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
|
/*
* Copyright © 1998 Keith Packard
*
* Permission to use, copy, modify, distribute, and sell this software and its
* documentation for any purpose is hereby granted without fee, provided that
* the above copyright notice appear in all copies and that both that
* copyright notice and this permission notice appear in supporting
* documentation, and that the name of Keith Packard not be used in
* advertising or publicity pertaining to distribution of the software without
* specific, written prior permission. Keith Packard makes no
* representations about the suitability of this software for any purpose. It
* is provided "as is" without express or implied warranty.
*
* KEITH PACKARD DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
* INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO
* EVENT SHALL KEITH PACKARD BE LIABLE FOR ANY SPECIAL, INDIRECT OR
* CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
* DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
* TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THIS SOFTWARE.
*/
#ifdef HAVE_DIX_CONFIG_H
#include <dix-config.h>
#endif
#include "fb.h"
/*
* Stipple masks are independent of bit/byte order as long
* as bitorder == byteorder. FB doesn't handle the case
* where these differ
*/
#define BitsMask(x,w) ((FB_ALLONES << ((x) & FB_MASK)) & \
(FB_ALLONES >> ((FB_UNIT - ((x) + (w))) & FB_MASK)))
#define Mask(x,w) BitsMask((x)*(w),(w))
#define SelMask(b,n,w) ((((b) >> n) & 1) * Mask(n,w))
#define C1(b,w) \
(SelMask(b,0,w))
#define C2(b,w) \
(SelMask(b,0,w) | \
SelMask(b,1,w))
#define C4(b,w) \
(SelMask(b,0,w) | \
SelMask(b,1,w) | \
SelMask(b,2,w) | \
SelMask(b,3,w))
#define C8(b,w) \
(SelMask(b,0,w) | \
SelMask(b,1,w) | \
SelMask(b,2,w) | \
SelMask(b,3,w) | \
SelMask(b,4,w) | \
SelMask(b,5,w) | \
SelMask(b,6,w) | \
SelMask(b,7,w))
static const FbBits fbStipple8Bits[256] = {
C8(0, 4), C8(1, 4), C8(2, 4), C8(3, 4), C8(4, 4), C8(5, 4),
C8(6, 4), C8(7, 4), C8(8, 4), C8(9, 4), C8(10, 4), C8(11, 4),
C8(12, 4), C8(13, 4), C8(14, 4), C8(15, 4), C8(16, 4), C8(17, 4),
C8(18, 4), C8(19, 4), C8(20, 4), C8(21, 4), C8(22, 4), C8(23, 4),
C8(24, 4), C8(25, 4), C8(26, 4), C8(27, 4), C8(28, 4), C8(29, 4),
C8(30, 4), C8(31, 4), C8(32, 4), C8(33, 4), C8(34, 4), C8(35, 4),
C8(36, 4), C8(37, 4), C8(38, 4), C8(39, 4), C8(40, 4), C8(41, 4),
C8(42, 4), C8(43, 4), C8(44, 4), C8(45, 4), C8(46, 4), C8(47, 4),
C8(48, 4), C8(49, 4), C8(50, 4), C8(51, 4), C8(52, 4), C8(53, 4),
C8(54, 4), C8(55, 4), C8(56, 4), C8(57, 4), C8(58, 4), C8(59, 4),
C8(60, 4), C8(61, 4), C8(62, 4), C8(63, 4), C8(64, 4), C8(65, 4),
C8(66, 4), C8(67, 4), C8(68, 4), C8(69, 4), C8(70, 4), C8(71, 4),
C8(72, 4), C8(73, 4), C8(74, 4), C8(75, 4), C8(76, 4), C8(77, 4),
C8(78, 4), C8(79, 4), C8(80, 4), C8(81, 4), C8(82, 4), C8(83, 4),
C8(84, 4), C8(85, 4), C8(86, 4), C8(87, 4), C8(88, 4), C8(89, 4),
C8(90, 4), C8(91, 4), C8(92, 4), C8(93, 4), C8(94, 4), C8(95, 4),
C8(96, 4), C8(97, 4), C8(98, 4), C8(99, 4), C8(100, 4), C8(101, 4),
C8(102, 4), C8(103, 4), C8(104, 4), C8(105, 4), C8(106, 4), C8(107, 4),
C8(108, 4), C8(109, 4), C8(110, 4), C8(111, 4), C8(112, 4), C8(113, 4),
C8(114, 4), C8(115, 4), C8(116, 4), C8(117, 4), C8(118, 4), C8(119, 4),
C8(120, 4), C8(121, 4), C8(122, 4), C8(123, 4), C8(124, 4), C8(125, 4),
C8(126, 4), C8(127, 4), C8(128, 4), C8(129, 4), C8(130, 4), C8(131, 4),
C8(132, 4), C8(133, 4), C8(134, 4), C8(135, 4), C8(136, 4), C8(137, 4),
C8(138, 4), C8(139, 4), C8(140, 4), C8(141, 4), C8(142, 4), C8(143, 4),
C8(144, 4), C8(145, 4), C8(146, 4), C8(147, 4), C8(148, 4), C8(149, 4),
C8(150, 4), C8(151, 4), C8(152, 4), C8(153, 4), C8(154, 4), C8(155, 4),
C8(156, 4), C8(157, 4), C8(158, 4), C8(159, 4), C8(160, 4), C8(161, 4),
C8(162, 4), C8(163, 4), C8(164, 4), C8(165, 4), C8(166, 4), C8(167, 4),
C8(168, 4), C8(169, 4), C8(170, 4), C8(171, 4), C8(172, 4), C8(173, 4),
C8(174, 4), C8(175, 4), C8(176, 4), C8(177, 4), C8(178, 4), C8(179, 4),
C8(180, 4), C8(181, 4), C8(182, 4), C8(183, 4), C8(184, 4), C8(185, 4),
C8(186, 4), C8(187, 4), C8(188, 4), C8(189, 4), C8(190, 4), C8(191, 4),
C8(192, 4), C8(193, 4), C8(194, 4), C8(195, 4), C8(196, 4), C8(197, 4),
C8(198, 4), C8(199, 4), C8(200, 4), C8(201, 4), C8(202, 4), C8(203, 4),
C8(204, 4), C8(205, 4), C8(206, 4), C8(207, 4), C8(208, 4), C8(209, 4),
C8(210, 4), C8(211, 4), C8(212, 4), C8(213, 4), C8(214, 4), C8(215, 4),
C8(216, 4), C8(217, 4), C8(218, 4), C8(219, 4), C8(220, 4), C8(221, 4),
C8(222, 4), C8(223, 4), C8(224, 4), C8(225, 4), C8(226, 4), C8(227, 4),
C8(228, 4), C8(229, 4), C8(230, 4), C8(231, 4), C8(232, 4), C8(233, 4),
C8(234, 4), C8(235, 4), C8(236, 4), C8(237, 4), C8(238, 4), C8(239, 4),
C8(240, 4), C8(241, 4), C8(242, 4), C8(243, 4), C8(244, 4), C8(245, 4),
C8(246, 4), C8(247, 4), C8(248, 4), C8(249, 4), C8(250, 4), C8(251, 4),
C8(252, 4), C8(253, 4), C8(254, 4), C8(255, 4),
};
static const FbBits fbStipple4Bits[16] = {
C4(0, 8), C4(1, 8), C4(2, 8), C4(3, 8), C4(4, 8), C4(5, 8),
C4(6, 8), C4(7, 8), C4(8, 8), C4(9, 8), C4(10, 8), C4(11, 8),
C4(12, 8), C4(13, 8), C4(14, 8), C4(15, 8),
};
static const FbBits fbStipple2Bits[4] = {
C2(0, 16), C2(1, 16), C2(2, 16), C2(3, 16),
};
static const FbBits fbStipple1Bits[2] = {
C1(0, 32), C1(1, 32),
};
#ifdef __clang__
/* shift overflow is intentional */
#pragma clang diagnostic ignored "-Wshift-overflow"
#endif
/*
* Example: srcX = 13 dstX = 8 (FB unit 32 dstBpp 8)
*
* **** **** **** **** **** **** **** ****
* ^
* ******** ******** ******** ********
* ^
* leftShift = 12
* rightShift = 20
*
* Example: srcX = 0 dstX = 8 (FB unit 32 dstBpp 8)
*
* **** **** **** **** **** **** **** ****
* ^
* ******** ******** ******** ********
* ^
*
* leftShift = 24
* rightShift = 8
*/
#define LoadBits {\
if (leftShift) { \
bitsRight = (src < srcEnd ? READ(src++) : 0); \
bits = (FbStipLeft (bitsLeft, leftShift) | \
FbStipRight(bitsRight, rightShift)); \
bitsLeft = bitsRight; \
} else \
bits = (src < srcEnd ? READ(src++) : 0); \
}
void
fbBltOne(FbStip * src, FbStride srcStride, /* FbStip units per scanline */
int srcX, /* bit position of source */
FbBits * dst, FbStride dstStride, /* FbBits units per scanline */
int dstX, /* bit position of dest */
int dstBpp, /* bits per destination unit */
int width, /* width in bits of destination */
int height, /* height in scanlines */
FbBits fgand, /* rrop values */
FbBits fgxor, FbBits bgand, FbBits bgxor)
{
const FbBits *fbBits;
FbBits *srcEnd;
int pixelsPerDst; /* dst pixels per FbBits */
int unitsPerSrc; /* src patterns per FbStip */
int leftShift, rightShift; /* align source with dest */
FbBits startmask, endmask; /* dest scanline masks */
FbStip bits = 0, bitsLeft, bitsRight; /* source bits */
FbStip left;
FbBits mask;
int nDst; /* dest longwords (w.o. end) */
int w;
int n, nmiddle;
int dstS; /* stipple-relative dst X coordinate */
Bool copy; /* accelerate dest-invariant */
Bool transparent; /* accelerate 0 nop */
int srcinc; /* source units consumed */
Bool endNeedsLoad = FALSE; /* need load for endmask */
int startbyte, endbyte;
if (dstBpp == 24) {
fbBltOne24(src, srcStride, srcX,
dst, dstStride, dstX, dstBpp,
width, height, fgand, fgxor, bgand, bgxor);
return;
}
/*
* Do not read past the end of the buffer!
*/
srcEnd = src + height * srcStride;
/*
* Number of destination units in FbBits == number of stipple pixels
* used each time
*/
pixelsPerDst = FB_UNIT / dstBpp;
/*
* Number of source stipple patterns in FbStip
*/
unitsPerSrc = FB_STIP_UNIT / pixelsPerDst;
copy = FALSE;
transparent = FALSE;
if (bgand == 0 && fgand == 0)
copy = TRUE;
else if (bgand == FB_ALLONES && bgxor == 0)
transparent = TRUE;
/*
* Adjust source and dest to nearest FbBits boundary
*/
src += srcX >> FB_STIP_SHIFT;
dst += dstX >> FB_SHIFT;
srcX &= FB_STIP_MASK;
dstX &= FB_MASK;
FbMaskBitsBytes(dstX, width, copy,
startmask, startbyte, nmiddle, endmask, endbyte);
/*
* Compute effective dest alignment requirement for
* source -- must align source to dest unit boundary
*/
dstS = dstX / dstBpp;
/*
* Compute shift constants for effective alignement
*/
if (srcX >= dstS) {
leftShift = srcX - dstS;
rightShift = FB_STIP_UNIT - leftShift;
}
else {
rightShift = dstS - srcX;
leftShift = FB_STIP_UNIT - rightShift;
}
/*
* Get pointer to stipple mask array for this depth
*/
fbBits = 0; /* unused */
switch (pixelsPerDst) {
case 8:
fbBits = fbStipple8Bits;
break;
case 4:
fbBits = fbStipple4Bits;
break;
case 2:
fbBits = fbStipple2Bits;
break;
case 1:
fbBits = fbStipple1Bits;
break;
default:
return;
}
/*
* Compute total number of destination words written, but
* don't count endmask
*/
nDst = nmiddle;
if (startmask)
nDst++;
dstStride -= nDst;
/*
* Compute total number of source words consumed
*/
srcinc = (nDst + unitsPerSrc - 1) / unitsPerSrc;
if (srcX > dstS)
srcinc++;
if (endmask) {
endNeedsLoad = nDst % unitsPerSrc == 0;
if (endNeedsLoad)
srcinc++;
}
srcStride -= srcinc;
/*
* Copy rectangle
*/
while (height--) {
w = nDst; /* total units across scanline */
n = unitsPerSrc; /* units avail in single stipple */
if (n > w)
n = w;
bitsLeft = 0;
if (srcX > dstS)
bitsLeft = READ(src++);
if (n) {
/*
* Load first set of stipple bits
*/
LoadBits;
/*
* Consume stipple bits for startmask
*/
if (startmask) {
mask = fbBits[FbLeftStipBits(bits, pixelsPerDst)];
if (mask || !transparent)
FbDoLeftMaskByteStippleRRop(dst, mask,
fgand, fgxor, bgand, bgxor,
startbyte, startmask);
bits = FbStipLeft(bits, pixelsPerDst);
dst++;
n--;
w--;
}
/*
* Consume stipple bits across scanline
*/
for (;;) {
w -= n;
if (copy) {
while (n--) {
mask = fbBits[FbLeftStipBits(bits, pixelsPerDst)];
WRITE(dst, FbOpaqueStipple(mask, fgxor, bgxor));
dst++;
bits = FbStipLeft(bits, pixelsPerDst);
}
}
else {
while (n--) {
left = FbLeftStipBits(bits, pixelsPerDst);
if (left || !transparent) {
mask = fbBits[left];
WRITE(dst, FbStippleRRop(READ(dst), mask, fgand,
fgxor, bgand, bgxor));
}
dst++;
bits = FbStipLeft(bits, pixelsPerDst);
}
}
if (!w)
break;
/*
* Load another set and reset number of available units
*/
LoadBits;
n = unitsPerSrc;
if (n > w)
n = w;
}
}
/*
* Consume stipple bits for endmask
*/
if (endmask) {
if (endNeedsLoad) {
LoadBits;
}
mask = fbBits[FbLeftStipBits(bits, pixelsPerDst)];
if (mask || !transparent)
FbDoRightMaskByteStippleRRop(dst, mask, fgand, fgxor,
bgand, bgxor, endbyte, endmask);
}
dst += dstStride;
src += srcStride;
}
}
/*
* Crufty macros to initialize the mask array, most of this
* is to avoid compile-time warnings about shift overflow
*/
#if BITMAP_BIT_ORDER == MSBFirst
#define Mask24Pos(x,r) ((x)*24-(r))
#else
#define Mask24Pos(x,r) ((x)*24-((r) ? 24 - (r) : 0))
#endif
#define Mask24Neg(x,r) (Mask24Pos(x,r) < 0 ? -Mask24Pos(x,r) : 0)
#define Mask24Check(x,r) (Mask24Pos(x,r) < 0 ? 0 : \
Mask24Pos(x,r) >= FB_UNIT ? 0 : Mask24Pos(x,r))
#define Mask24(x,r) (Mask24Pos(x,r) < FB_UNIT ? \
(Mask24Pos(x,r) < 0 ? \
0xffffffU >> Mask24Neg (x,r) : \
0xffffffU << Mask24Check(x,r)) : 0)
#define SelMask24(b,n,r) ((((b) >> n) & 1) * Mask24(n,r))
#define C2_24(b,r) \
(SelMask24(b,0,r) | \
SelMask24(b,1,r))
#define FbStip24Len 2
#if BITMAP_BIT_ORDER == MSBFirst
#define FbStip24New(rot) (1 + (rot == 0))
#else
#define FbStip24New(rot) (1 + (rot == 8))
#endif
const FbBits fbStipple24Bits[3][1 << FbStip24Len] = {
/* rotate 0 */
{
C2_24(0, 0), C2_24(1, 0), C2_24(2, 0), C2_24(3, 0),
},
/* rotate 8 */
{
C2_24(0, 8), C2_24(1, 8), C2_24(2, 8), C2_24(3, 8),
},
/* rotate 16 */
{
C2_24(0, 16), C2_24(1, 16), C2_24(2, 16), C2_24(3, 16),
}
};
#if BITMAP_BIT_ORDER == LSBFirst
#define FbMergeStip24Bits(left, right, new) \
(FbStipLeft (left, new) | FbStipRight ((right), (FbStip24Len - (new))))
#define FbMergePartStip24Bits(left, right, llen, rlen) \
(left | FbStipRight(right, llen))
#else
#define FbMergeStip24Bits(left, right, new) \
((FbStipLeft (left, new) & ((1 << FbStip24Len) - 1)) | right)
#define FbMergePartStip24Bits(left, right, llen, rlen) \
(FbStipLeft(left, rlen) | right)
#endif
#define fbFirstStipBits(len,stip) {\
int __len = (len); \
if (len <= remain) { \
stip = FbLeftStipBits(bits, len); \
} else { \
stip = FbLeftStipBits(bits, remain); \
bits = (src < srcEnd ? READ(src++) : 0); \
__len = (len) - remain; \
stip = FbMergePartStip24Bits(stip, FbLeftStipBits(bits, __len), \
remain, __len); \
remain = FB_STIP_UNIT; \
} \
bits = FbStipLeft (bits, __len); \
remain -= __len; \
}
#define fbInitStipBits(offset,len,stip) {\
bits = FbStipLeft (READ(src++),offset); \
remain = FB_STIP_UNIT - offset; \
fbFirstStipBits(len,stip); \
stip = FbMergeStip24Bits (0, stip, len); \
}
#define fbNextStipBits(rot,stip) {\
int __new = FbStip24New(rot); \
FbStip __right; \
fbFirstStipBits(__new, __right); \
stip = FbMergeStip24Bits (stip, __right, __new); \
rot = FbNext24Rot (rot); \
}
/*
* Use deep mask tables that incorporate rotation, pull
* a variable number of bits out of the stipple and
* reuse the right bits as needed for the next write
*
* Yes, this is probably too much code, but most 24-bpp screens
* have no acceleration so this code is used for stipples, copyplane
* and text
*/
void
fbBltOne24(FbStip * srcLine, FbStride srcStride, /* FbStip units per scanline */
int srcX, /* bit position of source */
FbBits * dst, FbStride dstStride, /* FbBits units per scanline */
int dstX, /* bit position of dest */
int dstBpp, /* bits per destination unit */
int width, /* width in bits of destination */
int height, /* height in scanlines */
FbBits fgand, /* rrop values */
FbBits fgxor, FbBits bgand, FbBits bgxor)
{
FbStip *src, *srcEnd;
FbBits leftMask, rightMask, mask;
int nlMiddle, nl;
FbStip stip, bits;
int remain;
int dstS;
int firstlen;
int rot0, rot;
int nDst;
/*
* Do not read past the end of the buffer!
*/
srcEnd = srcLine + height * srcStride;
srcLine += srcX >> FB_STIP_SHIFT;
dst += dstX >> FB_SHIFT;
srcX &= FB_STIP_MASK;
dstX &= FB_MASK;
rot0 = FbFirst24Rot(dstX);
FbMaskBits(dstX, width, leftMask, nlMiddle, rightMask);
dstS = (dstX + 23) / 24;
firstlen = FbStip24Len - dstS;
nDst = nlMiddle;
if (leftMask)
nDst++;
dstStride -= nDst;
/* opaque copy */
if (bgand == 0 && fgand == 0) {
while (height--) {
rot = rot0;
src = srcLine;
srcLine += srcStride;
fbInitStipBits(srcX, firstlen, stip);
if (leftMask) {
mask = fbStipple24Bits[rot >> 3][stip];
WRITE(dst, (READ(dst) & ~leftMask) |
(FbOpaqueStipple(mask,
FbRot24(fgxor, rot), FbRot24(bgxor, rot))
& leftMask));
dst++;
fbNextStipBits(rot, stip);
}
nl = nlMiddle;
while (nl--) {
mask = fbStipple24Bits[rot >> 3][stip];
WRITE(dst, FbOpaqueStipple(mask,
FbRot24(fgxor, rot),
FbRot24(bgxor, rot)));
dst++;
fbNextStipBits(rot, stip);
}
if (rightMask) {
mask = fbStipple24Bits[rot >> 3][stip];
WRITE(dst, (READ(dst) & ~rightMask) |
(FbOpaqueStipple(mask,
FbRot24(fgxor, rot), FbRot24(bgxor, rot))
& rightMask));
}
dst += dstStride;
src += srcStride;
}
}
/* transparent copy */
else if (bgand == FB_ALLONES && bgxor == 0 && fgand == 0) {
while (height--) {
rot = rot0;
src = srcLine;
srcLine += srcStride;
fbInitStipBits(srcX, firstlen, stip);
if (leftMask) {
if (stip) {
mask = fbStipple24Bits[rot >> 3][stip] & leftMask;
WRITE(dst,
(READ(dst) & ~mask) | (FbRot24(fgxor, rot) & mask));
}
dst++;
fbNextStipBits(rot, stip);
}
nl = nlMiddle;
while (nl--) {
if (stip) {
mask = fbStipple24Bits[rot >> 3][stip];
WRITE(dst,
(READ(dst) & ~mask) | (FbRot24(fgxor, rot) & mask));
}
dst++;
fbNextStipBits(rot, stip);
}
if (rightMask) {
if (stip) {
mask = fbStipple24Bits[rot >> 3][stip] & rightMask;
WRITE(dst,
(READ(dst) & ~mask) | (FbRot24(fgxor, rot) & mask));
}
}
dst += dstStride;
}
}
else {
while (height--) {
rot = rot0;
src = srcLine;
srcLine += srcStride;
fbInitStipBits(srcX, firstlen, stip);
if (leftMask) {
mask = fbStipple24Bits[rot >> 3][stip];
WRITE(dst, FbStippleRRopMask(READ(dst), mask,
FbRot24(fgand, rot),
FbRot24(fgxor, rot),
FbRot24(bgand, rot),
FbRot24(bgxor, rot), leftMask));
dst++;
fbNextStipBits(rot, stip);
}
nl = nlMiddle;
while (nl--) {
mask = fbStipple24Bits[rot >> 3][stip];
WRITE(dst, FbStippleRRop(READ(dst), mask,
FbRot24(fgand, rot),
FbRot24(fgxor, rot),
FbRot24(bgand, rot),
FbRot24(bgxor, rot)));
dst++;
fbNextStipBits(rot, stip);
}
if (rightMask) {
mask = fbStipple24Bits[rot >> 3][stip];
WRITE(dst, FbStippleRRopMask(READ(dst), mask,
FbRot24(fgand, rot),
FbRot24(fgxor, rot),
FbRot24(bgand, rot),
FbRot24(bgxor, rot), rightMask));
}
dst += dstStride;
}
}
}
/*
* Not very efficient, but simple -- copy a single plane
* from an N bit image to a 1 bit image
*/
void
fbBltPlane(FbBits * src,
FbStride srcStride,
int srcX,
int srcBpp,
FbStip * dst,
FbStride dstStride,
int dstX,
int width,
int height,
FbStip fgand,
FbStip fgxor, FbStip bgand, FbStip bgxor, Pixel planeMask)
{
FbBits *s;
FbBits pm;
FbBits srcMask;
FbBits srcMaskFirst;
FbBits srcMask0 = 0;
FbBits srcBits;
FbStip dstBits;
FbStip *d;
FbStip dstMask;
FbStip dstMaskFirst;
FbStip dstUnion;
int w;
int wt;
int rot0;
if (!width)
return;
src += srcX >> FB_SHIFT;
srcX &= FB_MASK;
dst += dstX >> FB_STIP_SHIFT;
dstX &= FB_STIP_MASK;
w = width / srcBpp;
pm = fbReplicatePixel(planeMask, srcBpp);
if (srcBpp == 24) {
int tmpw = 24;
rot0 = FbFirst24Rot(srcX);
if (srcX + tmpw > FB_UNIT)
tmpw = FB_UNIT - srcX;
srcMaskFirst = FbRot24(pm, rot0) & FbBitsMask(srcX, tmpw);
}
else {
rot0 = 0;
srcMaskFirst = pm & FbBitsMask(srcX, srcBpp);
srcMask0 = pm & FbBitsMask(0, srcBpp);
}
dstMaskFirst = FbStipMask(dstX, 1);
while (height--) {
d = dst;
dst += dstStride;
s = src;
src += srcStride;
srcMask = srcMaskFirst;
if (srcBpp == 24)
srcMask0 = FbRot24(pm, rot0) & FbBitsMask(0, srcBpp);
srcBits = READ(s++);
dstMask = dstMaskFirst;
dstUnion = 0;
dstBits = 0;
wt = w;
while (wt--) {
if (!srcMask) {
srcBits = READ(s++);
if (srcBpp == 24)
srcMask0 = FbNext24Pix(srcMask0) & FbBitsMask(0, 24);
srcMask = srcMask0;
}
if (!dstMask) {
WRITE(d, FbStippleRRopMask(READ(d), dstBits,
fgand, fgxor, bgand, bgxor,
dstUnion));
d++;
dstMask = FbStipMask(0, 1);
dstUnion = 0;
dstBits = 0;
}
if (srcBits & srcMask)
dstBits |= dstMask;
dstUnion |= dstMask;
if (srcBpp == FB_UNIT)
srcMask = 0;
else
srcMask = FbScrRight(srcMask, srcBpp);
dstMask = FbStipRight(dstMask, 1);
}
if (dstUnion)
WRITE(d, FbStippleRRopMask(READ(d), dstBits,
fgand, fgxor, bgand, bgxor, dstUnion));
}
}
|