forked from libsdl-org/SDL_ttf
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathSDL_ttf.c
4289 lines (3711 loc) · 150 KB
/
SDL_ttf.c
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
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
SDL_ttf: A companion library to SDL for working with TrueType (tm) fonts
Copyright (C) 2001-2025 Sam Lantinga <slouken@libsdl.org>
This software is provided 'as-is', without any express or implied
warranty. In no event will the authors be held liable for any damages
arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely, subject to the following restrictions:
1. The origin of this software must not be misrepresented; you must not
claim that you wrote the original software. If you use this software
in a product, an acknowledgment in the product documentation would be
appreciated but is not required.
2. Altered source versions must be plainly marked as such, and must not be
misrepresented as being the original software.
3. This notice may not be removed or altered from any source distribution.
*/
#include "SDL.h"
#include "SDL_cpuinfo.h"
#include "SDL_endian.h"
#include "SDL_ttf.h"
#include <ft2build.h>
#include FT_FREETYPE_H
#include FT_OUTLINE_H
#include FT_STROKER_H
#include FT_GLYPH_H
#include FT_TRUETYPE_IDS_H
#include FT_IMAGE_H
/* Enable rendering with color
* Freetype may need to be compiled with FT_CONFIG_OPTION_USE_PNG */
#if defined(FT_HAS_COLOR)
# define TTF_USE_COLOR 1
#else
# define TTF_USE_COLOR 0
#endif
/* Enable Signed Distance Field rendering (requires latest FreeType version) */
#if defined(FT_RASTER_FLAG_SDF)
# define TTF_USE_SDF 1
#else
# define TTF_USE_SDF 0
#endif
#if TTF_USE_SDF
#include FT_MODULE_H
#endif
/* Enable HarfBuzz for Complex text rendering
* Freetype may need to be compiled with FT_CONFIG_OPTION_USE_HARFBUZZ */
#ifndef TTF_USE_HARFBUZZ
# define TTF_USE_HARFBUZZ 0
#endif
#if defined(SDL_BUILD_MAJOR_VERSION) && defined(SDL_COMPILE_TIME_ASSERT)
SDL_COMPILE_TIME_ASSERT(SDL_BUILD_MAJOR_VERSION,
SDL_TTF_MAJOR_VERSION == SDL_BUILD_MAJOR_VERSION);
SDL_COMPILE_TIME_ASSERT(SDL_BUILD_MINOR_VERSION,
SDL_TTF_MINOR_VERSION == SDL_BUILD_MINOR_VERSION);
SDL_COMPILE_TIME_ASSERT(SDL_BUILD_MICRO_VERSION,
SDL_TTF_PATCHLEVEL == SDL_BUILD_MICRO_VERSION);
#endif
#if defined(SDL_COMPILE_TIME_ASSERT)
SDL_COMPILE_TIME_ASSERT(SDL_TTF_MAJOR_VERSION_min, SDL_TTF_MAJOR_VERSION >= 0);
/* Limited only by the need to fit in SDL_version */
SDL_COMPILE_TIME_ASSERT(SDL_TTF_MAJOR_VERSION_max, SDL_TTF_MAJOR_VERSION <= 255);
SDL_COMPILE_TIME_ASSERT(SDL_TTF_MINOR_VERSION_min, SDL_TTF_MINOR_VERSION >= 0);
/* Limited only by the need to fit in SDL_version */
SDL_COMPILE_TIME_ASSERT(SDL_TTF_MINOR_VERSION_max, SDL_TTF_MINOR_VERSION <= 255);
SDL_COMPILE_TIME_ASSERT(SDL_TTF_PATCHLEVEL_min, SDL_TTF_PATCHLEVEL >= 0);
/* Limited by its encoding in SDL_VERSIONNUM and in the ABI versions */
SDL_COMPILE_TIME_ASSERT(SDL_TTF_PATCHLEVEL_max, SDL_TTF_PATCHLEVEL <= 99);
#endif
#if TTF_USE_HARFBUZZ
#include <hb.h>
#include <hb-ft.h>
/* Default configuration */
static hb_direction_t g_hb_direction = HB_DIRECTION_LTR;
static hb_script_t g_hb_script = HB_SCRIPT_UNKNOWN;
#endif
/* Harfbuzz */
int TTF_SetDirection(int direction) /* hb_direction_t */
{
#if TTF_USE_HARFBUZZ
g_hb_direction = direction;
return 0;
#else
(void) direction;
return -1;
#endif
}
int TTF_SetScript(int script) /* hb_script_t */
{
#if TTF_USE_HARFBUZZ
g_hb_script = script;
return 0;
#else
(void) script;
return -1;
#endif
}
/* Round glyph to 16 bytes width and use SSE2 instructions */
#if defined(__SSE2__)
# define HAVE_SSE2_INTRINSICS 1
#endif
/* Round glyph width to 16 bytes use NEON instructions */
#if 0 /*defined(__ARM_NEON)*/
# define HAVE_NEON_INTRINSICS 1
#endif
/* Round glyph width to 8 bytes */
#define HAVE_BLIT_GLYPH_64
/* Android armeabi-v7a doesn't like int64 (Maybe all other __ARM_ARCH < 7 ?),
* un-activate it, especially if NEON isn't detected */
#if defined(__ARM_ARCH)
# if __ARM_ARCH < 8
# if defined(HAVE_BLIT_GLYPH_64)
# undef HAVE_BLIT_GLYPH_64
# endif
# endif
#endif
/* Default: round glyph width to 4 bytes to copy them faster */
#define HAVE_BLIT_GLYPH_32
/* Use Duff's device to unroll loops */
//#define USE_DUFFS_LOOP
#if defined(HAVE_SSE2_INTRINSICS)
static SDL_INLINE int hasSSE2(void)
{
static int val = -1;
if (val != -1) {
return val;
}
val = SDL_HasSSE2();
return val;
}
#endif
#if defined(HAVE_NEON_INTRINSICS)
static SDL_INLINE int hasNEON(void)
{
static int val = -1;
if (val != -1) {
return val;
}
val = SDL_HasNEON();
return val;
}
#endif
/* FIXME: Right now we assume the gray-scale renderer Freetype is using
supports 256 shades of gray, but we should instead key off of num_grays
in the result FT_Bitmap after the FT_Render_Glyph() call. */
#define NUM_GRAYS 256
/* x offset = cos(((90.0-12)/360) * 2 * M_PI), or 12 degree angle */
/* same value as in FT_GlyphSlot_Oblique, fixed point 16.16 */
#define GLYPH_ITALICS 0x0366AL
/* Handy routines for converting from fixed point 26.6 */
#define FT_FLOOR(X) (int)(((X) & -64) / 64)
#define FT_CEIL(X) FT_FLOOR((X) + 63)
/* Handy routine for converting to fixed point 26.6 */
#define F26Dot6(X) ((X) << 6)
/* Faster divide by 255, with same result
* in range [0; 255]: (x + 1 + (x >> 8)) >> 8
* in range [-255; 0]: (x + 255 + (x >> 8)) >> 8 */
#define DIVIDE_BY_255_SIGNED(x, sign_val) (((x) + (sign_val) + ((x)>>8)) >> 8)
/* When x positive */
#define DIVIDE_BY_255(x) DIVIDE_BY_255_SIGNED(x, 1)
#define CACHED_METRICS 0x20
#define CACHED_BITMAP 0x01
#define CACHED_PIXMAP 0x02
#define CACHED_COLOR 0x04
#define CACHED_LCD 0x08
#define CACHED_SUBPIX 0x10
typedef struct {
unsigned char *buffer; /* aligned */
int left;
int top;
int width;
int rows;
int pitch;
int is_color;
} TTF_Image;
/* Cached glyph information */
typedef struct cached_glyph {
int stored;
FT_UInt index;
TTF_Image bitmap;
TTF_Image pixmap;
int sz_left;
int sz_top;
int sz_width;
int sz_rows;
int advance;
union {
/* TTF_HINTING_LIGHT_SUBPIXEL (only pixmap) */
struct {
int lsb_minus_rsb;
int translation;
} subpixel;
/* Other hinting */
struct {
int rsb_delta;
int lsb_delta;
} kerning_smart;
};
} c_glyph;
/* Internal buffer to store positions computed by TTF_Size_Internal()
* for rendered string by Render_Line() */
typedef struct PosBuf {
FT_UInt index;
int x;
int y;
} PosBuf_t;
/* The structure used to hold internal font information */
struct TTF_Font {
/* Freetype2 maintains all sorts of useful info itself */
FT_Face face;
/* We'll cache these ourselves */
int height;
int ascent;
int descent;
int lineskip;
/* The font style */
int style;
int outline_val;
/* Whether kerning is desired */
int allow_kerning;
#if !TTF_USE_HARFBUZZ
int use_kerning;
#endif
/* Extra width in glyph bounds for text styles */
int glyph_overhang;
/* Information in the font for underlining */
int line_thickness;
int underline_top_row;
int strikethrough_top_row;
/* Cache for style-transformed glyphs */
c_glyph cache[256];
FT_UInt cache_index[128];
/* We are responsible for closing the font stream */
SDL_RWops *src;
int freesrc;
FT_Open_Args args;
/* Internal buffer to store positions computed by TTF_Size_Internal()
* for rendered string by Render_Line() */
PosBuf_t *pos_buf;
Uint32 pos_len;
Uint32 pos_max;
/* Hinting modes */
int ft_load_target;
int render_subpixel;
#if TTF_USE_HARFBUZZ
hb_font_t *hb_font;
/* If HB_SCRIPT_INVALID, use global default script */
hb_script_t hb_script;
/* If HB_DIRECTION_INVALID, use global default direction */
hb_direction_t hb_direction;
#endif
int render_sdf;
/* Extra layout setting for wrapped text */
int horizontal_align;
};
/* Tell if SDL_ttf has to handle the style */
#define TTF_HANDLE_STYLE_BOLD(font) ((font)->style & TTF_STYLE_BOLD)
#define TTF_HANDLE_STYLE_ITALIC(font) ((font)->style & TTF_STYLE_ITALIC)
#define TTF_HANDLE_STYLE_UNDERLINE(font) ((font)->style & TTF_STYLE_UNDERLINE)
#define TTF_HANDLE_STYLE_STRIKETHROUGH(font) ((font)->style & TTF_STYLE_STRIKETHROUGH)
/* Font styles that does not impact glyph drawing */
#define TTF_STYLE_NO_GLYPH_CHANGE (TTF_STYLE_UNDERLINE | TTF_STYLE_STRIKETHROUGH)
/* The FreeType font engine/library */
static FT_Library library = NULL;
static int TTF_initialized = 0;
static SDL_bool TTF_byteswapped = SDL_FALSE;
#define TTF_CHECK_INITIALIZED(errval) \
if (!TTF_initialized) { \
TTF_SetError("Library not initialized"); \
return errval; \
}
#define TTF_CHECK_POINTER(p, errval) \
if (!(p)) { \
TTF_SetError("Passed a NULL pointer"); \
return errval; \
}
typedef enum {
RENDER_SOLID = 0,
RENDER_SHADED,
RENDER_BLENDED,
RENDER_LCD
} render_mode_t;
typedef enum {
STR_UTF8 = 0,
STR_TEXT,
STR_UNICODE
} str_type_t;
static int TTF_initFontMetrics(TTF_Font *font);
static int TTF_Size_Internal(TTF_Font *font, const char *text, str_type_t str_type,
int *w, int *h, int *xstart, int *ystart, int measure_width, int *extent, int *count);
#define NO_MEASUREMENT \
0, NULL, NULL
static SDL_Surface* TTF_Render_Internal(TTF_Font *font, const char *text, str_type_t str_type,
SDL_Color fg, SDL_Color bg, render_mode_t render_mode);
static SDL_Surface* TTF_Render_Wrapped_Internal(TTF_Font *font, const char *text, str_type_t str_type,
SDL_Color fg, SDL_Color bg, Uint32 wrapLength, render_mode_t render_mode);
static SDL_INLINE int Find_GlyphByIndex(TTF_Font *font, FT_UInt idx,
int want_bitmap, int want_pixmap, int want_color, int want_lcd, int want_subpixel,
int translation, c_glyph **out_glyph, TTF_Image **out_image);
static void Flush_Cache(TTF_Font *font);
#if defined(USE_DUFFS_LOOP)
/* 4-times unrolled loop */
#define DUFFS_LOOP4(pixel_copy_increment, width) \
{ int n = (width+3)/4; \
switch (width & 3) { \
case 0: do { pixel_copy_increment; /* fallthrough */ \
case 3: pixel_copy_increment; /* fallthrough */ \
case 2: pixel_copy_increment; /* fallthrough */ \
case 1: pixel_copy_increment; /* fallthrough */ \
} while (--n > 0); \
} \
}
#else
/* Don't use Duff's device to unroll loops */
#define DUFFS_LOOP(pixel_copy_increment, width) \
{ int n; \
for ( n=width; n > 0; --n ) { \
pixel_copy_increment; \
} \
}
#define DUFFS_LOOP4(pixel_copy_increment, width) \
DUFFS_LOOP(pixel_copy_increment, width)
#endif
/* Blend colored glyphs */
static SDL_INLINE void BG_Blended_Color(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip, Uint8 fg_alpha)
{
const Uint32 *src = (Uint32 *)image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width;
Uint32 height = image->rows;
if (fg_alpha == 0) { /* SDL_ALPHA_OPAQUE */
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
*dst++ = *src++;
, width);
/* *INDENT-ON* */
src = (const Uint32 *)((const Uint8 *)src + srcskip);
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
} else {
Uint32 alpha;
Uint32 tmp;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
/* prevent misaligned load: tmp = *src++; */
/* eventually, we can expect the compiler to replace the memcpy call with something optimized */
memcpy(&tmp, src++, sizeof(tmp)); /* This should NOT be SDL_memcpy */
alpha = tmp >> 24;
tmp &= ~0xFF000000;
alpha = fg_alpha * alpha;
alpha = DIVIDE_BY_255(alpha) << 24;
*dst++ = tmp | alpha
, width);
/* *INDENT-ON* */
src = (const Uint32 *)((const Uint8 *)src + srcskip);
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
}
/* Blend with LCD rendering */
static SDL_INLINE void BG_Blended_LCD(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip, SDL_Color *fg)
{
const Uint32 *src = (Uint32 *)image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width;
Uint32 height = image->rows;
Uint32 tmp, bg;
Uint32 r, g, b;
Uint8 fg_r, fg_g, fg_b;
Uint8 bg_r, bg_g, bg_b;
Uint32 bg_a;
fg_r = fg->r;
fg_g = fg->g;
fg_b = fg->b;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
/* prevent misaligned load: tmp = *src++; */
memcpy(&tmp, src++, sizeof(tmp)); /* This should NOT be SDL_memcpy */
if (tmp) {
bg = *dst;
bg_a = bg & 0xff000000;
bg_r = (bg >> 16) & 0xff;
bg_g = (bg >> 8) & 0xff;
bg_b = (bg >> 0) & 0xff;
r = (tmp >> 16) & 0xff;
g = (tmp >> 8) & 0xff;
b = (tmp >> 0) & 0xff;
r = fg_r * r + bg_r * (255 - r) + 127;
r = DIVIDE_BY_255(r);
g = fg_g * g + bg_g * (255 - g) + 127;
g = DIVIDE_BY_255(g);
b = fg_b * b + bg_b * (255 - b) + 127;
b = DIVIDE_BY_255(b);
r <<= 16;
g <<= 8;
b <<= 0;
*dst = r | g | b | bg_a;
}
dst++;
, width);
/* *INDENT-ON* */
src = (const Uint32 *)((const Uint8 *)src + srcskip);
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
#if TTF_USE_SDF
/* Blended Opaque SDF */
static SDL_INLINE void BG_Blended_Opaque_SDF(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip)
{
const Uint8 *src = image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width;
Uint32 height = image->rows;
Uint32 s;
Uint32 d;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
d = *dst;
s = ((Uint32)*src++) << 24;
if (s > d) {
*dst = s;
}
dst++;
, width);
/* *INDENT-ON* */
src += srcskip;
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
/* Blended non-opaque SDF */
static SDL_INLINE void BG_Blended_SDF(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip, Uint8 fg_alpha)
{
const Uint8 *src = image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width;
Uint32 height = image->rows;
Uint32 s;
Uint32 d;
Uint32 tmp;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
d = *dst;
tmp = fg_alpha * (*src++);
s = DIVIDE_BY_255(tmp) << 24;
if (s > d) {
*dst = s;
}
dst++;
, width);
/* *INDENT-ON* */
src += srcskip;
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
#endif /* TTF_USE_SDF */
/* Blended Opaque */
static SDL_INLINE void BG_Blended_Opaque(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip)
{
const Uint8 *src = image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width;
Uint32 height = image->rows;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
*dst++ |= ((Uint32)*src++) << 24;
, width);
/* *INDENT-ON* */
src += srcskip;
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
/* Blended non-opaque */
static SDL_INLINE void BG_Blended(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip, Uint8 fg_alpha)
{
const Uint8 *src = image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width;
Uint32 height = image->rows;
Uint32 tmp;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
tmp = fg_alpha * (*src++);
*dst++ |= DIVIDE_BY_255(tmp) << 24;
, width);
/* *INDENT-ON* */
src += srcskip;
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
#if defined(HAVE_BLIT_GLYPH_32) || defined(HAVE_BLIT_GLYPH_64)
static SDL_INLINE void BG_Blended_Opaque_32(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip)
{
const Uint8 *src = image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width / 4;
Uint32 height = image->rows;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
*dst++ |= ((Uint32)*src++) << 24;
*dst++ |= ((Uint32)*src++) << 24;
*dst++ |= ((Uint32)*src++) << 24;
*dst++ |= ((Uint32)*src++) << 24;
, width);
/* *INDENT-ON* */
src += srcskip;
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
static SDL_INLINE void BG_Blended_32(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip, Uint8 fg_alpha)
{
const Uint8 *src = image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width / 4;
Uint32 height = image->rows;
Uint32 tmp0, tmp1, tmp2, tmp3;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
tmp0 = fg_alpha * (*src++);
tmp1 = fg_alpha * (*src++);
tmp2 = fg_alpha * (*src++);
tmp3 = fg_alpha * (*src++);
*dst++ |= DIVIDE_BY_255(tmp0) << 24;
*dst++ |= DIVIDE_BY_255(tmp1) << 24;
*dst++ |= DIVIDE_BY_255(tmp2) << 24;
*dst++ |= DIVIDE_BY_255(tmp3) << 24;
, width);
/* *INDENT-ON* */
src += srcskip;
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
#endif
#if defined(HAVE_SSE2_INTRINSICS)
/* Apply: alpha_table[i] = i << 24; */
static SDL_INLINE void BG_Blended_Opaque_SSE(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip)
{
const __m128i *src = (__m128i *)image->buffer;
__m128i *dst = (__m128i *)destination;
Uint32 width = image->width / 16;
Uint32 height = image->rows;
__m128i s, s0, s1, s2, s3, d0, d1, d2, d3, r0, r1, r2, r3, L, H;
const __m128i zero = _mm_setzero_si128();
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
/* Read 16 Uint8 at once and put into 4 __m128i */
s = _mm_loadu_si128(src); // load unaligned
d0 = _mm_load_si128(dst); // load
d1 = _mm_load_si128(dst + 1); // load
d2 = _mm_load_si128(dst + 2); // load
d3 = _mm_load_si128(dst + 3); // load
L = _mm_unpacklo_epi8(zero, s);
H = _mm_unpackhi_epi8(zero, s);
s0 = _mm_unpacklo_epi8(zero, L);
s1 = _mm_unpackhi_epi8(zero, L);
s2 = _mm_unpacklo_epi8(zero, H);
s3 = _mm_unpackhi_epi8(zero, H);
// already shifted by 24
r0 = _mm_or_si128(d0, s0); // or
r1 = _mm_or_si128(d1, s1); // or
r2 = _mm_or_si128(d2, s2); // or
r3 = _mm_or_si128(d3, s3); // or
_mm_store_si128(dst, r0); // store
_mm_store_si128(dst + 1, r1); // store
_mm_store_si128(dst + 2, r2); // store
_mm_store_si128(dst + 3, r3); // store
dst += 4;
src += 1;
, width);
/* *INDENT-ON* */
src = (const __m128i *)((const Uint8 *)src + srcskip);
dst = (__m128i *)((Uint8 *)dst + dstskip);
}
}
static SDL_INLINE void BG_Blended_SSE(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip, Uint8 fg_alpha)
{
const __m128i *src = (__m128i *)image->buffer;
__m128i *dst = (__m128i *)destination;
Uint32 width = image->width / 16;
Uint32 height = image->rows;
const __m128i alpha = _mm_set1_epi16(fg_alpha);
const __m128i one = _mm_set1_epi16(1);
const __m128i zero = _mm_setzero_si128();
__m128i s, s0, s1, s2, s3, d0, d1, d2, d3, r0, r1, r2, r3, L, H, Ls8, Hs8;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
/* Read 16 Uint8 at once and put into 4 __m128i */
s = _mm_loadu_si128(src); // load unaligned
d0 = _mm_load_si128(dst); // load
d1 = _mm_load_si128(dst + 1); // load
d2 = _mm_load_si128(dst + 2); // load
d3 = _mm_load_si128(dst + 3); // load
L = _mm_unpacklo_epi8(s, zero); // interleave, no shifting
H = _mm_unpackhi_epi8(s, zero); // enough room to multiply
/* Apply: alpha_table[i] = ((i * fg.a / 255) << 24; */
/* Divide by 255 is done as: (x + 1 + (x >> 8)) >> 8 */
L = _mm_mullo_epi16(L, alpha); // x := i * fg.a
H = _mm_mullo_epi16(H, alpha);
Ls8 = _mm_srli_epi16(L, 8); // x >> 8
Hs8 = _mm_srli_epi16(H, 8);
L = _mm_add_epi16(L, one); // x + 1
H = _mm_add_epi16(H, one);
L = _mm_add_epi16(L, Ls8); // x + 1 + (x >> 8)
H = _mm_add_epi16(H, Hs8);
L = _mm_srli_epi16(L, 8); // ((x + 1 + (x >> 8)) >> 8
H = _mm_srli_epi16(H, 8);
L = _mm_slli_epi16(L, 8); // shift << 8, so we're prepared
H = _mm_slli_epi16(H, 8); // to have final format << 24
s0 = _mm_unpacklo_epi8(zero, L);
s1 = _mm_unpackhi_epi8(zero, L);
s2 = _mm_unpacklo_epi8(zero, H);
s3 = _mm_unpackhi_epi8(zero, H);
// already shifted by 24
r0 = _mm_or_si128(d0, s0); // or
r1 = _mm_or_si128(d1, s1); // or
r2 = _mm_or_si128(d2, s2); // or
r3 = _mm_or_si128(d3, s3); // or
_mm_store_si128(dst, r0); // store
_mm_store_si128(dst + 1, r1); // store
_mm_store_si128(dst + 2, r2); // store
_mm_store_si128(dst + 3, r3); // store
dst += 4;
src += 1;
, width);
/* *INDENT-ON* */
src = (const __m128i *)((const Uint8 *)src + srcskip);
dst = (__m128i *)((Uint8 *)dst + dstskip);
}
}
#endif
#if defined(HAVE_NEON_INTRINSICS)
/* Apply: alpha_table[i] = i << 24; */
static SDL_INLINE void BG_Blended_Opaque_NEON(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip)
{
const Uint32 *src = (Uint32 *)image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width / 16;
Uint32 height = image->rows;
uint32x4_t s, d0, d1, d2, d3, r0, r1, r2, r3;
uint8x16x2_t sx, sx01, sx23;
const uint8x16_t zero = vdupq_n_u8(0);
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
/* Read 4 Uint32 and put 16 Uint8 into uint32x4x2_t (uint8x16x2_t)
* takes advantage of vzipq_u8 which produces two lanes */
s = vld1q_u32(src); // load
d0 = vld1q_u32(dst); // load
d1 = vld1q_u32(dst + 4); // load
d2 = vld1q_u32(dst + 8); // load
d3 = vld1q_u32(dst + 12); // load
sx = vzipq_u8(zero, (uint8x16_t)s); // interleave
sx01 = vzipq_u8(zero, sx.val[0]); // interleave
sx23 = vzipq_u8(zero, sx.val[1]); // interleave
// already shifted by 24
r0 = vorrq_u32(d0, (uint32x4_t)sx01.val[0]); // or
r1 = vorrq_u32(d1, (uint32x4_t)sx01.val[1]); // or
r2 = vorrq_u32(d2, (uint32x4_t)sx23.val[0]); // or
r3 = vorrq_u32(d3, (uint32x4_t)sx23.val[1]); // or
vst1q_u32(dst, r0); // store
vst1q_u32(dst + 4, r1); // store
vst1q_u32(dst + 8, r2); // store
vst1q_u32(dst + 12, r3); // store
dst += 16;
src += 4;
, width);
/* *INDENT-ON* */
src = (const Uint32 *)((const Uint8 *)src + srcskip);
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
/* Non-opaque, computes alpha blending on the fly */
static SDL_INLINE void BG_Blended_NEON(const TTF_Image *image, Uint32 *destination, Sint32 srcskip, Uint32 dstskip, Uint8 fg_alpha)
{
const Uint32 *src = (Uint32 *)image->buffer;
Uint32 *dst = destination;
Uint32 width = image->width / 16;
Uint32 height = image->rows;
uint32x4_t s, d0, d1, d2, d3, r0, r1, r2, r3;
uint16x8_t Ls8, Hs8;
uint8x16x2_t sx, sx01, sx23;
uint16x8x2_t sxm;
const uint16x8_t alpha = vmovq_n_u16(fg_alpha);
const uint16x8_t one = vmovq_n_u16(1);
const uint8x16_t zero = vdupq_n_u8(0);
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
/* Read 4 Uint32 and put 16 Uint8 into uint32x4x2_t (uint8x16x2_t)
* takes advantage of vzipq_u8 which produces two lanes */
s = vld1q_u32(src); // load
d0 = vld1q_u32(dst); // load
d1 = vld1q_u32(dst + 4); // load
d2 = vld1q_u32(dst + 8); // load
d3 = vld1q_u32(dst + 12); // load
sx = vzipq_u8((uint8x16_t)s, zero); // interleave, no shifting
// enough room to multiply
/* Apply: alpha_table[i] = ((i * fg.a / 255) << 24; */
/* Divide by 255 is done as: (x + 1 + (x >> 8)) >> 8 */
sxm.val[0] = vmulq_u16((uint16x8_t)sx.val[0], alpha); // x := i * fg.a
sxm.val[1] = vmulq_u16((uint16x8_t)sx.val[1], alpha);
Ls8 = vshrq_n_u16(sxm.val[0], 8); // x >> 8
Hs8 = vshrq_n_u16(sxm.val[1], 8);
sxm.val[0] = vaddq_u16(sxm.val[0], one); // x + 1
sxm.val[1] = vaddq_u16(sxm.val[1], one);
sxm.val[0] = vaddq_u16(sxm.val[0], Ls8); // x + 1 + (x >> 8)
sxm.val[1] = vaddq_u16(sxm.val[1], Hs8);
sxm.val[0] = vshrq_n_u16(sxm.val[0], 8); // ((x + 1 + (x >> 8)) >> 8
sxm.val[1] = vshrq_n_u16(sxm.val[1], 8);
sxm.val[0] = vshlq_n_u16(sxm.val[0], 8); // shift << 8, so we're prepared
sxm.val[1] = vshlq_n_u16(sxm.val[1], 8); // to have final format << 24
sx01 = vzipq_u8(zero, (uint8x16_t)sxm.val[0]); // interleave
sx23 = vzipq_u8(zero, (uint8x16_t)sxm.val[1]); // interleave
// already shifted by 24
r0 = vorrq_u32(d0, (uint32x4_t)sx01.val[0]); // or
r1 = vorrq_u32(d1, (uint32x4_t)sx01.val[1]); // or
r2 = vorrq_u32(d2, (uint32x4_t)sx23.val[0]); // or
r3 = vorrq_u32(d3, (uint32x4_t)sx23.val[1]); // or
vst1q_u32(dst, r0); // store
vst1q_u32(dst + 4, r1); // store
vst1q_u32(dst + 8, r2); // store
vst1q_u32(dst + 12, r3); // store
dst += 16;
src += 4;
, width);
/* *INDENT-ON* */
src = (const Uint32 *)((const Uint8 *)src + srcskip);
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
#endif
static SDL_INLINE void BG(const TTF_Image *image, Uint8 *destination, Sint32 srcskip, Uint32 dstskip)
{
const Uint8 *src = image->buffer;
Uint8 *dst = destination;
Uint32 width = image->width;
Uint32 height = image->rows;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
*dst++ |= *src++;
, width);
/* *INDENT-ON* */
src += srcskip;
dst += dstskip;
}
}
#if defined(HAVE_BLIT_GLYPH_64)
static SDL_INLINE void BG_64(const TTF_Image *image, Uint8 *destination, Sint32 srcskip, Uint32 dstskip)
{
const Uint64 *src = (Uint64 *)image->buffer;
Uint64 *dst = (Uint64 *)destination;
Uint32 width = image->width / 8;
Uint32 height = image->rows;
Uint64 tmp;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
/* prevent misaligned load: *dst++ |= *src++; */
memcpy(&tmp, src++, sizeof(tmp)); /* This should NOT be SDL_memcpy */
*dst++ |= tmp;
, width);
/* *INDENT-ON* */
src = (const Uint64 *)((const Uint8 *)src + srcskip);
dst = (Uint64 *)((Uint8 *)dst + dstskip);
}
}
#elif defined(HAVE_BLIT_GLYPH_32)
static SDL_INLINE void BG_32(const TTF_Image *image, Uint8 *destination, Sint32 srcskip, Uint32 dstskip)
{
const Uint32 *src = (Uint32 *)image->buffer;
Uint32 *dst = (Uint32 *)destination;
Uint32 width = image->width / 4;
Uint32 height = image->rows;
Uint32 tmp;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
/* prevent misaligned load: *dst++ |= *src++; */
memcpy(&tmp, src++, sizeof(tmp)); /* This should NOT be SDL_memcpy */
*dst++ |= tmp;
, width);
/* *INDENT-ON* */
src = (const Uint32 *)((const Uint8 *)src + srcskip);
dst = (Uint32 *)((Uint8 *)dst + dstskip);
}
}
#endif
#if defined(HAVE_SSE2_INTRINSICS)
static SDL_INLINE void BG_SSE(const TTF_Image *image, Uint8 *destination, Sint32 srcskip, Uint32 dstskip)
{
const __m128i *src = (__m128i *)image->buffer;
__m128i *dst = (__m128i *)destination;
Uint32 width = image->width / 16;
Uint32 height = image->rows;
__m128i s, d, r;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
s = _mm_loadu_si128(src); // load unaligned
d = _mm_load_si128(dst); // load
r = _mm_or_si128(d, s); // or
_mm_store_si128(dst, r); // store
src += 1;
dst += 1;
, width);
/* *INDENT-ON* */
src = (const __m128i *)((const Uint8 *)src + srcskip);
dst = (__m128i *)((Uint8 *)dst + dstskip);
}
}
#endif
#if defined(HAVE_NEON_INTRINSICS)
static SDL_INLINE void BG_NEON(const TTF_Image *image, Uint8 *destination, Sint32 srcskip, Uint32 dstskip)
{
const Uint8 *src = image->buffer;
Uint8 *dst = destination;
Uint32 width = image->width / 16;
Uint32 height = image->rows;
uint8x16_t s, d, r;
while (height--) {
/* *INDENT-OFF* */
DUFFS_LOOP4(
s = vld1q_u8(src); // load
d = vld1q_u8(dst); // load
r = vorrq_u8(d, s); // or
vst1q_u8(dst, r); // store
src += 16;
dst += 16;
, width);