Linux Audio

Check our new training course

Embedded Linux Audio

Check our new training course
with Creative Commons CC-BY-SA
lecture materials

Bootlin logo

Elixir Cross Referencer

Loading...
  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
/*
 *  linux/drivers/video/fbcon.h -- Low level frame buffer based console driver
 *
 *	Copyright (C) 1997 Geert Uytterhoeven
 *
 *  This file is subject to the terms and conditions of the GNU General Public
 *  License.  See the file COPYING in the main directory of this archive
 *  for more details.
 */

#ifndef _VIDEO_FBCON_H
#define _VIDEO_FBCON_H

#include <linux/config.h>
#include <linux/types.h>
#include <linux/console_struct.h>
#include <linux/vt_buffer.h>

#include <asm/io.h>


    /*                                  
     *  `switch' for the Low Level Operations
     */
 
struct display_switch {                                                
    void (*setup)(struct display *p);
    void (*bmove)(struct display *p, int sy, int sx, int dy, int dx,
		  int height, int width);
    /* for clear, conp may be NULL, which means use a blanking (black) color */
    void (*clear)(struct vc_data *conp, struct display *p, int sy, int sx,
		  int height, int width);
    void (*putc)(struct vc_data *conp, struct display *p, int c, int yy,
    		 int xx);
    void (*putcs)(struct vc_data *conp, struct display *p, const unsigned short *s,
		  int count, int yy, int xx);     
    void (*revc)(struct display *p, int xx, int yy);
    void (*cursor)(struct display *p, int mode, int xx, int yy);
    int  (*set_font)(struct display *p, int width, int height);
    void (*clear_margins)(struct vc_data *conp, struct display *p,
			  int bottom_only);
    unsigned int fontwidthmask;      /* 1 at (1 << (width - 1)) if width is supported */
}; 

extern struct display_switch fbcon_dummy;

   /*
    *    This is the interface between the low-level console driver and the
    *    low-level frame buffer device
    */

struct display {
    /* Filled in by the frame buffer device */

    struct fb_var_screeninfo var;   /* variable infos. yoffset and vmode */
                                    /* are updated by fbcon.c */
    struct fb_cmap cmap;            /* colormap */
    char *screen_base;              /* pointer to top of virtual screen */    
                                    /* (virtual address) */
    int visual;
    int type;                       /* see FB_TYPE_* */
    int type_aux;                   /* Interleave for interleaved Planes */
    u_short ypanstep;               /* zero if no hardware ypan */
    u_short ywrapstep;              /* zero if no hardware ywrap */
    u_long line_length;             /* length of a line in bytes */
    u_short can_soft_blank;         /* zero if no hardware blanking */
    u_short inverse;                /* != 0 text black on white as default */
    struct display_switch *dispsw;  /* low level operations */
    void *dispsw_data;              /* optional dispsw helper data */

#if 0
    struct fb_fix_cursorinfo fcrsr;
    struct fb_var_cursorinfo *vcrsr;
    struct fb_cursorstate crsrstate;
#endif

    /* Filled in by the low-level console driver */

    struct vc_data *conp;           /* pointer to console data */
    struct fb_info *fb_info;        /* frame buffer for this console */
    int vrows;                      /* number of virtual rows */
    unsigned short cursor_x;        /* current cursor position */
    unsigned short cursor_y;
    int fgcol;                      /* text colors */
    int bgcol;
    u_long next_line;               /* offset to one line below */
    u_long next_plane;              /* offset to next plane */
    u_char *fontdata;               /* Font associated to this display */
    unsigned short _fontheightlog;
    unsigned short _fontwidthlog;
    unsigned short _fontheight;
    unsigned short _fontwidth;
    int userfont;                   /* != 0 if fontdata kmalloc()ed */
    u_short scrollmode;             /* Scroll Method */
    short yscroll;                  /* Hardware scrolling */
    unsigned char fgshift, bgshift;
    unsigned short charmask;        /* 0xff or 0x1ff */
};

/* drivers/video/fbcon.c */
extern struct display fb_display[MAX_NR_CONSOLES];
extern char con2fb_map[MAX_NR_CONSOLES];
extern int PROC_CONSOLE(const struct fb_info *info);
extern void set_con2fb_map(int unit, int newidx);
extern int set_all_vcs(int fbidx, struct fb_ops *fb,
		       struct fb_var_screeninfo *var, struct fb_info *info);

#define fontheight(p) ((p)->_fontheight)
#define fontheightlog(p) ((p)->_fontheightlog)

#ifdef CONFIG_FBCON_FONTWIDTH8_ONLY

/* fontwidth w is supported by dispsw */
#define FONTWIDTH(w)	(1 << ((8) - 1))
/* fontwidths w1-w2 inclusive are supported by dispsw */
#define FONTWIDTHRANGE(w1,w2)	FONTWIDTH(8)

#define fontwidth(p) (8)
#define fontwidthlog(p) (0)

#else

/* fontwidth w is supported by dispsw */
#define FONTWIDTH(w)	(1 << ((w) - 1))
/* fontwidths w1-w2 inclusive are supported by dispsw */
#define FONTWIDTHRANGE(w1,w2)	(FONTWIDTH(w2+1) - FONTWIDTH(w1))

#define fontwidth(p) ((p)->_fontwidth)
#define fontwidthlog(p) ((p)->_fontwidthlog)

#endif

    /*
     *  Attribute Decoding
     */

/* Color */
#define attr_fgcol(p,s)    \
	(((s) >> ((p)->fgshift)) & 0x0f)
#define attr_bgcol(p,s)    \
	(((s) >> ((p)->bgshift)) & 0x0f)
#define	attr_bgcol_ec(p,conp) \
	((conp) ? (((conp)->vc_video_erase_char >> ((p)->bgshift)) & 0x0f) : 0)

/* Monochrome */
#define attr_bold(p,s) \
	((s) & 0x200)
#define attr_reverse(p,s) \
	(((s) & 0x800) ^ ((p)->inverse ? 0x800 : 0))
#define attr_underline(p,s) \
	((s) & 0x400)
#define attr_blink(p,s) \
	((s) & 0x8000)
	
    /*
     *  Scroll Method
     */
     
/* Internal flags */
#define __SCROLL_YPAN		0x001
#define __SCROLL_YWRAP		0x002
#define __SCROLL_YMOVE		0x003
#define __SCROLL_YREDRAW	0x004
#define __SCROLL_YMASK		0x00f
#define __SCROLL_YFIXED		0x010
#define __SCROLL_YNOMOVE	0x020
#define __SCROLL_YPANREDRAW	0x040
#define __SCROLL_YNOPARTIAL	0x080

/* Only these should be used by the drivers */
/* Which one should you use? If you have a fast card and slow bus,
   then probably just 0 to indicate fbcon should choose between
   YWRAP/YPAN+MOVE/YMOVE. On the other side, if you have a fast bus
   and even better if your card can do fonting (1->8/32bit painting),
   you should consider either SCROLL_YREDRAW (if your card is
   able to do neither YPAN/YWRAP), or SCROLL_YNOMOVE.
   The best is to test it with some real life scrolling (usually, not
   all lines on the screen are filled completely with non-space characters,
   and REDRAW performs much better on such lines, so don't cat a file
   with every line covering all screen columns, it would not be the right
   benchmark).
 */
#define SCROLL_YREDRAW		(__SCROLL_YFIXED|__SCROLL_YREDRAW)
#define SCROLL_YNOMOVE		(__SCROLL_YNOMOVE|__SCROLL_YPANREDRAW)

/* SCROLL_YNOPARTIAL, used in combination with the above, is for video
   cards which can not handle using panning to scroll a portion of the
   screen without excessive flicker.  Panning will only be used for
   whole screens.
 */
/* Namespace consistency */
#define SCROLL_YNOPARTIAL	__SCROLL_YNOPARTIAL


#if defined(__sparc__)

/* We map all of our framebuffers such that big-endian accesses
 * are what we want, so the following is sufficient.
 */

#define fb_readb sbus_readb
#define fb_readw sbus_readw
#define fb_readl sbus_readl
#define fb_writeb sbus_writeb
#define fb_writew sbus_writew
#define fb_writel sbus_writel
#define fb_memset sbus_memset_io

#elif defined(__i386__) || defined(__alpha__)

#define fb_readb __raw_readb
#define fb_readw __raw_readw
#define fb_readl __raw_readl
#define fb_writeb __raw_writeb
#define fb_writew __raw_writew
#define fb_writel __raw_writel
#define fb_memset memset_io

#else

#define fb_readb(addr) (*(volatile u8 *) (addr))
#define fb_readw(addr) (*(volatile u16 *) (addr))
#define fb_readl(addr) (*(volatile u32 *) (addr))
#define fb_writeb(b,addr) (*(volatile u8 *) (addr) = (b))
#define fb_writew(b,addr) (*(volatile u16 *) (addr) = (b))
#define fb_writel(b,addr) (*(volatile u32 *) (addr) = (b))
#define fb_memset memset

#endif


extern void fbcon_redraw_clear(struct vc_data *, struct display *, int, int, int, int);
extern void fbcon_redraw_bmove(struct display *, int, int, int, int, int, int);


/* ================================================================= */
/*                      Utility Assembler Functions                  */
/* ================================================================= */


#if defined(__mc68000__)

/* ====================================================================== */

/* Those of a delicate disposition might like to skip the next couple of
 * pages.
 *
 * These functions are drop in replacements for memmove and
 * memset(_, 0, _). However their five instances add at least a kilobyte
 * to the object file. You have been warned.
 *
 * Not a great fan of assembler for the sake of it, but I think
 * that these routines are at least 10 times faster than their C
 * equivalents for large blits, and that's important to the lowest level of
 * a graphics driver. Question is whether some scheme with the blitter
 * would be faster. I suspect not for simple text system - not much
 * asynchrony.
 *
 * Code is very simple, just gruesome expansion. Basic strategy is to
 * increase data moved/cleared at each step to 16 bytes to reduce
 * instruction per data move overhead. movem might be faster still
 * For more than 15 bytes, we try to align the write direction on a
 * longword boundary to get maximum speed. This is even more gruesome.
 * Unaligned read/write used requires 68020+ - think this is a problem?
 *
 * Sorry!
 */


/* ++roman: I've optimized Robert's original versions in some minor
 * aspects, e.g. moveq instead of movel, let gcc choose the registers,
 * use movem in some places...
 * For other modes than 1 plane, lots of more such assembler functions
 * were needed (e.g. the ones using movep or expanding color values).
 */

/* ++andreas: more optimizations:
   subl #65536,d0 replaced by clrw d0; subql #1,d0 for dbcc
   addal is faster than addaw
   movep is rather expensive compared to ordinary move's
   some functions rewritten in C for clarity, no speed loss */

static __inline__ void *fb_memclear_small(void *s, size_t count)
{
   if (!count)
      return(0);

   __asm__ __volatile__(
         "lsrl   #1,%1 ; jcc 1f ; moveb %2,%0@-\n\t"
      "1: lsrl   #1,%1 ; jcc 1f ; movew %2,%0@-\n\t"
      "1: lsrl   #1,%1 ; jcc 1f ; movel %2,%0@-\n\t"
      "1: lsrl   #1,%1 ; jcc 1f ; movel %2,%0@- ; movel %2,%0@-\n\t"
      "1:"
         : "=a" (s), "=d" (count)
         : "d" (0), "0" ((char *)s+count), "1" (count)
   );
   __asm__ __volatile__(
         "subql  #1,%1 ; jcs 3f\n\t"
	 "movel %2,%%d4; movel %2,%%d5; movel %2,%%d6\n\t"
      "2: moveml %2/%%d4/%%d5/%%d6,%0@-\n\t"
         "dbra %1,2b\n\t"
      "3:"
         : "=a" (s), "=d" (count)
         : "d" (0), "0" (s), "1" (count)
	 : "d4", "d5", "d6"
  );

   return(0);
}


static __inline__ void *fb_memclear(void *s, size_t count)
{
   if (!count)
      return(0);

   if (count < 16) {
      __asm__ __volatile__(
            "lsrl   #1,%1 ; jcc 1f ; clrb %0@+\n\t"
         "1: lsrl   #1,%1 ; jcc 1f ; clrw %0@+\n\t"
         "1: lsrl   #1,%1 ; jcc 1f ; clrl %0@+\n\t"
         "1: lsrl   #1,%1 ; jcc 1f ; clrl %0@+ ; clrl %0@+\n\t"
         "1:"
            : "=a" (s), "=d" (count)
            : "0" (s), "1" (count)
     );
   } else {
      long tmp;
      __asm__ __volatile__(
            "movel %1,%2\n\t"
            "lsrl   #1,%2 ; jcc 1f ; clrb %0@+ ; subqw #1,%1\n\t"
            "lsrl   #1,%2 ; jcs 2f\n\t"  /* %0 increased=>bit 2 switched*/
            "clrw   %0@+  ; subqw  #2,%1 ; jra 2f\n\t"
         "1: lsrl   #1,%2 ; jcc 2f\n\t"
            "clrw   %0@+  ; subqw  #2,%1\n\t"
         "2: movew %1,%2; lsrl #2,%1 ; jeq 6f\n\t"
            "lsrl   #1,%1 ; jcc 3f ; clrl %0@+\n\t"
         "3: lsrl   #1,%1 ; jcc 4f ; clrl %0@+ ; clrl %0@+\n\t"
         "4: subql  #1,%1 ; jcs 6f\n\t"
         "5: clrl %0@+; clrl %0@+ ; clrl %0@+ ; clrl %0@+\n\t"
            "dbra %1,5b   ; clrw %1; subql #1,%1; jcc 5b\n\t"
         "6: movew %2,%1; btst #1,%1 ; jeq 7f ; clrw %0@+\n\t"
         "7:            ; btst #0,%1 ; jeq 8f ; clrb %0@+\n\t"
         "8:"
            : "=a" (s), "=d" (count), "=d" (tmp)
            : "0" (s), "1" (count)
     );
   }

   return(0);
}


static __inline__ void *fb_memset255(void *s, size_t count)
{
   if (!count)
      return(0);

   __asm__ __volatile__(
         "lsrl   #1,%1 ; jcc 1f ; moveb %2,%0@-\n\t"
      "1: lsrl   #1,%1 ; jcc 1f ; movew %2,%0@-\n\t"
      "1: lsrl   #1,%1 ; jcc 1f ; movel %2,%0@-\n\t"
      "1: lsrl   #1,%1 ; jcc 1f ; movel %2,%0@- ; movel %2,%0@-\n\t"
      "1:"
         : "=a" (s), "=d" (count)
         : "d" (-1), "0" ((char *)s+count), "1" (count)
   );
   __asm__ __volatile__(
         "subql  #1,%1 ; jcs 3f\n\t"
	 "movel %2,%%d4; movel %2,%%d5; movel %2,%%d6\n\t"
      "2: moveml %2/%%d4/%%d5/%%d6,%0@-\n\t"
         "dbra %1,2b\n\t"
      "3:"
         : "=a" (s), "=d" (count)
         : "d" (-1), "0" (s), "1" (count)
	 : "d4", "d5", "d6"
  );

   return(0);
}


static __inline__ void *fb_memmove(void *d, const void *s, size_t count)
{
   if (d < s) {
      if (count < 16) {
         __asm__ __volatile__(
               "lsrl   #1,%2 ; jcc 1f ; moveb %1@+,%0@+\n\t"
            "1: lsrl   #1,%2 ; jcc 1f ; movew %1@+,%0@+\n\t"
            "1: lsrl   #1,%2 ; jcc 1f ; movel %1@+,%0@+\n\t"
            "1: lsrl   #1,%2 ; jcc 1f ; movel %1@+,%0@+ ; movel %1@+,%0@+\n\t"
            "1:"
               : "=a" (d), "=a" (s), "=d" (count)
               : "0" (d), "1" (s), "2" (count)
        );
      } else {
         long tmp;
         __asm__ __volatile__(
               "movel  %0,%3\n\t"
               "lsrl   #1,%3 ; jcc 1f ; moveb %1@+,%0@+ ; subqw #1,%2\n\t"
               "lsrl   #1,%3 ; jcs 2f\n\t"  /* %0 increased=>bit 2 switched*/
               "movew  %1@+,%0@+  ; subqw  #2,%2 ; jra 2f\n\t"
            "1: lsrl   #1,%3 ; jcc 2f\n\t"
               "movew  %1@+,%0@+  ; subqw  #2,%2\n\t"
            "2: movew  %2,%-; lsrl #2,%2 ; jeq 6f\n\t"
               "lsrl   #1,%2 ; jcc 3f ; movel %1@+,%0@+\n\t"
            "3: lsrl   #1,%2 ; jcc 4f ; movel %1@+,%0@+ ; movel %1@+,%0@+\n\t"
            "4: subql  #1,%2 ; jcs 6f\n\t"
            "5: movel  %1@+,%0@+;movel %1@+,%0@+\n\t"
               "movel  %1@+,%0@+;movel %1@+,%0@+\n\t"
               "dbra   %2,5b ; clrw %2; subql #1,%2; jcc 5b\n\t"
            "6: movew  %+,%2; btst #1,%2 ; jeq 7f ; movew %1@+,%0@+\n\t"
            "7:              ; btst #0,%2 ; jeq 8f ; moveb %1@+,%0@+\n\t"
            "8:"
               : "=a" (d), "=a" (s), "=d" (count), "=d" (tmp)
               : "0" (d), "1" (s), "2" (count)
        );
      }
   } else {
      if (count < 16) {
         __asm__ __volatile__(
               "lsrl   #1,%2 ; jcc 1f ; moveb %1@-,%0@-\n\t"
            "1: lsrl   #1,%2 ; jcc 1f ; movew %1@-,%0@-\n\t"
            "1: lsrl   #1,%2 ; jcc 1f ; movel %1@-,%0@-\n\t"
            "1: lsrl   #1,%2 ; jcc 1f ; movel %1@-,%0@- ; movel %1@-,%0@-\n\t"
            "1:"
               : "=a" (d), "=a" (s), "=d" (count)
               : "0" ((char *) d + count), "1" ((char *) s + count), "2" (count)
        );
      } else {
         long tmp;
         __asm__ __volatile__(
               "movel %0,%3\n\t"
               "lsrl   #1,%3 ; jcc 1f ; moveb %1@-,%0@- ; subqw #1,%2\n\t"
               "lsrl   #1,%3 ; jcs 2f\n\t"  /* %0 increased=>bit 2 switched*/
               "movew  %1@-,%0@-  ; subqw  #2,%2 ; jra 2f\n\t"
            "1: lsrl   #1,%3 ; jcc 2f\n\t"
               "movew  %1@-,%0@-  ; subqw  #2,%2\n\t"
            "2: movew %2,%-; lsrl #2,%2 ; jeq 6f\n\t"
               "lsrl   #1,%2 ; jcc 3f ; movel %1@-,%0@-\n\t"
            "3: lsrl   #1,%2 ; jcc 4f ; movel %1@-,%0@- ; movel %1@-,%0@-\n\t"
            "4: subql  #1,%2 ; jcs 6f\n\t"
            "5: movel %1@-,%0@-;movel %1@-,%0@-\n\t"
               "movel %1@-,%0@-;movel %1@-,%0@-\n\t"
               "dbra %2,5b ; clrw %2; subql #1,%2; jcc 5b\n\t"
            "6: movew %+,%2; btst #1,%2 ; jeq 7f ; movew %1@-,%0@-\n\t"
            "7:              ; btst #0,%2 ; jeq 8f ; moveb %1@-,%0@-\n\t"
            "8:"
               : "=a" (d), "=a" (s), "=d" (count), "=d" (tmp)
               : "0" ((char *) d + count), "1" ((char *) s + count), "2" (count)
        );
      }
   }

   return(0);
}


/* ++andreas: Simple and fast version of memmove, assumes size is
   divisible by 16, suitable for moving the whole screen bitplane */
static __inline__ void fast_memmove(char *dst, const char *src, size_t size)
{
  if (!size)
    return;
  if (dst < src)
    __asm__ __volatile__
      ("1:"
       "  moveml %0@+,%/d0/%/d1/%/a0/%/a1\n"
       "  moveml %/d0/%/d1/%/a0/%/a1,%1@\n"
       "  addql #8,%1; addql #8,%1\n"
       "  dbra %2,1b\n"
       "  clrw %2; subql #1,%2\n"
       "  jcc 1b"
       : "=a" (src), "=a" (dst), "=d" (size)
       : "0" (src), "1" (dst), "2" (size / 16 - 1)
       : "d0", "d1", "a0", "a1", "memory");
  else
    __asm__ __volatile__
      ("1:"
       "  subql #8,%0; subql #8,%0\n"
       "  moveml %0@,%/d0/%/d1/%/a0/%/a1\n"
       "  moveml %/d0/%/d1/%/a0/%/a1,%1@-\n"
       "  dbra %2,1b\n"
       "  clrw %2; subql #1,%2\n"
       "  jcc 1b"
       : "=a" (src), "=a" (dst), "=d" (size)
       : "0" (src + size), "1" (dst + size), "2" (size / 16 - 1)
       : "d0", "d1", "a0", "a1", "memory");
}

#elif defined(CONFIG_SUN4)

/* You may think that I'm crazy and that I should use generic
   routines.  No, I'm not: sun4's framebuffer crashes if we std
   into it, so we cannot use memset.  */

static __inline__ void *sun4_memset(void *s, char val, size_t count)
{
    int i;
    for(i=0; i<count;i++)
        ((char *) s) [i] = val;
    return s;
}

static __inline__ void *fb_memset255(void *s, size_t count)
{
    return sun4_memset(s, 255, count);
}

static __inline__ void *fb_memclear(void *s, size_t count)
{
    return sun4_memset(s, 0, count);
}

static __inline__ void *fb_memclear_small(void *s, size_t count)
{
    return sun4_memset(s, 0, count);
}

/* To be honest, this is slow_memmove :). But sun4 is crappy, so what we can do. */
static __inline__ void fast_memmove(void *d, const void *s, size_t count)
{
    int i;
    if (d<s) {
	for (i=0; i<count; i++)
	    ((char *) d)[i] = ((char *) s)[i];
    } else
	for (i=0; i<count; i++)
	    ((char *) d)[count-i-1] = ((char *) s)[count-i-1];
}

static __inline__ void *fb_memmove(char *dst, const char *src, size_t size)
{
    fast_memmove(dst, src, size);
    return dst;
}

#else

static __inline__ void *fb_memclear_small(void *s, size_t count)
{
    char *xs = (char *) s;

    while (count--)
	fb_writeb(0, xs++);

    return s;
}

static __inline__ void *fb_memclear(void *s, size_t count)
{
    unsigned long xs = (unsigned long) s;

    if (count < 8)
	goto rest;

    if (xs & 1) {
	fb_writeb(0, xs++);
	count--;
    }
    if (xs & 2) {
	fb_writew(0, xs);
	xs += 2;
	count -= 2;
    }
    while (count > 3) {
	fb_writel(0, xs);
	xs += 4;
	count -= 4;
    }
rest:
    while (count--)
	fb_writeb(0, xs++);

    return s;
}

static __inline__ void *fb_memset255(void *s, size_t count)
{
    unsigned long xs = (unsigned long) s;

    if (count < 8)
	goto rest;

    if (xs & 1) {
	fb_writeb(0xff, xs++);
	count--;
    }
    if (xs & 2) {
	fb_writew(0xffff, xs);
	xs += 2;
	count -= 2;
    }
    while (count > 3) {
	fb_writel(0xffffffff, xs);
	xs += 4;
	count -= 4;
    }
rest:
    while (count--)
	fb_writeb(0xff, xs++);

    return s;
}

#if defined(__i386__)

static __inline__ void fast_memmove(void *d, const void *s, size_t count)
{
  int d0, d1, d2, d3;
    if (d < s) {
__asm__ __volatile__ (
	"cld\n\t"
	"shrl $1,%%ecx\n\t"
	"jnc 1f\n\t"
	"movsb\n"
	"1:\tshrl $1,%%ecx\n\t"
	"jnc 2f\n\t"
	"movsw\n"
	"2:\trep\n\t"
	"movsl"
	: "=&c" (d0), "=&D" (d1), "=&S" (d2)
	:"0"(count),"1"((long)d),"2"((long)s)
	:"memory");
    } else {
__asm__ __volatile__ (
	"std\n\t"
	"shrl $1,%%ecx\n\t"
	"jnc 1f\n\t"
	"movb 3(%%esi),%%al\n\t"
	"movb %%al,3(%%edi)\n\t"
	"decl %%esi\n\t"
	"decl %%edi\n"
	"1:\tshrl $1,%%ecx\n\t"
	"jnc 2f\n\t"
	"movw 2(%%esi),%%ax\n\t"
	"movw %%ax,2(%%edi)\n\t"
	"decl %%esi\n\t"
	"decl %%edi\n\t"
	"decl %%esi\n\t"
	"decl %%edi\n"
	"2:\trep\n\t"
	"movsl\n\t"
	"cld"
	: "=&c" (d0), "=&D" (d1), "=&S" (d2), "=&a" (d3)
	:"0"(count),"1"(count-4+(long)d),"2"(count-4+(long)s)
	:"memory");
    }
}

static __inline__ void *fb_memmove(char *dst, const char *src, size_t size)
{
    fast_memmove(dst, src, size);
    return dst;
}

#else /* !__i386__ */

    /*
     *  Anyone who'd like to write asm functions for other CPUs?
     *   (Why are these functions better than those from include/asm/string.h?)
     */

static __inline__ void *fb_memmove(void *d, const void *s, size_t count)
{
    unsigned long dst, src;

    if (d < s) {
	dst = (unsigned long) d;
	src = (unsigned long) s;

	if ((count < 8) || ((dst ^ src) & 3))
	    goto restup;

	if (dst & 1) {
	    fb_writeb(fb_readb(src++), dst++);
	    count--;
	}
	if (dst & 2) {
	    fb_writew(fb_readw(src), dst);
	    src += 2;
	    dst += 2;
	    count -= 2;
	}
	while (count > 3) {
	    fb_writel(fb_readl(src), dst);
	    src += 4;
	    dst += 4;
	    count -= 4;
	}

    restup:
	while (count--)
	    fb_writeb(fb_readb(src++), dst++);
    } else {
	dst = (unsigned long) d + count;
	src = (unsigned long) s + count;

	if ((count < 8) || ((dst ^ src) & 3))
	    goto restdown;

	if (dst & 1) {
	    src--;
	    dst--;
	    count--;
	    fb_writeb(fb_readb(src), dst);
	}
	if (dst & 2) {
	    src -= 2;
	    dst -= 2;
	    count -= 2;
	    fb_writew(fb_readw(src), dst);
	}
	while (count > 3) {
	    src -= 4;
	    dst -= 4;
	    count -= 4;
	    fb_writel(fb_readl(src), dst);
	}

    restdown:
	while (count--) {
	    src--;
	    dst--;
	    fb_writeb(fb_readb(src), dst);
	}
    }

    return d;
}

static __inline__ void fast_memmove(char *d, const char *s, size_t count)
{
    unsigned long dst, src;

    if (d < s) {
	dst = (unsigned long) d;
	src = (unsigned long) s;

	if ((count < 8) || ((dst ^ src) & 3))
	    goto restup;

	if (dst & 1) {
	    fb_writeb(fb_readb(src++), dst++);
	    count--;
	}
	if (dst & 2) {
	    fb_writew(fb_readw(src), dst);
	    src += 2;
	    dst += 2;
	    count -= 2;
	}
	while (count > 3) {
	    fb_writel(fb_readl(src), dst);
	    src += 4;
	    dst += 4;
	    count -= 4;
	}

    restup:
	while (count--)
	    fb_writeb(fb_readb(src++), dst++);
    } else {
	dst = (unsigned long) d + count;
	src = (unsigned long) s + count;

	if ((count < 8) || ((dst ^ src) & 3))
	    goto restdown;

	if (dst & 1) {
	    src--;
	    dst--;
	    count--;
	    fb_writeb(fb_readb(src), dst);
	}
	if (dst & 2) {
	    src -= 2;
	    dst -= 2;
	    count -= 2;
	    fb_writew(fb_readw(src), dst);
	}
	while (count > 3) {
	    src -= 4;
	    dst -= 4;
	    count -= 4;
	    fb_writel(fb_readl(src), dst);
	}

    restdown:
	while (count--) {
	    src--;
	    dst--;
	    fb_writeb(fb_readb(src), dst);
	}
    }
}

#endif /* !__i386__ */

#endif /* !__mc68000__ */

#endif /* _VIDEO_FBCON_H */