUFFS_LOOP
slouken@0
  1949
			DUFFS_LOOP(
icculus@1162
  1950
				DISEMBLE_RGB(src, srcbpp, srcfmt, Pixel,
slouken@0
  1951
								sR, sG, sB);
slouken@0
  1952
				if ( 1 ) {
slouken@0
  1953
				  	/* Pack RGB into 8bit pixel */
slouken@0
  1954
				  	*dst = map[((sR>>5)<<(3+2))|
slouken@0
  1955
						   ((sG>>5)<<(2))  |
slouken@0
  1956
						   ((sB>>6)<<(0))  ];
slouken@0
  1957
				}
slouken@0
  1958
				dst++;
slouken@0
  1959
				src += srcbpp;
slouken@0
  1960
			, width);
slouken@0
  1961
#else
slouken@0
  1962
			for ( c=width; c; --c ) {
icculus@1162
  1963
				DISEMBLE_RGB(src, srcbpp, srcfmt, Pixel,
slouken@0
  1964
								sR, sG, sB);
slouken@0
  1965
				if ( 1 ) {
slouken@0
  1966
				  	/* Pack RGB into 8bit pixel */
slouken@0
  1967
				  	*dst = map[((sR>>5)<<(3+2))|
slouken@0
  1968
						   ((sG>>5)<<(2))  |
slouken@0
  1969
						   ((sB>>6)<<(0))  ];
slouken@0
  1970
				}
slouken@0
  1971
				dst++;
slouken@0
  1972
				src += srcbpp;
slouken@0
  1973
			}
slouken@0
  1974
#endif /* USE_DUFFS_LOOP */
slouken@0
  1975
			src += srcskip;
slouken@0
  1976
			dst += dstskip;
slouken@0
  1977
		}
slouken@0
  1978
	}
slouken@0
  1979
}
icculus@1232
  1980
icculus@1232
  1981
/* blits 32 bit RGB<->RGBA with both surfaces having the same R,G,B fields */
icculus@1232
  1982
static void Blit4to4MaskAlpha(SDL_BlitInfo *info)
icculus@1232
  1983
{
icculus@1232
  1984
	int width = info->d_width;
icculus@1232
  1985
	int height = info->d_height;
icculus@1232
  1986
	Uint32 *src = (Uint32 *)info->s_pixels;
icculus@1232
  1987
	int srcskip = info->s_skip;
icculus@1232
  1988
	Uint32 *dst = (Uint32 *)info->d_pixels;
icculus@1232
  1989
	int dstskip = info->d_skip;
icculus@1232
  1990
	SDL_PixelFormat *srcfmt = info->src;
icculus@1232
  1991
	SDL_PixelFormat *dstfmt = info->dst;
icculus@1232
  1992
icculus@1232
  1993
	if (dstfmt->Amask) {
icculus@1232
  1994
		/* RGB->RGBA, SET_ALPHA */
icculus@1232
  1995
		Uint32 mask = (srcfmt->alpha >> dstfmt->Aloss) << dstfmt->Ashift;
icculus@1232
  1996
icculus@1232
  1997
		while ( height-- ) {
icculus@1232
  1998
			DUFFS_LOOP(
icculus@1232
  1999
			{
icculus@1232
  2000
				*dst = *src | mask;
icculus@1232
  2001
				++dst;
icculus@1232
  2002
				++src;
icculus@1232
  2003
			},
icculus@1232
  2004
			width);
icculus@1232
  2005
			src = (Uint32*)((Uint8*)src + srcskip);
icculus@1232
  2006
			dst = (Uint32*)((Uint8*)dst + dstskip);
icculus@1232
  2007
		}
icculus@1232
  2008
	} else {
icculus@1232
  2009
		/* RGBA->RGB, NO_ALPHA */
icculus@1232
  2010
		Uint32 mask = srcfmt->Rmask | srcfmt->Gmask | srcfmt->Bmask;
icculus@1232
  2011
icculus@1232
  2012
		while ( height-- ) {
icculus@1232
  2013
			DUFFS_LOOP(
icculus@1232
  2014
			{
icculus@1232
  2015
				*dst = *src & mask;
icculus@1232
  2016
				++dst;
icculus@1232
  2017
				++src;
icculus@1232
  2018
			},
icculus@1232
  2019
			width);
icculus@1232
  2020
			src = (Uint32*)((Uint8*)src + srcskip);
icculus@1232
  2021
			dst = (Uint32*)((Uint8*)dst + dstskip);
icculus@1232
  2022
		}
icculus@1232
  2023
	}
icculus@1232
  2024
}
icculus@1232
  2025
slouken@0
  2026
static void BlitNtoN(SDL_BlitInfo *info)
slouken@0
  2027
{
slouken@0
  2028
	int width = info->d_width;
slouken@0
  2029
	int height = info->d_height;
slouken@0
  2030
	Uint8 *src = info->s_pixels;
slouken@0
  2031
	int srcskip = info->s_skip;
slouken@0
  2032
	Uint8 *dst = info->d_pixels;
slouken@0
  2033
	int dstskip = info->d_skip;
slouken@0
  2034
	SDL_PixelFormat *srcfmt = info->src;
slouken@0
  2035
	int srcbpp = srcfmt->BytesPerPixel;
slouken@0
  2036
	SDL_PixelFormat *dstfmt = info->dst;
slouken@0
  2037
	int dstbpp = dstfmt->BytesPerPixel;
slouken@431
  2038
	unsigned alpha = dstfmt->Amask ? srcfmt->alpha : 0;
slouken@0
  2039
slouken@0
  2040
	while ( height-- ) {
slouken@0
  2041
		DUFFS_LOOP(
slouken@0
  2042
		{
icculus@1162
  2043
		        Uint32 Pixel;
slouken@0
  2044
			unsigned sR;
slouken@0
  2045
			unsigned sG;
slouken@0
  2046
			unsigned sB;
icculus@1162
  2047
			DISEMBLE_RGB(src, srcbpp, srcfmt, Pixel, sR, sG, sB);
slouken@0
  2048
			ASSEMBLE_RGBA(dst, dstbpp, dstfmt, sR, sG, sB, alpha);
slouken@0
  2049
			dst += dstbpp;
slouken@0
  2050
			src += srcbpp;
slouken@0
  2051
		},
slouken@0
  2052
		width);
slouken@0
  2053
		src += srcskip;
slouken@0
  2054
		dst += dstskip;
slouken@0
  2055
	}
slouken@0
  2056
}
slouken@0
  2057
slouken@0
  2058
static void BlitNtoNCopyAlpha(SDL_BlitInfo *info)
slouken@0
  2059
{
slouken@0
  2060
	int width = info->d_width;
slouken@0
  2061
	int height = info->d_height;
slouken@0
  2062
	Uint8 *src = info->s_pixels;
slouken@0
  2063
	int srcskip = info->s_skip;
slouken@0
  2064
	Uint8 *dst = info->d_pixels;
slouken@0
  2065
	int dstskip = info->d_skip;
slouken@0
  2066
	SDL_PixelFormat *srcfmt = info->src;
slouken@0
  2067
	int srcbpp = srcfmt->BytesPerPixel;
slouken@0
  2068
	SDL_PixelFormat *dstfmt = info->dst;
slouken@0
  2069
	int dstbpp = dstfmt->BytesPerPixel;
slouken@0
  2070
	int c;
slouken@0
  2071
slouken@0
  2072
	/* FIXME: should map alpha to [0..255] correctly! */
slouken@0
  2073
	while ( height-- ) {
slouken@0
  2074
		for ( c=width; c; --c ) {
icculus@1162
  2075
		        Uint32 Pixel;
slouken@0
  2076
			unsigned sR, sG, sB, sA;
icculus@1162
  2077
			DISEMBLE_RGBA(src, srcbpp, srcfmt, Pixel,
slouken@0
  2078
				      sR, sG, sB, sA);
slouken@0
  2079
			ASSEMBLE_RGBA(dst, dstbpp, dstfmt,
slouken@0
  2080
				      sR, sG, sB, sA);
slouken@0
  2081
			dst += dstbpp;
slouken@0
  2082
			src += srcbpp;
slouken@0
  2083
		}
slouken@0
  2084
		src += srcskip;
slouken@0
  2085
		dst += dstskip;
slouken@0
  2086
	}
slouken@0
  2087
}
slouken@0
  2088
slouken@0
  2089
static void BlitNto1Key(SDL_BlitInfo *info)
slouken@0
  2090
{
slouken@0
  2091
	int width = info->d_width;
slouken@0
  2092
	int height = info->d_height;
slouken@0
  2093
	Uint8 *src = info->s_pixels;
slouken@0
  2094
	int srcskip = info->s_skip;
slouken@0
  2095
	Uint8 *dst = info->d_pixels;
slouken@0
  2096
	int dstskip = info->d_skip;
slouken@0
  2097
	SDL_PixelFormat *srcfmt = info->src;
slouken@91
  2098
	const Uint8 *palmap = info->table;
slouken@0
  2099
	Uint32 ckey = srcfmt->colorkey;
slouken@0
  2100
	Uint32 rgbmask = ~srcfmt->Amask;
slouken@0
  2101
	int srcbpp;
icculus@1162
  2102
	Uint32 Pixel;
slouken@0
  2103
	Uint8  sR, sG, sB;
slouken@0
  2104
slouken@0
  2105
	/* Set up some basic variables */
slouken@0
  2106
	srcbpp = srcfmt->BytesPerPixel;
slouken@0
  2107
	ckey &= rgbmask;
slouken@0
  2108
slouken@0
  2109
	if ( palmap == NULL ) {
slouken@0
  2110
		while ( height-- ) {
slouken@0
  2111
			DUFFS_LOOP(
slouken@0
  2112
			{
icculus@1162
  2113
				DISEMBLE_RGB(src, srcbpp, srcfmt, Pixel,
slouken@0
  2114
								sR, sG, sB);
icculus@1162
  2115
				if ( (Pixel & rgbmask) != ckey ) {
slouken@0
  2116
				  	/* Pack RGB into 8bit pixel */
slouken@0
  2117
				  	*dst = ((sR>>5)<<(3+2))|
slouken@0
  2118
						((sG>>5)<<(2)) |
slouken@0
  2119
						((sB>>6)<<(0)) ;
slouken@0
  2120
				}
slouken@0
  2121
				dst++;
slouken@0
  2122
				src += srcbpp;
slouken@0
  2123
			},
slouken@0
  2124
			width);
slouken@0
  2125
			src += srcskip;
slouken@0
  2126
			dst += dstskip;
slouken@0
  2127
		}
slouken@0
  2128
	} else {
slouken@0
  2129
		while ( height-- ) {
slouken@0
  2130
			DUFFS_LOOP(
slouken@0
  2131
			{
icculus@1162
  2132
				DISEMBLE_RGB(src, srcbpp, srcfmt, Pixel,
slouken@0
  2133
								sR, sG, sB);
icculus@1162
  2134
				if ( (Pixel & rgbmask) != ckey ) {
slouken@0
  2135
				  	/* Pack RGB into 8bit pixel */
slouken@0
  2136
				  	*dst = palmap[((sR>>5)<<(3+2))|
slouken@0
  2137
							((sG>>5)<<(2))  |
slouken@0
  2138
							((sB>>6)<<(0))  ];
slouken@0
  2139
				}
slouken@0
  2140
				dst++;
slouken@0
  2141
				src += srcbpp;
slouken@0
  2142
			},
slouken@0
  2143
			width);
slouken@0
  2144
			src += srcskip;
slouken@0
  2145
			dst += dstskip;
slouken@0
  2146
		}
slouken@0
  2147
	}
slouken@0
  2148
}
slouken@0
  2149
slouken@0
  2150
static void Blit2to2Key(SDL_BlitInfo *info)
slouken@0
  2151
{
slouken@0
  2152
	int width = info->d_width;
slouken@0
  2153
	int height = info->d_height;
slouken@0
  2154
	Uint16 *srcp = (Uint16 *)info->s_pixels;
slouken@0
  2155
	int srcskip = info->s_skip;
slouken@0
  2156
	Uint16 *dstp = (Uint16 *)info->d_pixels;
slouken@0
  2157
	int dstskip = info->d_skip;
slouken@0
  2158
	Uint32 ckey = info->src->colorkey;
slouken@0
  2159
	Uint32 rgbmask = ~info->src->Amask;
slouken@0
  2160
slouken@0
  2161
	/* Set up some basic variables */
slouken@0
  2162
        srcskip /= 2;
slouken@0
  2163
        dstskip /= 2;
slouken@0
  2164
	ckey &= rgbmask;
slouken@0
  2165
slouken@0
  2166
	while ( height-- ) {
slouken@0
  2167
		DUFFS_LOOP(
slouken@0
  2168
		{
slouken@0
  2169
			if ( (*srcp & rgbmask) != ckey ) {
slouken@0
  2170
				*dstp = *srcp;
slouken@0
  2171
			}
slouken@0
  2172
			dstp++;
slouken@0
  2173
			srcp++;
slouken@0
  2174
		},
slouken@0
  2175
		width);
slouken@0
  2176
		srcp += srcskip;
slouken@0
  2177
		dstp += dstskip;
slouken@0
  2178
	}
slouken@0
  2179
}
slouken@0
  2180
slouken@0
  2181
static void BlitNtoNKey(SDL_BlitInfo *info)
slouken@0
  2182
{
slouken@0
  2183
	int width = info->d_width;
slouken@0
  2184
	int height = info->d_height;
slouken@0
  2185
	Uint8 *src = info->s_pixels;
slouken@0
  2186
	int srcskip = info->s_skip;
slouken@0
  2187
	Uint8 *dst = info->d_pixels;
slouken@0
  2188
	int dstskip = info->d_skip;
slouken@0
  2189
	Uint32 ckey = info->src->colorkey;
slouken@0
  2190
	SDL_PixelFormat *srcfmt = info->src;
slouken@0
  2191
	SDL_PixelFormat *dstfmt = info->dst;
slouken@0
  2192
	int srcbpp = srcfmt->BytesPerPixel;
slouken@0
  2193
	int dstbpp = dstfmt->BytesPerPixel;
slouken@431
  2194
	unsigned alpha = dstfmt->Amask ? srcfmt->alpha : 0;
slouken@0
  2195
slouken@0
  2196
	while ( height-- ) {
slouken@0
  2197
		DUFFS_LOOP(
slouken@0
  2198
		{
icculus@1162
  2199
		        Uint32 Pixel;
slouken@0
  2200
			unsigned sR;
slouken@0
  2201
			unsigned sG;
slouken@0
  2202
			unsigned sB;
icculus@1162
  2203
			RETRIEVE_RGB_PIXEL(src, srcbpp, Pixel);
icculus@1162
  2204
			if ( Pixel != ckey ) {
icculus@1162
  2205
			        RGB_FROM_PIXEL(Pixel, srcfmt, sR, sG, sB);
slouken@0
  2206
				ASSEMBLE_RGBA(dst, dstbpp, dstfmt,
slouken@0
  2207
					      sR, sG, sB, alpha);
slouken@0
  2208
			}
slouken@0
  2209
			dst += dstbpp;
slouken@0
  2210
			src += srcbpp;
slouken@0
  2211
		},
slouken@0
  2212
		width);
slouken@0
  2213
		src += srcskip;
slouken@0
  2214
		dst += dstskip;
slouken@0
  2215
	}
slouken@0
  2216
}
slouken@0
  2217
slouken@0
  2218
static void BlitNtoNKeyCopyAlpha(SDL_BlitInfo *info)
slouken@0
  2219
{
slouken@0
  2220
	int width = info->d_width;
slouken@0
  2221
	int height = info->d_height;
slouken@0
  2222
	Uint8 *src = info->s_pixels;
slouken@0
  2223
	int srcskip = info->s_skip;
slouken@0
  2224
	Uint8 *dst = info->d_pixels;
slouken@0
  2225
	int dstskip = info->d_skip;
slouken@0
  2226
	Uint32 ckey = info->src->colorkey;
slouken@0
  2227
	SDL_PixelFormat *srcfmt = info->src;
slouken@0
  2228
	SDL_PixelFormat *dstfmt = info->dst;
slouken@0
  2229
	Uint32 rgbmask = ~srcfmt->Amask;
slouken@0
  2230
slouken@0
  2231
	Uint8 srcbpp;
slouken@0
  2232
	Uint8 dstbpp;
icculus@1162
  2233
	Uint32 Pixel;
slouken@0
  2234
	Uint8  sR, sG, sB, sA;
slouken@0
  2235
slouken@0
  2236
	/* Set up some basic variables */
slouken@0
  2237
	srcbpp = srcfmt->BytesPerPixel;
slouken@0
  2238
	dstbpp = dstfmt->BytesPerPixel;
slouken@0
  2239
	ckey &= rgbmask;
slouken@0
  2240
slouken@0
  2241
	/* FIXME: should map alpha to [0..255] correctly! */
slouken@0
  2242
	while ( height-- ) {
slouken@0
  2243
		DUFFS_LOOP(
slouken@0
  2244
		{
icculus@1162
  2245
			DISEMBLE_RGBA(src, srcbpp, srcfmt, Pixel,
slouken@0
  2246
				      sR, sG, sB, sA);
icculus@1162
  2247
			if ( (Pixel & rgbmask) != ckey ) {
slouken@0
  2248
				  ASSEMBLE_RGBA(dst, dstbpp, dstfmt,
slouken@0
  2249
						sR, sG, sB, sA);
slouken@0
  2250
			}
slouken@0
  2251
			dst += dstbpp;
slouken@0
  2252
			src += srcbpp;
slouken@0
  2253
		},
slouken@0
  2254
		width);
slouken@0
  2255
		src += srcskip;
slouken@0
  2256
		dst += dstskip;
slouken@0
  2257
	}
slouken@0
  2258
}
slouken@0
  2259
slouken@0
  2260
/* Normal N to N optimized blitters */
slouken@0
  2261
struct blit_table {
slouken@0
  2262
	Uint32 srcR, srcG, srcB;
slouken@0
  2263
	int dstbpp;
slouken@0
  2264
	Uint32 dstR, dstG, dstB;
icculus@1047
  2265
	Uint32 blit_features;
slouken@0
  2266
	void *aux_data;
slouken@0
  2267
	SDL_loblit blitfunc;
icculus@1047
  2268
	enum { NO_ALPHA=1, SET_ALPHA=2, COPY_ALPHA=4 } alpha;
slouken@0
  2269
};
slouken@91
  2270
static const struct blit_table normal_blit_1[] = {
slouken@0
  2271
	/* Default for 8-bit RGB source, an invalid combination */
slouken@0
  2272
	{ 0,0,0, 0, 0,0,0, 0, NULL, NULL },
slouken@0
  2273
};
slouken@91
  2274
static const struct blit_table normal_blit_2[] = {
slouken@1361
  2275
#if SDL_HERMES_BLITTERS
slouken@0
  2276
    { 0x0000F800,0x000007E0,0x0000001F, 2, 0x0000001F,0x000007E0,0x0000F800,
slouken@0
  2277
      0, ConvertX86p16_16BGR565, ConvertX86, NO_ALPHA },
slouken@0
  2278
    { 0x0000F800,0x000007E0,0x0000001F, 2, 0x00007C00,0x000003E0,0x0000001F,
slouken@0
  2279
      0, ConvertX86p16_16RGB555, ConvertX86, NO_ALPHA },
slouken@0
  2280
    { 0x0000F800,0x000007E0,0x0000001F, 2, 0x0000001F,0x000003E0,0x00007C00,
slouken@0
  2281
      0, ConvertX86p16_16BGR555, ConvertX86, NO_ALPHA },
slouken@1361
  2282
#elif SDL_ALTIVEC_BLITTERS
icculus@1047
  2283
    /* has-altivec */
icculus@1047
  2284
    { 0x0000F800,0x000007E0,0x0000001F, 4, 0x00000000,0x00000000,0x00000000,
icculus@1047
  2285
      2, NULL, Blit_RGB565_32Altivec, NO_ALPHA | COPY_ALPHA | SET_ALPHA },
icculus@1139
  2286
    { 0x00007C00,0x000003E0,0x0000001F, 4, 0x00000000,0x00000000,0x00000000,
icculus@1139
  2287
      2, NULL, Blit_RGB555_32Altivec, NO_ALPHA | COPY_ALPHA | SET_ALPHA },
icculus@1047
  2288
#endif
slouken@0
  2289
    { 0x0000F800,0x000007E0,0x0000001F, 4, 0x00FF0000,0x0000FF00,0x000000FF,
slouken@0
  2290
      0, NULL, Blit_RGB565_ARGB8888, SET_ALPHA },
slouken@0
  2291
    { 0x0000F800,0x000007E0,0x0000001F, 4, 0x000000FF,0x0000FF00,0x00FF0000,
slouken@0
  2292
      0, NULL, Blit_RGB565_ABGR8888, SET_ALPHA },
slouken@0
  2293
    { 0x0000F800,0x000007E0,0x0000001F, 4, 0xFF000000,0x00FF0000,0x0000FF00,
slouken@0
  2294
      0, NULL, Blit_RGB565_RGBA8888, SET_ALPHA },
slouken@0
  2295
    { 0x0000F800,0x000007E0,0x0000001F, 4, 0x0000FF00,0x00FF0000,0xFF000000,
slouken@0
  2296
      0, NULL, Blit_RGB565_BGRA8888, SET_ALPHA },
slouken@0
  2297
slouken@0
  2298
    /* Default for 16-bit RGB source, used if no other blitter matches */
slouken@0
  2299
    { 0,0,0, 0, 0,0,0, 0, NULL, BlitNtoN, 0 }
slouken@0
  2300
};
slouken@91
  2301
static const struct blit_table normal_blit_3[] = {
slouken@0
  2302
	/* Default for 24-bit RGB source, never optimized */
slouken@0
  2303
    { 0,0,0, 0, 0,0,0, 0, NULL, BlitNtoN, 0 }
slouken@0
  2304
};
slouken@91
  2305
static const struct blit_table normal_blit_4[] = {
slouken@1361
  2306
#if SDL_HERMES_BLITTERS
slouken@0
  2307
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x0000F800,0x000007E0,0x0000001F,
slouken@739
  2308
      1, ConvertMMXpII32_16RGB565, ConvertMMX, NO_ALPHA },
slouken@0
  2309
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x0000F800,0x000007E0,0x0000001F,
slouken@0
  2310
      0, ConvertX86p32_16RGB565, ConvertX86, NO_ALPHA },
slouken@0
  2311
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x0000001F,0x000007E0,0x0000F800,
slouken@739
  2312
      1, ConvertMMXpII32_16BGR565, ConvertMMX, NO_ALPHA },
slouken@0
  2313
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x0000001F,0x000007E0,0x0000F800,
slouken@0
  2314
      0, ConvertX86p32_16BGR565, ConvertX86, NO_ALPHA },
slouken@0
  2315
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x00007C00,0x000003E0,0x0000001F,
slouken@739
  2316
      1, ConvertMMXpII32_16RGB555, ConvertMMX, NO_ALPHA },
slouken@0
  2317
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x00007C00,0x000003E0,0x0000001F,
slouken@0
  2318
      0, ConvertX86p32_16RGB555, ConvertX86, NO_ALPHA },
slouken@0
  2319
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x0000001F,0x000003E0,0x00007C00,
slouken@739
  2320
      1, ConvertMMXpII32_16BGR555, ConvertMMX, NO_ALPHA },
slouken@0
  2321
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x0000001F,0x000003E0,0x00007C00,
slouken@0
  2322
      0, ConvertX86p32_16BGR555, ConvertX86, NO_ALPHA },
slouken@0
  2323
    { 0x00FF0000,0x0000FF00,0x000000FF, 3, 0x00FF0000,0x0000FF00,0x000000FF,
slouken@0
  2324
      0, ConvertX86p32_24RGB888, ConvertX86, NO_ALPHA },
slouken@0
  2325
    { 0x00FF0000,0x0000FF00,0x000000FF, 3, 0x000000FF,0x0000FF00,0x00FF0000,
slouken@0
  2326
      0, ConvertX86p32_24BGR888, ConvertX86, NO_ALPHA },
slouken@0
  2327
    { 0x00FF0000,0x0000FF00,0x000000FF, 4, 0x000000FF,0x0000FF00,0x00FF0000,
slouken@0
  2328
      0, ConvertX86p32_32BGR888, ConvertX86, NO_ALPHA },
slouken@0
  2329
    { 0x00FF0000,0x0000FF00,0x000000FF, 4, 0xFF000000,0x00FF0000,0x0000FF00,
slouken@0
  2330
      0, ConvertX86p32_32RGBA888, ConvertX86, NO_ALPHA },
slouken@0
  2331
    { 0x00FF0000,0x0000FF00,0x000000FF, 4, 0x0000FF00,0x00FF0000,0xFF000000,
slouken@0
  2332
      0, ConvertX86p32_32BGRA888, ConvertX86, NO_ALPHA },
slouken@0
  2333
#else
slouken@1361
  2334
#if SDL_ALTIVEC_BLITTERS
icculus@1047
  2335
    /* has-altivec | dont-use-prefetch */
icculus@1047
  2336
    { 0x00000000,0x00000000,0x00000000, 4, 0x00000000,0x00000000,0x00000000,
icculus@1047
  2337
      6, NULL, ConvertAltivec32to32_noprefetch, NO_ALPHA | COPY_ALPHA | SET_ALPHA },
icculus@1047
  2338
    /* has-altivec */
icculus@1047
  2339
    { 0x00000000,0x00000000,0x00000000, 4, 0x00000000,0x00000000,0x00000000,
icculus@1047
  2340
      2, NULL, ConvertAltivec32to32_prefetch, NO_ALPHA | COPY_ALPHA | SET_ALPHA },
icculus@1047
  2341
    /* has-altivec */
icculus@1047
  2342
    { 0x00000000,0x00000000,0x00000000, 2, 0x0000F800,0x000007E0,0x0000001F,
icculus@1047
  2343
      2, NULL, Blit_RGB888_RGB565Altivec, NO_ALPHA },
icculus@1047
  2344
#endif
slouken@0
  2345
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x0000F800,0x000007E0,0x0000001F,
slouken@0
  2346
      0, NULL, Blit_RGB888_RGB565, NO_ALPHA },
slouken@0
  2347
    { 0x00FF0000,0x0000FF00,0x000000FF, 2, 0x00007C00,0x000003E0,0x0000001F,
slouken@0
  2348
      0, NULL, Blit_RGB888_RGB555, NO_ALPHA },
slouken@0
  2349
#endif
slouken@0
  2350
	/* Default for 32-bit RGB source, used if no other blitter matches */
slouken@0
  2351
	{ 0,0,0, 0, 0,0,0, 0, NULL, BlitNtoN, 0 }
slouken@0
  2352
};
slouken@91
  2353
static const struct blit_table *normal_blit[] = {
slouken@0
  2354
	normal_blit_1, normal_blit_2, normal_blit_3, normal_blit_4
slouken@0
  2355
};
slouken@0
  2356
icculus@1047
  2357
/* Mask matches table, or table entry is zero */
icculus@1047
  2358
#define MASKOK(x, y) (((x) == (y)) || ((y) == 0x00000000))
icculus@1047
  2359
slouken@0
  2360
SDL_loblit SDL_CalculateBlitN(SDL_Surface *surface, int blit_index)
slouken@0
  2361
{
slouken@0
  2362
	struct private_swaccel *sdata;
slouken@0
  2363
	SDL_PixelFormat *srcfmt;
slouken@0
  2364
	SDL_PixelFormat *dstfmt;
slouken@91
  2365
	const struct blit_table *table;
slouken@0
  2366
	int which;
slouken@0
  2367
	SDL_loblit blitfun;
slouken@0
  2368
slouken@0
  2369
	/* Set up data for choosing the blit */
slouken@0
  2370
	sdata = surface->map->sw_data;
slouken@0
  2371
	srcfmt = surface->format;
slouken@0
  2372
	dstfmt = surface->map->dst->format;
slouken@0
  2373
slouken@0
  2374
	if ( blit_index & 2 ) {
slouken@0
  2375
	        /* alpha or alpha+colorkey */
slouken@0
  2376
	        return SDL_CalculateAlphaBlit(surface, blit_index);
slouken@0
  2377
	}
slouken@0
  2378
slouken@0
  2379
	/* We don't support destinations less than 8-bits */
slouken@0
  2380
	if ( dstfmt->BitsPerPixel < 8 ) {
slouken@0
  2381
		return(NULL);
slouken@0
  2382
	}
slouken@0
  2383
	
slouken@0
  2384
	if(blit_index == 1) {
slouken@0
  2385
	    /* colorkey blit: Here we don't have too many options, mostly
slouken@0
  2386
	       because RLE is the preferred fast way to deal with this.
slouken@0
  2387
	       If a particular case turns out to be useful we'll add it. */
slouken@0
  2388
slouken@0
  2389
	    if(srcfmt->BytesPerPixel == 2
slouken@0
  2390
	       && surface->map->identity)
slouken@0
  2391
		return Blit2to2Key;
slouken@0
  2392
	    else if(dstfmt->BytesPerPixel == 1)
slouken@0
  2393
		return BlitNto1Key;
slouken@0
  2394
	    else {
slouken@1361
  2395
#if SDL_ALTIVEC_BLITTERS
icculus@1047
  2396
        if((srcfmt->BytesPerPixel == 4) && (dstfmt->BytesPerPixel == 4) && SDL_HasAltiVec()) {
icculus@1047
  2397
            return Blit32to32KeyAltivec;
icculus@1047
  2398
        } else
icculus@1047
  2399
#endif
icculus@1047
  2400
slouken@0
  2401
		if(srcfmt->Amask && dstfmt->Amask)
slouken@0
  2402
		    return BlitNtoNKeyCopyAlpha;
slouken@0
  2403
		else
slouken@0
  2404
		    return BlitNtoNKey;
slouken@0
  2405
	    }
slouken@0
  2406
	}
slouken@0
  2407
slouken@0
  2408
	blitfun = NULL;
slouken@0
  2409
	if ( dstfmt->BitsPerPixel == 8 ) {
slouken@0
  2410
		/* We assume 8-bit destinations are palettized */
slouken@0
  2411
		if ( (srcfmt->BytesPerPixel == 4) &&
slouken@0
  2412
		     (srcfmt->Rmask == 0x00FF0000) &&
slouken@0
  2413
		     (srcfmt->Gmask == 0x0000FF00) &&
slouken@0
  2414
		     (srcfmt->Bmask == 0x000000FF) ) {
slouken@0
  2415
			if ( surface->map->table ) {
slouken@0
  2416
				blitfun = Blit_RGB888_index8_map;
slouken@0
  2417
			} else {
slouken@1361
  2418
#if SDL_HERMES_BLITTERS
slouken@0
  2419
				sdata->aux_data = ConvertX86p32_8RGB332;
slouken@0
  2420
				blitfun = ConvertX86;
slouken@0
  2421
#else
slouken@0
  2422
				blitfun = Blit_RGB888_index8;
slouken@0
  2423
#endif
slouken@0
  2424
			}
slouken@0
  2425
		} else {
slouken@0
  2426
			blitfun = BlitNto1;
slouken@0
  2427
		}
slouken@0
  2428
	} else {
slouken@0
  2429
		/* Now the meat, choose the blitter we want */
icculus@1047
  2430
		int a_need = 0;
slouken@0
  2431
		if(dstfmt->Amask)
slouken@0
  2432
		    a_need = srcfmt->Amask ? COPY_ALPHA : SET_ALPHA;
slouken@0
  2433
		table = normal_blit[srcfmt->BytesPerPixel-1];
icculus@1047
  2434
		for ( which=0; table[which].dstbpp; ++which ) {
icculus@1047
  2435
			if ( MASKOK(srcfmt->Rmask, table[which].srcR) &&
icculus@1047
  2436
			    MASKOK(srcfmt->Gmask, table[which].srcG) &&
icculus@1047
  2437
			    MASKOK(srcfmt->Bmask, table[which].srcB) &&
icculus@1047
  2438
			    MASKOK(dstfmt->Rmask, table[which].dstR) &&
icculus@1047
  2439
			    MASKOK(dstfmt->Gmask, table[which].dstG) &&
icculus@1047
  2440
			    MASKOK(dstfmt->Bmask, table[which].dstB) &&
icculus@1047
  2441
			    dstfmt->BytesPerPixel == table[which].dstbpp &&
icculus@1047
  2442
			    (a_need & table[which].alpha) == a_need &&
icculus@1047
  2443
			    ((table[which].blit_features & GetBlitFeatures()) == table[which].blit_features) )
slouken@0
  2444
				break;
slouken@0
  2445
		}
slouken@0
  2446
		sdata->aux_data = table[which].aux_data;
slouken@0
  2447
		blitfun = table[which].blitfunc;
icculus@1232
  2448
icculus@1232
  2449
		if(blitfun == BlitNtoN) {  /* default C fallback catch-all. Slow! */
icculus@1232
  2450
			/* Fastpath C fallback: 32bit RGB<->RGBA blit with matching RGB */
icculus@1232
  2451
			if ( srcfmt->BytesPerPixel == 4 && dstfmt->BytesPerPixel == 4 &&
icculus@1232
  2452
			     srcfmt->Rmask == dstfmt->Rmask &&
icculus@1232
  2453
			     srcfmt->Gmask == dstfmt->Gmask &&
icculus@1232
  2454
			     srcfmt->Bmask == dstfmt->Bmask ) {
icculus@1232
  2455
				blitfun = Blit4to4MaskAlpha;
icculus@1232
  2456
			} else if ( a_need == COPY_ALPHA ) {
icculus@1232
  2457
			    blitfun = BlitNtoNCopyAlpha;
icculus@1232
  2458
			}
icculus@1232
  2459
		}
slouken@0
  2460
	}
slouken@0
  2461
slouken@0
  2462
#ifdef DEBUG_ASM
slouken@1361
  2463
#if SDL_HERMES_BLITTERS
slouken@0
  2464
	if ( blitfun == ConvertMMX )
slouken@0
  2465
		fprintf(stderr, "Using mmx blit\n");
slouken@0
  2466
	else
slouken@0
  2467
	if ( blitfun == ConvertX86 )
slouken@0
  2468
		fprintf(stderr, "Using asm blit\n");
slouken@0
  2469
	else
slouken@0
  2470
#endif
icculus@1224
  2471
	if ( (blitfun == BlitNtoN) || (blitfun == BlitNto1) )
slouken@0
  2472
		fprintf(stderr, "Using C blit\n");
slouken@0
  2473
	else
slouken@0
  2474
		fprintf(stderr, "Using optimized C blit\n");
slouken@0
  2475
#endif /* DEBUG_ASM */
slouken@0
  2476
slouken@0
  2477
	return(blitfun);
slouken@0
  2478
}