Repository navigation
Expand file tree
/
Copy pathZoom.cpp
More file actions
935 lines (806 loc) · 23 KB
/
Copy pathZoom.cpp
File metadata and controls
935 lines (806 loc) · 23 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
/*
* Copyright 2010-2016 OpenXcom Developers.
*
* This file is part of OpenXcom.
*
* OpenXcom is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* OpenXcom is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with OpenXcom. If not, see <http://www.gnu.org/licenses/>.
*/
#include "Zoom.h"
#include "Surface.h"
#include "Logger.h"
#include "Options.h"
#include "Screen.h"
#include "OpenGL.h"
// Scale2X
#include "Scalers/scalebit.h"
// HQX
#include "Scalers/hqx.h"
// xBRZ
#include "Scalers/xbrz.h"
#if (_MSC_VER >= 1400) || (defined(__MINGW32__) && defined(__SSE2__))
#ifndef __SSE2__
#define __SSE2__ true
#endif
// probably Visual Studio (or Intel C++ which should also work)
#include <intrin.h>
#endif
#ifdef __GNUC__
#if (__i386__ || __x86_64__)
#include <cpuid.h>
#endif
#endif
#ifdef __SSE2__
#include <emmintrin.h> // for SSE2 intrinsics; see http://msdn.microsoft.com/en-us/library/has3d153%28v=vs.71%29.aspx
#endif
namespace OpenXcom
{
/**
* Optimized 8-bit zoomer for resizing by a factor of 2. Doesn't flip.
* Used internally by _zoomSurfaceY() below.
* source and dest. widths must be multiples of 8 bytes for 64-bit access
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @return 0 for success or -1 for error.
*/
/*
static int zoomSurface2X_64bit(SDL_Surface *src, SDL_Surface *dst)
{
Uint64 dataSrc;
Uint64 dataDst;
Uint8 *pixelSrc = (Uint8*)src->pixels;
Uint8 *pixelDstRow = (Uint8*)dst->pixels;
int sx, sy;
static bool proclaimed = false;
if (!proclaimed)
{
proclaimed = true;
Log(LOG_INFO) << "Using somewhat fast 2X zoom routine.";
}
for (sy = 0; sy < src->h; ++sy, pixelDstRow += dst->pitch*2)
{
Uint64 *pixelDst = (Uint64*)pixelDstRow;
Uint64 *pixelDst2 = (Uint64*)(pixelDstRow + dst->pitch);
for (sx = 0; sx < src->w; sx += 8, pixelSrc += 8)
{
dataSrc = *((Uint64*) pixelSrc);
// boo
(void)SDL_SwapLE64(dataSrc);
*/
/* expanded form of data shift:
dataDst = (dataSrc & 0xFF) | ((dataSrc & 0xFF) << 8) |
((dataSrc & 0xFF00 ) << 8) | ((dataSrc & 0xFF00)) << 16) |
((dataSrc & 0xFF0000) << 16) | ((dataSrc & 0xFF0000) << 24) |
((dataSrc & 0xFF000000) << 24) | ((dataSrc & 0xFF000000) << 32);
*/
// compact form, combining terms with equal multipliers (shifts)
/*
dataDst = (dataSrc & 0xFF) | ((dataSrc & 0xFFFF) << 8) |
((dataSrc & 0xFFFF00) << 16) |
((dataSrc & 0xFFFF0000) << 24) |
((dataSrc & 0xFF000000) << 32);
*pixelDst = dataDst;
*pixelDst2 = dataDst;
pixelDst++; // forward 8 bytes!
pixelDst2++;
dataSrc >>= 32;
dataDst = (dataSrc & 0xFF) | ((dataSrc & 0xFFFF) << 8) |
((dataSrc & 0xFFFF00) << 16) |
((dataSrc & 0xFFFF0000) << 24) |
((dataSrc & 0xFF000000) << 32);
*pixelDst = dataDst;
*pixelDst2 = dataDst;
pixelDst++; // 8 bytes again
pixelDst2++;
}
}
return 0;
}
*/
#if defined(__WORDSIZE) && (__WORDSIZE == 64) || defined(SIZE_MAX) && (SIZE_MAX > 0xFFFFFFFF)
#else
/**
* Optimized 8-bit zoomer for resizing by a factor of 2. Doesn't flip.
* 32-bit version for sad old x86 chips which run out of registers
* with the 64-bit version.
* Used internally by _zoomSurfaceY() below.
* source and dest. widths must be multiples of 4 bytes for 32-bit access
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @return 0 for success or -1 for error.
*/
/*
static int zoomSurface2X_32bit(SDL_Surface *src, SDL_Surface *dst)
{
Uint32 dataSrc;
Uint32 dataDst;
Uint8 *pixelSrc = (Uint8*)src->pixels;
Uint8 *pixelDstRow = (Uint8*)dst->pixels;
int sx, sy;
static bool proclaimed = false;
if (!proclaimed)
{
proclaimed = true;
Log(LOG_INFO) << "Using 32-bit 2X zoom routine.";
}
for (sy = 0; sy < src->h; ++sy, pixelDstRow += dst->pitch*2)
{
Uint32 *pixelDst = (Uint32*)pixelDstRow;
Uint32 *pixelDst2 = (Uint32*)(pixelDstRow + dst->pitch);
for (sx = 0; sx < src->w; sx += 4, pixelSrc += 4)
{
dataSrc = *((Uint32*) pixelSrc);
// boo
dataSrc = SDL_SwapLE32(dataSrc);
dataDst = SDL_SwapLE32( (dataSrc & 0xFF) | ((dataSrc & 0xFFFF) << 8) |
((dataSrc & 0xFF00) << 16) );
*pixelDst = dataDst;
*pixelDst2 = dataDst;
pixelDst++; // forward 4 bytes!
pixelDst2++;
dataSrc >>= 16;
dataDst = SDL_SwapLE32( (dataSrc & 0xFF) | ((dataSrc & 0xFFFF) << 8) |
((dataSrc & 0xFF00) << 16) );
*pixelDst = dataDst;
*pixelDst2 = dataDst;
pixelDst++; // forward 4 bytes!
pixelDst2++;
}
}
return 0;
}
*/
#endif
/**
* Optimized 8-bit zoomer for resizing by a factor of 4. Doesn't flip.
* Used internally by _zoomSurfaceY() below.
* source and dest. widths must be multiples of 8 bytes for 64-bit access
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @return 0 for success or -1 for error.
*/
/*
static int zoomSurface4X_64bit(SDL_Surface *src, SDL_Surface *dst)
{
Uint64 dataSrc;
Uint64 dataDst;
Uint8 *pixelSrc = (Uint8*)src->pixels;
Uint8 *pixelDstRow = (Uint8*)dst->pixels;
int sx, sy;
static bool proclaimed = false;
if (!proclaimed)
{
proclaimed = true;
Log(LOG_INFO) << "Using modestly fast 4X zoom routine.";
}
for (sy = 0; sy < src->h; ++sy, pixelDstRow += dst->pitch*4)
{
Uint8 *pixelDst = pixelDstRow;
for (sx = 0; sx < src->w; sx += 8, pixelSrc += 8)
{
dataSrc = *((Uint64*) pixelSrc);
// boo
(void)SDL_SwapLE64(dataSrc);
*/
/* expanded form of data shift:
dataDst = (dataSrc & 0xFF) | ((dataSrc & 0xFF) << 8) |
((dataSrc & 0xFF) << 16 | ((datasrc & 0xFF) << 24) |
((dataSrc & 0xFF00 ) << 24) | ((dataSrc & 0xFF00) << 32) |
((dataSrc & 0xFF00 ) << 40) | ((dataSrc & 0xFF00) << 48) ;
*/
/*
for (int i = 0; i < 4; ++i)
{
// compact form, combining terms with equal multipliers (shifts)
dataDst = (dataSrc & 0xFF) | ((dataSrc & 0xFF) << 8) |
((dataSrc & 0xFF) << 16) |
((dataSrc & 0xFFFF ) << 24) | ((dataSrc & 0xFF00) << 32) |
((dataSrc & 0xFF00 ) << 40) | ((dataSrc & 0xFF00) << 48) ;
*((Uint64*)pixelDst) = dataDst;
*((Uint64*)(pixelDst + dst->pitch)) = dataDst;
*((Uint64*)(pixelDst + dst->pitch*2)) = dataDst;
*((Uint64*)(pixelDst + dst->pitch*3)) = dataDst;
pixelDst+=8; // forward 8 bytes!
dataSrc >>= 16;
}
}
}
return 0;
}
*/
#if defined(__WORDSIZE) && (__WORDSIZE == 64) || defined(SIZE_MAX) && (SIZE_MAX > 0xFFFFFFFF)
#else
/**
* Optimized 8-bit zoomer for resizing by a factor of 4. Doesn't flip.
* 32-bit version.
* Used internally by _zoomSurfaceY() below.
* source and dest. widths must be multiples of 4 bytes for 32-bit access
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @return 0 for success or -1 for error.
*/
/*
static int zoomSurface4X_32bit(SDL_Surface *src, SDL_Surface *dst)
{
Uint32 dataSrc;
Uint32 dataDst;
Uint8 *pixelSrc = (Uint8*)src->pixels;
Uint8 *pixelDstRow = (Uint8*)dst->pixels;
int sx, sy;
static bool proclaimed = false;
if (!proclaimed)
{
proclaimed = true;
Log(LOG_INFO) << "Using 32-bit 4X zoom routine.";
}
for (sy = 0; sy < src->h; ++sy, pixelDstRow += dst->pitch*4)
{
Uint32 *pixelDst = (Uint32*)pixelDstRow;
Uint32 *pixelDst2 = (Uint32*)(pixelDstRow + dst->pitch);
Uint32 *pixelDst3 = (Uint32*)(pixelDstRow + 2*dst->pitch);
Uint32 *pixelDst4 = (Uint32*)(pixelDstRow + 3*dst->pitch);
for (sx = 0; sx < src->w; sx += 4, pixelSrc += 4)
{
dataSrc = *((Uint32*) pixelSrc);
// boo
dataSrc = SDL_SwapLE32(dataSrc);
for (int i = 0; i < 4; ++i)
{
dataDst = SDL_SwapLE32( (dataSrc & 0xFF) | ((dataSrc & 0xFF) << 8) |
((dataSrc & 0xFF) << 16) | ((dataSrc & 0xFF ) << 24) );
*pixelDst = dataDst;
*pixelDst2 = dataDst;
*pixelDst3 = dataDst;
*pixelDst4 = dataDst;
pixelDst++; // forward 4 bytes!
pixelDst2++;
pixelDst3++;
pixelDst4++;
dataSrc >>= 8;
}
}
}
return 0;
}
*/
#endif
/**
* Optimized 8-bit zoomer for resizing by a factor of 4. Doesn't flip.
* 32-bit version.
* Used internally by _zoomSurfaceY() below.
* source and dest. widths must be multiples of 4 bytes for 32-bit access
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @return 0 for success or -1 for error.
*/
/*
static int zoomSurface2X_XAxis_32bit(SDL_Surface *src, SDL_Surface *dst)
{
Uint32 dataSrc;
Uint32 dataDst;
Uint8 *pixelSrc;
Uint8 *pixelDstRow = (Uint8*)dst->pixels;
Uint8 *pixelSrcRow = (Uint8*)src->pixels;
int sx;
int dsty;
static bool proclaimed = false;
static Uint32 *say = 0;
Uint32 *csay;
int csy;
if (!proclaimed)
{
proclaimed = true;
Log(LOG_INFO) << "Using mediocre scaling routine due to screen height.";
}
if ((say = (Uint32 *) realloc(say, (dst->h + 1) * sizeof(Uint32))) == NULL) {
say = 0;
return (-1);
}
csy = 0;
csay = say;
for (int y = 0; y < dst->h; y++) {
csy += src->h;
*csay = 0;
while (csy >= dst->h) {
csy -= dst->h;
(*csay)++;
}
(*csay) *= src->pitch;
csay++;
}
for (dsty = 0; dsty < dst->h; ++dsty, pixelDstRow += dst->pitch)
{
if (!say[dsty]) continue;
Uint32 *pixelDst = (Uint32*)pixelDstRow;
pixelSrc = pixelSrcRow;
pixelSrcRow += say[dsty];
for (sx = 0; sx < src->w; sx += 4, pixelSrc += 4)
{
dataSrc = *((Uint32*) pixelSrc);
// boo
dataSrc = SDL_SwapLE32(dataSrc);
for (int i = 0; i < 2; ++i)
{
dataDst = SDL_SwapLE32( (dataSrc & 0xFF) | ((dataSrc & 0xFFFF) << 8) |
((dataSrc & 0xFF00) << 16) );
int j = 0;
do
{
if (dsty + j >= dst->h) break;
*(pixelDst + (dst->pitch/sizeof(Uint32))*j) = dataDst;
} while (say[dsty + ++j] == 0); // fill in all relevant rows
dataSrc >>= 16;
pixelDst++; // forward 4 bytes!
}
}
}
return 0;
}
*/
/**
* Optimized 8-bit zoomer for resizing by a factor of 2. Doesn't flip.
* 32-bit version.
* Used internally by _zoomSurfaceY() below.
* source and dest. widths must be multiples of 4 bytes for 32-bit access
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @return 0 for success or -1 for error.
*/
/*
static int zoomSurface4X_XAxis_32bit(SDL_Surface *src, SDL_Surface *dst)
{
Uint32 dataSrc;
Uint32 dataDst;
Uint8 *pixelSrc;
Uint8 *pixelDstRow = (Uint8*)dst->pixels;
Uint8 *pixelSrcRow = (Uint8*)src->pixels;
int sx;
int dsty;
static bool proclaimed = false;
static Uint32 *say = 0;
Uint32 *csay;
int csy;
if (!proclaimed)
{
proclaimed = true;
Log(LOG_INFO) << "Using mediocre scaling routine due to screen height.";
}
if ((say = (Uint32 *) realloc(say, (dst->h + 1) * sizeof(Uint32))) == NULL) {
say = 0;
return (-1);
}
csy = 0;
csay = say;
for (int y = 0; y < dst->h; y++) {
csy += src->h;
*csay = 0;
while (csy >= dst->h) {
csy -= dst->h;
(*csay)++;
}
(*csay) *= src->pitch;
csay++;
}
for (dsty = 0; dsty < dst->h; ++dsty, pixelDstRow += dst->pitch)
{
if (!say[dsty]) continue;
Uint32 *pixelDst = (Uint32*)pixelDstRow;
pixelSrc = pixelSrcRow;
pixelSrcRow += say[dsty];
for (sx = 0; sx < src->w; sx += 4, pixelSrc += 4)
{
dataSrc = *((Uint32*) pixelSrc);
// boo
dataSrc = SDL_SwapLE32(dataSrc);
for (int i = 0; i < 4; ++i)
{
dataDst = SDL_SwapLE32( (dataSrc & 0xFF) | ((dataSrc & 0xFF) << 8) |
((dataSrc & 0xFF) << 16) | ((dataSrc & 0xFF ) << 24) );
int j = 0;
do
{
if (dsty + j >= dst->h) break;
*(pixelDst + (dst->pitch/sizeof(Uint32))*j) = dataDst;
} while (say[dsty + ++j] == 0); // fill in all relevant rows
dataSrc >>= 8;
pixelDst++; // forward 4 bytes!
}
}
}
return 0;
}
*/
#ifdef __SSE2__
/**
* Optimized 8-bit zoomer for resizing by a factor of 4. Doesn't flip.
* Used internally by _zoomSurfaceY() below.
* This is an SSE2 version written with Intel intrinsics.
* source and dest. widths must be multiples of 16 bytes for 128-bit access
* and it would help if they were aligned properly... :(
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @return 0 for success or -1 for error.
*/
/*
static int zoomSurface4X_SSE2(SDL_Surface *src, SDL_Surface *dst)
{
__m128i dataSrc;
__m128i dataDst;
Uint8 *pixelSrc = (Uint8*)src->pixels;
Uint8 *pixelDstRow = (Uint8*)dst->pixels;
int sx, sy;
static bool proclaimed = false;
if (!proclaimed)
{
proclaimed = true;
Log(LOG_INFO) << "Using SSE2 4X zoom routine.";
}
for (sy = 0; sy < src->h; ++sy, pixelDstRow += dst->pitch*4)
{
__m128i *pixelDst = (__m128i*)pixelDstRow;
__m128i *pixelDst2 = (__m128i*)((Uint8*)pixelDstRow + dst->pitch);
__m128i *pixelDst3 = (__m128i*)((Uint8*)pixelDstRow + dst->pitch*2);
__m128i *pixelDst4 = (__m128i*)((Uint8*)pixelDstRow + dst->pitch*3);
for (sx = 0; sx < src->w; sx += 16, pixelSrc += 16)
{
dataSrc = *((__m128i*) pixelSrc);
__m128i halfDone = _mm_unpacklo_epi8(dataSrc, dataSrc);
dataDst = _mm_unpacklo_epi8(halfDone, halfDone);
*/
/* #define WRITE_DST if ((char*)pixelDst4 + 128 > (char*)dst->pixels+(dst->w*dst->pitch)) { Log(LOG_ERROR) << "HELL"; exit(0); } \ */
#define WRITE_DST *(pixelDst++) = dataDst; \
*(pixelDst2++) = dataDst; \
*(pixelDst3++) = dataDst; \
*(pixelDst4++) = dataDst; \
/*
WRITE_DST;
dataDst = _mm_unpackhi_epi8(halfDone, halfDone);
WRITE_DST;
halfDone = _mm_unpackhi_epi8(dataSrc, dataSrc);
dataDst = _mm_unpacklo_epi8(halfDone, halfDone);
WRITE_DST;
dataDst = _mm_unpackhi_epi8(halfDone, halfDone);
WRITE_DST;
}
}
return 0;
}
*/
/**
* Optimized 8-bit zoomer for resizing by a factor of 2. Doesn't flip.
* Used internally by _zoomSurfaceY() below.
* This is an SSE2 version written with Intel intrinsics.
* source and dest. widths must be multiples of 16 bytes for 128-bit access
* and it would help if they were aligned properly... :(
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @return 0 for success or -1 for error.
*/
/*
static int zoomSurface2X_SSE2(SDL_Surface *src, SDL_Surface *dst)
{
__m128i dataSrc;
__m128i dataDst;
Uint8 *pixelSrc = (Uint8*)src->pixels;
Uint8 *pixelDstRow = (Uint8*)dst->pixels;
int sx, sy;
static bool proclaimed = false;
if (!proclaimed)
{
proclaimed = true;
Log(LOG_INFO) << "Using SSE2 2X zoom routine.";
}
for (sy = 0; sy < src->h; ++sy, pixelDstRow += dst->pitch*2)
{
__m128i *pixelDst = (__m128i*)pixelDstRow;
__m128i *pixelDst2 = (__m128i*)((Uint8*)pixelDstRow + dst->pitch);
for (sx = 0; sx < src->w; sx += 16, pixelSrc += 16)
{
dataSrc = *((__m128i*) pixelSrc);
dataDst = _mm_unpacklo_epi8(dataSrc, dataSrc);
#undef WRITE_DST
#define WRITE_DST *(pixelDst++) = dataDst; \
*(pixelDst2++) = dataDst; \
WRITE_DST;
dataDst = _mm_unpackhi_epi8(dataSrc, dataSrc);
WRITE_DST;
}
}
return 0;
}
*/
/**
* Checks the SSE2 feature bit returned by the CPUID instruction
* @return Does the CPU support SSE2?
*/
bool Zoom::haveSSE2()
{
#ifdef __GNUC__
unsigned int CPUInfo[4] = {0, 0, 0, 0};
__get_cpuid(1, CPUInfo, CPUInfo+1, CPUInfo+2, CPUInfo+3);
#elif _WIN32
int CPUInfo[4];
__cpuid(CPUInfo, 1);
#else
unsigned int CPUInfo[4] = {0, 0, 0, 0};
#endif
return (CPUInfo[3] & 0x04000000) ? true : false;
}
#endif
/**
* Wrapper around various software and OpenGL screen buffer pushing functions which zoom.
* Basically called just from Screen::flip()
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @param topBlackBand Size of top black band in pixels (letterboxing).
* @param bottomBlackBand Size of bottom black band in pixels (letterboxing).
* @param leftBlackBand Size of left black band in pixels (letterboxing).
* @param rightBlackBand Size of right black band in pixels (letterboxing).
* @param glOut OpenGL output.
*/
void Zoom::flipWithZoom(SDL_Surface *src, SDL_Surface *dst, int topBlackBand, int bottomBlackBand, int leftBlackBand, int rightBlackBand, OpenGL *glOut)
{
int dstWidth = dst->w - leftBlackBand - rightBlackBand;
int dstHeight = dst->h - topBlackBand - bottomBlackBand;
if (Screen::useOpenGL())
{
#ifndef __NO_OPENGL
if (glOut->buffer_surface)
{
SDL_BlitSurface(src, 0, glOut->buffer_surface->getSurface(), 0); // TODO; this is less than ideal...
glOut->refresh(glOut->linear, glOut->iwidth, glOut->iheight, dst->w, dst->h, topBlackBand, bottomBlackBand, leftBlackBand, rightBlackBand);
SDL_GL_SwapBuffers();
}
#endif
}
else if (topBlackBand <= 0 && bottomBlackBand <= 0 && leftBlackBand <= 0 && rightBlackBand <= 0)
{
_zoomSurfaceY(src, dst, 0, 0);
}
else if (dstWidth == src->w && dstHeight == src->h)
{
SDL_Rect dstrect = {(Sint16)leftBlackBand, (Sint16)topBlackBand, (Uint16)src->w, (Uint16)src->h};
SDL_BlitSurface(src, NULL, dst, &dstrect);
}
else
{
SDL_Surface *tmp = SDL_CreateRGBSurface(dst->flags, dstWidth, dstHeight, dst->format->BitsPerPixel, 0, 0, 0, 0);
_zoomSurfaceY(src, tmp, 0, 0);
if (src->format->palette != NULL)
{
SDL_SetPalette(tmp, SDL_LOGPAL|SDL_PHYSPAL, src->format->palette->colors, 0, src->format->palette->ncolors);
}
SDL_Rect dstrect = {(Sint16)leftBlackBand, (Sint16)topBlackBand, (Uint16)tmp->w, (Uint16)tmp->h};
SDL_BlitSurface(tmp, NULL, dst, &dstrect);
SDL_FreeSurface(tmp);
}
}
/**
* Internal 8-bit Zoomer without smoothing.
* Source code originally from SDL_gfx (LGPL) with permission by author.
*
* Zooms 8bit palette/Y 'src' surface to 'dst' surface.
* Assumes src and dst surfaces are of 8-bit depth.
* Assumes dst surface was allocated with the correct dimensions.
*
* @param src The surface to zoom (input).
* @param dst The zoomed surface (output).
* @param flipx Flag indicating if the image should be horizontally flipped.
* @param flipy Flag indicating if the image should be vertically flipped.
* @return 0 for success or -1 for error.
*/
int Zoom::_zoomSurfaceY(SDL_Surface * src, SDL_Surface * dst, int flipx, int flipy)
{
int x, y;
static Uint32 *sax, *say;
Uint32 *csax, *csay;
int csx, csy;
Uint8 *sp, *dp, *csp;
int dgap;
static bool proclaimed = false;
if (Screen::use32bitScaler())
{
if (Options::useXBRZFilter)
{
// check the resolution to see which scale we need
for (size_t factor = 2; factor <= 6; factor++)
{
if (dst->w == src->w * (int)factor && dst->h == src->h * (int)factor)
{
xbrz::scale(factor, (uint32_t*)src->pixels, (uint32_t*)dst->pixels, src->w, src->h, xbrz::RGB);
return 0;
}
}
}
if (Options::useHQXFilter)
{
static bool initDone = false;
if (!initDone)
{
hqxInit();
initDone = true;
}
// HQX_API void HQX_CALLCONV hq2x_32_rb( uint32_t * src, uint32_t src_rowBytes, uint32_t * dest, uint32_t dest_rowBytes, int width, int height );
if (dst->w == src->w * 2 && dst->h == src->h * 2)
{
hq2x_32_rb((uint32_t*)src->pixels, src->pitch, (uint32_t*)dst->pixels, dst->pitch, src->w, src->h);
return 0;
}
if (dst->w == src->w * 3 && dst->h == src->h * 3)
{
hq3x_32_rb((uint32_t*)src->pixels, src->pitch, (uint32_t*)dst->pixels, dst->pitch, src->w, src->h);
return 0;
}
if (dst->w == src->w * 4 && dst->h == src->h * 4)
{
hq4x_32_rb((uint32_t*)src->pixels, src->pitch, (uint32_t*)dst->pixels, dst->pitch, src->w, src->h);
return 0;
}
}
}
if (Options::useScaleFilter)
{
// check the resolution to see which of scale2x, scale3x, etc. we need
for (size_t factor = 2; factor <= 4; factor++)
{
if (dst->w == src->w * (int)factor && dst->h == src->h * (int)factor && !scale_precondition(factor, src->format->BytesPerPixel, src->w, src->h))
{
scale(factor, dst->pixels, dst->pitch, src->pixels, src->pitch, src->format->BytesPerPixel, src->w, src->h);
return 0;
}
}
}
// if we're scaling by a factor of 2 or 4, try to use a more efficient function
/*
if (src->format->BytesPerPixel == 1 && dst->format->BytesPerPixel == 1)
{
#ifdef __SSE2__
static bool _haveSSE2 = haveSSE2();
if (_haveSSE2 &&
!((ptrdiff_t)src->pixels % 16) &&
!((ptrdiff_t)dst->pixels % 16)) // alignment check
{
if (dst->w == src->w * 2 && dst->h == src->h * 2) return zoomSurface2X_SSE2(src, dst);
else if (dst->w == src->w * 4 && dst->h == src->h * 4) return zoomSurface4X_SSE2(src, dst);
} else
{
static bool complained = false;
if (!complained)
{
complained = true;
Log(LOG_ERROR) << "Misaligned surface buffers.";
}
}
#endif
// __WORDSIZE is defined on Linux, SIZE_MAX on Windows
#if defined(__WORDSIZE) && (__WORDSIZE == 64) || defined(SIZE_MAX) && (SIZE_MAX > 0xFFFFFFFF)
if (dst->w == src->w * 2 && dst->h == src->h * 2) return zoomSurface2X_64bit(src, dst);
else if (dst->w == src->w * 4 && dst->h == src->h * 4) return zoomSurface4X_64bit(src, dst);
#else
if (sizeof(void *) == 8)
{
if (dst->w == src->w * 2 && dst->h == src->h * 2) return zoomSurface2X_64bit(src, dst);
else if (dst->w == src->w * 4 && dst->h == src->h * 4) return zoomSurface4X_64bit(src, dst);
}
else
{
if (dst->w == src->w * 2 && dst->h == src->h * 2) return zoomSurface2X_32bit(src, dst);
else if (dst->w == src->w * 4 && dst->h == src->h * 4) return zoomSurface4X_32bit(src, dst);
}
#endif
// maybe X is scaled by 2 or 4 but not Y?
if (dst->w == src->w * 4) return zoomSurface4X_XAxis_32bit(src, dst);
else if (dst->w == src->w * 2) return zoomSurface2X_XAxis_32bit(src, dst);
}
*/
if (!proclaimed)
{
Log(LOG_INFO) << "Using software scaling routine. For best results, try an OpenGL filter.";
proclaimed = true;
}
/*
* Allocate memory for row increments
*/
if ((sax = (Uint32 *) realloc(sax, (dst->w + 1) * sizeof(Uint32))) == NULL) {
sax = 0;
return (-1);
}
if ((say = (Uint32 *) realloc(say, (dst->h + 1) * sizeof(Uint32))) == NULL) {
say = 0;
//free(sax);
return (-1);
}
/*
* Pointer setup
*/
sp = csp = (Uint8 *) src->pixels;
dp = (Uint8 *) dst->pixels;
dgap = dst->pitch - dst->w;
if (flipx) csp += (src->w-1);
if (flipy) csp = ( (Uint8*)csp + src->pitch*(src->h-1) );
/*
* Precalculate row increments
*/
csx = 0;
csax = sax;
for (x = 0; x < dst->w; x++) {
csx += src->w;
*csax = 0;
while (csx >= dst->w) {
csx -= dst->w;
(*csax)++;
}
(*csax) *= (flipx ? -1 : 1);
csax++;
}
csy = 0;
csay = say;
for (y = 0; y < dst->h; y++) {
csy += src->h;
*csay = 0;
while (csy >= dst->h) {
csy -= dst->h;
(*csay)++;
}
(*csay) *= src->pitch * (flipy ? -1 : 1);
csay++;
}
/*
* Draw
*/
csay = say;
for (y = 0; y < dst->h; y++) {
csax = sax;
sp = csp;
for (x = 0; x < dst->w; x++) {
/*
* Draw
*/
*dp = *sp;
/*
* Advance source pointers
*/
sp += (*csax);
csax++;
/*
* Advance destination pointer
*/
dp++;
}
/*
* Advance source pointer (for row)
*/
csp += (*csay);
csay++;
/*
* Advance destination pointers
*/
dp += dgap;
}
/*
* Never remove temp arrays
*/
//free(sax);
//free(say);
return 0;
}
}